Deploy ClimaIQ Kisan app, model weights, and dataset CSV

.gitignore

@@ -0,0 +1,14 @@
+__pycache__/
+*.py[cod]
+.pytest_cache/
+.DS_Store
+.env
+.venv/
+venv/
+_hf_clone/
+
+# Local-only (not synced to HF Space by sync_hf_space.sh)
+*.docx
+*.ipynb
+word files/
+pdf files/

.streamlit/config.toml

@@ -0,0 +1,9 @@
+[theme]
+primaryColor = "#2E7D32"
+backgroundColor = "#F9F6F0"
+secondaryBackgroundColor = "#FFFFFF"
+textColor = "#1f2a20"
+font = "sans-serif"
+
+[client]
+showSidebarNavigation = false

Dockerfile

@@ -1,20 +0,0 @@
-FROM python:3.13.5-slim
-
-WORKDIR /app
-
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    curl \
-    git \
-    && rm -rf /var/lib/apt/lists/*
-
-COPY requirements.txt ./
-COPY src/ ./src/
-
-RUN pip3 install -r requirements.txt
-
-EXPOSE 8501
-
-HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
-
-ENTRYPOINT ["streamlit", "run", "src/streamlit_app.py", "--server.port=8501", "--server.address=0.0.0.0"]

README.md

@@ -1,19 +1,29 @@
 ---
-title: Climaiq Kisan
-emoji: 🚀
-colorFrom: red
-colorTo: red
-sdk: docker
-app_port: 8501
-tags:
-- streamlit
+title: ClimaIQ Kisan
+emoji: 🌾
+colorFrom: green
+colorTo: gray
+sdk: streamlit
+app_file: app.py
 pinned: false
-short_description: Climate-Adjusted Credit Risk Model For Agrticulture Lending
+license: apache-2.0
 ---
 
-# Welcome to Streamlit!
+# ClimaIQ Kisan
 
-Edit `/src/streamlit_app.py` to customize this app to your heart's desire. :heart:
+Climate-adjusted credit scoring for Indian agricultural lending: **Kisan view**, **loan officer** scorecard and PDF-ready report, and **portfolio stress** scenarios. Bilingual (English / Hindi) narratives via **Google Gemma (cloud)** or **local Ollama**.
 
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+## Dataset (in this Space)
+
+`data/climaiq_india_agricultural_credit_1000.csv` — 1,000 synthetic loans (Maharashtra / Punjab, crops, SPI, rainfall deficit, engineered features, default label, ClimaIQ score and risk band), aligned with the same generator and model as the app.
+
+## Optional secret
+
+Add **`GEMMA_API_KEY`** (Google AI Studio) under Space **Settings → Secrets** for cloud Gemma. Without it, open the app and use **Local (Ollama)** only when you run Streamlit on your own machine; the hosted Space still runs the **ClimaIQ model** and charts without the key.
+
+## Local run
+
+```bash
+pip install -r requirements.txt
+streamlit run app.py
+```

app.py

@@ -0,0 +1,1189 @@
+import html
+import io
+import os
+from typing import Dict, List, Optional, Tuple
+
+import numpy as np
+import pandas as pd
+import plotly.express as px
+import plotly.graph_objects as go
+import streamlit as st
+
+from climaiq_engine import CROP_WATER_MAP, load_model, predict_single, run_stress_test
+from climaiq_gemma import (
+    explain_for_farmer,
+    explain_for_officer,
+    explain_portfolio_stress,
+    get_client,
+    ollama_verify_connection,
+)
+from climaiq_report_pdf import build_officer_assessment_pdf, split_report_sections
+
+
+st.set_page_config(page_title="ClimaIQ Kisan", page_icon="🌾", layout="wide")
+
+
+COLORS = {
+    "primary": "#2E7D32",
+    "accent": "#FF8F00",
+    "danger": "#C62828",
+    "safe": "#1B5E20",
+    "bg": "#F9F6F0",
+    "ink": "#1f2a20",
+}
+
+LANG_TEXT = {
+    "English": {
+        "app_subtitle": "Climate intelligence for agricultural lending",
+        "hero_title": "ClimaIQ Kisan",
+        "hero_sub": "One climate-adjusted credit model with three views: farmer, loan officer, and portfolio stress.",
+        "tab_kisan_hint": "Enter your details and climate signals, then get a plain-language score and guidance.",
+        "tab_officer_hint": "Compact borrower inputs, scorecard, risk drivers, and an AI report for credit decisions.",
+        "tab_portfolio_hint": "Simulate a book of loans and compare losses under four drought scenarios.",
+        "empty_kisan": "Fill the form on the left, then tap “Know Your Score”. Your gauge, default likelihood, and Gemma explanation appear on the right.",
+        "empty_officer": "Open the assessment panel, adjust fields, then run “Run ClimaIQ assessment” to see the scorecard and downloadable report.",
+        "empty_portfolio": "Set portfolio parameters, then run the stress test to see scenario losses and default rates.",
+        "your_info": "Your information",
+        "crop": "Crop Type",
+        "state": "State",
+        "land": "Land (acres)",
+        "income": "Annual Income (₹ Lakhs)",
+        "loan": "Loan Amount (₹ Lakhs)",
+        "default_hist": "Previous Default?",
+        "weather": "Current Climate Condition",
+        "rain_def": "Rainfall Deficit (%)",
+        "spi": "Drought Severity (SPI)",
+        "drought_years": "Consecutive Drought Years",
+        "cta": "Know Your Score",
+        "score_title": "Your ClimaIQ Score",
+        "prob": "Default Probability",
+        "actions": "What You Can Do",
+        "farmer_expl": "Plain-language summary (Gemma 4)",
+        "section_profile": "Profile & loan",
+        "section_climate": "This season’s climate",
+        "narrative_source_cloud": "This explanation came from Google Gemma in the cloud.",
+        "narrative_source_ollama": "This explanation came from Gemma on your computer (Ollama).",
+        "ollama_panel_title": "##### Local inference with Ollama",
+        "ollama_panel_intro": "Enter the address and model name that your Ollama install uses.",
+        "ollama_base_label": "Base URL",
+        "ollama_model_label": "Model name",
+        "ollama_model_help": "Must match a name from ollama list (e.g. gemma3:4b).",
+        "ollama_test_btn": "Test connection",
+        "ollama_setup_md": (
+            "**Setup**\n\n"
+            "1. Keep Ollama running (for example `brew services start ollama` on a Mac, or the Ollama app).\n\n"
+            "2. Pull a model, for example **`ollama pull gemma3:4b`**, then type the same name under **Model name**. "
+            "For a smaller, faster model, run **`ollama pull gemma2:2b`** and use **`gemma2:2b`**.\n\n"
+            "3. Run **`ollama list`** in Terminal. Your model must appear before narratives will work.\n\n"
+            "4. Tap **Test connection** below. You should see a green confirmation when Ollama and the model are ready.\n\n"
+            "**Demo:** Turn Wi‑Fi on and generate an officer report, then turn Wi‑Fi off. Responses still run on your machine."
+        ),
+        "sidebar_mode_cloud": "Cloud (Google AI Studio)",
+        "sidebar_mode_ollama": "Local (Ollama, offline)",
+        "sidebar_gemma_caption": (
+            "Cloud suits a Hugging Face Spaces demo well. For Ollama, run ollama serve on your machine, "
+            "then you can turn Wi‑Fi off and narratives still run. When Local is on, use the bordered panel under the title for URL and model."
+        ),
+        "sidebar_gemma_cloud_ok": "Gemma: cloud (Google AI Studio), API key present.",
+        "sidebar_gemma_cloud_missing": (
+            "Gemma: cloud unavailable (no GEMMA_API_KEY). Use Local (Ollama) for offline narratives, or add a Space secret."
+        ),
+        "sidebar_gemma_local_verified": "Locally verified. Gemma on Ollama: {model} at {base}.",
+        "sidebar_gemma_local_unverified": "Gemma on Ollama: {model} at {base}.",
+        "hide_sidebar_label": "Hide side panel",
+        "hide_sidebar_help": "Gives charts and reports more horizontal space. Uncheck to open language and Gemma settings again.",
+    },
+    "Hindi": {
+        "app_subtitle": "कृ��ि ऋण के लिए जलवायु आधारित समझ",
+        "hero_title": "ClimaIQ Kisan",
+        "hero_sub": "एक ही मॉडल, तीन दृष्टि: किसान, ऋण अधिकारी, और पोर्टफोलियो तनाव परीक्षण।",
+        "tab_kisan_hint": "अपनी जानकारी और मौसम भरें, फिर स्कोर और सरल सलाह देखें।",
+        "tab_officer_hint": "संक्षिप्त फॉर्म, स्कोरकार्ड, जोखिम कारक और AI रिपोर्ट।",
+        "tab_portfolio_hint": "ऋणों की संख्या सेट करें और चार सूखा परिदृश्यों में नुकसान की तुलना करें।",
+        "empty_kisan": "बाईं ओर फॉर्म भरें, फिर “अपना स्कोर जानें” दबाएँ। दाईं ओर गेज, संभावना और सरल व्याख्या दिखेगी।",
+        "empty_officer": "पैनल खोलें, फील्ड बदलें, फिर मूल्यांकन बटन चलाएँ; स्कोरकार्ड और रिपोर्ट नीचे दिखेगी।",
+        "empty_portfolio": "पोर्टफोलियो सेट करके स्ट्रेस टेस्ट चलाएँ; परिदृश्य और चार्ट नीचे दिखेंगे।",
+        "your_info": "आपकी जानकारी",
+        "crop": "फसल",
+        "state": "राज्य",
+        "land": "ज़मीन (एकड़)",
+        "income": "सालाना आय (₹ लाख)",
+        "loan": "ऋण राशि (₹ लाख)",
+        "default_hist": "पहले डिफ़ॉल्ट?",
+        "weather": "इस मौसम का हाल",
+        "rain_def": "बारिश की कमी (%)",
+        "spi": "सूखा गंभीरता (SPI)",
+        "drought_years": "लगातार सूखा साल",
+        "cta": "अपना स्कोर जानें",
+        "score_title": "आपका ClimaIQ स्कोर",
+        "prob": "डिफ़ॉल्ट संभावना",
+        "actions": "आप क्या कर सकते हैं",
+        "farmer_expl": "सरल भाषा में सारांश (Gemma 4)",
+        "section_profile": "प्रोफाइल और ऋण",
+        "section_climate": "इस मौसम की स्थिति",
+        "narrative_source_cloud": "यह सारांश Google Gemma (क्लाउड) से आया है।",
+        "narrative_source_ollama": "यह सारांश आपके कंप्यूटर पर Gemma (Ollama) से आया है।",
+        "ollama_panel_title": "##### स्थानीय अनुमान: Ollama कनेक्शन",
+        "ollama_panel_intro": "वह पता और मॉडल का नाम भरें जो आपके Ollama में सेट है।",
+        "ollama_base_label": "आधार URL",
+        "ollama_model_label": "मॉडल का नाम",
+        "ollama_model_help": "`ollama list` में दिखा नाम यहाँ लिखें (उदा. gemma3:4b)।",
+        "ollama_test_btn": "कनेक्शन जाँचें",
+        "ollama_setup_md": (
+            "**सेटअप**\n\n"
+            "1. Ollama चालू रखें (Mac पर उदा. `brew services start ollama`, या Ollama ऐप)।\n\n"
+            "2. मॉडल खींचें, उदा. **`ollama pull gemma3:4b`**, फिर वही नाम **मॉडल का नाम** में लिखें। "
+            "छोटा/तेज़ विकल्प: **`ollama pull gemma2:2b`** और फिर **`gemma2:2b`** इस्तेमाल करें।\n\n"
+            "3. टर्मिनल में **`ollama list`** चलाएँ; बयान (नैरेटिव) से पहले आपका मॉडल सूची में होना चाहिए।\n\n"
+            "4. नीचे **कनेक्शन जाँचें** दबाएँ। Ollama और मॉडल तैयार हों तो हरी पुष्टि दिखेगी।\n\n"
+            "**डेमो:** Wi‑Fi चालू कर ऋण अधिकारी रिपोर्ट बनाएँ, फिर Wi‑Fi बंद करें; जवाब फिर भी आपके कंप्यूटर पर चलेगा।"
+        ),
+        "sidebar_mode_cloud": "क्लाउड (Google AI Studio)",
+        "sidebar_mode_ollama": "स्थानीय (Ollama, ऑफ़लाइन)",
+        "sidebar_gemma_caption": (
+            "क्लाउड Hugging Face Spaces डेमो के लिए अच्छा है। Ollama के लिए अपने कंप्यूटर पर ollama serve चलाएँ, "
+            "फिर Wi‑Fi बंद करने पर भी बयान चल सकते हैं। स्थानीय मोड में URL और मॉडल शीर्षक के नीचे वाले बॉक्स में भरें।"
+        ),
+        "sidebar_gemma_cloud_ok": "Gemma: क्लाउड (Google AI Studio), API कुंजी मिली।",
+        "sidebar_gemma_cloud_missing": (
+            "Gemma: क्लाउड उपलब्ध नहीं (GEMMA_API_KEY नहीं)। ऑफ़लाइन बयान के लिए स्थानीय (Ollama) चुनें, या Space में रहस्य जोड़ें।"
+        ),
+        "sidebar_gemma_local_verified": "स्थानीय जाँच सफल। Ollama पर Gemma: {model}, पता {base}।",
+        "sidebar_gemma_local_unverified": "Ollama पर Gemma: {model}, पता {base}।",
+        "hide_sidebar_label": "साइड पैनल छुपाएँ",
+        "hide_sidebar_help": "चार्ट और रिपोर्ट को चौड़ाई मिलेगी। भाषा व Gemma सेटिंग वापस लाने के लिए अनचेक करें।",
+    },
+}
+
+SPI_LABELS = {
+    "English": {0: "Severe Drought", 1: "Moderate Drought", 2: "Normal"},
+    "Hindi": {0: "गंभीर सूखा", 1: "मध्यम सूखा", 2: "सामान्य"},
+}
+
+DROUGHT_OPTIONS = [0, 1, 2, 3]
+DROUGHT_LABEL = {0: "0", 1: "1", 2: "2", 3: "3+"}
+
+def format_gemma_inference_mode(mode: str) -> str:
+    """Stable format_func for sidebar mode radio (lambdas here cause extra reruns / double clicks)."""
+    lang = st.session_state.get("language", "English")
+    if mode == "cloud":
+        return LANG_TEXT[lang]["sidebar_mode_cloud"]
+    return LANG_TEXT[lang]["sidebar_mode_ollama"]
+
+
+def format_drought_years(value: int) -> str:
+    return DROUGHT_LABEL[value]
+
+
+OFFICER_UI = {
+    "form_title": "Borrower assessment",
+    "form_help": "Same ClimaIQ engine as the farmer view. Field labels stay in English for branch and audit use.",
+    "expander": "Assessment inputs",
+    "run": "Run ClimaIQ assessment",
+    "empty": "Adjust inputs in the panel above, then run the assessment. Results persist while you explore other tabs.",
+    "scorecard": "Scorecard",
+    "lbl_score": "Credit score",
+    "lbl_risk": "Risk band",
+    "lbl_default": "Default likelihood",
+    "risk_drivers": "Top risk drivers",
+    "risk_drivers_caption": "Relative contribution to model risk for this profile (longer bar = stronger upward pressure on default likelihood).",
+    "ai_report": "AI assessment report",
+    "report_lang": "Report language",
+    "download": "Download PDF report",
+    "download_txt_fallback": "Download narrative (.txt)",
+    "micro_caption": "Driver strength at a glance (normalized to the strongest driver in the top three).",
+    "pdf_hint": "The PDF includes the scorecard table plus this narrative, ready for credit file notes.",
+    "moratorium_note": "If approved, consider a moratorium clause if SPI falls below −1.5 during the loan term.",
+    "rec_action": "Recommended action",
+}
+
+
+def narr_html(body: str) -> str:
+    """Escape user/model text for safe insertion inside HTML blocks."""
+    return f"<div class='soft-note narr-body'>{html.escape(body)}</div>"
+
+
+def render_officer_report_ui(report_text: str) -> None:
+    """Render Gemma officer narrative with clear section headings (1. … 2. …)."""
+    for head, body in split_report_sections(report_text):
+        parts = []
+        if head:
+            parts.append(f"<div class='report-head'>{html.escape(head)}</div>")
+        if body:
+            parts.append(f"<div class='report-body'>{html.escape(body)}</div>")
+        if parts:
+            st.markdown(
+                "<div class='report-block'>" + "".join(parts) + "</div>",
+                unsafe_allow_html=True,
+            )
+
+
+def render_driver_micro_bars(result: Dict) -> None:
+    drivers = result.get("top_risk_drivers") or []
+    if not drivers:
+        return
+    dlist = drivers[:3]
+    mx = max(abs(float(d["contribution"])) for d in dlist) or 1e-9
+    st.caption(OFFICER_UI["micro_caption"])
+    for d in dlist:
+        pct = min(100, int(100 * abs(float(d["contribution"])) / mx))
+        name = html.escape(str(d.get("display_name", "")))
+        st.markdown(
+            f"<div class='microbar-wrap'><span class='microbar-label'>{name}</span>"
+            f"<div class='microbar-track'><div class='microbar-fill' style='width:{pct}%'></div></div></div>",
+            unsafe_allow_html=True,
+        )
+
+
+def inject_styles(hide_sidebar: bool = False) -> None:
+    sidebar_hide_css = ""
+    if hide_sidebar:
+        sidebar_hide_css = """
+            section[data-testid="stSidebar"] {
+                display: none !important;
+            }
+            [data-testid="stSidebarCollapsedControl"] {
+                display: none !important;
+            }
+            .main .block-container {
+                max-width: min(1420px, 98vw) !important;
+                padding-left: 1.75rem !important;
+                padding-right: 1.75rem !important;
+            }
+        """
+    st.markdown(
+        f"""
+        <style>
+            @import url('https://fonts.googleapis.com/css2?family=DM+Sans:ital,opsz,wght@0,9..40,400;0,9..40,500;0,9..40,600;0,9..40,700;1,9..40,400&display=swap');
+            html, body, [class*="css"]  {{
+                font-family: "DM Sans", "Segoe UI", system-ui, sans-serif;
+            }}
+            .stApp {{
+                background: linear-gradient(180deg, #f9f6f0 0%, #f3efe6 100%);
+                color: {COLORS["ink"]};
+            }}
+            .main .block-container {{
+                padding-top: 1.25rem;
+                padding-bottom: 2.25rem;
+                max-width: 1180px;
+            }}
+            [data-testid="stSidebar"] {{
+                background: #123019;
+                border-right: 1px solid #204d2b;
+                color: #f6fff5;
+            }}
+            /* Do NOT use sidebar * {{ color }} — it breaks light widgets (white text on white inputs). */
+            [data-testid="stSidebar"] h1,
+            [data-testid="stSidebar"] h2,
+            [data-testid="stSidebar"] h3,
+            [data-testid="stSidebar"] p,
+            [data-testid="stSidebar"] .stMarkdown,
+            [data-testid="stSidebar"] [data-testid="stMarkdownContainer"] p {{
+                color: #f6fff5;
+            }}
+            [data-testid="stSidebar"] [data-testid="stHeading"] {{
+                color: #f6fff5;
+            }}
+            [data-testid="stSidebar"] [data-testid="stCaption"] {{
+                color: #c8e6c9 !important;
+            }}
+            [data-testid="stSidebar"] .stRadio [role="radiogroup"] {{
+                gap: 0.35rem;
+                align-items: center;
+            }}
+            [data-testid="stSidebar"] .stRadio label {{
+                background: rgba(255,255,255,0.06);
+                border-radius: 8px;
+                padding: 0.28rem 0.55rem;
+                color: #f6fff5 !important;
+                cursor: pointer;
+                user-select: none;
+            }}
+            [data-testid="stSidebar"] .stRadio input {{
+                cursor: pointer;
+            }}
+            .main .stRadio [role="radiogroup"] label {{
+                cursor: pointer;
+                user-select: none;
+            }}
+            [data-testid="stSidebar"] hr {{
+                border-color: rgba(255, 255, 255, 0.18) !important;
+            }}
+            /* Do not style sidebar text_input / expander via CSS — Streamlit Base Web uses
+               scoped rules that beat generic overrides. Ollama fields render in main instead. */
+            [data-testid="stSidebar"] code,
+            [data-testid="stSidebar"] pre {{
+                background: rgba(255, 255, 255, 0.12) !important;
+                color: #f4fff6 !important;
+                border: 1px solid rgba(255, 255, 255, 0.22);
+                padding: 0.12rem 0.4rem;
+                border-radius: 6px;
+                font-size: 0.88em;
+            }}
+            .header-wrap {{
+                background: linear-gradient(135deg, #ffffff 0%, #f5f8f3 100%);
+                border: 1px solid #e1eadf;
+                border-radius: 16px;
+                padding: 1rem 1.25rem;
+                margin-bottom: 1rem;
+            }}
+            .header-title {{
+                font-size: 1.45rem;
+                font-weight: 700;
+                color: #142818;
+                letter-spacing: -0.02em;
+                margin-bottom: 0.2rem;
+            }}
+            .header-sub {{
+                color: #4a5c50;
+                font-size: 0.98rem;
+                line-height: 1.5;
+                max-width: 52rem;
+            }}
+            .tab-hint {{
+                color: #5a6d61;
+                font-size: 0.9rem;
+                line-height: 1.45;
+                margin: -0.2rem 0 1rem 0;
+            }}
+            .panel-title {{
+                font-weight: 600;
+                font-size: 0.82rem;
+                text-transform: uppercase;
+                letter-spacing: 0.06em;
+                color: #3d5244;
+                margin-bottom: 0.5rem;
+            }}
+            .section-label {{
+                font-size: 0.8rem;
+                font-weight: 600;
+                color: {COLORS["primary"]};
+                margin: 0.35rem 0 0.5rem 0;
+            }}
+            .hero-metric {{
+                text-align: center;
+                padding: 0.85rem 0.6rem;
+                border-radius: 12px;
+                background: #ffffff;
+                border: 1px solid #e5e8df;
+                margin-top: 0.2rem;
+                margin-bottom: 0.5rem;
+            }}
+            .risk-pill {{
+                display: inline-block;
+                padding: 0.2rem 0.65rem;
+                border-radius: 999px;
+                font-weight: 700;
+                font-size: 0.86rem;
+            }}
+            .chip-high {{ background: #fde8e8; color: {COLORS["danger"]}; }}
+            .chip-medium {{ background: #fff4de; color: {COLORS["accent"]}; }}
+            .chip-low {{ background: #e8f5e9; color: {COLORS["safe"]}; }}
+            .action-chip {{
+                border: 1px solid #e5e8df;
+                border-radius: 12px;
+                padding: 0.7rem 0.6rem;
+                background: #fff;
+                min-height: 74px;
+                text-align: center;
+                font-weight: 600;
+                font-size: 0.9rem;
+            }}
+            .soft-note {{
+                background: #ffffff;
+                border: 1px solid #e5e8df;
+                border-left: 4px solid {COLORS["primary"]};
+                border-radius: 12px;
+                padding: 0.75rem 1rem;
+                margin-bottom: 0.65rem;
+            }}
+            .narr-body {{
+                line-height: 1.58;
+                font-size: 0.95rem;
+                color: #2a382e;
+                white-space: pre-line;
+            }}
+            .score-tile {{
+                background: #fff;
+                border: 1px solid #e3e7de;
+                border-radius: 12px;
+                padding: 0.65rem;
+                text-align: center;
+            }}
+            .score-k {{
+                font-size: 0.78rem;
+                letter-spacing: 0.04em;
+                color: #5e7566;
+                margin-bottom: 0.15rem;
+            }}
+            .score-v {{
+                font-size: 1.25rem;
+                font-weight: 700;
+                color: #203429;
+            }}
+            .small-muted {{
+                color: #6f6f6f;
+                font-size: 0.9rem;
+            }}
+            .stButton > button {{
+                border-radius: 10px;
+                border: 0;
+                font-weight: 600;
+                box-shadow: 0 1px 0 rgba(0,0,0,.06);
+            }}
+            [data-baseweb="tab-list"] button {{
+                font-weight: 600;
+                font-size: 0.95rem;
+            }}
+            div[data-testid="stExpander"] summary {{
+                font-weight: 600;
+            }}
+            .report-block {{
+                background: #fff;
+                border: 1px solid #e3e8e0;
+                border-radius: 12px;
+                padding: 0.85rem 1rem 0.95rem 1rem;
+                margin-bottom: 0.75rem;
+                border-left: 4px solid {COLORS["primary"]};
+            }}
+            .report-head {{
+                font-weight: 700;
+                font-size: 1rem;
+                color: {COLORS["safe"]};
+                margin-bottom: 0.45rem;
+                padding-bottom: 0.25rem;
+                border-bottom: 1px solid #c8e6c9;
+            }}
+            .report-body {{
+                line-height: 1.62;
+                font-size: 0.95rem;
+                color: #2a382e;
+                white-space: pre-line;
+            }}
+            .microbar-wrap {{ margin-bottom: 0.5rem; }}
+            .microbar-label {{
+                font-size: 0.8rem;
+                color: #3d5244;
+                display: block;
+                margin-bottom: 3px;
+            }}
+            .microbar-track {{
+                height: 9px;
+                background: #e8ebe4;
+                border-radius: 99px;
+                overflow: hidden;
+            }}
+            .microbar-fill {{
+                height: 100%;
+                border-radius: 99px;
+                background: linear-gradient(90deg, {COLORS["primary"]}, {COLORS["danger"]});
+            }}
+            {sidebar_hide_css}
+        </style>
+        """,
+        unsafe_allow_html=True,
+    )
+
+
+def init_state():
+    if "model" not in st.session_state or "scaler" not in st.session_state:
+        model, scaler = load_model()
+        st.session_state.model = model
+        st.session_state.scaler = scaler
+    if "language" not in st.session_state:
+        st.session_state.language = "Hindi"
+    if "last_result" not in st.session_state:
+        st.session_state.last_result = None
+    if "last_officer_result" not in st.session_state:
+        st.session_state.last_officer_result = None
+    if "last_stress_result" not in st.session_state:
+        st.session_state.last_stress_result = None
+    if "gemma_inference_mode" not in st.session_state:
+        st.session_state.gemma_inference_mode = os.environ.get("GEMMA_INFERENCE_MODE", "cloud")
+    if "ollama_base_url" not in st.session_state:
+        st.session_state.ollama_base_url = os.environ.get("OLLAMA_BASE_URL", "http://127.0.0.1:11434")
+    if "ollama_model" not in st.session_state:
+        st.session_state.ollama_model = os.environ.get("OLLAMA_MODEL", "gemma3:4b")
+    if "ollama_last_verify_ok" not in st.session_state:
+        st.session_state.ollama_last_verify_ok = False
+    if "ollama_last_verify_msg" not in st.session_state:
+        st.session_state.ollama_last_verify_msg = None
+    if "ui_hide_sidebar" not in st.session_state:
+        st.session_state.ui_hide_sidebar = False
+
+
+def _invalidate_gemma_outputs() -> None:
+    """Clear cached Gemma strings when switching cloud ↔ Ollama."""
+    lr = st.session_state.get("last_result")
+    if lr and isinstance(lr, dict) and "explanations" in lr:
+        lr["explanations"] = {}
+        lr["explanation_sources"] = {}
+    lo = st.session_state.get("last_officer_result")
+    if lo and isinstance(lo, dict) and "reports" in lo:
+        lo["reports"] = {}
+        lo["report_sources"] = {}
+    ls = st.session_state.get("last_stress_result")
+    if ls and isinstance(ls, dict) and "narratives" in ls:
+        ls["narratives"] = {}
+        ls["narrative_sources"] = {}
+    st.session_state.ollama_last_verify_ok = False
+    st.session_state.ollama_last_verify_msg = None
+
+
+def narrative_source_caption(mode: str, label_lang: str) -> str:
+    """Short attribution for which Gemma backend produced the visible narrative."""
+    if mode == "ollama":
+        return LANG_TEXT[label_lang]["narrative_source_ollama"]
+    return LANG_TEXT[label_lang]["narrative_source_cloud"]
+
+
+def get_inference_config() -> Tuple[str, str, str]:
+    mode = st.session_state.get("gemma_inference_mode", "cloud")
+    base = (st.session_state.get("ollama_base_url") or "http://127.0.0.1:11434").strip().rstrip("/")
+    model = (st.session_state.get("ollama_model") or "gemma3:4b").strip()
+    return mode, base, model
+
+
+def render_ollama_connection_main() -> None:
+    """
+    Ollama URL + model must live in the MAIN area, not the sidebar.
+    Streamlit’s Base Web inputs in the dark sidebar inherit unreadable contrast
+    no matter how much global CSS we add; scoped widget styles win.
+    """
+    if st.session_state.get("gemma_inference_mode") != "ollama":
+        return
+    ui = LANG_TEXT[st.session_state.language]
+    with st.container(border=True):
+        st.markdown(ui["ollama_panel_title"])
+        st.caption(ui["ollama_panel_intro"])
+        st.markdown(ui["ollama_setup_md"])
+        c1, c2 = st.columns(2)
+        with c1:
+            st.text_input(ui["ollama_base_label"], key="ollama_base_url")
+        with c2:
+            st.text_input(
+                ui["ollama_model_label"],
+                key="ollama_model",
+                help=ui["ollama_model_help"],
+            )
+        btn_col, _ = st.columns([1, 3])
+        with btn_col:
+            if st.button(ui["ollama_test_btn"], type="secondary", use_container_width=True, key="ollama_test_btn"):
+                base = (st.session_state.get("ollama_base_url") or "").strip().rstrip("/")
+                model = (st.session_state.get("ollama_model") or "").strip()
+                ok, msg = ollama_verify_connection(
+                    base_url=base or None,
+                    model=model or None,
+                )
+                st.session_state.ollama_last_verify_ok = ok
+                st.session_state.ollama_last_verify_msg = msg
+        msg = st.session_state.get("ollama_last_verify_msg")
+        if msg:
+            if st.session_state.get("ollama_last_verify_ok"):
+                st.success(msg)
+            else:
+                st.error(msg)
+
+
+def plotly_theme(fig):
+    fig.update_layout(
+        plot_bgcolor="rgba(0,0,0,0)",
+        paper_bgcolor="rgba(0,0,0,0)",
+        font=dict(family="Inter, Segoe UI, sans-serif", color="#24352a"),
+        margin=dict(l=12, r=12, t=20, b=20),
+    )
+    return fig
+
+
+def get_gemma_client_safe():
+    key = os.environ.get("GEMMA_API_KEY")
+    if not key:
+        return None
+    try:
+        return get_client(key)
+    except Exception:
+        return None
+
+
+def build_farmer_input(prefix: str, compact: bool = False, label_lang: Optional[str] = None) -> Dict:
+    lang = label_lang if label_lang is not None else st.session_state.language
+    text = LANG_TEXT[lang]
+    crops = list(CROP_WATER_MAP.keys())
+    states = ["Maharashtra", "Punjab"]
+
+    if compact:
+        c1, c2 = st.columns(2)
+        with c1:
+            crop = st.selectbox(text["crop"], crops, key=f"{prefix}_crop")
+            state = st.selectbox(text["state"], states, key=f"{prefix}_state")
+            land = st.slider(text["land"], 0.5, 20.0, 3.5, 0.5, key=f"{prefix}_land")
+            income = st.slider(text["income"], 1.0, 15.0, 3.0, 0.5, key=f"{prefix}_income")
+        with c2:
+            loan = st.slider(text["loan"], 0.5, 8.0, 2.0, 0.1, key=f"{prefix}_loan")
+            prev_default = st.radio(
+                text["default_hist"], ["No", "Yes"], horizontal=True, key=f"{prefix}_default"
+            )
+            rain_def = st.slider(text["rain_def"], 0, 60, 20, 1, key=f"{prefix}_rain")
+            spi = st.slider(text["spi"], -3.0, 2.0, -1.0, 0.1, key=f"{prefix}_spi")
+            drought = st.selectbox(
+                text["drought_years"],
+                DROUGHT_OPTIONS,
+                format_func=format_drought_years,
+                key=f"{prefix}_drought",
+            )
+    else:
+        st.markdown(f"<div class='section-label'>{html.escape(text['section_profile'])}</div>", unsafe_allow_html=True)
+        crop = st.selectbox(text["crop"], crops, key=f"{prefix}_crop")
+        state = st.selectbox(text["state"], states, key=f"{prefix}_state")
+        land = st.slider(text["land"], 0.5, 20.0, 3.5, 0.5, key=f"{prefix}_land")
+        income = st.slider(text["income"], 1.0, 15.0, 3.0, 0.5, key=f"{prefix}_income")
+        loan = st.slider(text["loan"], 0.5, 8.0, 2.0, 0.1, key=f"{prefix}_loan")
+        prev_default = st.radio(
+            text["default_hist"], ["No", "Yes"], horizontal=True, key=f"{prefix}_default"
+        )
+        st.markdown("---")
+        st.markdown(f"<div class='section-label'>{html.escape(text['section_climate'])}</div>", unsafe_allow_html=True)
+        rain_def = st.slider(text["rain_def"], 0, 60, 20, 1, key=f"{prefix}_rain")
+        spi = st.slider(text["spi"], -3.0, 2.0, -1.0, 0.1, key=f"{prefix}_spi")
+        spi_label_idx = 0 if spi <= -1.5 else 1 if spi <= -0.5 else 2
+        st.caption(SPI_LABELS[lang][spi_label_idx])
+        drought = st.selectbox(
+            text["drought_years"],
+            DROUGHT_OPTIONS,
+            format_func=format_drought_years,
+            key=f"{prefix}_drought",
+        )
+
+    return {
+        "age": 42,
+        "land_size_acres": land,
+        "annual_income_lakhs": income,
+        "loan_amount_lakhs": loan,
+        "previous_defaults": 1 if prev_default == "Yes" else 0,
+        "crop_type": crop,
+        "state": state,
+        "rainfall_deficit_pct": -float(rain_def),
+        "spi": float(spi),
+        "consecutive_drought_years": int(drought),
+    }
+
+
+def risk_chip_class(risk_band: str) -> str:
+    if "High" in risk_band:
+        return "chip-high"
+    if "Medium" in risk_band:
+        return "chip-medium"
+    return "chip-low"
+
+
+def risk_color_band(risk_band: str) -> str:
+    if "High" in risk_band:
+        return COLORS["danger"]
+    if "Medium" in risk_band:
+        return COLORS["accent"]
+    return COLORS["safe"]
+
+
+def build_score_gauge(score: int, risk_band: str):
+    color = risk_color_band(risk_band)
+    fig = go.Figure(
+        go.Indicator(
+            mode="gauge+number",
+            value=score,
+            number={"font": {"size": 44, "color": color}},
+            title={"text": "ClimaIQ Score (300-850)", "font": {"size": 15}},
+            gauge={
+                "axis": {"range": [300, 850], "tickwidth": 1},
+                "bar": {"color": color, "thickness": 0.34},
+                "steps": [
+                    {"range": [300, 600], "color": "#fde8e8"},
+                    {"range": [600, 650], "color": "#fff4de"},
+                    {"range": [650, 850], "color": "#e7f7ea"},
+                ],
+                "borderwidth": 0,
+            },
+        )
+    )
+    fig.update_layout(height=275)
+    return plotly_theme(fig)
+
+
+def extract_action_cards(text: str, language: str) -> List[str]:
+    fallback = (
+        ["🌱 Fasal Badlein", "💧 Sinchai Sudharen", "🛡️ Bima Karwain"]
+        if language == "Hindi"
+        else ["🌱 Shift Crop Mix", "💧 Improve Irrigation", "🛡️ Take Insurance"]
+    )
+    if not text:
+        return fallback
+    lines = [ln.strip(" -•\t") for ln in text.splitlines() if ln.strip()]
+    picks = [
+        ln for ln in lines if any(k in ln.lower() for k in ["insurance", "crop", "irrig", "बीमा", "फसल", "सिंच"])
+    ]
+    return (picks[:3] if len(picks) >= 3 else fallback)[:3]
+
+
+def try_farmer_explanation(client, farmer_input, result, language: str) -> str:
+    mode, ollama_base, ollama_model = get_inference_config()
+    if mode == "ollama":
+        try:
+            return explain_for_farmer(
+                farmer_input,
+                result,
+                language=language,
+                client=None,
+                inference_mode="ollama",
+                ollama_base=ollama_base,
+                ollama_model=ollama_model,
+            )
+        except Exception as ex:
+            return (
+                f"Ollama error: {ex}\n\n"
+                "Check that `ollama serve` is running and the model is pulled "
+                f"(e.g. `ollama pull {ollama_model}`)."
+            )
+    if client is None:
+        return (
+            "Cloud mode needs GEMMA_API_KEY (Google AI Studio), or switch sidebar to "
+            "Local (Ollama) for offline narratives."
+            if language == "English"
+            else "क्लाउड मोड के लिए GEMMA_API_KEY चाहिए, या साइडबार में Local (Ollama) चुनें।"
+        )
+    try:
+        return explain_for_farmer(
+            farmer_input,
+            result,
+            language=language,
+            client=client,
+            inference_mode="cloud",
+        )
+    except Exception as ex:
+        return f"Gemma response unavailable: {ex}"
+
+
+def render_score_tiles(result):
+    u = OFFICER_UI
+    c1, c2, c3 = st.columns(3)
+    with c1:
+        st.markdown(
+            f"<div class='score-tile'><div class='score-k'>{html.escape(u['lbl_score'])}</div>"
+            f"<div class='score-v'>{result['credit_score']} <span style='font-size:0.75rem;font-weight:500;color:#5e7566'>/ 850</span></div></div>",
+            unsafe_allow_html=True,
+        )
+    with c2:
+        st.markdown(
+            f"<div class='score-tile'><div class='score-k'>{html.escape(u['lbl_risk'])}</div>"
+            f"<div class='score-v'>{html.escape(str(result['risk_band']))}</div></div>",
+            unsafe_allow_html=True,
+        )
+    with c3:
+        st.markdown(
+            f"<div class='score-tile'><div class='score-k'>{html.escape(u['lbl_default'])}</div>"
+            f"<div class='score-v'>{result['default_probability']}%</div></div>",
+            unsafe_allow_html=True,
+        )
+
+
+def tab_kisan(client):
+    lang = st.session_state.language
+    text = LANG_TEXT[lang]
+    st.markdown(f"<div class='tab-hint'>{html.escape(text['tab_kisan_hint'])}</div>", unsafe_allow_html=True)
+    left, right = st.columns([1.05, 1], gap="large")
+
+    with left:
+        st.markdown(f"<div class='panel-title'>{html.escape(text['your_info'])}</div>", unsafe_allow_html=True)
+        with st.container(border=True):
+            farmer_input = build_farmer_input("kisan", compact=False)
+            run = st.button(text["cta"], type="primary", use_container_width=True)
+        if run:
+            with st.spinner("Computing your ClimaIQ score…"):
+                result = predict_single(farmer_input, st.session_state.model, st.session_state.scaler)
+                explanation = try_farmer_explanation(client, farmer_input, result, lang)
+            mode_run, _, _ = get_inference_config()
+            st.session_state.last_result = {
+                "input": farmer_input,
+                "result": result,
+                "explanations": {lang: explanation},
+                "explanation_sources": {lang: mode_run},
+            }
+            st.success("Done. Your results are on the right.")
+
+    with right:
+        st.markdown(f"<div class='panel-title'>{html.escape(text['score_title'])}</div>", unsafe_allow_html=True)
+        if not st.session_state.last_result:
+            st.markdown(narr_html(text["empty_kisan"]), unsafe_allow_html=True)
+            return
+        stored = st.session_state.last_result
+        result = stored["result"]
+        if lang not in stored["explanations"]:
+            with st.spinner("Updating explanation for the selected language…"):
+                stored["explanations"][lang] = try_farmer_explanation(client, stored["input"], result, lang)
+                stored.setdefault("explanation_sources", {})[lang] = get_inference_config()[0]
+        explanation = stored["explanations"][lang]
+        expl_mode = stored.get("explanation_sources", {}).get(lang, get_inference_config()[0])
+        cards = extract_action_cards(explanation, lang)
+
+        st.plotly_chart(build_score_gauge(result["credit_score"], result["risk_band"]), use_container_width=True)
+        st.markdown(
+            f"<div class='hero-metric'><span class='risk-pill {risk_chip_class(result['risk_band'])}'>"
+            f"{html.escape(str(result['risk_band']))}</span><br>"
+            f"<b>{html.escape(text['prob'])}: {result['default_probability']}%</b></div>",
+            unsafe_allow_html=True,
+        )
+        st.markdown(f"**{text['farmer_expl']}**")
+        st.caption(narrative_source_caption(expl_mode, lang))
+        st.markdown(narr_html(explanation), unsafe_allow_html=True)
+        st.markdown(f"**{text['actions']}**")
+        a1, a2, a3 = st.columns(3)
+        for col, item in zip([a1, a2, a3], cards):
+            with col:
+                st.markdown(
+                    f"<div class='action-chip'>{html.escape(item)}</div>",
+                    unsafe_allow_html=True,
+                )
+
+
+def tab_officer(client):
+    u = OFFICER_UI
+    st.markdown(f"<div class='panel-title'>{html.escape(u['form_title'])}</div>", unsafe_allow_html=True)
+    st.markdown(f"<div class='tab-hint'>{html.escape(u['form_help'])}</div>", unsafe_allow_html=True)
+    with st.expander(u["expander"], expanded=True):
+        borrower_input = build_farmer_input("officer", compact=True, label_lang="English")
+        run = st.button(u["run"], type="primary", use_container_width=True)
+    if run:
+        with st.spinner("Running ClimaIQ assessment…"):
+            result = predict_single(borrower_input, st.session_state.model, st.session_state.scaler)
+        st.session_state.last_officer_result = {
+            "input": borrower_input,
+            "result": result,
+            "reports": {},
+            "report_sources": {},
+        }
+        st.success("Assessment complete. Scorecard is below.")
+
+    if not st.session_state.last_officer_result:
+        st.markdown(narr_html(u["empty"]), unsafe_allow_html=True)
+        return
+
+    stored = st.session_state.last_officer_result
+    result = stored["result"]
+    st.markdown(f"<div class='panel-title'>{html.escape(u['scorecard'])}</div>", unsafe_allow_html=True)
+    render_score_tiles(result)
+
+    rec = html.escape(str(result["recommended_action"]))
+    note = html.escape(u["moratorium_note"])
+    st.markdown(
+        f"<div class='soft-note narr-body'><b>{html.escape(u['rec_action'])}:</b> {rec}<br><br>{note}</div>",
+        unsafe_allow_html=True,
+    )
+
+    st.markdown(f"<div class='panel-title'>{html.escape(u['risk_drivers'])}</div>", unsafe_allow_html=True)
+    st.caption(u["risk_drivers_caption"])
+    df = pd.DataFrame(result["top_risk_drivers"])
+    df["impact"] = df["contribution"].abs()
+    df = df.sort_values("impact", ascending=True)
+    bar = px.bar(
+        df,
+        x="impact",
+        y="display_name",
+        orientation="h",
+        color="impact",
+        color_continuous_scale=["#2E7D32", "#FF8F00", "#C62828"],
+    )
+    bar.update_layout(height=295, coloraxis_showscale=False, xaxis_title="Relative impact", yaxis_title="")
+    st.plotly_chart(plotly_theme(bar), use_container_width=True)
+
+    st.markdown(f"<div class='panel-title'>{html.escape(u['ai_report'])}</div>", unsafe_allow_html=True)
+    rep_lang = st.radio(u["report_lang"], ["English", "Hindi"], horizontal=True, key="officer_rep_lang")
+    if rep_lang not in stored["reports"]:
+        mode, ollama_base, ollama_model = get_inference_config()
+        if mode == "ollama":
+            try:
+                with st.spinner("Generating assessment report (Ollama)…"):
+                    stored["reports"][rep_lang] = explain_for_officer(
+                        stored["input"],
+                        result,
+                        language=rep_lang,
+                        client=None,
+                        inference_mode="ollama",
+                        ollama_base=ollama_base,
+                        ollama_model=ollama_model,
+                    )
+            except Exception as ex:
+                stored["reports"][rep_lang] = (
+                    f"Ollama error: {ex}\n\nEnsure `ollama serve` is running and "
+                    f"`ollama pull {ollama_model}` has been run on this machine."
+                )
+        elif client is None:
+            stored["reports"][rep_lang] = (
+                "Cloud mode needs a `GEMMA_API_KEY` secret on this Space, or switch the sidebar to "
+                "Local (Ollama) for offline AI reports. Scorecard and charts above still work."
+            )
+        else:
+            try:
+                with st.spinner("Generating assessment report…"):
+                    stored["reports"][rep_lang] = explain_for_officer(
+                        stored["input"],
+                        result,
+                        language=rep_lang,
+                        client=client,
+                        inference_mode="cloud",
+                    )
+            except Exception as ex:
+                stored["reports"][rep_lang] = f"Gemma response unavailable: {ex}"
+        stored.setdefault("report_sources", {})[rep_lang] = get_inference_config()[0]
+    report_text = stored["reports"][rep_lang]
+    rep_mode = stored.get("report_sources", {}).get(rep_lang, get_inference_config()[0])
+    st.caption(narrative_source_caption(rep_mode, rep_lang))
+    render_driver_micro_bars(result)
+    render_officer_report_ui(report_text)
+
+    st.caption(u["pdf_hint"])
+    try:
+        pdf_bytes = build_officer_assessment_pdf(report_text, stored["input"], result, rep_lang)
+        st.download_button(
+            u["download"],
+            data=pdf_bytes,
+            file_name=f"climaiq_officer_report_{rep_lang.lower()}.pdf",
+            mime="application/pdf",
+            use_container_width=True,
+        )
+    except Exception as ex:
+        st.warning(
+            f"PDF build failed ({ex}). First run may need network access to cache fonts; try again, or use the text fallback."
+        )
+        st.download_button(
+            u["download_txt_fallback"],
+            data=io.BytesIO(report_text.encode("utf-8")),
+            file_name=f"climaiq_officer_report_{rep_lang.lower()}.txt",
+            mime="text/plain",
+            use_container_width=True,
+        )
+
+
+def build_portfolio(n_loans: int, avg_loan: float, dominant_crop: str):
+    crops = list(CROP_WATER_MAP.keys())
+    probs_map = {
+        "Cotton": [0.5 if c == "Cotton" else 0.5 / (len(crops) - 1) for c in crops],
+        "Rice": [0.5 if c == "Rice" else 0.5 / (len(crops) - 1) for c in crops],
+        "Wheat": [0.5 if c == "Wheat" else 0.5 / (len(crops) - 1) for c in crops],
+        "Mixed": [1 / len(crops)] * len(crops),
+    }
+    crop_probs = probs_map[dominant_crop]
+    portfolio = []
+    for _ in range(n_loans):
+        crop = np.random.choice(crops, p=crop_probs)
+        portfolio.append(
+            {
+                "age": int(np.random.randint(28, 62)),
+                "land_size_acres": round(np.random.uniform(1, 10), 2),
+                "annual_income_lakhs": round(np.random.uniform(1.5, 10), 2),
+                "loan_amount_lakhs": round(max(0.5, np.random.normal(avg_loan, 0.6)), 2),
+                "previous_defaults": int(np.random.choice([0, 1], p=[0.82, 0.18])),
+                "crop_type": crop,
+                "state": str(np.random.choice(["Maharashtra", "Punjab"], p=[0.6, 0.4])),
+                "rainfall_deficit_pct": 0.0,
+                "spi": 0.0,
+                "consecutive_drought_years": 0,
+            }
+        )
+    return portfolio
+
+
+def tab_portfolio(client):
+    lang = st.session_state.language
+    text = LANG_TEXT[lang]
+    st.markdown(f"<div class='tab-hint'>{html.escape(text['tab_portfolio_hint'])}</div>", unsafe_allow_html=True)
+    st.markdown("<div class='panel-title'>Portfolio configuration</div>", unsafe_allow_html=True)
+    with st.container(border=True):
+        c1, c2, c3 = st.columns(3)
+        with c1:
+            n_loans = st.number_input("Number of loans to simulate", min_value=20, max_value=2000, value=100)
+        with c2:
+            avg_loan = st.number_input("Average loan size (₹ lakhs)", min_value=0.5, max_value=8.0, value=3.0, step=0.1)
+        with c3:
+            recovery_rate = st.slider("Recovery rate (%)", 0, 80, 30, 1)
+        dominant_crop = st.radio(
+            "Dominant crop in portfolio",
+            ["Cotton", "Rice", "Wheat", "Mixed"],
+            horizontal=True,
+            key="portfolio_dominant_crop",
+        )
+        run = st.button("Run stress test", type="primary", use_container_width=True)
+
+    if run:
+        with st.spinner("Simulating portfolio across climate scenarios…"):
+            portfolio = build_portfolio(int(n_loans), float(avg_loan), dominant_crop)
+            stress = run_stress_test(st.session_state.model, st.session_state.scaler, portfolio)
+        rr = recovery_rate / 100
+        for row in stress:
+            row["total_loss_lakhs"] = round(row["avg_default_pct"] / 100 * avg_loan * (1 - rr) * n_loans, 2)
+        st.session_state.last_stress_result = {"stress": stress, "narratives": {}, "narrative_sources": {}}
+        st.success("Stress test complete. Charts are below.")
+
+    if not st.session_state.last_stress_result:
+        st.markdown(narr_html(text["empty_portfolio"]), unsafe_allow_html=True)
+        return
+
+    stress = st.session_state.last_stress_result["stress"]
+    df = pd.DataFrame(stress)
+    order = ["Normal Monsoon", "Moderate Drought", "Severe Drought", "Back-to-Back Drought"]
+    df["scenario"] = pd.Categorical(df["scenario"], order, ordered=True)
+    df = df.sort_values("scenario")
+
+    st.markdown("<div class='panel-title'>Estimated loss by scenario (₹ lakhs)</div>", unsafe_allow_html=True)
+    st.caption(
+        "Illustrative portfolio simulation: loss scales with count of loans, average ticket, recovery rate, and modelled default rates."
+    )
+    bar = px.bar(
+        df,
+        x="total_loss_lakhs",
+        y="scenario",
+        orientation="h",
+        text="total_loss_lakhs",
+        color="scenario",
+        color_discrete_map={
+            "Normal Monsoon": "#2E7D32",
+            "Moderate Drought": "#FFB300",
+            "Severe Drought": "#E53935",
+            "Back-to-Back Drought": "#7F0000",
+        },
+    )
+    bar.update_layout(showlegend=False, yaxis_title="", xaxis_title="Estimated loss (₹ lakhs)", height=330)
+    st.plotly_chart(plotly_theme(bar), use_container_width=True)
+
+    st.markdown("<div class='panel-title'>Average default rate by scenario</div>", unsafe_allow_html=True)
+    st.caption("The step up from “normal” to stressed monsoons shows how climate shocks can move the book non-linearly.")
+    line = px.line(df, x="scenario", y="avg_default_pct", markers=True, color_discrete_sequence=[COLORS["danger"]])
+    line.update_layout(yaxis_title="Portfolio average default (%)", xaxis_title="", height=300)
+    st.plotly_chart(plotly_theme(line), use_container_width=True)
+
+    st.markdown("<div class='panel-title'>Portfolio risk narrative</div>", unsafe_allow_html=True)
+    rep_lang = st.radio("Narrative language", ["English", "Hindi"], horizontal=True, key="port_lang")
+    narratives = st.session_state.last_stress_result["narratives"]
+    if rep_lang not in narratives:
+        mode, ollama_base, ollama_model = get_inference_config()
+        if mode == "ollama":
+            try:
+                with st.spinner("Drafting portfolio narrative (Ollama)…"):
+                    narratives[rep_lang] = explain_portfolio_stress(
+                        stress,
+                        language=rep_lang,
+                        client=None,
+                        inference_mode="ollama",
+                        ollama_base=ollama_base,
+                        ollama_model=ollama_model,
+                    )
+            except Exception as ex:
+                narratives[rep_lang] = f"Ollama error: {ex}"
+        elif client is None:
+            narratives[rep_lang] = (
+                "Set `GEMMA_API_KEY` for cloud narratives, or use Local (Ollama) in the sidebar. "
+                "Figures above are already computed from your ClimaIQ engine."
+            )
+        else:
+            try:
+                with st.spinner("Drafting portfolio narrative…"):
+                    narratives[rep_lang] = explain_portfolio_stress(
+                        stress, language=rep_lang, client=client, inference_mode="cloud"
+                    )
+            except Exception as ex:
+                narratives[rep_lang] = f"Gemma response unavailable: {ex}"
+        st.session_state.last_stress_result.setdefault("narrative_sources", {})[rep_lang] = get_inference_config()[0]
+    narr_mode = st.session_state.last_stress_result.get("narrative_sources", {}).get(
+        rep_lang, get_inference_config()[0]
+    )
+    st.caption(narrative_source_caption(narr_mode, rep_lang))
+    st.markdown(narr_html(narratives[rep_lang]), unsafe_allow_html=True)
+
+    normal_loss = float(df[df["scenario"] == "Normal Monsoon"]["total_loss_lakhs"].iloc[0])
+    worst_loss = float(df["total_loss_lakhs"].max())
+    gap = round(worst_loss - normal_loss, 2)
+    buf = int(gap // 100 * 100)
+    cap_lines = (
+        f"<b>Capital buffer indicator</b><br><br>"
+        f"Baseline (normal monsoon) loss estimate: ₹{normal_loss} lakhs<br>"
+        f"Incremental stress to worst scenario shown: ₹{gap} lakhs<br>"
+        f"Rule-of-thumb buffer to discuss with risk: ₹{buf} lakhs+"
+    )
+    st.markdown(f"<div class='soft-note narr-body'>{cap_lines}</div>", unsafe_allow_html=True)
+
+
+def main():
+    init_state()
+    inject_styles(hide_sidebar=bool(st.session_state.get("ui_hide_sidebar", False)))
+    client = get_gemma_client_safe()
+    text = LANG_TEXT[st.session_state.language]
+    _, layout_toggle = st.columns([5.5, 1.5])
+    with layout_toggle:
+        st.checkbox(
+            text["hide_sidebar_label"],
+            key="ui_hide_sidebar",
+            help=text["hide_sidebar_help"],
+        )
+
+    with st.sidebar:
+        st.markdown("## ClimaIQ Kisan")
+        st.markdown(
+            f"<div class='small-muted' style='color:#c8e6c9;line-height:1.45'>{html.escape(text['app_subtitle'])}</div>",
+            unsafe_allow_html=True,
+        )
+        st.divider()
+        st.markdown("### Language / भाषा")
+        st.radio(
+            "Display language",
+            ["English", "Hindi"],
+            horizontal=True,
+            label_visibility="collapsed",
+            key="language",
+        )
+        text = LANG_TEXT[st.session_state.language]
+        st.caption("Applies to farmer-facing labels and Gemma prompts tied to that view.")
+        st.divider()
+        st.markdown("### Gemma inference")
+        st.radio(
+            "Mode",
+            ["cloud", "ollama"],
+            format_func=format_gemma_inference_mode,
+            key="gemma_inference_mode",
+            on_change=_invalidate_gemma_outputs,
+        )
+        st.caption(text["sidebar_gemma_caption"])
+        st.divider()
+        mode_now = st.session_state.get("gemma_inference_mode", "cloud")
+        if mode_now == "ollama":
+            ok_local = st.session_state.get("ollama_last_verify_ok")
+            model = st.session_state.get("ollama_model", "")
+            base = st.session_state.get("ollama_base_url", "")
+            if ok_local:
+                st.caption(text["sidebar_gemma_local_verified"].format(model=model, base=base))
+            else:
+                st.caption(text["sidebar_gemma_local_unverified"].format(model=model, base=base))
+        elif client is not None:
+            st.caption(text["sidebar_gemma_cloud_ok"])
+        else:
+            st.caption(text["sidebar_gemma_cloud_missing"])
+        st.divider()
+        st.markdown("### About")
+        st.caption("ClimaIQ engine + Gemma 4 explanations.")
+        st.caption("Model AUC (hold-out): 0.804")
+
+    text = LANG_TEXT[st.session_state.language]
+    st.markdown(
+        f"<div class='header-wrap'><div class='header-title'>{html.escape(text['hero_title'])}</div>"
+        f"<div class='header-sub'>{html.escape(text['hero_sub'])}</div></div>",
+        unsafe_allow_html=True,
+    )
+    render_ollama_connection_main()
+    tab1, tab2, tab3 = st.tabs(["Kisan view", "Loan officer view", "Portfolio stress test"])
+
+    with tab1:
+        tab_kisan(client)
+    with tab2:
+        tab_officer(client)
+    with tab3:
+        tab_portfolio(client)
+
+
+if __name__ == "__main__":
+    main()

climaiq_engine.py

@@ -0,0 +1,450 @@
+"""
+ClimaIQ Engine — Climate-Adjusted Credit Risk Model
+Core prediction module for ClimaIQ Kisan
+Author: Krishna Dahale
+"""
+
+import numpy as np
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+from sklearn.linear_model import LogisticRegression
+import joblib
+import os
+import warnings
+warnings.filterwarnings("ignore")
+
+# ─── Feature Definitions ───────────────────────────────────────────────────────
+
+TRADITIONAL_FEATURES = [
+    "Age", "Land_Size_Acres", "Annual_Income_Lakhs",
+    "Loan_Amount_Lakhs", "Debt_to_Income_Ratio",
+    "Previous_Defaults", "Land_Productivity"
+]
+
+CLIMATE_FEATURES = [
+    "Rainfall_Deficit_Pct", "SPI", "Consecutive_Drought_Years"
+]
+
+ENHANCED_FEATURES = [
+    "State_Maharashtra", "Crop_Water_Intensive", "Crop_Drought_Interaction"
+]
+
+ALL_FEATURES = TRADITIONAL_FEATURES + CLIMATE_FEATURES + ENHANCED_FEATURES
+
+CROP_WATER_MAP = {
+    "Rice": 1.0,
+    "Sugarcane": 1.0,
+    "Cotton": 0.5,
+    "Wheat": 0.2,
+    "Millets": 0.0
+}
+
+FEATURE_DISPLAY_NAMES = {
+    "Crop_Drought_Interaction": "Crop × Drought Severity",
+    "Rainfall_Deficit_Pct": "Rainfall Deficit",
+    "SPI": "Drought Severity (SPI)",
+    "Consecutive_Drought_Years": "Consecutive Drought Years",
+    "Previous_Defaults": "Previous Default History",
+    "Debt_to_Income_Ratio": "Debt-to-Income Ratio",
+    "State_Maharashtra": "Geographic Risk (Maharashtra)",
+    "Crop_Water_Intensive": "Crop Water Dependency",
+    "Annual_Income_Lakhs": "Annual Income",
+    "Loan_Amount_Lakhs": "Loan Amount",
+    "Land_Size_Acres": "Land Size",
+    "Age": "Borrower Age",
+    "Land_Productivity": "Land Productivity"
+}
+
+# ─── Data Generation ───────────────────────────────────────────────────────────
+
+def generate_training_data(n_loans=5000, seed=42):
+    """Generate synthetic agricultural loan dataset."""
+    np.random.seed(seed)
+
+    df = pd.DataFrame({
+        "Loan_ID": range(1, n_loans + 1),
+        "State": np.random.choice(["Maharashtra", "Punjab"], n_loans, p=[0.6, 0.4]),
+        "Age": np.random.normal(45, 12, n_loans).clip(25, 70),
+        "Land_Size_Acres": np.random.exponential(3, n_loans).clip(0.5, 20),
+        "Annual_Income_Lakhs": np.random.lognormal(2.5, 0.8, n_loans).clip(1, 15),
+        "Loan_Amount_Lakhs": np.random.uniform(0.5, 8, n_loans),
+        "Previous_Defaults": np.random.choice([0, 1], n_loans, p=[0.85, 0.15]),
+        "Crop_Type": np.random.choice(
+            ["Rice", "Cotton", "Wheat", "Sugarcane", "Millets"],
+            n_loans,
+            p=[0.25, 0.20, 0.25, 0.15, 0.15],
+        ),
+    })
+
+    df["Rainfall_Deficit_Pct"] = np.random.normal(-12, 15, n_loans).clip(-60, 20)
+    df["SPI"] = np.random.normal(-0.3, 1.0, n_loans).clip(-3, 2)
+    df["Consecutive_Drought_Years"] = np.random.choice([0, 1, 2, 3], n_loans, p=[0.5, 0.3, 0.15, 0.05])
+    df["Debt_to_Income_Ratio"] = df["Loan_Amount_Lakhs"] / df["Annual_Income_Lakhs"]
+    df["Land_Productivity"] = df["Annual_Income_Lakhs"] / df["Land_Size_Acres"]
+
+    def generate_default(row):
+        p = 0.08
+        if row["Previous_Defaults"] == 1: p *= 4
+        if row["Debt_to_Income_Ratio"] > 0.5: p *= 2
+        elif row["Debt_to_Income_Ratio"] > 0.3: p *= 1.5
+        if row["Rainfall_Deficit_Pct"] < -25: p *= 2.5
+        elif row["Rainfall_Deficit_Pct"] < -10: p *= 1.4
+        if row["SPI"] < -1.5: p *= 2
+        elif row["SPI"] < -1.0: p *= 1.3
+        if row["Consecutive_Drought_Years"] >= 2: p *= 2.5
+        if row["Crop_Type"] in ["Rice", "Sugarcane"] and row["Rainfall_Deficit_Pct"] < -15: p *= 1.6
+        if row["Age"] < 30 or row["Age"] > 60: p *= 1.2
+        if row["State"] == "Maharashtra": p *= 1.3
+        return 1 if np.random.rand() < min(p, 0.95) else 0
+
+    df["Default"] = df.apply(generate_default, axis=1)
+
+    # Feature engineering
+    df["State_Maharashtra"] = (df["State"] == "Maharashtra").astype(int)
+    df["Crop_Water_Intensive"] = df["Crop_Type"].map(CROP_WATER_MAP)
+    df["Crop_Drought_Interaction"] = (
+        df["Crop_Water_Intensive"] * df["Rainfall_Deficit_Pct"].clip(upper=0).abs()
+    )
+
+    return df
+
+
+# ─── Model Training ────────────────────────────────────────────────────────────
+
+def train_model():
+    """Train the climate-adjusted credit risk model. Returns model and scaler."""
+    df = generate_training_data()
+
+    X = df[ALL_FEATURES]
+    y = df["Default"]
+
+    X_train, _, y_train, _ = train_test_split(
+        X, y, test_size=0.3, stratify=y, random_state=42
+    )
+
+    scaler = StandardScaler()
+    X_train_scaled = scaler.fit_transform(X_train)
+
+    model = LogisticRegression(max_iter=1000, class_weight="balanced")
+    model.fit(X_train_scaled, y_train)
+
+    return model, scaler
+
+
+# ─── Save / Load ───────────────────────────────────────────────────────────────
+
+def save_model(model, scaler, model_path="climaiq_model.pkl", scaler_path="climaiq_scaler.pkl"):
+    joblib.dump(model, model_path)
+    joblib.dump(scaler, scaler_path)
+
+
+def load_model(model_path="climaiq_model.pkl", scaler_path="climaiq_scaler.pkl"):
+    """Load saved model and scaler. Trains fresh if files not found."""
+    if os.path.exists(model_path) and os.path.exists(scaler_path):
+        model = joblib.load(model_path)
+        scaler = joblib.load(scaler_path)
+    else:
+        model, scaler = train_model()
+        save_model(model, scaler, model_path, scaler_path)
+    return model, scaler
+
+
+# ─── Feature Engineering for Single Input ─────────────────────────────────────
+
+def engineer_single_input(data: dict) -> pd.DataFrame:
+    """
+    Convert raw farmer input dict into model-ready feature DataFrame.
+
+    Expected keys:
+        age, land_size_acres, annual_income_lakhs, loan_amount_lakhs,
+        previous_defaults, crop_type, state,
+        rainfall_deficit_pct, spi, consecutive_drought_years
+    """
+    debt_to_income = data["loan_amount_lakhs"] / data["annual_income_lakhs"]
+    land_productivity = data["annual_income_lakhs"] / data["land_size_acres"]
+    state_mh = 1 if data["state"] == "Maharashtra" else 0
+    crop_water = CROP_WATER_MAP.get(data["crop_type"], 0.5)
+    crop_drought = crop_water * abs(min(data["rainfall_deficit_pct"], 0))
+
+    features = {
+        "Age": data["age"],
+        "Land_Size_Acres": data["land_size_acres"],
+        "Annual_Income_Lakhs": data["annual_income_lakhs"],
+        "Loan_Amount_Lakhs": data["loan_amount_lakhs"],
+        "Debt_to_Income_Ratio": debt_to_income,
+        "Previous_Defaults": data["previous_defaults"],
+        "Land_Productivity": land_productivity,
+        "Rainfall_Deficit_Pct": data["rainfall_deficit_pct"],
+        "SPI": data["spi"],
+        "Consecutive_Drought_Years": data["consecutive_drought_years"],
+        "State_Maharashtra": state_mh,
+        "Crop_Water_Intensive": crop_water,
+        "Crop_Drought_Interaction": crop_drought,
+    }
+
+    return pd.DataFrame([features])[ALL_FEATURES]
+
+
+# ─── Credit Score Calculation ──────────────────────────────────────────────────
+
+def compute_credit_score(model, scaler, feature_row: pd.DataFrame,
+                          base_score=650, pdo=50) -> int:
+    """Convert default probability into 300–850 credit score."""
+    coef = model.coef_[0]
+    factor = pdo / np.log(2)
+    offset = base_score - factor * np.log(20)
+
+    z = scaler.transform(feature_row)[0]
+    score = offset + np.sum(z * (-factor * coef))
+    return int(np.clip(score, 300, 850))
+
+
+def get_risk_band(score: int) -> str:
+    if score >= 700:
+        return "Very Low Risk"
+    elif score >= 650:
+        return "Low Risk"
+    elif score >= 600:
+        return "Medium Risk"
+    else:
+        return "High Risk"
+
+
+# ─── Risk Driver Extraction ────────────────────────────────────────────────────
+
+def get_top_risk_drivers(model, scaler, feature_row: pd.DataFrame, top_n=3) -> list:
+    """
+    Return the top N features driving default risk for this borrower.
+    Each driver is a dict: {feature, display_name, contribution, direction}
+    """
+    z = scaler.transform(feature_row)[0]
+    coef = model.coef_[0]
+
+    # Contribution = standardized value × coefficient
+    # Positive contribution = increases default probability = increases risk
+    contributions = z * coef
+
+    driver_df = pd.DataFrame({
+        "feature": ALL_FEATURES,
+        "contribution": contributions
+    }).sort_values("contribution", ascending=False)
+
+    # Top risk-increasing drivers only
+    top_drivers = driver_df.head(top_n)
+
+    result = []
+    for _, row in top_drivers.iterrows():
+        result.append({
+            "feature": row["feature"],
+            "display_name": FEATURE_DISPLAY_NAMES.get(row["feature"], row["feature"]),
+            "contribution": round(row["contribution"], 4),
+            "direction": "increases risk" if row["contribution"] > 0 else "reduces risk"
+        })
+
+    return result
+
+
+# ─── Main Prediction Function ──────────────────────────────────────────────────
+
+def predict_single(data: dict, model, scaler) -> dict:
+    """
+    Full prediction pipeline for a single farmer/borrower.
+
+    Returns:
+        default_probability  : float (0–100)
+        credit_score         : int (300–850)
+        risk_band            : str
+        top_risk_drivers     : list of dicts
+        recommended_action   : str
+    """
+    feature_row = engineer_single_input(data)
+
+    scaled = scaler.transform(feature_row)
+    default_prob = model.predict_proba(scaled)[0][1] * 100
+
+    credit_score = compute_credit_score(model, scaler, feature_row)
+    risk_band = get_risk_band(credit_score)
+    top_drivers = get_top_risk_drivers(model, scaler, feature_row)
+
+    # Recommended action based on risk band
+    action_map = {
+        "Very Low Risk": "Auto-Approve",
+        "Low Risk": "Approve",
+        "Medium Risk": "Manual Review",
+        "High Risk": "Decline or Offer Climate-Linked Premium Rate"
+    }
+
+    return {
+        "default_probability": round(default_prob, 2),
+        "credit_score": credit_score,
+        "risk_band": risk_band,
+        "top_risk_drivers": top_drivers,
+        "recommended_action": action_map[risk_band]
+    }
+
+
+# ─── Stress Test ───────────────────────────────────────────────────────────────
+
+def run_stress_test(model, scaler, base_portfolio: list) -> list:
+    """
+    Run 4 climate scenarios across a list of farmer input dicts.
+    Returns list of scenario result dicts.
+    """
+    scenarios = {
+        "Normal Monsoon":     {"rainfall_deficit_pct": 0,   "spi": 0,    "consecutive_drought_years": 0},
+        "Moderate Drought":   {"rainfall_deficit_pct": -20, "spi": -1.2, "consecutive_drought_years": 1},
+        "Severe Drought":     {"rainfall_deficit_pct": -35, "spi": -1.8, "consecutive_drought_years": 1},
+        "Back-to-Back Drought": {"rainfall_deficit_pct": -25, "spi": -1.5, "consecutive_drought_years": 2},
+    }
+
+    avg_loan = 3.0
+    recovery = 0.30
+    results = []
+
+    for scenario_name, overrides in scenarios.items():
+        probs = []
+        for farmer in base_portfolio:
+            stressed = {**farmer, **overrides}
+            result = predict_single(stressed, model, scaler)
+            probs.append(result["default_probability"] / 100)
+
+        avg_default_pct = np.mean(probs) * 100
+        total_loss = np.mean(probs) * avg_loan * (1 - recovery) * len(base_portfolio)
+
+        results.append({
+            "scenario": scenario_name,
+            "avg_default_pct": round(avg_default_pct, 2),
+            "total_loss_lakhs": round(total_loss, 2),
+            "portfolio_size": len(base_portfolio)
+        })
+
+    return results
+
+
+# ─── Kaggle / research export ─────────────────────────────────────────────────
+
+KAGGLE_DATASET_COLUMNS = [
+    "loan_id",
+    "state",
+    "age",
+    "land_size_acres",
+    "annual_income_lakhs",
+    "loan_amount_lakhs",
+    "crop_type",
+    "previous_defaults",
+    "rainfall_deficit_pct",
+    "spi",
+    "consecutive_drought_years",
+    "debt_to_income_ratio",
+    "land_productivity",
+    "crop_water_intensity",
+    "crop_drought_interaction",
+    "state_maharashtra",
+    "default",
+    "climaiq_score",
+    "risk_band",
+]
+
+
+def _simplify_risk_band(label: str) -> str:
+    """Kaggle-friendly labels: High / Medium / Low / Very Low."""
+    return label.replace(" Risk", "").strip()
+
+
+def _training_row_to_farmer_dict(row: pd.Series) -> dict:
+    """Map a generate_training_data row to the dict expected by predict_single."""
+    return {
+        "age": int(round(float(row["Age"]))),
+        "land_size_acres": float(row["Land_Size_Acres"]),
+        "annual_income_lakhs": float(row["Annual_Income_Lakhs"]),
+        "loan_amount_lakhs": float(row["Loan_Amount_Lakhs"]),
+        "previous_defaults": int(row["Previous_Defaults"]),
+        "crop_type": str(row["Crop_Type"]),
+        "state": str(row["State"]),
+        "rainfall_deficit_pct": float(row["Rainfall_Deficit_Pct"]),
+        "spi": float(row["SPI"]),
+        "consecutive_drought_years": int(row["Consecutive_Drought_Years"]),
+    }
+
+
+def build_kaggle_export_dataframe(
+    n_rows: int = 1000,
+    seed: int = 42,
+    model=None,
+    scaler=None,
+) -> pd.DataFrame:
+    """
+    Build a tabular dataset aligned with the ClimaIQ engine: raw inputs, engineered
+    features, synthetic default target (causal logic in generate_training_data), and
+    ClimaIQ credit score / risk band from the trained logistic model.
+
+    Intended for Kaggle / research (e.g. *ClimaIQ — Climate-Adjusted Agricultural
+    Credit Risk Dataset (India)*).
+    """
+    if model is None or scaler is None:
+        model, scaler = load_model()
+
+    src = generate_training_data(n_rows, seed=seed)
+    records = []
+
+    for _, row in src.iterrows():
+        farmer = _training_row_to_farmer_dict(row)
+        pred = predict_single(farmer, model, scaler)
+        records.append(
+            {
+                "loan_id": int(row["Loan_ID"]),
+                "state": str(row["State"]),
+                "age": farmer["age"],
+                "land_size_acres": round(farmer["land_size_acres"], 3),
+                "annual_income_lakhs": round(farmer["annual_income_lakhs"], 3),
+                "loan_amount_lakhs": round(farmer["loan_amount_lakhs"], 3),
+                "crop_type": farmer["crop_type"],
+                "previous_defaults": farmer["previous_defaults"],
+                "rainfall_deficit_pct": round(farmer["rainfall_deficit_pct"], 3),
+                "spi": round(farmer["spi"], 3),
+                "consecutive_drought_years": farmer["consecutive_drought_years"],
+                "debt_to_income_ratio": round(float(row["Debt_to_Income_Ratio"]), 4),
+                "land_productivity": round(float(row["Land_Productivity"]), 4),
+                "crop_water_intensity": round(float(row["Crop_Water_Intensive"]), 4),
+                "crop_drought_interaction": round(float(row["Crop_Drought_Interaction"]), 4),
+                "state_maharashtra": int(row["State_Maharashtra"]),
+                "default": int(row["Default"]),
+                "climaiq_score": int(pred["credit_score"]),
+                "risk_band": _simplify_risk_band(str(pred["risk_band"])),
+            }
+        )
+
+    out = pd.DataFrame(records)
+    return out[KAGGLE_DATASET_COLUMNS]
+
+
+# ─── Quick Test ────────────────────────────────────────────────────────────────
+
+if __name__ == "__main__":
+    print("Training ClimaIQ model...")
+    model, scaler = load_model()
+    print("Model ready.\n")
+
+    sample_farmer = {
+        "age": 42,
+        "land_size_acres": 3.5,
+        "annual_income_lakhs": 3.0,
+        "loan_amount_lakhs": 2.0,
+        "previous_defaults": 0,
+        "crop_type": "Cotton",
+        "state": "Maharashtra",
+        "rainfall_deficit_pct": -35.0,
+        "spi": -1.8,
+        "consecutive_drought_years": 1
+    }
+
+    result = predict_single(sample_farmer, model, scaler)
+
+    print(f"Credit Score     : {result['credit_score']}")
+    print(f"Risk Band        : {result['risk_band']}")
+    print(f"Default Prob     : {result['default_probability']}%")
+    print(f"Recommended      : {result['recommended_action']}")
+    print(f"\nTop Risk Drivers:")
+    for d in result["top_risk_drivers"]:
+        print(f"  - {d['display_name']} ({d['direction']})")

climaiq_gemma.py

@@ -0,0 +1,407 @@
+"""
+ClimaIQ Gemma Layer — Bilingual Explanation Engine
+Converts ClimaIQ model output into plain language for farmers and loan officers.
+
+Inference modes:
+  - cloud: Google AI Studio / GenAI API (GEMMA_API_KEY, model GEMMA_MODEL_CLOUD)
+  - ollama: local Ollama HTTP API at OLLAMA_BASE_URL (default http://127.0.0.1:11434)
+Author: Krishna Dahale
+"""
+
+import json
+import os
+import urllib.error
+import urllib.request
+from typing import Optional, Tuple
+
+from google import genai
+
+# ─── Client Setup ──────────────────────────────────────────────────────────────
+
+def get_client(api_key: str = None):
+    key = api_key or os.environ.get("GEMMA_API_KEY")
+    if not key:
+        raise ValueError("API key not found. Set GEMMA_API_KEY environment variable.")
+    return genai.Client(api_key=key)
+
+
+GEMMA_MODEL_CLOUD = "gemma-4-26b-a4b-it"
+# Backwards compatibility
+GEMMA_MODEL = GEMMA_MODEL_CLOUD
+
+DEFAULT_OLLAMA_BASE = os.environ.get("OLLAMA_BASE_URL", "http://127.0.0.1:11434").rstrip("/")
+DEFAULT_OLLAMA_MODEL = os.environ.get("OLLAMA_MODEL", "gemma3:4b")
+
+
+def _ollama_installed_name_matches(installed: str, requested: str) -> bool:
+    """True if Ollama tags name matches user model (e.g. gemma3:4b vs gemma3:4b:latest)."""
+    installed = (installed or "").strip()
+    requested = (requested or "").strip()
+    if not requested:
+        return True
+    if installed == requested:
+        return True
+    if installed.startswith(requested + ":"):
+        return True
+    return False
+
+
+def ollama_verify_connection(
+    base_url: Optional[str] = None,
+    model: Optional[str] = None,
+    timeout: int = 12,
+) -> Tuple[bool, str]:
+    """
+    GET /api/tags — confirms Ollama is reachable and the chosen model is installed.
+    Returns (success, human-readable message). No LLM inference run.
+    """
+    base = (base_url or DEFAULT_OLLAMA_BASE).rstrip("/")
+    want = (model or DEFAULT_OLLAMA_MODEL).strip()
+    url = f"{base}/api/tags"
+    try:
+        with urllib.request.urlopen(url, timeout=timeout) as resp:
+            data = json.loads(resp.read().decode("utf-8"))
+    except urllib.error.HTTPError as e:
+        body = e.read().decode("utf-8", errors="replace")[:300] if e.fp else ""
+        return False, f"Ollama returned HTTP {e.code} at {url}. {body}"
+    except Exception as e:
+        return False, f"Cannot reach Ollama at {base}. Start the server (e.g. ollama serve), then try again. ({e})"
+
+    models = data.get("models") or []
+    names = [m.get("name", "") for m in models if m.get("name")]
+    if not names:
+        hint = f"Run: ollama pull {want}" if want else "Run: ollama pull <model>"
+        return False, f"Ollama responded but no models are installed yet. {hint}"
+
+    if want:
+        hits = [n for n in names if _ollama_installed_name_matches(n, want)]
+        if not hits:
+            preview = ", ".join(names[:10])
+            suffix = " …" if len(names) > 10 else ""
+            return (
+                False,
+                f"Ollama is running, but model '{want}' is not installed. "
+                f"You have: {preview}{suffix}. Run: ollama pull {want}",
+            )
+        return True, f"Connected locally at {base}. Model '{hits[0]}' is ready, so you can use offline narratives."
+
+    return True, f"Connected at {base}. {len(names)} model(s) installed."
+
+
+def ollama_generate(
+    prompt: str,
+    base_url: Optional[str] = None,
+    model: Optional[str] = None,
+    timeout: int = 300,
+) -> str:
+    """
+    Call Ollama's /api/chat (non-streaming). Works offline once Ollama is running.
+    """
+    base = (base_url or DEFAULT_OLLAMA_BASE).rstrip("/")
+    m = model or DEFAULT_OLLAMA_MODEL
+    url = f"{base}/api/chat"
+    payload = json.dumps(
+        {
+            "model": m,
+            "messages": [{"role": "user", "content": prompt}],
+            "stream": False,
+        }
+    ).encode("utf-8")
+    req = urllib.request.Request(
+        url,
+        data=payload,
+        headers={"Content-Type": "application/json"},
+        method="POST",
+    )
+    try:
+        with urllib.request.urlopen(req, timeout=timeout) as resp:
+            data = json.loads(resp.read().decode("utf-8"))
+    except urllib.error.HTTPError as e:
+        body = e.read().decode("utf-8", errors="replace") if e.fp else ""
+        raise RuntimeError(f"Ollama HTTP {e.code}: {body[:500]}") from e
+    except urllib.error.URLError as e:
+        raise RuntimeError(
+            f"Cannot reach Ollama at {base}. Is `ollama serve` running? ({e.reason})"
+        ) from e
+    msg = data.get("message") or {}
+    text = (msg.get("content") or "").strip()
+    if not text and data.get("error"):
+        raise RuntimeError(str(data["error"]))
+    return text
+
+
+def generate_explanation_text(
+    prompt: str,
+    *,
+    inference_mode: str = "cloud",
+    cloud_client=None,
+    ollama_base: Optional[str] = None,
+    ollama_model: Optional[str] = None,
+) -> str:
+    """Single entry: cloud (Google GenAI) or local (Ollama)."""
+    if inference_mode == "ollama":
+        return ollama_generate(prompt, base_url=ollama_base, model=ollama_model)
+    if cloud_client is None:
+        raise ValueError("Cloud mode requires a Google GenAI client and GEMMA_API_KEY.")
+    response = cloud_client.models.generate_content(
+        model=GEMMA_MODEL_CLOUD,
+        contents=prompt,
+    )
+    return response.text
+
+# ─── Prompt Templates ──────────────────────────────────────────────────────────
+
+def _build_farmer_prompt(farmer_data: dict, result: dict, language: str) -> str:
+    drivers = "\n".join([f"- {d['display_name']}" for d in result["top_risk_drivers"]])
+
+    if language == "Hindi":
+        return f"""
+आप एक सहानुभूतिपूर्ण कृषि ऋण सलाहकार हैं जो किसानों को सरल हिंदी में समझाते हैं।
+
+किसान की जानकारी:
+- फसल: {farmer_data['crop_type']}
+- राज्य: {farmer_data['state']}
+- वार्षिक आय: ₹{farmer_data['annual_income_lakhs']} लाख
+- मांगा गया ऋण: ₹{farmer_data['loan_amount_lakhs']} लाख
+- इस मौसम में बारिश की कमी: {abs(farmer_data['rainfall_deficit_pct'])}%
+- सूखे की गंभीरता (SPI): {farmer_data['spi']}
+- लगातार सूखे के वर्ष: {farmer_data['consecutive_drought_years']}
+
+ClimaIQ परिणाम:
+- क्रेडिट स्कोर: {result['credit_score']} (300–850 में से)
+- जोखिम श्रेणी: {result['risk_band']}
+- डिफ़ॉल्ट संभावना: {result['default_probability']}%
+- मुख्य जोखिम कारण:
+{drivers}
+
+निम्नलिखित प्रारूप में 4-5 वाक्यों में उत्तर दें:
+1. किसान को सरल भाषा में बताएं कि उनका स्कोर क्यों कम/अधिक है
+2. मौसम और फसल का जोखिम पर क्या असर हुआ, यह समझाएं
+3. किसान 3 व्यावहारिक कदम क्या उठा सकते हैं (बीमा, वैकल्पिक फसल, सिंचाई आदि)
+केवल हिंदी में उत्तर दें। जटिल शब्दों से बचें।
+"""
+    else:
+        return f"""
+You are a helpful agricultural credit advisor explaining a loan assessment to a farmer in simple English.
+
+Farmer Profile:
+- Crop: {farmer_data['crop_type']}
+- State: {farmer_data['state']}
+- Annual Income: ₹{farmer_data['annual_income_lakhs']} lakhs
+- Loan Requested: ₹{farmer_data['loan_amount_lakhs']} lakhs
+- Rainfall Deficit this season: {abs(farmer_data['rainfall_deficit_pct'])}%
+- Drought Severity (SPI): {farmer_data['spi']}
+- Consecutive Drought Years: {farmer_data['consecutive_drought_years']}
+
+ClimaIQ Result:
+- Credit Score: {result['credit_score']} out of 850
+- Risk Band: {result['risk_band']}
+- Default Probability: {result['default_probability']}%
+- Key Risk Drivers:
+{drivers}
+
+In 4-5 simple sentences:
+1. Explain why their score is what it is
+2. Explain how the weather and crop type has affected their risk
+3. Give 3 practical steps they can take (insurance, alternate crops, irrigation, etc.)
+Use simple, non-technical language. Be empathetic and constructive.
+"""
+
+
+def _build_officer_prompt(farmer_data: dict, result: dict, language: str) -> str:
+    drivers = "\n".join([f"- {d['display_name']}: {d['direction']}" for d in result["top_risk_drivers"]])
+
+    if language == "Hindi":
+        return f"""
+आप एक वरिष्ठ कृषि ऋण अधिकारी हैं। नीचे दी गई जानकारी के आधार पर एक संक्षिप्त ऋण मूल्यांकन रिपोर्ट हिंदी में तैयार करें।
+
+किसान प्रोफाइल:
+- फसल: {farmer_data['crop_type']} | राज्य: {farmer_data['state']}
+- आयु: {farmer_data['age']} वर्ष | भूमि: {farmer_data['land_size_acres']} एकड़
+- वार्षिक आय: ₹{farmer_data['annual_income_lakhs']} लाख
+- ऋण राशि: ₹{farmer_data['loan_amount_lakhs']} लाख
+- पिछले डिफ़ॉल्ट: {'हाँ' if farmer_data['previous_defaults'] else 'नहीं'}
+- बारिश की कमी: {abs(farmer_data['rainfall_deficit_pct'])}%
+- SPI: {farmer_data['spi']} | लगातार सूखा: {farmer_data['consecutive_drought_years']} वर्ष
+
+ClimaIQ स्कोर: {result['credit_score']} | जोखिम: {result['risk_band']} | डिफ़ॉल्ट संभावना: {result['default_probability']}%
+सुझाई गई कार्रवाई: {result['recommended_action']}
+
+मुख्य जोखिम कारण:
+{drivers}
+
+निम्नलिखित शीर्षक ठीक इसी रूप में अलग पंक्ति पर लिखें, फिर एक खाली पंक्ति, फिर अनुच्छेद:
+1. ऋण मूल्यांकन सारांश
+2. जलवायु जोखिम विश्लेषण
+3. सुझाई गई कार्रवाई और शर्तें
+4. निगरानी बिंदु
+शीर्षकों पर मार्कडाउन बोल्ड न लगाएँ। प्रत्येक खंड में लगभग 2 वाक्य रखें।
+"""
+    else:
+        return f"""
+You are a senior agricultural loan officer. Prepare a concise loan assessment report based on the ClimaIQ analysis below.
+
+Farmer Profile:
+- Crop: {farmer_data['crop_type']} | State: {farmer_data['state']}
+- Age: {farmer_data['age']} | Land: {farmer_data['land_size_acres']} acres
+- Annual Income: ₹{farmer_data['annual_income_lakhs']} lakhs
+- Loan Requested: ₹{farmer_data['loan_amount_lakhs']} lakhs
+- Previous Defaults: {'Yes' if farmer_data['previous_defaults'] else 'No'}
+- Rainfall Deficit: {abs(farmer_data['rainfall_deficit_pct'])}%
+- SPI: {farmer_data['spi']} | Consecutive Drought Years: {farmer_data['consecutive_drought_years']}
+
+ClimaIQ Score: {result['credit_score']} | Risk Band: {result['risk_band']} | Default Probability: {result['default_probability']}%
+Recommended Action: {result['recommended_action']}
+
+Top Risk Drivers:
+{drivers}
+
+Structure your report using these exact numbered headings, each on its own line, followed by a blank line, then the paragraph(s):
+1. Assessment Summary
+2. Climate Risk Analysis
+3. Recommended Action with conditions
+4. Monitoring flags if loan is approved
+Do not use markdown bold for the numbers. Keep each heading text exactly as shown (you may add detail after the blank line only).
+Be professional and specific. Reference the ClimaIQ score in your reasoning.
+"""
+
+
+def _build_portfolio_prompt(stress_results: list, language: str) -> str:
+    scenario_text = "\n".join([
+        f"- {r['scenario']}: {r['avg_default_pct']}% avg default, ₹{r['total_loss_lakhs']}L estimated loss"
+        for r in stress_results
+    ])
+
+    if language == "Hindi":
+        return f"""
+आप एक पोर्टफोलियो जोखिम प्रबंधक हैं। निम्नलिखित जलवायु तनाव परीक्षण परिणामों के आधार पर हिंदी में एक संक्षिप्त जोखिम रिपोर्ट तैयार करें।
+
+तनाव परीक्षण परिणाम ({stress_results[0]['portfolio_size']} ऋणों का पोर्टफोलियो):
+{scenario_text}
+
+निम्नलिखित पर रिपोर्ट दें:
+1. सामान्य बनाम सूखे की स्थिति में जोखिम वृद्धि
+2. सबसे गंभीर परिदृश्य का प्रभाव
+3. पूंजी बफर और प्रावधान की सिफारिश
+3-4 वाक्यों में, स्पष्ट और व्यावसायिक भाषा में।
+"""
+    else:
+        return f"""
+You are a portfolio risk manager at an agricultural lending institution.
+Based on the ClimaIQ stress test results below, write a concise risk narrative for senior management.
+
+Stress Test Results (Portfolio of {stress_results[0]['portfolio_size']} loans):
+{scenario_text}
+
+Cover:
+1. Risk escalation from normal to stressed conditions
+2. The non-linear jump — why losses multiply rather than add
+3. Capital buffer and provisioning recommendation
+4. Which drought scenario should trigger portfolio rebalancing
+Write in 4-5 sentences. Be specific about the numbers. Professional tone.
+"""
+
+
+# ─── Main Explanation Functions ────────────────────────────────────────────────
+
+def explain_for_farmer(
+    farmer_data: dict,
+    result: dict,
+    language: str = "Hindi",
+    client=None,
+    inference_mode: str = "cloud",
+    ollama_base: Optional[str] = None,
+    ollama_model: Optional[str] = None,
+) -> str:
+    """Generate farmer-facing explanation in Hindi or English."""
+    prompt = _build_farmer_prompt(farmer_data, result, language)
+    return generate_explanation_text(
+        prompt,
+        inference_mode=inference_mode,
+        cloud_client=client,
+        ollama_base=ollama_base,
+        ollama_model=ollama_model,
+    )
+
+
+def explain_for_officer(
+    farmer_data: dict,
+    result: dict,
+    language: str = "English",
+    client=None,
+    inference_mode: str = "cloud",
+    ollama_base: Optional[str] = None,
+    ollama_model: Optional[str] = None,
+) -> str:
+    """Generate loan officer assessment report in Hindi or English."""
+    prompt = _build_officer_prompt(farmer_data, result, language)
+    return generate_explanation_text(
+        prompt,
+        inference_mode=inference_mode,
+        cloud_client=client,
+        ollama_base=ollama_base,
+        ollama_model=ollama_model,
+    )
+
+
+def explain_portfolio_stress(
+    stress_results: list,
+    language: str = "English",
+    client=None,
+    inference_mode: str = "cloud",
+    ollama_base: Optional[str] = None,
+    ollama_model: Optional[str] = None,
+) -> str:
+    """Generate portfolio-level stress test narrative."""
+    prompt = _build_portfolio_prompt(stress_results, language)
+    return generate_explanation_text(
+        prompt,
+        inference_mode=inference_mode,
+        cloud_client=client,
+        ollama_base=ollama_base,
+        ollama_model=ollama_model,
+    )
+
+
+# ─── Quick Test ────────────────────────────────────────────────────────────────
+
+if __name__ == "__main__":
+    import os
+    from climaiq_engine import load_model, predict_single, run_stress_test
+
+    API_KEY = input("Enter your Gemma API key: ").strip()
+    client = get_client(API_KEY)
+
+    model, scaler = load_model()
+
+    sample_farmer = {
+        "age": 42,
+        "land_size_acres": 3.5,
+        "annual_income_lakhs": 3.0,
+        "loan_amount_lakhs": 2.0,
+        "previous_defaults": 0,
+        "crop_type": "Cotton",
+        "state": "Maharashtra",
+        "rainfall_deficit_pct": -35.0,
+        "spi": -1.8,
+        "consecutive_drought_years": 1
+    }
+
+    result = predict_single(sample_farmer, model, scaler)
+
+    print("\n" + "="*60)
+    print("FARMER EXPLANATION (Hindi)")
+    print("="*60)
+    print(explain_for_farmer(sample_farmer, result, language="Hindi", client=client))
+
+    print("\n" + "="*60)
+    print("OFFICER REPORT (English)")
+    print("="*60)
+    print(explain_for_officer(sample_farmer, result, language="English", client=client))
+
+    print("\n" + "="*60)
+    print("PORTFOLIO STRESS TEST NARRATIVE")
+    print("="*60)
+    portfolio = [sample_farmer] * 50
+    stress = run_stress_test(model, scaler, portfolio)
+    print(explain_portfolio_stress(stress, language="English", client=client))

climaiq_report_pdf.py

@@ -0,0 +1,242 @@
+"""
+ClimaIQ — Loan officer assessment PDF (A4, branded header, Unicode via Noto fonts).
+"""
+
+from __future__ import annotations
+
+import io
+import os
+import re
+import tempfile
+import urllib.request
+from datetime import datetime, timezone
+from typing import Dict, List, Optional, Tuple
+
+from fpdf import FPDF
+
+PRIMARY = (46, 125, 50)
+PRIMARY_DARK = (27, 94, 32)
+INK = (31, 42, 32)
+MUTED = (90, 109, 102)
+
+NOTO_SANS_URL = (
+    "https://raw.githubusercontent.com/googlefonts/noto-fonts/main/hinted/ttf/"
+    "NotoSans/NotoSans-Regular.ttf"
+)
+NOTO_DEVA_URL = (
+    "https://raw.githubusercontent.com/googlefonts/noto-fonts/main/hinted/ttf/"
+    "NotoSansDevanagari/NotoSansDevanagari-Regular.ttf"
+)
+
+
+def _font_cache_dir() -> str:
+    d = os.path.join(tempfile.gettempdir(), "climaiq_fonts")
+    os.makedirs(d, exist_ok=True)
+    return d
+
+
+def _ensure_font(url: str, filename: str) -> Optional[str]:
+    path = os.path.join(_font_cache_dir(), filename)
+    if os.path.isfile(path) and os.path.getsize(path) > 10000:
+        return path
+    try:
+        req = urllib.request.Request(url, headers={"User-Agent": "ClimaIQ-Kisan/1.0"})
+        with urllib.request.urlopen(req, timeout=45) as resp:
+            data = resp.read()
+        with open(path, "wb") as f:
+            f.write(data)
+        return path if os.path.isfile(path) and os.path.getsize(path) > 10000 else None
+    except Exception:
+        return None
+
+
+def _font_path_for_language(language: str) -> Optional[str]:
+    if language == "Hindi":
+        return _ensure_font(NOTO_DEVA_URL, "NotoSansDevanagari-Regular.ttf")
+    return _ensure_font(NOTO_SANS_URL, "NotoSans-Regular.ttf")
+
+
+def split_report_sections(text: str) -> List[Tuple[Optional[str], str]]:
+    """
+    Split officer report into (numbered heading or None, body) tuples.
+    Headings are lines like '1. Assessment Summary' at the start of a block.
+    """
+    text = (text or "").strip()
+    if not text:
+        return []
+    matches = list(re.finditer(r"(?m)^\d+\.\s+[^\n]+", text))
+    if not matches:
+        return [(None, text)]
+    out: List[Tuple[Optional[str], str]] = []
+    if matches[0].start() > 0:
+        pre = text[: matches[0].start()].strip()
+        if pre:
+            out.append((None, pre))
+    for i, m in enumerate(matches):
+        start = m.start()
+        end = matches[i + 1].start() if i + 1 < len(matches) else len(text)
+        chunk = text[start:end].strip()
+        first_nl = chunk.find("\n")
+        if first_nl == -1:
+            out.append((chunk.strip(), ""))
+        else:
+            title = chunk[:first_nl].strip()
+            body = chunk[first_nl + 1 :].strip()
+            out.append((title, body))
+    return out
+
+
+def _ascii_fallback(s: str) -> str:
+    return "".join(c if ord(c) < 128 else "?" for c in str(s))
+
+
+def build_officer_assessment_pdf(
+    report_text: str,
+    farmer_data: Dict,
+    result: Dict,
+    language: str,
+) -> bytes:
+    font_path = _font_path_for_language(language)
+    use_core = font_path is None
+
+    pdf = FPDF(orientation="P", unit="mm", format="A4")
+    pdf.set_auto_page_break(auto=True, margin=18)
+    pdf.set_margins(18, 18, 18)
+    pdf.add_page()
+
+    if font_path and not use_core:
+        try:
+            pdf.add_font("ClimaIQ", "", font_path)
+            pdf.set_font("ClimaIQ", "", 11)
+        except Exception:
+            use_core = True
+    if use_core:
+        pdf.set_font("Helvetica", "", 10)
+
+    def set_size(sz: float) -> None:
+        if use_core:
+            pdf.set_font("Helvetica", "", sz)
+        else:
+            pdf.set_font("ClimaIQ", "", sz)
+
+    def cell_txt(w: float, h: float, txt: str, **kwargs) -> None:
+        t = _ascii_fallback(txt) if use_core else str(txt)
+        pdf.cell(w, h, t[:500], **kwargs)
+
+    def multi_txt(w: float, h: float, txt: str) -> None:
+        t = _ascii_fallback(txt) if use_core else str(txt)
+        pdf.multi_cell(w, h, t)
+
+    # Banner
+    pdf.set_fill_color(*PRIMARY)
+    pdf.rect(0, 0, 210, 16, "F")
+    pdf.set_text_color(255, 255, 255)
+    set_size(13)
+    pdf.set_xy(18, 5)
+    title = "Loan assessment report" if language != "Hindi" else "ऋण मूल्यांकन रिपोर्ट"
+    cell_txt(0, 8, title, ln=True)
+    pdf.set_text_color(*INK)
+
+    # Meta
+    pdf.set_xy(18, 22)
+    set_size(9)
+    pdf.set_text_color(*MUTED)
+    ts = datetime.now(timezone.utc).strftime("%Y-%m-%d %H:%M UTC")
+    lang_lbl = "Language" if language != "Hindi" else "भाषा"
+    pdf.cell(95, 5, f"{lang_lbl}: {language}" if use_core else f"{lang_lbl}: {language}")
+    pdf.cell(0, 5, ts, ln=True)
+    pdf.set_text_color(*INK)
+    pdf.ln(3)
+
+    # Scorecard title
+    set_size(10)
+    pdf.set_text_color(*PRIMARY_DARK)
+    sc_title = "ClimaIQ scorecard" if language != "Hindi" else "ClimaIQ स्कोरकार्ड"
+    cell_txt(0, 6, sc_title, ln=True)
+    pdf.set_text_color(*INK)
+    pdf.ln(1)
+
+    col_w = (210 - 36) / 2
+    pdf.set_fill_color(245, 248, 243)
+    set_size(9)
+
+    if language == "Hindi":
+        rows = [
+            ("क्रेडिट स्कोर", f"{result['credit_score']} / 850"),
+            ("जोखिम श्रेणी", str(result["risk_band"])),
+            ("डिफ़ॉल्ट संभावना", f"{result['default_probability']}%"),
+            ("सुझाई कार्रवाई", str(result["recommended_action"])[:110]),
+            ("फसल", str(farmer_data.get("crop_type", ""))),
+            ("राज्य", str(farmer_data.get("state", ""))),
+            ("ऋण (लाख ₹)", str(farmer_data.get("loan_amount_lakhs", ""))),
+            ("बारिश कमी %", str(abs(float(farmer_data.get("rainfall_deficit_pct", 0))))),
+            ("SPI", str(farmer_data.get("spi", ""))),
+        ]
+    else:
+        rows = [
+            ("Credit score", f"{result['credit_score']} / 850"),
+            ("Risk band", str(result["risk_band"])),
+            ("Default probability", f"{result['default_probability']}%"),
+            ("Recommended action", str(result["recommended_action"])[:110]),
+            ("Crop", str(farmer_data.get("crop_type", ""))),
+            ("State", str(farmer_data.get("state", ""))),
+            ("Loan (Rs lakhs)", str(farmer_data.get("loan_amount_lakhs", ""))),
+            ("Rain deficit %", str(abs(float(farmer_data.get("rainfall_deficit_pct", 0))))),
+            ("SPI", str(farmer_data.get("spi", ""))),
+        ]
+
+    for a, b in rows:
+        pdf.set_x(18)
+        cell_txt(col_w - 2, 7, a, border="B", fill=True)
+        cell_txt(col_w + 2, 7, b, border="B", ln=True, fill=True)
+    pdf.ln(4)
+
+    # Narrative
+    set_size(10)
+    pdf.set_text_color(*PRIMARY)
+    narr_h = "AI narrative" if language != "Hindi" else "AI विवरण"
+    cell_txt(0, 6, narr_h, ln=True)
+    pdf.set_text_color(*INK)
+    pdf.ln(2)
+
+    usable_w = 210 - 36
+    for head, body in split_report_sections(report_text):
+        if head:
+            set_size(10.5)
+            pdf.set_x(18)
+            multi_txt(usable_w, 5.5, head)
+            pdf.ln(1)
+        if body:
+            set_size(10)
+            pdf.set_x(18)
+            multi_txt(usable_w, 5.2, body)
+            pdf.ln(3)
+
+    # Footer
+    pdf.set_y(-24)
+    pdf.set_draw_color(200, 210, 200)
+    yf = pdf.get_y()
+    pdf.line(18, yf, 192, yf)
+    pdf.ln(2)
+    set_size(8)
+    pdf.set_text_color(*MUTED)
+    foot = (
+        "ClimaIQ Kisan — climate-adjusted agricultural credit intelligence. "
+        "For decision support only; not a substitute for policy and compliance review."
+    )
+    if language == "Hindi":
+        foot = (
+            "ClimaIQ Kisan — कृषि ऋण हेतु जलवायु-समायोजित विश्लेषण। "
+            "केवल निर्णय सहायता; नीति व अनुपालन का स्थान नहीं लेता।"
+        )
+    pdf.set_x(18)
+    multi_txt(usable_w, 4, foot)
+
+    out = pdf.output(dest="S")
+    if isinstance(out, str):
+        return out.encode("latin-1", errors="replace")
+    return bytes(out)
+
+
+def pdf_to_bytesio(report_text: str, farmer_data: Dict, result: Dict, language: str) -> io.BytesIO:
+    return io.BytesIO(build_officer_assessment_pdf(report_text, farmer_data, result, language))

data/climaiq_india_agricultural_credit_1000.csv

@@ -0,0 +1,1001 @@
+loan_id,state,age,land_size_acres,annual_income_lakhs,loan_amount_lakhs,crop_type,previous_defaults,rainfall_deficit_pct,spi,consecutive_drought_years,debt_to_income_ratio,land_productivity,crop_water_intensity,crop_drought_interaction,state_maharashtra,default,climaiq_score,risk_band
+1,Maharashtra,47,10.677,15.0,0.604,Wheat,0,-22.749,-1.765,3,0.0403,1.4049,0.2,4.5497,1,1,375,High
+2,Punjab,29,1.207,8.45,5.37,Wheat,0,0.144,-0.968,0,0.6355,6.998,0.2,0.0,0,0,512,High
+3,Punjab,50,1.974,11.772,5.517,Rice,0,-4.037,-1.298,0,0.4686,5.9649,1.0,4.0366,0,0,497,High
+4,Maharashtra,52,0.655,12.761,6.49,Cotton,0,0.872,-2.054,0,0.5086,19.4879,0.5,0.0,1,1,468,High
+5,Maharashtra,52,2.831,15.0,7.496,Cotton,0,-13.847,-0.205,0,0.4997,5.2988,0.5,6.9235,1,1,460,High
+6,Maharashtra,58,0.988,2.019,0.651,Wheat,0,-2.961,1.506,3,0.3225,2.0425,0.2,0.5923,1,0,398,High
+7,Maharashtra,55,0.696,7.498,1.653,Wheat,1,-18.685,2.0,1,0.2205,10.7766,0.2,3.737,1,0,300,High
+8,Punjab,51,2.181,14.426,7.146,Rice,0,18.807,-0.203,1,0.4954,6.6157,1.0,0.0,0,0,530,High
+9,Punjab,44,0.5,15.0,3.939,Sugarcane,0,-24.188,0.817,1,0.2626,30.0,1.0,24.1882,0,0,463,High
+10,Punjab,25,0.5,8.219,4.734,Sugarcane,0,19.579,-0.068,0,0.576,16.4385,1.0,0.0,0,0,580,High
+11,Maharashtra,50,0.742,2.715,5.473,Cotton,0,-11.477,0.199,0,2.0158,3.6599,0.5,5.7383,1,0,422,High
+12,Punjab,47,0.5,15.0,5.593,Rice,0,-32.633,-0.642,1,0.3729,30.0,1.0,32.6327,0,1,384,High
+13,Punjab,48,0.5,7.328,7.488,Sugarcane,0,8.521,-0.105,1,1.0218,14.6561,1.0,0.0,0,0,469,High
+14,Maharashtra,30,1.241,4.703,7.977,Millets,0,-10.111,0.237,0,1.6961,3.7888,0.0,0.0,1,0,433,High
+15,Maharashtra,32,4.867,7.393,5.901,Rice,0,0.767,-0.878,1,0.7982,1.5189,1.0,0.0,1,0,434,High
+16,Maharashtra,58,2.546,10.511,2.654,Cotton,1,6.358,-0.377,0,0.2525,4.1279,0.5,0.0,1,0,357,High
+17,Maharashtra,45,2.156,7.51,7.098,Rice,0,-6.931,-0.358,0,0.9452,3.4825,1.0,6.9307,1,0,436,High
+18,Maharashtra,53,1.042,2.087,0.869,Wheat,0,-20.884,0.526,0,0.4165,2.002,0.2,4.1767,1,0,457,High
+19,Maharashtra,45,8.054,15.0,2.233,Rice,1,2.636,-1.366,0,0.1489,1.8625,1.0,0.0,1,1,371,High
+20,Maharashtra,45,1.518,15.0,7.23,Cotton,0,-11.989,-0.352,2,0.482,9.8841,0.5,5.9943,1,1,388,High
+21,Punjab,56,0.5,15.0,2.421,Wheat,0,-1.712,1.612,2,0.1614,30.0,0.2,0.3424,0,0,535,High
+22,Maharashtra,39,2.88,4.903,2.155,Wheat,0,-44.306,0.864,0,0.4395,1.7026,0.2,8.8612,1,1,417,High
+23,Maharashtra,46,0.5,10.369,1.536,Cotton,0,-5.295,1.049,1,0.1481,20.7377,0.5,2.6476,1,0,510,High
+24,Maharashtra,39,1.27,12.119,6.943,Cotton,1,-21.579,-0.276,1,0.5729,9.5397,0.5,10.7894,1,1,300,High
+25,Maharashtra,40,2.124,15.0,4.25,Wheat,1,12.8,0.021,0,0.2834,7.0613,0.2,0.0,1,0,396,High
+26,Punjab,41,6.22,15.0,5.562,Rice,0,-2.654,0.702,0,0.3708,2.4117,1.0,2.6537,0,0,566,High
+27,Maharashtra,48,2.041,6.771,2.293,Rice,0,-8.402,1.26,3,0.3386,3.3176,1.0,8.4021,1,1,384,High
+28,Maharashtra,39,3.635,13.008,6.185,Millets,0,-4.705,-2.252,1,0.4755,3.5791,0.0,0.0,1,0,421,High
+29,Maharashtra,60,14.808,15.0,6.2,Rice,0,-26.977,-1.615,3,0.4133,1.013,1.0,26.9766,1,1,300,High
+30,Maharashtra,34,0.5,15.0,2.844,Sugarcane,0,12.349,1.009,0,0.1896,30.0,1.0,0.0,1,0,613,Medium
+31,Punjab,43,0.964,15.0,3.58,Millets,0,17.871,-0.577,1,0.2387,15.566,0.0,0.0,0,0,568,High
+32,Maharashtra,40,1.505,15.0,2.073,Cotton,1,-6.106,-0.972,0,0.1382,9.9636,0.5,3.0531,1,1,354,High
+33,Maharashtra,62,1.644,13.027,7.824,Cotton,1,11.856,-1.554,0,0.6006,7.9259,0.5,0.0,1,1,304,High
+34,Punjab,47,1.588,15.0,5.342,Wheat,0,-20.537,-0.855,1,0.3561,9.4453,0.2,4.1074,0,0,452,High
+35,Punjab,57,7.145,15.0,7.509,Millets,0,-23.957,-1.702,0,0.5006,2.0993,0.0,0.0,0,1,452,High
+36,Punjab,27,3.756,6.117,2.775,Millets,0,-11.386,1.791,0,0.4537,1.6288,0.0,0.0,0,0,557,High
+37,Maharashtra,48,2.809,15.0,6.364,Sugarcane,0,-5.476,-0.322,0,0.4243,5.3404,1.0,5.4759,1,0,492,High
+38,Maharashtra,56,1.11,15.0,6.244,Rice,0,-17.91,-0.358,0,0.4163,13.5079,1.0,17.9098,1,1,441,High
+39,Punjab,46,5.208,8.547,5.687,Wheat,0,-3.933,0.288,1,0.6654,1.641,0.2,0.7867,0,1,477,High
+40,Maharashtra,58,9.307,8.301,7.742,Cotton,0,-7.404,0.662,0,0.9327,0.8919,0.5,3.7021,1,0,458,High
+41,Maharashtra,39,5.164,15.0,3.452,Wheat,0,-26.975,0.646,0,0.2301,2.9045,0.2,5.3949,1,0,502,High
+42,Maharashtra,62,0.5,15.0,1.48,Rice,0,-4.218,0.774,0,0.0987,30.0,1.0,4.2181,1,0,566,High
+43,Maharashtra,70,3.035,14.265,5.538,Rice,1,0.953,-0.931,0,0.3882,4.7001,1.0,0.0,1,0,320,High
+44,Punjab,41,0.5,15.0,5.97,Sugarcane,0,-9.428,-0.598,1,0.398,30.0,1.0,9.428,0,0,472,High
+45,Maharashtra,40,0.894,15.0,4.813,Wheat,0,5.29,0.469,0,0.3209,16.7855,0.2,0.0,1,0,561,High
+46,Punjab,62,0.5,15.0,2.051,Cotton,0,-30.261,-0.827,1,0.1367,30.0,0.5,15.1305,0,0,444,High
+47,Maharashtra,64,2.777,14.19,4.283,Cotton,0,-4.981,-0.305,0,0.3019,5.1088,0.5,2.4904,1,0,507,High
+48,Maharashtra,39,3.481,15.0,2.841,Millets,1,-29.554,0.099,0,0.1894,4.3094,0.0,0.0,1,1,323,High
+49,Maharashtra,40,0.5,5.333,3.706,Sugarcane,0,-28.711,0.485,2,0.6949,10.6667,1.0,28.7112,1,1,314,High
+50,Maharashtra,42,1.453,15.0,5.339,Cotton,0,-21.464,-2.078,1,0.3559,10.3253,0.5,10.732,1,0,389,High
+51,Punjab,29,1.828,15.0,6.089,Rice,0,-26.131,0.415,0,0.4059,8.2069,1.0,26.1309,0,1,469,High
+52,Punjab,34,1.382,10.232,2.242,Cotton,0,-20.22,-0.534,2,0.2191,7.4062,0.5,10.11,0,0,419,High
+53,Punjab,33,0.5,6.184,3.54,Millets,0,-15.212,0.407,0,0.5725,12.3671,0.0,0.0,0,0,511,High
+54,Punjab,36,1.629,7.231,2.519,Wheat,0,0.557,0.477,2,0.3483,4.4394,0.2,0.0,0,0,479,High
+55,Maharashtra,45,4.173,5.095,1.99,Wheat,0,-16.817,1.308,1,0.3905,1.221,0.2,3.3635,1,0,454,High
+56,Punjab,48,0.5,6.503,6.375,Wheat,0,-35.784,1.053,1,0.9805,13.005,0.2,7.1569,0,0,391,High
+57,Maharashtra,64,0.5,9.058,7.319,Cotton,0,5.101,1.047,3,0.8081,18.115,0.5,0.0,1,0,378,High
+58,Maharashtra,33,1.314,3.957,3.277,Sugarcane,0,-24.556,1.144,0,0.828,3.0106,1.0,24.5564,1,1,422,High
+59,Maharashtra,57,8.528,12.336,6.135,Wheat,0,-12.881,-0.009,1,0.4973,1.4465,0.2,2.5762,1,0,435,High
+60,Maharashtra,42,3.313,15.0,7.227,Rice,0,-5.302,-0.268,0,0.4818,4.5276,1.0,5.302,1,0,488,High
+61,Maharashtra,44,3.406,5.893,6.846,Rice,1,-9.006,0.573,0,1.1617,1.7303,1.0,9.0058,1,1,300,High
+62,Maharashtra,53,1.348,15.0,1.871,Wheat,1,4.413,0.432,0,0.1247,11.1286,0.2,0.0,1,1,405,High
+63,Punjab,32,2.702,15.0,7.672,Millets,0,-4.815,-0.609,0,0.5115,5.5521,0.0,0.0,0,0,521,High
+64,Maharashtra,50,0.5,15.0,3.325,Cotton,0,-24.92,0.105,1,0.2217,30.0,0.5,12.4598,1,0,442,High
+65,Maharashtra,47,3.033,7.197,3.044,Millets,0,20.0,-0.335,1,0.423,2.3734,0.0,0.0,1,0,506,High
+66,Maharashtra,51,7.335,15.0,0.968,Sugarcane,0,-9.085,-1.154,1,0.0645,2.0449,1.0,9.085,1,0,480,High
+67,Maharashtra,48,2.845,5.086,4.146,Rice,0,-14.21,0.99,0,0.8153,1.7878,1.0,14.2101,1,0,450,High
+68,Punjab,70,6.207,15.0,1.318,Rice,0,2.459,-0.245,0,0.0878,2.4165,1.0,0.0,0,0,593,High
+69,Maharashtra,37,3.862,13.978,2.364,Rice,0,-8.258,1.749,2,0.1691,3.6195,1.0,8.2582,1,1,481,High
+70,Punjab,39,0.5,15.0,2.877,Sugarcane,0,-6.012,-1.925,0,0.1918,30.0,1.0,6.0116,0,0,534,High
+71,Punjab,38,6.982,4.757,6.566,Rice,1,20.0,-0.857,1,1.3805,0.6812,1.0,0.0,0,1,305,High
+72,Maharashtra,38,0.5,5.953,7.066,Cotton,0,-16.127,-0.267,1,1.1869,11.9066,0.5,8.0634,1,0,371,High
+73,Maharashtra,37,2.29,15.0,2.768,Wheat,0,-16.766,0.837,1,0.1846,6.5488,0.2,3.3533,1,0,496,High
+74,Punjab,59,0.5,5.705,6.94,Sugarcane,0,-10.251,0.099,1,1.2165,11.4098,1.0,10.2513,0,0,405,High
+75,Punjab,62,0.5,15.0,6.929,Rice,0,4.073,0.01,2,0.4619,30.0,1.0,0.0,0,0,467,High
+76,Punjab,38,1.645,4.081,6.235,Cotton,1,-34.598,-1.427,0,1.5279,2.4811,0.5,17.299,0,1,300,High
+77,Punjab,35,1.049,15.0,1.6,Wheat,0,-41.483,-1.457,2,0.1067,14.3014,0.2,8.2967,0,1,394,High
+78,Maharashtra,51,1.997,4.545,5.677,Wheat,0,-20.807,0.023,0,1.2492,2.2763,0.2,4.1613,1,0,412,High
+79,Maharashtra,38,2.583,15.0,7.588,Sugarcane,0,-36.509,-1.664,0,0.5058,5.8082,1.0,36.5088,1,1,340,High
+80,Maharashtra,53,0.5,15.0,5.619,Rice,0,-18.372,-0.832,0,0.3746,30.0,1.0,18.3725,1,0,443,High
+81,Punjab,47,0.5,6.977,2.987,Sugarcane,0,-4.463,0.201,1,0.4282,13.9535,1.0,4.4625,0,0,485,High
+82,Punjab,27,2.454,15.0,4.254,Wheat,0,-25.361,-1.282,0,0.2836,6.1132,0.2,5.0722,0,1,494,High
+83,Maharashtra,64,1.261,15.0,5.343,Wheat,0,-33.402,1.305,0,0.3562,11.8955,0.2,6.6805,1,0,464,High
+84,Maharashtra,67,3.919,7.916,4.329,Wheat,0,-10.96,0.924,0,0.5469,2.0201,0.2,2.192,1,1,482,High
+85,Maharashtra,38,3.169,5.437,1.4,Cotton,0,0.938,-1.871,2,0.2574,1.7155,0.5,0.0,1,0,407,High
+86,Maharashtra,40,5.605,2.534,2.84,Rice,0,-37.425,-1.232,2,1.1208,0.4522,1.0,37.4246,1,1,300,High
+87,Punjab,48,3.538,15.0,6.518,Wheat,0,0.303,-2.586,1,0.4345,4.24,0.2,0.0,0,0,458,High
+88,Punjab,49,1.686,3.484,6.961,Wheat,0,0.26,-1.083,1,1.9979,2.0664,0.2,0.0,0,0,411,High
+89,Punjab,53,3.353,13.148,1.53,Wheat,0,2.311,0.466,0,0.1163,3.9212,0.2,0.0,0,0,600,Medium
+90,Maharashtra,69,0.966,9.867,7.643,Cotton,0,6.277,-0.714,0,0.7746,10.2113,0.5,0.0,1,0,465,High
+91,Maharashtra,43,1.097,15.0,2.966,Millets,0,20.0,0.161,2,0.1977,13.6711,0.0,0.0,1,0,522,High
+92,Punjab,35,4.667,9.567,5.467,Wheat,0,-17.341,-1.822,0,0.5715,2.0498,0.2,3.4683,0,0,458,High
+93,Punjab,28,1.774,9.36,6.139,Rice,0,-10.356,0.498,0,0.6558,5.2761,1.0,10.3564,0,1,498,High
+94,Maharashtra,36,4.804,15.0,6.59,Cotton,0,-18.247,-0.855,0,0.4393,3.1223,0.5,9.1233,1,1,454,High
+95,Punjab,45,5.185,15.0,7.593,Sugarcane,0,-11.97,-0.288,0,0.5062,2.8929,1.0,11.9703,0,0,488,High
+96,Maharashtra,67,5.847,15.0,2.847,Sugarcane,0,10.993,-0.393,0,0.1898,2.5656,1.0,0.0,1,1,564,High
+97,Maharashtra,39,7.454,15.0,6.939,Wheat,0,-32.175,-1.572,1,0.4626,2.0125,0.2,6.435,1,0,372,High
+98,Maharashtra,48,1.692,3.566,1.488,Millets,0,-24.916,-1.453,0,0.4171,2.1083,0.0,0.0,1,0,423,High
+99,Maharashtra,45,1.152,15.0,5.788,Cotton,0,8.658,0.668,1,0.3859,13.0209,0.5,0.0,1,0,518,High
+100,Maharashtra,59,2.618,15.0,6.065,Cotton,0,3.301,0.099,0,0.4043,5.729,0.5,0.0,1,0,525,High
+101,Maharashtra,70,1.392,5.401,5.583,Millets,0,-17.447,0.75,0,1.0336,3.8811,0.0,0.0,1,0,437,High
+102,Punjab,39,2.757,10.022,2.392,Millets,0,0.816,0.025,0,0.2387,3.635,0.0,0.0,0,0,569,High
+103,Maharashtra,39,3.668,11.805,7.048,Wheat,0,-6.882,-0.157,1,0.597,3.2182,0.2,1.3764,1,0,439,High
+104,Maharashtra,58,3.498,10.94,1.766,Millets,0,10.504,-2.026,0,0.1615,3.1273,0.0,0.0,1,0,523,High
+105,Punjab,53,1.408,15.0,0.902,Rice,0,-10.882,-0.57,1,0.0602,10.6544,1.0,10.8824,0,0,509,High
+106,Maharashtra,67,0.548,15.0,5.881,Millets,0,3.384,-0.327,1,0.3921,27.3892,0.0,0.0,1,1,479,High
+107,Maharashtra,52,1.689,15.0,4.066,Cotton,0,20.0,1.653,1,0.2711,8.8802,0.5,0.0,1,0,574,High
+108,Punjab,41,0.5,4.264,6.868,Rice,0,-3.029,0.387,0,1.6107,8.5281,1.0,3.0285,0,0,478,High
+109,Maharashtra,52,6.624,15.0,3.375,Rice,0,-11.26,1.236,0,0.225,2.2643,1.0,11.2601,1,0,532,High
+110,Maharashtra,58,1.273,15.0,1.762,Cotton,0,-1.163,0.298,0,0.1175,11.7797,0.5,0.5816,1,0,564,High
+111,Maharashtra,55,0.503,15.0,6.773,Cotton,0,-17.479,-1.011,1,0.4516,29.801,0.5,8.7396,1,0,404,High
+112,Maharashtra,51,0.5,15.0,4.617,Cotton,1,20.0,-0.183,0,0.3078,30.0,0.5,0.0,1,0,403,High
+113,Punjab,58,3.114,15.0,1.919,Rice,1,-9.299,-0.064,2,0.1279,4.8172,1.0,9.2989,0,1,300,High
+114,Punjab,59,3.039,15.0,5.903,Cotton,0,-16.094,-1.239,2,0.3935,4.9354,0.5,8.047,0,0,395,High
+115,Punjab,62,1.249,14.872,4.333,Wheat,0,-24.535,-0.684,0,0.2914,11.9039,0.2,4.9071,0,0,488,High
+116,Punjab,53,0.5,15.0,5.032,Cotton,0,4.458,-0.506,0,0.3355,30.0,0.5,0.0,0,0,564,High
+117,Punjab,43,1.581,13.68,3.957,Cotton,0,1.008,1.306,1,0.2892,8.6525,0.5,0.0,0,0,555,High
+118,Maharashtra,47,1.118,12.418,6.715,Sugarcane,0,-19.46,-2.381,0,0.5408,11.1022,1.0,19.4603,1,1,381,High
+119,Punjab,59,3.391,9.197,6.722,Millets,0,-3.353,0.906,0,0.7309,2.7118,0.0,0.0,0,0,517,High
+120,Maharashtra,35,2.793,15.0,5.82,Sugarcane,0,-34.628,-1.237,2,0.388,5.3715,1.0,34.6283,1,1,300,High
+121,Punjab,49,1.36,6.33,1.291,Cotton,0,-9.225,-1.672,0,0.2039,4.6526,0.5,4.6127,0,1,491,High
+122,Punjab,40,0.737,15.0,6.791,Sugarcane,0,-5.209,0.88,2,0.4528,20.3448,1.0,5.2087,0,1,465,High
+123,Maharashtra,45,13.278,15.0,5.533,Rice,0,13.61,0.432,1,0.3688,1.1297,1.0,0.0,1,0,536,High
+124,Maharashtra,60,1.815,3.977,5.728,Rice,0,-23.562,0.955,1,1.4402,2.1907,1.0,23.5623,1,1,342,High
+125,Maharashtra,47,3.497,15.0,3.222,Cotton,1,-31.175,0.501,0,0.2148,4.2895,0.5,15.5876,1,1,300,High
+126,Maharashtra,46,0.5,2.27,7.095,Sugarcane,0,-25.431,1.76,0,3.1248,4.5409,1.0,25.4314,1,0,369,High
+127,Punjab,29,1.994,15.0,5.94,Rice,0,-32.049,0.259,1,0.396,7.5218,1.0,32.0485,0,1,404,High
+128,Punjab,54,0.5,15.0,1.344,Rice,0,8.05,-0.456,0,0.0896,30.0,1.0,0.0,0,0,610,Medium
+129,Maharashtra,53,2.112,7.585,3.558,Rice,1,11.23,0.043,0,0.4692,3.5906,1.0,0.0,1,1,351,High
+130,Maharashtra,70,9.974,13.448,7.148,Millets,0,-7.197,0.943,0,0.5315,1.3483,0.0,0.0,1,0,501,High
+131,Maharashtra,41,1.455,15.0,0.854,Wheat,0,6.319,-1.079,0,0.0569,10.3126,0.2,0.0,1,0,573,High
+132,Maharashtra,48,0.5,8.129,7.711,Cotton,0,-9.991,-0.952,0,0.9486,16.2574,0.5,4.9953,1,0,419,High
+133,Maharashtra,48,0.5,5.222,5.594,Wheat,0,-13.427,-0.609,1,1.0712,10.4439,0.2,2.6853,1,0,385,High
+134,Maharashtra,64,1.473,15.0,5.235,Wheat,0,-16.996,-2.405,0,0.349,10.1833,0.2,3.3991,1,1,435,High
+135,Punjab,44,0.523,15.0,2.529,Wheat,0,-16.656,-0.162,0,0.1686,28.6725,0.2,3.3312,0,1,551,High
+136,Maharashtra,48,0.5,12.219,2.366,Rice,0,-35.217,-2.186,1,0.1936,24.4371,1.0,35.217,1,1,331,High
+137,Maharashtra,52,4.234,15.0,7.565,Millets,1,-8.036,-0.247,3,0.5044,3.5424,0.0,0.0,1,1,300,High
+138,Punjab,47,1.84,15.0,2.737,Cotton,0,-12.605,0.565,0,0.1825,8.1506,0.5,6.3025,0,0,560,High
+139,Maharashtra,40,1.024,6.726,6.316,Wheat,0,-0.82,-0.777,0,0.9391,6.5668,0.2,0.1639,1,0,459,High
+140,Punjab,47,6.91,8.263,3.448,Rice,0,-10.597,1.377,1,0.4173,1.1957,1.0,10.5971,0,1,490,High
+141,Punjab,58,0.5,15.0,1.544,Cotton,0,-14.441,-1.761,2,0.103,30.0,0.5,7.2203,0,0,440,High
+142,Maharashtra,33,9.365,15.0,0.58,Cotton,1,-25.068,0.891,1,0.0387,1.6017,0.5,12.5339,1,1,318,High
+143,Maharashtra,47,1.13,15.0,4.582,Wheat,0,-10.918,0.375,1,0.3055,13.2686,0.2,2.1835,1,0,478,High
+144,Maharashtra,37,6.579,6.026,7.818,Rice,0,-0.926,-2.016,0,1.2974,0.9161,1.0,0.9264,1,0,419,High
+145,Maharashtra,59,2.771,15.0,3.269,Sugarcane,0,-29.15,2.0,1,0.2179,5.4132,1.0,29.1501,1,0,432,High
+146,Maharashtra,27,5.258,7.966,3.179,Rice,0,-10.688,-0.364,0,0.399,1.515,1.0,10.6879,1,0,477,High
+147,Punjab,38,12.408,9.317,3.081,Wheat,0,-8.504,-1.177,1,0.3306,0.7509,0.2,1.7009,0,1,481,High
+148,Maharashtra,50,1.021,15.0,4.404,Sugarcane,0,-22.771,-1.707,1,0.2936,14.6952,1.0,22.771,1,1,379,High
+149,Maharashtra,64,9.743,15.0,3.805,Wheat,0,-20.818,-0.52,3,0.2537,1.5396,0.2,4.1636,1,0,361,High
+150,Maharashtra,44,1.48,15.0,7.099,Sugarcane,0,-8.541,-0.427,0,0.4733,10.1328,1.0,8.5408,1,0,472,High
+151,Punjab,38,1.461,15.0,4.291,Wheat,0,-29.001,-0.197,0,0.2861,10.2677,0.2,5.8003,0,0,500,High
+152,Maharashtra,68,1.248,15.0,4.308,Cotton,0,-6.751,-0.598,1,0.2872,12.0163,0.5,3.3756,1,0,458,High
+153,Maharashtra,28,2.339,12.614,1.538,Sugarcane,1,-25.43,0.333,1,0.1219,5.3934,1.0,25.4303,1,1,300,High
+154,Maharashtra,25,0.502,15.0,7.871,Cotton,0,3.943,-0.688,1,0.5247,29.8673,0.5,0.0,1,0,475,High
+155,Punjab,50,2.42,1.61,2.375,Wheat,0,20.0,-2.235,0,1.4749,0.6653,0.2,0.0,0,1,510,High
+156,Maharashtra,39,2.341,9.609,1.05,Sugarcane,0,-24.911,-0.449,0,0.1093,4.1052,1.0,24.9114,1,0,444,High
+157,Punjab,33,4.305,14.777,5.474,Cotton,0,-3.987,-2.667,1,0.3704,3.4323,0.5,1.9934,0,0,462,High
+158,Punjab,54,5.381,4.851,4.965,Cotton,0,-22.901,0.054,0,1.0235,0.9015,0.5,11.4506,0,0,437,High
+159,Maharashtra,48,1.742,15.0,3.614,Wheat,1,3.614,-1.279,2,0.2409,8.6117,0.2,0.0,1,1,300,High
+160,Punjab,38,1.08,10.344,3.666,Sugarcane,0,-30.979,-0.137,0,0.3544,9.5776,1.0,30.9789,0,0,434,High
+161,Maharashtra,30,0.901,15.0,6.388,Millets,1,-4.791,1.014,1,0.4259,16.6552,0.0,0.0,1,1,322,High
+162,Punjab,55,0.649,15.0,1.968,Wheat,0,-25.095,-1.249,1,0.1312,23.1005,0.2,5.019,0,0,467,High
+163,Punjab,53,0.5,8.393,6.652,Wheat,0,-15.105,0.761,1,0.7926,16.7853,0.2,3.021,0,0,443,High
+164,Maharashtra,44,1.258,10.798,5.366,Millets,1,-27.465,-0.733,0,0.4969,8.5853,0.0,0.0,1,1,300,High
+165,Maharashtra,67,0.945,11.483,3.642,Rice,0,0.335,1.453,2,0.3172,12.1491,1.0,0.0,1,1,465,High
+166,Punjab,32,10.181,8.488,3.935,Rice,1,-24.555,-0.665,0,0.4636,0.8337,1.0,24.5547,0,1,300,High
+167,Maharashtra,27,2.447,14.24,2.425,Sugarcane,0,-3.721,-1.14,1,0.1703,5.8183,1.0,3.7205,1,0,489,High
+168,Maharashtra,37,1.28,6.642,6.141,Sugarcane,1,-17.161,0.387,0,0.9246,5.1896,1.0,17.1615,1,1,300,High
+169,Maharashtra,44,2.606,4.931,4.347,Millets,0,-13.717,-0.736,0,0.8815,1.8922,0.0,0.0,1,1,441,High
+170,Maharashtra,48,0.5,15.0,2.906,Cotton,0,-9.784,-1.141,0,0.1938,30.0,0.5,4.8918,1,0,509,High
+171,Punjab,42,1.761,15.0,4.97,Rice,0,-23.427,-0.778,0,0.3314,8.5169,1.0,23.4274,0,1,461,High
+172,Maharashtra,49,2.952,6.402,7.968,Cotton,0,-8.844,-0.345,3,1.2447,2.1683,0.5,4.4222,1,0,300,High
+173,Maharashtra,30,2.014,15.0,3.764,Rice,0,10.955,1.042,0,0.251,7.4495,1.0,0.0,1,0,598,High
+174,Maharashtra,62,1.541,7.35,7.405,Cotton,0,-23.671,-2.234,1,1.0075,4.7686,0.5,11.8357,1,1,303,High
+175,Punjab,44,15.273,14.917,3.088,Sugarcane,0,5.577,-0.277,3,0.207,0.9767,1.0,0.0,0,0,481,High
+176,Maharashtra,58,6.369,15.0,0.91,Rice,0,10.471,0.325,0,0.0607,2.3553,1.0,0.0,1,0,602,Medium
+177,Punjab,49,2.93,11.859,2.032,Sugarcane,0,-24.242,-0.107,1,0.1713,4.0479,1.0,24.2423,0,0,440,High
+178,Maharashtra,50,2.528,15.0,3.64,Cotton,0,-33.826,1.01,0,0.2427,5.9346,0.5,16.9129,1,0,465,High
+179,Punjab,52,2.908,8.06,2.147,Wheat,0,-7.471,-0.374,0,0.2664,2.7715,0.2,1.4941,0,0,526,High
+180,Maharashtra,50,0.674,10.411,0.502,Rice,0,-20.211,-0.485,0,0.0482,15.4379,1.0,20.2111,1,0,468,High
+181,Maharashtra,53,1.508,10.333,7.363,Cotton,0,16.029,0.292,2,0.7125,6.8507,0.5,0.0,1,1,437,High
+182,Maharashtra,61,0.5,6.704,6.829,Rice,0,-11.478,-0.201,0,1.0186,13.4084,1.0,11.4779,1,0,413,High
+183,Punjab,47,1.936,10.576,6.067,Millets,0,-24.033,-0.69,1,0.5737,5.4622,0.0,0.0,0,1,421,High
+184,Punjab,54,2.349,3.529,1.759,Sugarcane,1,-29.207,0.216,1,0.4983,1.5022,1.0,29.2074,0,1,300,High
+185,Maharashtra,44,0.775,8.224,1.403,Wheat,0,4.349,-0.953,2,0.1706,10.6072,0.2,0.0,1,0,444,High
+186,Punjab,62,9.975,9.7,1.007,Rice,0,-20.568,-0.911,0,0.1038,0.9724,1.0,20.5678,0,0,478,High
+187,Punjab,37,7.206,9.925,5.93,Millets,0,-11.343,-0.828,1,0.5975,1.3774,0.0,0.0,0,0,453,High
+188,Maharashtra,67,3.842,10.042,7.466,Cotton,0,2.337,1.053,1,0.7435,2.6138,0.5,0.0,1,0,455,High
+189,Maharashtra,45,2.287,11.595,3.729,Rice,0,-19.264,-3.0,1,0.3216,5.0706,1.0,19.2639,1,1,357,High
+190,Maharashtra,28,6.12,15.0,1.394,Wheat,0,-12.689,0.096,2,0.0929,2.451,0.2,2.5378,1,0,473,High
+191,Maharashtra,47,0.5,15.0,4.343,Millets,0,-12.249,-0.266,2,0.2895,30.0,0.0,0.0,1,0,433,High
+192,Punjab,37,4.689,7.244,3.116,Rice,0,-1.75,-1.981,0,0.4302,1.5448,1.0,1.7501,0,0,501,High
+193,Punjab,55,0.5,4.653,3.311,Rice,0,-3.139,0.01,0,0.7117,9.3053,1.0,3.1388,0,0,505,High
+194,Punjab,37,4.743,5.293,5.16,Millets,0,-15.302,0.176,0,0.9749,1.116,0.0,0.0,0,1,483,High
+195,Maharashtra,40,0.968,8.25,5.89,Rice,1,-11.045,0.29,0,0.714,8.5236,1.0,11.0453,1,1,300,High
+196,Maharashtra,25,6.289,9.193,1.33,Rice,0,-17.577,-2.952,0,0.1446,1.4618,1.0,17.5767,1,1,430,High
+197,Punjab,40,8.649,6.58,0.767,Wheat,0,-34.164,1.066,0,0.1166,0.7607,0.2,6.8328,0,1,505,High
+198,Punjab,25,0.5,4.319,7.957,Millets,1,-8.136,-1.038,0,1.8423,8.6387,0.0,0.0,0,1,300,High
+199,Punjab,26,1.863,8.458,2.245,Wheat,0,16.383,0.387,3,0.2655,4.5388,0.2,0.0,0,0,485,High
+200,Punjab,54,11.888,14.085,0.904,Millets,0,-25.256,0.441,0,0.0642,1.1848,0.0,0.0,0,1,561,High
+201,Punjab,54,1.982,15.0,1.924,Wheat,0,-23.595,-1.177,2,0.1283,7.57,0.2,4.7191,0,1,429,High
+202,Maharashtra,50,5.975,7.8,0.792,Rice,1,-25.061,0.36,0,0.1016,1.3053,1.0,25.0609,1,1,300,High
+203,Maharashtra,33,2.666,8.755,3.439,Millets,1,-1.131,-0.685,1,0.3928,3.2845,0.0,0.0,1,1,300,High
+204,Punjab,44,1.412,5.805,6.518,Cotton,0,-5.273,-2.371,2,1.1229,4.112,0.5,2.6364,0,1,352,High
+205,Punjab,45,1.01,11.88,3.059,Cotton,0,-14.378,0.868,0,0.2575,11.7657,0.5,7.189,0,0,541,High
+206,Maharashtra,31,0.682,6.18,3.831,Wheat,0,-25.77,-1.075,0,0.62,9.0671,0.2,5.1541,1,0,419,High
+207,Maharashtra,63,4.304,15.0,5.567,Cotton,0,-10.614,-1.154,0,0.3711,3.4852,0.5,5.3072,1,1,467,High
+208,Punjab,56,1.466,2.919,4.317,Rice,0,-35.469,-0.703,0,1.4787,1.9914,1.0,35.4693,0,1,345,High
+209,Maharashtra,42,2.148,9.137,6.961,Rice,1,-9.849,-0.028,2,0.7619,4.2538,1.0,9.8486,1,0,300,High
+210,Maharashtra,45,2.034,15.0,6.995,Cotton,0,1.073,-0.357,2,0.4664,7.3756,0.5,0.0,1,1,428,High
+211,Maharashtra,48,2.583,14.113,0.574,Sugarcane,0,-30.768,0.248,0,0.0407,5.4634,1.0,30.768,1,0,461,High
+212,Punjab,25,6.02,15.0,6.335,Wheat,0,-34.988,-1.639,0,0.4224,2.4916,0.2,6.9976,0,1,446,High
+213,Punjab,42,11.849,15.0,3.956,Millets,0,11.692,-1.042,0,0.2637,1.2659,0.0,0.0,0,0,591,High
+214,Maharashtra,37,1.571,5.456,7.916,Cotton,0,-21.176,-0.416,0,1.4509,3.4739,0.5,10.5882,1,0,381,High
+215,Punjab,33,5.272,11.287,4.124,Cotton,0,5.039,1.598,1,0.3654,2.1407,0.5,0.0,0,0,561,High
+216,Maharashtra,42,4.338,2.942,7.902,Rice,1,-14.31,-0.414,0,2.6861,0.6781,1.0,14.31,1,1,300,High
+217,Maharashtra,67,2.557,11.422,6.159,Rice,1,-14.429,0.785,1,0.5393,4.4675,1.0,14.4293,1,1,300,High
+218,Punjab,53,9.374,6.256,2.365,Cotton,0,-5.424,-0.261,0,0.378,0.6674,0.5,2.7119,0,0,524,High
+219,Punjab,38,0.671,15.0,4.538,Sugarcane,1,-25.26,-0.784,0,0.3025,22.3477,1.0,25.26,0,1,300,High
+220,Punjab,52,0.5,7.849,7.941,Millets,0,-10.013,-0.547,1,1.0117,15.6977,0.0,0.0,0,0,418,High
+221,Punjab,62,5.772,11.096,6.931,Millets,0,-17.722,-0.468,0,0.6247,1.9225,0.0,0.0,0,1,469,High
+222,Maharashtra,56,1.735,7.327,1.23,Rice,0,-31.807,0.304,3,0.1679,4.2232,1.0,31.8074,1,1,300,High
+223,Maharashtra,46,5.631,15.0,5.798,Millets,0,-6.938,1.075,2,0.3865,2.6637,0.0,0.0,1,0,453,High
+224,Maharashtra,37,6.707,9.422,3.092,Cotton,1,-15.95,0.06,0,0.3282,1.4047,0.5,7.9748,1,1,308,High
+225,Maharashtra,53,0.5,15.0,0.552,Rice,0,-14.355,0.887,0,0.0368,30.0,1.0,14.3554,1,0,544,High
+226,Maharashtra,50,6.449,15.0,3.526,Millets,0,-30.152,-1.265,2,0.2351,2.3261,0.0,0.0,1,1,381,High
+227,Punjab,56,1.784,15.0,2.261,Wheat,0,11.871,-0.715,2,0.1507,8.4064,0.2,0.0,0,0,517,High
+228,Maharashtra,53,2.143,15.0,6.642,Rice,0,1.151,-1.024,2,0.4428,7.0005,1.0,0.0,1,1,415,High
+229,Punjab,58,2.955,15.0,3.03,Cotton,0,2.462,0.014,0,0.202,5.0757,0.5,0.0,0,0,583,High
+230,Punjab,39,7.827,2.606,6.706,Cotton,0,-16.582,-0.659,3,2.5733,0.333,0.5,8.2908,0,1,300,High
+231,Punjab,61,0.5,9.746,0.504,Sugarcane,0,-37.938,0.43,0,0.0518,19.4927,1.0,37.9377,0,1,439,High
+232,Maharashtra,47,1.944,15.0,2.809,Wheat,0,17.294,-1.231,1,0.1872,7.7175,0.2,0.0,1,1,529,High
+233,Maharashtra,70,3.492,11.715,2.309,Millets,0,-12.916,-0.257,2,0.1971,3.3553,0.0,0.0,1,0,418,High
+234,Maharashtra,37,3.848,3.958,2.252,Rice,0,-23.232,0.387,0,0.5689,1.0287,1.0,23.232,1,1,424,High
+235,Maharashtra,66,3.538,11.682,1.011,Cotton,0,-5.748,-2.365,2,0.0866,3.3018,0.5,2.874,1,0,404,High
+236,Punjab,47,0.5,15.0,1.903,Wheat,0,5.507,1.822,0,0.1269,30.0,0.2,0.0,0,1,647,Medium
+237,Maharashtra,37,1.067,15.0,4.667,Wheat,0,-12.346,-0.583,3,0.3111,14.0524,0.2,2.4692,1,1,379,High
+238,Maharashtra,39,1.333,11.593,2.685,Cotton,0,7.179,-2.325,0,0.2316,8.6979,0.5,0.0,1,0,513,High
+239,Punjab,41,4.896,15.0,3.63,Sugarcane,0,4.245,-0.142,3,0.242,3.064,1.0,0.0,0,0,466,High
+240,Maharashtra,50,0.98,9.087,3.327,Millets,1,-28.146,-0.267,0,0.3661,9.2687,0.0,0.0,1,0,300,High
+241,Punjab,51,0.71,3.117,7.069,Millets,0,-22.107,-0.382,1,2.2679,4.3912,0.0,0.0,0,0,372,High
+242,Punjab,38,9.471,5.815,7.29,Sugarcane,0,2.363,0.691,1,1.2536,0.614,1.0,0.0,0,0,475,High
+243,Punjab,45,0.5,15.0,4.197,Millets,1,-52.689,0.447,0,0.2798,30.0,0.0,0.0,0,1,300,High
+244,Maharashtra,70,18.386,9.785,2.976,Wheat,0,-14.242,-0.002,0,0.3042,0.5322,0.2,2.8484,1,0,495,High
+245,Maharashtra,66,3.389,10.027,1.301,Sugarcane,0,-33.462,0.15,1,0.1297,2.959,1.0,33.4625,1,1,369,High
+246,Punjab,50,5.289,9.584,7.651,Cotton,0,-19.148,-0.404,0,0.7982,1.8121,0.5,9.5742,0,0,441,High
+247,Maharashtra,45,1.047,15.0,0.952,Cotton,1,-16.482,1.475,1,0.0634,14.3259,0.5,8.2412,1,0,339,High
+248,Punjab,46,0.5,15.0,2.732,Cotton,1,-17.374,-1.779,0,0.1822,30.0,0.5,8.6868,0,1,330,High
+249,Punjab,52,4.017,15.0,2.735,Wheat,0,-20.284,-0.16,3,0.1824,3.7344,0.2,4.0568,0,1,410,High
+250,Punjab,33,5.39,10.727,2.903,Cotton,0,-28.457,-0.235,0,0.2706,1.9902,0.5,14.2286,0,1,481,High
+251,Maharashtra,42,4.047,15.0,1.732,Millets,0,20.0,0.513,0,0.1155,3.7067,0.0,0.0,1,0,624,Medium
+252,Maharashtra,25,0.5,3.621,6.374,Cotton,1,16.012,-1.144,0,1.7603,7.2415,0.5,0.0,1,1,300,High
+253,Punjab,50,4.2,15.0,1.419,Rice,1,-29.431,-2.044,0,0.0946,3.5712,1.0,29.431,0,1,300,High
+254,Maharashtra,53,4.395,4.786,5.262,Wheat,0,11.732,-0.143,0,1.0993,1.0891,0.2,0.0,1,1,493,High
+255,Maharashtra,39,3.224,14.489,1.429,Rice,0,-17.992,-0.019,0,0.0986,4.4948,1.0,17.992,1,0,503,High
+256,Maharashtra,64,4.359,5.064,1.029,Rice,0,-10.062,1.19,2,0.2032,1.1618,1.0,10.0622,1,1,415,High
+257,Punjab,30,5.611,7.606,7.321,Rice,0,-12.949,-1.57,1,0.9625,1.3555,1.0,12.9491,0,0,387,High
+258,Punjab,27,2.853,6.235,1.926,Wheat,0,-21.896,0.855,1,0.3089,2.1857,0.2,4.3792,0,1,477,High
+259,Maharashtra,48,0.5,7.491,6.659,Cotton,0,2.758,2.0,0,0.889,14.982,0.5,0.0,1,0,518,High
+260,Maharashtra,58,2.006,7.915,4.585,Sugarcane,0,-10.112,-0.678,1,0.5793,3.9452,1.0,10.1117,1,0,399,High
+261,Punjab,65,0.5,7.857,1.084,Sugarcane,0,-19.148,-1.591,0,0.138,15.7137,1.0,19.1483,0,1,452,High
+262,Punjab,39,1.57,15.0,7.666,Cotton,1,-31.569,-0.707,1,0.5111,9.5515,0.5,15.7844,0,1,300,High
+263,Maharashtra,58,1.568,5.033,2.352,Cotton,0,-25.404,-0.286,0,0.4673,3.2097,0.5,12.702,1,1,417,High
+264,Maharashtra,45,0.5,14.533,6.947,Cotton,0,-36.621,0.264,0,0.478,29.0663,0.5,18.3104,1,1,413,High
+265,Punjab,43,1.287,15.0,1.766,Sugarcane,0,-11.817,-0.891,0,0.1177,11.6558,1.0,11.8166,0,0,536,High
+266,Punjab,56,0.5,8.08,6.905,Rice,1,-29.95,1.255,2,0.8545,16.1608,1.0,29.95,0,1,300,High
+267,Punjab,53,4.954,6.86,0.599,Rice,0,-15.983,-1.415,0,0.0874,1.3846,1.0,15.9834,0,0,474,High
+268,Punjab,26,8.852,10.131,4.338,Sugarcane,0,8.552,-2.039,1,0.4282,1.1445,1.0,0.0,0,1,497,High
+269,Maharashtra,63,0.5,15.0,6.271,Wheat,0,5.989,0.807,0,0.4181,30.0,0.2,0.0,1,0,546,High
+270,Maharashtra,62,9.311,14.229,7.511,Rice,0,-12.484,-2.575,1,0.5279,1.5282,1.0,12.4842,1,0,364,High
+271,Punjab,37,2.218,7.965,7.319,Rice,0,-13.306,-0.504,2,0.9189,3.5909,1.0,13.3063,0,1,361,High
+272,Punjab,50,1.068,15.0,6.647,Millets,0,16.198,2.0,0,0.4431,14.0428,0.0,0.0,0,0,618,Medium
+273,Punjab,51,0.5,5.044,7.161,Millets,1,-4.189,-0.312,0,1.4198,10.0877,0.0,0.0,0,0,300,High
+274,Punjab,48,2.083,15.0,1.994,Sugarcane,0,-16.917,-1.421,2,0.1329,7.2005,1.0,16.9171,0,1,420,High
+275,Maharashtra,38,4.747,15.0,2.631,Wheat,0,-33.39,-0.357,1,0.1754,3.1598,0.2,6.6781,1,1,435,High
+276,Maharashtra,37,3.684,9.523,2.667,Wheat,0,-6.092,-0.351,0,0.2801,2.5853,0.2,1.2184,1,0,511,High
+277,Punjab,45,0.5,6.228,3.328,Millets,0,0.44,-0.139,1,0.5343,12.4569,0.0,0.0,0,0,490,High
+278,Punjab,59,0.5,5.993,3.45,Millets,0,-21.475,0.362,1,0.5756,11.9868,0.0,0.0,0,1,443,High
+279,Punjab,52,1.547,15.0,4.584,Rice,0,-7.632,-1.339,0,0.3056,9.697,1.0,7.632,0,0,510,High
+280,Punjab,41,1.05,15.0,1.699,Sugarcane,0,-17.454,1.388,2,0.1133,14.287,1.0,17.4536,0,0,479,High
+281,Punjab,54,0.793,7.317,5.684,Rice,0,-14.76,-0.055,0,0.7769,9.2219,1.0,14.7605,0,0,450,High
+282,Maharashtra,25,0.99,15.0,1.757,Rice,0,-49.317,0.427,0,0.1171,15.1561,1.0,49.3173,1,1,400,High
+283,Maharashtra,59,4.881,2.294,2.827,Wheat,0,-15.327,-0.928,1,1.2324,0.47,0.2,3.0653,1,1,382,High
+284,Maharashtra,25,7.945,7.632,4.275,Cotton,0,2.541,0.986,0,0.5602,0.9606,0.5,0.0,1,0,540,High
+285,Maharashtra,41,1.558,11.884,6.473,Millets,1,-21.471,-0.557,0,0.5447,7.6269,0.0,0.0,1,1,300,High
+286,Maharashtra,32,7.097,5.884,6.046,Rice,0,-11.212,-0.022,1,1.0275,0.8291,1.0,11.2121,1,0,398,High
+287,Maharashtra,29,1.164,5.759,5.342,Wheat,0,-23.339,-1.233,2,0.9275,4.9495,0.2,4.6679,1,1,323,High
+288,Maharashtra,59,1.941,7.14,1.668,Millets,0,-6.593,-1.297,2,0.2336,3.6782,0.0,0.0,1,0,397,High
+289,Maharashtra,39,0.769,15.0,4.603,Millets,0,-15.248,0.388,3,0.3069,19.5144,0.0,0.0,1,0,397,High
+290,Maharashtra,49,3.069,10.487,1.156,Rice,0,-26.089,-0.744,2,0.1102,3.4172,1.0,26.0886,1,1,352,High
+291,Maharashtra,44,11.587,2.033,3.33,Cotton,1,7.988,-1.154,0,1.6382,0.1754,0.5,0.0,1,1,304,High
+292,Maharashtra,51,2.775,2.233,3.601,Wheat,0,1.934,-0.943,0,1.6126,0.8047,0.2,0.0,1,0,457,High
+293,Punjab,46,1.329,7.497,0.799,Rice,0,-21.074,0.399,2,0.1065,5.6427,1.0,21.0736,0,1,410,High
+294,Maharashtra,30,3.131,15.0,2.555,Millets,0,-4.533,-0.929,0,0.1703,4.7911,0.0,0.0,1,0,541,High
+295,Maharashtra,57,0.5,1.353,7.667,Millets,0,-3.939,-0.418,1,5.6688,2.7051,0.0,0.0,1,0,352,High
+296,Maharashtra,39,6.585,8.168,7.652,Rice,0,-2.365,-0.285,1,0.9369,1.2403,1.0,2.3655,1,0,419,High
+297,Punjab,26,2.098,7.996,3.142,Rice,0,-37.414,-2.217,0,0.393,3.8115,1.0,37.4138,0,1,368,High
+298,Maharashtra,40,1.79,15.0,0.823,Rice,0,3.099,-1.771,2,0.0549,8.38,1.0,0.0,1,1,471,High
+299,Punjab,63,2.645,8.953,1.843,Millets,1,-20.545,0.036,1,0.2059,3.3853,0.0,0.0,0,0,300,High
+300,Maharashtra,55,2.941,15.0,3.442,Rice,0,-17.367,-1.037,0,0.2294,5.1007,1.0,17.3674,1,0,460,High
+301,Maharashtra,41,0.5,13.64,7.608,Cotton,0,20.0,0.757,3,0.5578,27.28,0.5,0.0,1,0,441,High
+302,Maharashtra,41,3.443,2.216,1.559,Cotton,0,-7.472,0.667,0,0.7036,0.6435,0.5,3.7359,1,0,491,High
+303,Maharashtra,41,0.831,15.0,6.226,Wheat,0,-17.795,0.419,0,0.4151,18.0517,0.2,3.559,1,0,490,High
+304,Punjab,70,3.755,14.473,1.315,Rice,0,-45.079,1.52,1,0.0909,3.8547,1.0,45.0794,0,0,404,High
+305,Punjab,50,5.187,15.0,7.859,Rice,0,8.846,-0.383,2,0.524,2.8919,1.0,0.0,0,1,464,High
+306,Punjab,50,4.888,15.0,3.362,Millets,1,-32.178,0.349,1,0.2241,3.0686,0.0,0.0,0,1,301,High
+307,Maharashtra,57,2.412,2.3,6.314,Millets,1,-27.052,-0.884,1,2.7459,0.9533,0.0,0.0,1,1,300,High
+308,Maharashtra,48,2.203,15.0,3.069,Sugarcane,1,-16.97,2.0,1,0.2046,6.809,1.0,16.9703,1,1,309,High
+309,Punjab,42,0.5,9.68,5.208,Rice,0,9.246,1.684,0,0.538,19.3596,1.0,0.0,0,0,587,High
+310,Maharashtra,43,4.48,15.0,1.94,Cotton,0,-0.401,-0.288,2,0.1293,3.3485,0.5,0.2007,1,0,481,High
+311,Maharashtra,44,0.95,11.747,1.595,Wheat,0,-14.368,0.087,2,0.1358,12.3661,0.2,2.8735,1,1,438,High
+312,Maharashtra,45,2.06,8.676,7.535,Cotton,0,-7.376,1.059,0,0.8685,4.2125,0.5,3.6878,1,0,470,High
+313,Maharashtra,54,1.003,7.722,7.382,Millets,0,-21.006,-0.447,0,0.956,7.6965,0.0,0.0,1,1,410,High
+314,Punjab,46,0.5,15.0,7.156,Wheat,0,-9.905,-0.957,3,0.4771,30.0,0.2,1.981,0,0,381,High
+315,Punjab,54,2.979,3.619,5.191,Rice,0,-10.9,0.574,2,1.4342,1.2148,1.0,10.8999,0,0,379,High
+316,Punjab,44,0.5,15.0,6.562,Millets,0,2.532,-0.222,2,0.4375,30.0,0.0,0.0,0,0,473,High
+317,Maharashtra,46,4.143,8.732,1.018,Millets,0,2.508,-0.117,2,0.1166,2.1079,0.0,0.0,1,1,464,High
+318,Maharashtra,25,1.146,4.933,3.011,Rice,0,-4.59,-1.134,1,0.6104,4.3032,1.0,4.5903,1,1,427,High
+319,Maharashtra,56,0.5,8.499,2.853,Wheat,0,-17.435,-2.018,1,0.3356,16.9987,0.2,3.487,1,0,394,High
+320,Maharashtra,49,2.147,15.0,5.267,Cotton,0,-0.873,-0.064,1,0.3512,6.9866,0.5,0.4363,1,0,485,High
+321,Maharashtra,57,0.5,7.939,3.17,Cotton,0,-13.38,0.605,0,0.3993,15.8771,0.5,6.6898,1,0,481,High
+322,Punjab,25,0.97,15.0,1.614,Rice,0,-28.229,-0.526,1,0.1076,15.4704,1.0,28.2287,0,1,452,High
+323,Punjab,70,3.683,6.789,4.974,Wheat,0,-18.049,1.136,1,0.7326,1.8435,0.2,3.6099,0,0,442,High
+324,Maharashtra,43,0.5,15.0,6.608,Millets,0,-10.732,-1.004,1,0.4405,30.0,0.0,0.0,1,0,442,High
+325,Punjab,58,5.485,15.0,3.366,Wheat,0,1.653,-1.355,1,0.2244,2.7345,0.2,0.0,0,0,514,High
+326,Punjab,33,0.5,15.0,1.492,Rice,0,-14.753,-0.367,0,0.0995,30.0,1.0,14.7529,0,0,548,High
+327,Maharashtra,52,0.85,10.586,1.975,Wheat,0,-30.138,-0.383,0,0.1866,12.4489,0.2,6.0276,1,1,457,High
+328,Punjab,32,4.052,15.0,2.281,Cotton,0,-11.149,0.47,0,0.1521,3.7022,0.5,5.5744,0,0,576,High
+329,Maharashtra,38,1.141,13.177,2.749,Millets,0,-10.298,-1.075,0,0.2086,11.5527,0.0,0.0,1,0,509,High
+330,Maharashtra,68,0.5,15.0,5.321,Cotton,0,-9.212,1.068,0,0.3547,30.0,0.5,4.6062,1,0,518,High
+331,Maharashtra,43,1.34,6.855,5.007,Rice,1,1.798,1.667,3,0.7305,5.1156,1.0,0.0,1,1,300,High
+332,Punjab,48,0.946,15.0,5.721,Rice,0,3.284,-0.339,0,0.3814,15.8604,1.0,0.0,0,0,556,High
+333,Maharashtra,55,5.549,10.425,3.219,Sugarcane,0,-20.913,0.95,0,0.3087,1.8788,1.0,20.9132,1,1,463,High
+334,Maharashtra,51,1.128,11.573,6.68,Cotton,0,-21.242,-1.356,1,0.5772,10.2615,0.5,10.6211,1,0,365,High
+335,Maharashtra,47,4.667,13.11,2.031,Sugarcane,0,-27.603,0.775,1,0.1549,2.8093,1.0,27.603,1,0,424,High
+336,Maharashtra,49,6.673,14.906,4.018,Rice,0,-1.518,0.747,0,0.2695,2.2337,1.0,1.5178,1,0,554,High
+337,Punjab,70,1.707,11.101,6.507,Cotton,0,9.732,0.563,1,0.5862,6.5049,0.5,0.0,0,1,504,High
+338,Punjab,44,7.222,14.466,1.851,Wheat,1,-16.726,-0.316,1,0.128,2.003,0.2,3.3452,0,1,334,High
+339,Maharashtra,47,1.421,15.0,1.324,Millets,0,-15.621,-3.0,0,0.0883,10.5548,0.0,0.0,1,1,482,High
+340,Maharashtra,58,9.979,15.0,7.128,Millets,0,10.002,-0.942,0,0.4752,1.5031,0.0,0.0,1,1,517,High
+341,Maharashtra,58,0.5,15.0,3.072,Rice,0,-4.417,-0.926,2,0.2048,30.0,1.0,4.4166,1,0,437,High
+342,Maharashtra,59,3.485,8.462,5.629,Cotton,1,-5.456,-1.278,2,0.6652,2.4282,0.5,2.7282,1,1,300,High
+343,Maharashtra,53,2.043,5.981,0.69,Rice,0,-29.864,-0.178,1,0.1154,2.9283,1.0,29.8637,1,0,366,High
+344,Maharashtra,31,1.471,15.0,5.675,Wheat,0,-1.819,-0.719,0,0.3784,10.1947,0.2,0.3638,1,0,517,High
+345,Maharashtra,65,3.005,15.0,2.871,Wheat,0,-25.937,-1.125,0,0.1914,4.992,0.2,5.1874,1,0,461,High
+346,Punjab,31,3.651,15.0,0.918,Millets,0,-29.01,-1.359,1,0.0612,4.1088,0.0,0.0,0,0,486,High
+347,Punjab,49,0.5,7.514,5.074,Millets,0,-17.202,-1.016,0,0.6752,15.0287,0.0,0.0,0,0,464,High
+348,Maharashtra,36,0.548,10.139,2.306,Cotton,0,-20.787,0.821,0,0.2274,18.5014,0.5,10.3934,1,0,496,High
+349,Maharashtra,44,3.745,3.282,3.401,Wheat,0,-15.324,-0.961,1,1.0363,0.8763,0.2,3.0648,1,0,388,High
+350,Punjab,49,3.293,8.9,1.104,Millets,0,-33.615,-0.815,0,0.1241,2.7024,0.0,0.0,0,0,482,High
+351,Maharashtra,49,10.148,15.0,1.604,Sugarcane,1,2.382,-0.757,1,0.1069,1.4781,1.0,0.0,1,0,348,High
+352,Punjab,50,4.294,8.388,2.735,Rice,0,4.915,0.039,2,0.326,1.9533,1.0,0.0,0,0,478,High
+353,Punjab,64,9.034,15.0,1.173,Cotton,1,-41.21,0.156,0,0.0782,1.6604,0.5,20.6051,0,1,304,High
+354,Maharashtra,50,3.637,10.46,7.331,Cotton,0,2.677,-3.0,2,0.7009,2.8758,0.5,0.0,1,1,350,High
+355,Maharashtra,42,1.062,15.0,0.545,Rice,0,-24.158,0.722,0,0.0364,14.1284,1.0,24.1579,1,1,505,High
+356,Punjab,57,0.5,4.864,4.129,Rice,0,0.855,1.136,2,0.8489,9.7279,1.0,0.0,0,0,450,High
+357,Maharashtra,59,4.567,13.042,1.424,Rice,0,-13.027,-1.35,0,0.1092,2.8555,1.0,13.0267,1,0,479,High
+358,Maharashtra,30,3.1,1.113,2.808,Wheat,0,-11.568,-1.272,0,2.5225,0.3591,0.2,2.3137,1,0,429,High
+359,Punjab,52,0.5,10.491,7.445,Cotton,0,-14.776,-0.75,1,0.7097,20.9813,0.5,7.3882,0,0,413,High
+360,Maharashtra,53,1.339,3.306,1.713,Rice,0,6.182,-1.056,0,0.5181,2.4685,1.0,0.0,1,0,490,High
+361,Maharashtra,41,9.426,15.0,7.744,Rice,0,-30.594,-0.076,2,0.5162,1.5914,1.0,30.5938,1,1,315,High
+362,Punjab,62,2.082,15.0,2.043,Sugarcane,0,-2.882,-2.247,0,0.1362,7.2051,1.0,2.8824,0,0,532,High
+363,Maharashtra,43,1.701,15.0,6.062,Millets,0,4.52,-1.418,0,0.4041,8.8196,0.0,0.0,1,0,508,High
+364,Maharashtra,47,1.836,15.0,4.6,Wheat,0,-1.255,1.239,1,0.3067,8.1708,0.2,0.251,1,0,518,High
+365,Punjab,43,0.703,5.702,7.134,Cotton,0,20.0,-1.228,0,1.251,8.1122,0.5,0.0,0,0,507,High
+366,Maharashtra,45,1.38,15.0,4.655,Millets,1,-19.934,0.943,0,0.3103,10.8699,0.0,0.0,1,1,338,High
+367,Punjab,32,1.385,15.0,4.509,Rice,0,2.36,0.055,1,0.3006,10.8286,1.0,0.0,0,0,541,High
+368,Punjab,28,0.5,10.082,1.966,Rice,0,-13.591,-2.109,1,0.195,20.1641,1.0,13.5914,0,0,446,High
+369,Maharashtra,64,4.377,15.0,1.747,Cotton,0,6.501,0.335,2,0.1164,3.4266,0.5,0.0,1,1,499,High
+370,Maharashtra,35,1.616,5.028,7.374,Rice,0,-12.115,-0.418,3,1.4666,3.1109,1.0,12.1153,1,1,300,High
+371,Maharashtra,33,5.181,6.299,2.959,Cotton,1,-17.818,-0.755,0,0.4697,1.2158,0.5,8.9091,1,1,300,High
+372,Maharashtra,25,5.698,7.487,3.291,Millets,0,7.591,-0.055,1,0.4395,1.3138,0.0,0.0,1,0,498,High
+373,Maharashtra,37,0.715,7.979,6.199,Wheat,0,-19.373,0.175,0,0.7769,11.1634,0.2,3.8747,1,0,442,High
+374,Punjab,29,3.213,5.232,3.041,Cotton,0,-33.191,0.408,1,0.5813,1.6282,0.5,16.5954,0,1,406,High
+375,Maharashtra,65,1.918,15.0,7.64,Millets,0,-19.753,0.878,0,0.5094,7.8224,0.0,0.0,1,1,472,High
+376,Maharashtra,57,6.365,9.904,6.034,Rice,1,-2.715,0.108,3,0.6093,1.556,1.0,2.7145,1,1,300,High
+377,Punjab,37,0.729,15.0,4.981,Rice,0,8.881,-1.676,1,0.332,20.5748,1.0,0.0,0,1,516,High
+378,Maharashtra,70,3.398,7.719,7.107,Wheat,0,-10.333,-0.774,3,0.9206,2.2718,0.2,2.0666,1,1,300,High
+379,Punjab,57,2.808,2.839,0.872,Wheat,0,7.773,2.0,2,0.3071,1.0113,0.2,0.0,0,1,506,High
+380,Maharashtra,41,1.05,15.0,2.428,Rice,0,-32.373,0.424,2,0.1619,14.2863,1.0,32.3726,1,1,367,High
+381,Maharashtra,25,0.5,2.638,3.988,Cotton,0,-18.649,2.0,2,1.5114,5.2769,0.5,9.3245,1,0,383,High
+382,Punjab,70,3.164,11.532,2.202,Wheat,0,-16.329,1.095,0,0.191,3.6451,0.2,3.2657,0,1,542,High
+383,Punjab,28,4.025,4.077,6.079,Wheat,0,-13.718,-0.058,0,1.4911,1.0129,0.2,2.7437,0,0,464,High
+384,Punjab,25,1.138,15.0,7.394,Wheat,1,-10.739,0.001,0,0.4929,13.1838,0.2,2.1477,0,1,347,High
+385,Punjab,57,3.105,15.0,3.678,Cotton,0,-1.636,-0.337,0,0.2452,4.8301,0.5,0.8182,0,0,561,High
+386,Punjab,70,1.508,13.258,3.066,Rice,0,-18.672,-2.038,1,0.2313,8.7905,1.0,18.6721,0,0,410,High
+387,Maharashtra,62,3.747,15.0,7.712,Cotton,0,-9.915,-0.282,1,0.5141,4.0032,0.5,4.9573,1,1,423,High
+388,Maharashtra,52,0.666,6.19,3.836,Rice,1,-12.677,0.911,1,0.6198,9.2876,1.0,12.6768,1,1,300,High
+389,Punjab,52,6.628,15.0,4.168,Millets,0,-25.888,-0.9,3,0.2779,2.2632,0.0,0.0,0,0,378,High
+390,Punjab,55,1.017,14.294,3.137,Wheat,0,-13.705,0.112,2,0.2195,14.0615,0.2,2.741,0,0,462,High
+391,Punjab,54,1.376,15.0,7.322,Cotton,0,-14.155,0.552,0,0.4881,10.9045,0.5,7.0774,0,0,504,High
+392,Maharashtra,48,0.5,15.0,4.767,Sugarcane,0,-14.476,-1.21,0,0.3178,30.0,1.0,14.4762,1,1,462,High
+393,Maharashtra,46,0.5,15.0,6.216,Wheat,0,-11.524,-0.751,0,0.4144,30.0,0.2,2.3047,1,1,484,High
+394,Punjab,44,2.176,15.0,6.589,Cotton,0,-10.865,-1.318,0,0.4392,6.8929,0.5,5.4324,0,0,490,High
+395,Maharashtra,36,0.934,3.242,3.064,Sugarcane,0,-18.484,-1.122,1,0.9451,3.4698,1.0,18.4844,1,1,357,High
+396,Punjab,42,5.414,5.652,3.753,Millets,0,7.091,-1.588,1,0.664,1.0439,0.0,0.0,0,0,472,High
+397,Punjab,25,0.5,11.043,6.864,Wheat,0,-29.542,-1.2,1,0.6215,22.0868,0.2,5.9084,0,1,399,High
+398,Maharashtra,44,1.43,13.127,3.224,Rice,1,-22.815,1.119,2,0.2456,9.1817,1.0,22.8146,1,1,300,High
+399,Punjab,33,1.235,4.932,6.455,Wheat,1,20.0,-1.215,2,1.3086,3.9955,0.2,0.0,0,0,300,High
+400,Punjab,32,0.5,15.0,7.557,Sugarcane,0,-27.57,-0.174,0,0.5038,30.0,1.0,27.5699,0,1,440,High
+401,Maharashtra,47,0.5,15.0,5.174,Rice,0,-32.306,-0.803,1,0.3449,30.0,1.0,32.3063,1,0,357,High
+402,Punjab,62,1.601,15.0,6.062,Rice,0,-0.983,0.879,0,0.4041,9.3663,1.0,0.9828,0,0,557,High
+403,Maharashtra,56,2.037,12.905,6.538,Millets,0,-43.157,1.011,0,0.5066,6.3356,0.0,0.0,1,0,429,High
+404,Punjab,26,1.554,10.28,6.838,Rice,0,-15.529,0.564,0,0.6651,6.6154,1.0,15.5291,0,0,478,High
+405,Maharashtra,33,2.271,5.689,3.463,Sugarcane,0,20.0,0.006,0,0.6087,2.5048,1.0,0.0,1,0,546,High
+406,Punjab,56,2.713,12.962,0.884,Wheat,0,-35.174,-0.531,0,0.0682,4.7783,0.2,7.0348,0,0,493,High
+407,Maharashtra,33,0.5,14.972,3.514,Cotton,0,-41.729,-1.492,0,0.2347,29.9446,0.5,20.8647,1,0,408,High
+408,Maharashtra,42,1.871,4.511,5.553,Rice,0,-34.488,1.644,1,1.2309,2.4107,1.0,34.4877,1,1,328,High
+409,Punjab,52,9.931,15.0,3.878,Sugarcane,0,1.475,-1.06,1,0.2585,1.5104,1.0,0.0,0,0,521,High
+410,Maharashtra,33,2.196,10.76,6.117,Cotton,0,4.872,0.364,0,0.5685,4.8999,0.5,0.0,1,0,522,High
+411,Maharashtra,46,3.395,2.648,3.177,Wheat,0,-19.689,0.76,0,1.2,0.7799,0.2,3.9378,1,1,448,High
+412,Punjab,29,1.121,6.121,5.261,Wheat,0,-1.752,-0.855,2,0.8595,5.4586,0.2,0.3504,0,1,416,High
+413,Punjab,38,4.462,8.751,1.898,Sugarcane,0,8.854,0.738,0,0.2169,1.9613,1.0,0.0,0,0,599,High
+414,Maharashtra,49,4.447,15.0,6.12,Millets,0,14.81,0.464,2,0.408,3.3728,0.0,0.0,1,0,482,High
+415,Punjab,26,2.21,15.0,4.593,Wheat,0,-15.695,1.428,1,0.3062,6.7873,0.2,3.139,0,0,526,High
+416,Maharashtra,50,11.191,4.185,2.03,Cotton,1,-46.522,-0.769,0,0.4851,0.374,0.5,23.261,1,1,300,High
+417,Maharashtra,45,0.5,15.0,2.689,Millets,0,-24.313,-1.586,0,0.1793,30.0,0.0,0.0,1,0,482,High
+418,Maharashtra,52,0.5,3.533,6.115,Wheat,0,-18.402,-0.805,0,1.7307,7.0661,0.2,3.6804,1,0,390,High
+419,Punjab,48,4.411,15.0,4.12,Cotton,0,-15.884,-0.091,0,0.2746,3.4005,0.5,7.942,0,0,524,High
+420,Punjab,61,4.938,8.357,2.12,Rice,1,-14.818,-0.242,0,0.2537,1.6925,1.0,14.8179,0,0,312,High
+421,Punjab,47,0.5,11.303,7.755,Sugarcane,1,4.783,-0.693,1,0.686,22.6065,1.0,0.0,0,1,300,High
+422,Punjab,40,0.926,15.0,5.863,Cotton,0,7.116,-2.209,0,0.3909,16.1981,0.5,0.0,0,0,532,High
+423,Maharashtra,46,0.5,4.35,7.679,Wheat,1,-14.575,0.17,1,1.765,8.7008,0.2,2.915,1,1,300,High
+424,Maharashtra,52,1.041,3.984,3.703,Wheat,0,4.477,1.672,1,0.9295,3.8254,0.2,0.0,1,0,481,High
+425,Maharashtra,46,4.45,7.638,4.063,Rice,0,-6.635,0.164,0,0.532,1.7163,1.0,6.6351,1,0,481,High
+426,Maharashtra,45,2.189,15.0,7.235,Wheat,0,-11.032,0.309,0,0.4823,6.8519,0.2,2.2065,1,0,490,High
+427,Maharashtra,37,1.284,3.613,4.438,Rice,0,1.598,0.272,0,1.2283,2.8148,1.0,0.0,1,0,485,High
+428,Punjab,37,1.394,1.264,5.805,Cotton,1,-10.916,0.097,0,4.5923,0.9065,0.5,5.4581,0,1,300,High
+429,Maharashtra,28,0.5,8.492,2.391,Wheat,0,18.867,0.632,1,0.2816,16.9831,0.2,0.0,1,0,547,High
+430,Maharashtra,66,1.069,15.0,3.85,Cotton,1,-33.576,0.095,0,0.2567,14.0296,0.5,16.7881,1,1,300,High
+431,Maharashtra,30,3.119,15.0,1.426,Sugarcane,0,-19.728,-0.613,1,0.0951,4.8088,1.0,19.7276,1,0,452,High
+432,Punjab,37,10.971,8.41,2.045,Sugarcane,0,-13.898,-1.096,0,0.2431,0.7666,1.0,13.8978,0,0,496,High
+433,Punjab,36,5.633,8.766,0.507,Sugarcane,0,-5.408,-0.549,0,0.0578,1.5561,1.0,5.4082,0,0,552,High
+434,Maharashtra,56,0.5,15.0,0.794,Rice,0,7.019,0.014,1,0.053,30.0,1.0,0.0,1,0,552,High
+435,Maharashtra,41,6.865,2.73,7.83,Rice,0,-12.366,-0.616,1,2.8683,0.3976,1.0,12.366,1,0,337,High
+436,Punjab,60,4.586,8.928,2.318,Wheat,0,-33.333,0.014,0,0.2596,1.9466,0.2,6.6667,0,0,470,High
+437,Maharashtra,37,4.548,14.187,5.475,Millets,1,0.638,0.139,0,0.3859,3.1198,0.0,0.0,1,1,358,High
+438,Punjab,48,1.837,15.0,6.793,Sugarcane,0,-0.676,0.533,1,0.4529,8.1635,1.0,0.6756,0,0,510,High
+439,Punjab,35,1.523,8.102,4.63,Sugarcane,0,-41.128,0.983,0,0.5714,5.3217,1.0,41.1279,0,0,395,High
+440,Maharashtra,70,1.083,12.523,1.649,Cotton,0,-12.057,-0.222,0,0.1316,11.5679,0.5,6.0285,1,0,503,High
+441,Maharashtra,31,0.5,1.664,5.961,Cotton,0,-35.852,-1.608,1,3.5818,3.3284,0.5,17.9261,1,1,300,High
+442,Punjab,49,0.777,7.195,5.002,Wheat,1,-21.97,-0.645,0,0.6952,9.2574,0.2,4.394,0,1,300,High
+443,Maharashtra,53,0.849,15.0,5.984,Rice,0,-25.542,-0.256,1,0.399,17.6576,1.0,25.5421,1,1,379,High
+444,Maharashtra,50,1.985,5.552,6.273,Rice,0,0.139,-2.136,1,1.1299,2.7964,1.0,0.0,1,1,382,High
+445,Punjab,43,4.128,12.771,7.816,Millets,0,-10.248,1.089,1,0.612,3.0935,0.0,0.0,0,0,483,High
+446,Punjab,43,1.926,15.0,4.802,Rice,0,-31.284,-0.158,1,0.3201,7.787,1.0,31.2839,0,0,404,High
+447,Punjab,46,0.5,9.262,3.062,Rice,0,1.201,0.081,1,0.3306,18.5248,1.0,0.0,0,0,516,High
+448,Punjab,43,9.495,13.962,5.363,Rice,0,-33.507,1.066,1,0.3841,1.4705,1.0,33.5071,0,1,414,High
+449,Maharashtra,57,8.58,5.64,1.013,Millets,0,-7.048,-0.93,2,0.1797,0.6573,0.0,0.0,1,1,410,High
+450,Maharashtra,70,4.614,10.326,7.23,Wheat,0,-24.21,0.274,0,0.7002,2.2381,0.2,4.842,1,1,419,High
+451,Punjab,38,14.243,15.0,1.395,Wheat,0,-17.485,-0.41,0,0.093,1.0532,0.2,3.497,0,0,566,High
+452,Maharashtra,29,2.359,13.813,2.959,Millets,0,-4.436,-0.484,1,0.2142,5.8554,0.0,0.0,1,0,498,High
+453,Maharashtra,44,9.872,7.62,6.618,Cotton,0,-4.178,-1.072,1,0.8685,0.7719,0.5,2.0891,1,1,410,High
+454,Punjab,70,0.5,14.576,4.98,Wheat,0,8.055,1.366,0,0.3416,29.1519,0.2,0.0,0,0,597,High
+455,Maharashtra,35,1.366,15.0,3.452,Rice,0,-13.234,-0.676,2,0.2302,10.9841,1.0,13.2338,1,0,411,High
+456,Maharashtra,65,0.766,2.363,4.051,Rice,0,-7.403,0.399,0,1.7141,3.0846,1.0,7.4032,1,0,437,High
+457,Maharashtra,65,0.654,15.0,6.909,Sugarcane,1,-18.566,-2.175,1,0.4606,22.9412,1.0,18.5661,1,0,300,High
+458,Maharashtra,38,0.5,9.309,3.05,Sugarcane,0,-9.27,0.88,1,0.3276,18.6183,1.0,9.2702,1,1,469,High
+459,Maharashtra,52,2.922,15.0,7.022,Cotton,0,-5.408,0.752,3,0.4682,5.1337,0.5,2.7039,1,0,387,High
+460,Maharashtra,65,4.56,15.0,1.161,Cotton,1,-13.475,-0.203,0,0.0774,3.2891,0.5,6.7373,1,1,345,High
+461,Maharashtra,40,2.59,4.066,6.326,Wheat,0,-39.056,-0.527,1,1.5558,1.5697,0.2,7.8113,1,1,312,High
+462,Punjab,43,0.5,6.906,6.857,Rice,0,-7.028,0.1,0,0.9928,13.8128,1.0,7.0279,0,0,475,High
+463,Punjab,38,5.0,9.671,1.864,Rice,0,-6.532,1.989,0,0.1927,1.9343,1.0,6.5316,0,0,585,High
+464,Maharashtra,33,3.031,6.508,3.728,Rice,0,-22.203,-2.063,0,0.5728,2.1468,1.0,22.2026,1,1,382,High
+465,Punjab,37,1.474,15.0,1.741,Cotton,0,-31.266,0.482,1,0.1161,10.1772,0.5,15.6329,0,0,477,High
+466,Maharashtra,30,3.364,6.139,5.8,Sugarcane,0,10.138,-1.212,0,0.9448,1.8249,1.0,0.0,1,0,482,High
+467,Maharashtra,45,0.903,7.811,4.515,Cotton,1,-4.628,-0.32,0,0.578,8.651,0.5,2.3142,1,0,302,High
+468,Punjab,36,1.27,14.347,5.265,Rice,0,-15.436,-0.548,1,0.367,11.2944,1.0,15.4362,0,1,452,High
+469,Maharashtra,48,7.484,4.657,1.974,Wheat,0,-29.055,-0.728,1,0.4238,0.6222,0.2,5.8111,1,0,385,High
+470,Punjab,26,1.03,8.877,2.088,Millets,0,-24.327,-0.31,1,0.2352,8.6217,0.0,0.0,0,0,471,High
+471,Maharashtra,49,1.905,15.0,0.811,Rice,0,7.539,-1.434,2,0.0541,7.8758,1.0,0.0,1,0,483,High
+472,Punjab,55,6.626,9.334,2.916,Rice,0,-9.935,0.729,0,0.3123,1.4088,1.0,9.9351,0,0,528,High
+473,Maharashtra,25,3.69,11.306,4.698,Cotton,0,-20.259,-2.094,3,0.4156,3.0637,0.5,10.1294,1,1,306,High
+474,Punjab,49,0.5,7.977,6.935,Sugarcane,0,-4.355,0.24,2,0.8693,15.9536,1.0,4.3547,0,0,409,High
+475,Maharashtra,60,0.5,3.627,5.502,Wheat,0,-24.216,-0.627,0,1.5169,7.2541,0.2,4.8432,1,0,383,High
+476,Punjab,30,0.5,15.0,3.765,Sugarcane,1,10.513,-1.471,0,0.251,30.0,1.0,0.0,0,0,407,High
+477,Maharashtra,65,2.99,15.0,7.648,Millets,1,8.292,-0.098,0,0.5099,5.0163,0.0,0.0,1,1,337,High
+478,Maharashtra,50,1.781,12.363,5.894,Sugarcane,0,-18.999,-1.857,0,0.4768,6.9423,1.0,18.9986,1,0,399,High
+479,Punjab,37,0.5,14.588,7.476,Wheat,0,-21.43,-0.26,1,0.5125,29.1764,0.2,4.286,0,0,446,High
+480,Maharashtra,44,9.506,15.0,4.457,Wheat,0,-51.272,-1.723,1,0.2971,1.578,0.2,10.2544,1,0,347,High
+481,Punjab,52,2.263,4.413,2.442,Rice,1,-34.224,0.161,1,0.5533,1.9501,1.0,34.2237,0,1,300,High
+482,Punjab,46,0.831,15.0,0.896,Millets,0,-32.092,-1.224,1,0.0597,18.055,0.0,0.0,0,0,475,High
+483,Maharashtra,51,2.083,14.323,5.946,Cotton,0,-9.972,-0.751,0,0.4151,6.8759,0.5,4.9861,1,0,474,High
+484,Punjab,34,3.415,15.0,1.41,Rice,0,-16.894,-0.006,2,0.094,4.3927,1.0,16.8935,0,0,461,High
+485,Maharashtra,47,0.5,6.777,2.771,Wheat,0,-32.18,-1.073,0,0.4088,13.5545,0.2,6.436,1,1,411,High
+486,Punjab,28,0.964,15.0,4.494,Cotton,0,-16.217,-0.358,3,0.2996,15.5678,0.5,8.1087,0,0,402,High
+487,Maharashtra,44,4.935,15.0,4.733,Wheat,0,-4.42,-1.814,0,0.3155,3.0397,0.2,0.884,1,0,494,High
+488,Maharashtra,34,1.847,15.0,5.004,Wheat,0,-4.082,0.331,3,0.3336,8.1226,0.2,0.8165,1,1,414,High
+489,Punjab,36,2.373,14.691,1.747,Cotton,0,-40.495,1.465,0,0.1189,6.1913,0.5,20.2475,0,0,508,High
+490,Punjab,60,1.701,9.986,3.348,Sugarcane,0,-13.924,2.0,0,0.3353,5.8699,1.0,13.924,0,0,529,High
+491,Maharashtra,58,0.5,15.0,5.129,Cotton,0,-32.75,-0.209,0,0.3419,30.0,0.5,16.3751,1,0,430,High
+492,Punjab,52,0.544,13.63,7.774,Cotton,1,1.526,0.488,1,0.5703,25.0709,0.5,0.0,0,1,323,High
+493,Maharashtra,32,1.769,2.525,5.957,Wheat,0,-0.238,-2.081,2,2.3591,1.4273,0.2,0.0476,1,1,334,High
+494,Maharashtra,41,0.704,4.984,7.42,Wheat,1,-25.47,1.203,0,1.4886,7.0789,0.2,5.0941,1,1,300,High
+495,Maharashtra,60,0.5,10.499,6.217,Sugarcane,0,-23.491,0.476,2,0.5921,20.9989,1.0,23.4907,1,1,329,High
+496,Maharashtra,47,5.567,15.0,4.938,Cotton,0,-18.919,-0.784,1,0.3292,2.6945,0.5,9.4595,1,0,425,High
+497,Maharashtra,70,11.921,11.643,1.94,Wheat,0,-45.792,0.937,1,0.1666,0.9767,0.2,9.1584,1,1,408,High
+498,Maharashtra,50,4.727,15.0,5.5,Rice,0,-6.96,-0.554,2,0.3667,3.1731,1.0,6.9605,1,0,410,High
+499,Punjab,47,5.768,2.558,5.175,Millets,0,-16.093,-0.758,0,2.0234,0.4434,0.0,0.0,0,1,442,High
+500,Punjab,41,0.831,13.665,5.019,Cotton,0,2.553,0.175,0,0.3672,16.4421,0.5,0.0,0,0,568,High
+501,Punjab,47,9.704,2.845,4.173,Sugarcane,0,-28.476,-1.363,2,1.4668,0.2932,1.0,28.4761,0,1,300,High
+502,Maharashtra,43,0.658,15.0,4.465,Rice,0,-21.287,0.722,2,0.2977,22.799,1.0,21.2868,1,1,395,High
+503,Maharashtra,53,9.074,11.296,3.008,Rice,0,-34.598,-1.356,3,0.2663,1.2448,1.0,34.5979,1,1,300,High
+504,Punjab,56,15.788,15.0,4.395,Wheat,1,-27.064,-1.35,2,0.293,0.9501,0.2,5.4129,0,1,300,High
+505,Punjab,36,3.731,6.104,1.982,Millets,0,-29.73,-1.21,0,0.3247,1.6357,0.0,0.0,0,0,464,High
+506,Maharashtra,29,11.913,15.0,6.539,Wheat,1,2.94,0.46,2,0.4359,1.2592,0.2,0.0,1,1,300,High
+507,Punjab,25,2.529,15.0,1.893,Millets,0,-22.474,-0.917,0,0.1262,5.9319,0.0,0.0,0,1,541,High
+508,Punjab,51,0.901,15.0,1.137,Wheat,0,-10.137,-1.648,3,0.0758,16.6397,0.2,2.0275,0,1,424,High
+509,Punjab,32,1.723,10.983,3.772,Wheat,0,-17.711,-0.906,0,0.3434,6.3731,0.2,3.5423,0,0,499,High
+510,Punjab,25,2.701,15.0,5.437,Rice,0,-31.042,-0.332,1,0.3625,5.5536,1.0,31.042,0,0,404,High
+511,Punjab,50,0.5,6.92,3.783,Sugarcane,0,-14.525,-1.481,1,0.5466,13.841,1.0,14.5245,0,0,404,High
+512,Maharashtra,70,2.921,15.0,2.576,Rice,0,-3.013,-1.452,0,0.1717,5.1347,1.0,3.0135,1,0,508,High
+513,Punjab,45,11.918,15.0,4.686,Millets,0,-30.116,1.811,0,0.3124,1.2586,0.0,0.0,0,0,547,High
+514,Punjab,55,0.633,15.0,3.103,Millets,0,0.137,-1.161,2,0.2068,23.7131,0.0,0.0,0,1,479,High
+515,Maharashtra,46,4.719,6.796,7.497,Millets,0,-24.746,-0.623,2,1.1031,1.44,0.0,0.0,1,0,317,High
+516,Maharashtra,44,7.155,15.0,7.42,Wheat,0,-19.245,-0.667,0,0.4947,2.0966,0.2,3.849,1,0,454,High
+517,Maharashtra,56,8.631,4.33,4.268,Wheat,0,-4.04,-0.721,1,0.9857,0.5016,0.2,0.808,1,0,417,High
+518,Punjab,42,9.667,15.0,2.959,Sugarcane,0,-21.952,0.321,1,0.1973,1.5517,1.0,21.9515,0,0,475,High
+519,Punjab,48,2.211,15.0,6.028,Rice,0,-11.436,0.648,0,0.4019,6.7841,1.0,11.4364,0,0,520,High
+520,Maharashtra,49,11.357,2.73,0.775,Wheat,0,-16.72,-0.606,0,0.2841,0.2403,0.2,3.3441,1,0,464,High
+521,Maharashtra,37,4.248,4.809,4.061,Rice,0,-10.059,-1.536,2,0.8443,1.1322,1.0,10.0595,1,1,342,High
+522,Maharashtra,57,0.529,9.713,3.022,Wheat,0,-47.711,0.974,1,0.3112,18.3603,0.2,9.5422,1,0,382,High
+523,Punjab,43,1.944,9.577,7.404,Rice,0,-39.445,-0.395,1,0.7731,4.9263,1.0,39.4452,0,1,310,High
+524,Punjab,36,3.801,4.629,0.59,Rice,0,-2.76,1.43,1,0.1275,1.2179,1.0,2.7605,0,0,533,High
+525,Punjab,51,0.852,15.0,4.646,Cotton,1,3.558,-2.546,0,0.3097,17.5968,0.5,0.0,0,1,349,High
+526,Maharashtra,34,3.07,14.897,6.055,Cotton,0,-25.657,-0.105,0,0.4064,4.8524,0.5,12.8287,1,0,451,High
+527,Maharashtra,45,3.295,10.429,4.138,Wheat,0,-15.66,1.619,0,0.3967,3.1648,0.2,3.1321,1,0,511,High
+528,Maharashtra,45,0.533,6.655,1.141,Rice,0,-7.579,-0.653,0,0.1714,12.4925,1.0,7.5792,1,0,487,High
+529,Maharashtra,58,2.498,15.0,7.793,Wheat,0,-1.602,-0.525,1,0.5196,6.0052,0.2,0.3203,1,0,445,High
+530,Punjab,51,4.43,15.0,4.385,Rice,0,-42.415,-0.004,2,0.2923,3.3858,1.0,42.4152,0,0,325,High
+531,Punjab,45,2.073,8.274,5.106,Wheat,0,-25.291,-0.859,0,0.6171,3.9918,0.2,5.0581,0,1,445,High
+532,Punjab,55,0.5,7.465,2.275,Wheat,0,-16.989,-0.612,0,0.3047,14.9306,0.2,3.3977,0,0,492,High
+533,Punjab,62,0.5,15.0,4.126,Wheat,0,-1.453,-0.638,1,0.2751,30.0,0.2,0.2906,0,0,513,High
+534,Maharashtra,52,1.325,9.148,3.719,Sugarcane,0,20.0,-0.927,2,0.4065,6.9046,1.0,0.0,1,1,454,High
+535,Punjab,45,7.819,10.897,1.062,Cotton,0,-15.42,0.929,1,0.0974,1.3937,0.5,7.71,0,1,518,High
+536,Punjab,29,0.779,15.0,1.296,Sugarcane,1,-33.105,0.161,2,0.0864,19.2569,1.0,33.1047,0,1,300,High
+537,Punjab,32,3.018,2.645,6.781,Cotton,0,-5.966,-0.049,1,2.5641,0.8762,0.5,2.983,0,0,418,High
+538,Maharashtra,41,0.753,15.0,2.299,Wheat,0,7.815,0.509,1,0.1532,19.9103,0.2,0.0,1,0,553,High
+539,Maharashtra,38,1.164,12.912,1.962,Wheat,0,-10.38,-0.564,0,0.152,11.0892,0.2,2.076,1,0,522,High
+540,Maharashtra,43,5.652,8.765,4.287,Rice,0,-25.6,-1.074,0,0.4891,1.5507,1.0,25.5997,1,0,392,High
+541,Punjab,46,3.915,11.343,6.266,Wheat,0,2.37,-0.903,0,0.5524,2.8971,0.2,0.0,0,1,518,High
+542,Punjab,51,0.5,11.822,0.965,Wheat,0,-2.047,-0.298,1,0.0816,23.6436,0.2,0.4094,0,0,537,High
+543,Maharashtra,44,1.68,3.05,4.826,Cotton,0,-27.391,-1.359,2,1.5825,1.8158,0.5,13.6956,1,1,300,High
+544,Punjab,51,0.5,15.0,1.394,Rice,0,-6.279,-0.8,2,0.0929,30.0,1.0,6.2786,0,1,482,High
+545,Punjab,46,1.967,12.592,0.773,Rice,0,2.959,-0.816,0,0.0614,6.4019,1.0,0.0,0,0,584,High
+546,Maharashtra,25,3.258,15.0,0.896,Millets,0,5.728,1.825,0,0.0597,4.6041,0.0,0.0,1,0,635,Medium
+547,Maharashtra,34,0.5,15.0,6.756,Wheat,0,-12.997,-3.0,0,0.4504,30.0,0.2,2.5993,1,0,438,High
+548,Punjab,43,1.023,5.382,1.385,Rice,0,-17.371,-0.48,0,0.2573,5.2623,1.0,17.3712,0,0,472,High
+549,Maharashtra,30,1.522,15.0,0.839,Cotton,0,6.409,1.426,1,0.0559,9.8575,0.5,0.0,1,1,586,High
+550,Punjab,52,7.56,9.834,3.784,Wheat,0,-30.152,-0.178,2,0.3848,1.3009,0.2,6.0305,0,1,389,High
+551,Maharashtra,63,14.997,15.0,6.83,Sugarcane,0,-5.421,0.453,3,0.4553,1.0002,1.0,5.4213,1,1,384,High
+552,Punjab,60,0.5,2.222,2.468,Rice,1,-32.55,-0.2,1,1.1109,4.4432,1.0,32.5504,0,1,300,High
+553,Maharashtra,42,4.294,11.254,3.663,Rice,0,3.981,-0.967,2,0.3255,2.6209,1.0,0.0,1,0,438,High
+554,Maharashtra,63,1.394,7.521,0.799,Millets,1,-15.871,-2.542,1,0.1063,5.3942,0.0,0.0,1,1,300,High
+555,Punjab,47,1.495,15.0,3.868,Rice,0,7.029,-0.095,0,0.2579,10.0334,1.0,0.0,0,0,588,High
+556,Maharashtra,41,4.211,15.0,4.832,Rice,0,-4.629,-1.076,1,0.3221,3.5625,1.0,4.6288,1,0,460,High
+557,Maharashtra,38,7.519,6.916,4.782,Millets,0,-30.348,-1.1,1,0.6914,0.9199,0.0,0.0,1,0,373,High
+558,Maharashtra,41,9.043,6.891,2.993,Wheat,1,9.369,-0.799,1,0.4343,0.762,0.2,0.0,1,0,306,High
+559,Punjab,40,2.582,11.15,2.866,Cotton,0,11.269,1.445,0,0.2571,4.3184,0.5,0.0,0,0,618,Medium
+560,Punjab,47,1.325,5.946,1.3,Rice,0,1.234,2.0,0,0.2186,4.4861,1.0,0.0,0,0,590,High
+561,Punjab,47,4.647,15.0,3.251,Rice,0,-23.424,-0.083,0,0.2167,3.2278,1.0,23.4238,0,0,491,High
+562,Maharashtra,45,0.867,9.067,1.246,Rice,0,-16.59,-0.765,3,0.1374,10.4572,1.0,16.5902,1,1,340,High
+563,Maharashtra,50,2.491,1.191,6.249,Millets,0,-9.819,-1.477,0,5.2493,0.478,0.0,0.0,1,0,379,High
+564,Maharashtra,59,1.332,9.026,7.747,Cotton,1,-22.556,0.169,1,0.8583,6.7755,0.5,11.2778,1,1,300,High
+565,Maharashtra,56,3.207,5.308,7.776,Sugarcane,1,13.385,1.483,1,1.4649,1.6551,1.0,0.0,1,1,300,High
+566,Maharashtra,27,0.825,3.303,6.988,Millets,0,-9.835,-0.547,0,2.1153,4.0048,0.0,0.0,1,0,421,High
+567,Maharashtra,25,0.638,4.527,5.006,Cotton,1,-52.55,-2.679,1,1.1058,7.0948,0.5,26.2751,1,0,300,High
+568,Maharashtra,56,7.512,13.297,5.756,Rice,0,-6.91,0.37,2,0.4329,1.7701,1.0,6.9097,1,0,415,High
+569,Punjab,29,0.5,15.0,2.639,Wheat,0,-17.222,0.685,0,0.1759,30.0,0.2,3.4445,0,0,572,High
+570,Maharashtra,42,2.115,15.0,5.231,Millets,0,-13.124,-1.886,2,0.3487,7.0908,0.0,0.0,1,1,389,High
+571,Maharashtra,31,0.749,7.499,3.917,Sugarcane,0,-17.529,-0.124,0,0.5224,10.0168,1.0,17.5293,1,1,441,High
+572,Punjab,25,0.5,15.0,4.115,Rice,1,0.106,0.151,0,0.2743,30.0,1.0,0.0,0,1,414,High
+573,Maharashtra,51,0.5,8.438,6.183,Wheat,0,4.795,-0.373,2,0.7328,16.8757,0.2,0.0,1,1,405,High
+574,Punjab,54,0.578,6.989,2.614,Wheat,0,-7.754,-1.091,2,0.3741,12.0931,0.2,1.5508,0,0,418,High
+575,Punjab,38,6.047,4.838,2.885,Millets,0,-17.09,0.716,3,0.5964,0.8001,0.0,0.0,0,1,390,High
+576,Punjab,25,0.996,3.0,7.435,Rice,0,-12.62,-0.703,1,2.4785,3.0123,1.0,12.6201,0,0,372,High
+577,Maharashtra,38,9.015,8.918,0.922,Wheat,0,-11.706,-1.785,2,0.1034,0.9892,0.2,2.3412,1,1,409,High
+578,Punjab,50,2.614,13.825,4.142,Millets,1,-24.036,-2.349,2,0.2996,5.2884,0.0,0.0,0,1,300,High
+579,Maharashtra,27,1.721,11.276,7.346,Cotton,0,-3.316,0.961,2,0.6515,6.5507,0.5,1.658,1,1,425,High
+580,Maharashtra,47,2.603,8.734,5.078,Wheat,0,-5.099,0.317,1,0.5814,3.3552,0.2,1.0199,1,1,450,High
+581,Maharashtra,45,2.181,5.717,4.598,Rice,0,11.107,-0.212,0,0.8043,2.6208,1.0,0.0,1,0,504,High
+582,Punjab,52,4.266,15.0,2.373,Cotton,0,6.83,-0.74,2,0.1582,3.5161,0.5,0.0,0,0,508,High
+583,Maharashtra,46,0.996,4.247,3.075,Cotton,0,-22.285,-0.829,0,0.7239,4.2647,0.5,11.1426,1,0,409,High
+584,Maharashtra,33,1.306,15.0,6.89,Millets,0,0.304,-1.376,1,0.4594,11.4814,0.0,0.0,1,0,455,High
+585,Punjab,59,6.736,5.464,4.694,Cotton,0,-26.264,-0.731,0,0.8591,0.8112,0.5,13.1322,0,1,417,High
+586,Maharashtra,43,8.782,6.673,4.361,Wheat,0,-1.561,-2.42,1,0.6535,0.7599,0.2,0.3122,1,1,410,High
+587,Maharashtra,45,6.692,3.768,1.247,Wheat,0,-19.749,-1.036,0,0.3309,0.563,0.2,3.9498,1,0,446,High
+588,Maharashtra,34,1.631,8.157,0.62,Rice,0,-27.349,-0.611,0,0.076,5.0002,1.0,27.3488,1,0,431,High
+589,Maharashtra,40,4.547,15.0,7.728,Wheat,0,-7.799,0.31,2,0.5152,3.2986,0.2,1.5599,1,0,416,High
+590,Maharashtra,34,1.941,15.0,3.327,Sugarcane,0,-52.119,-0.34,0,0.2218,7.7291,1.0,52.1185,1,0,352,High
+591,Maharashtra,43,2.065,15.0,5.024,Millets,0,-38.265,0.495,1,0.3349,7.2649,0.0,0.0,1,0,423,High
+592,Punjab,66,0.687,15.0,1.098,Wheat,1,-14.662,1.049,0,0.0732,21.833,0.2,2.9325,0,1,404,High
+593,Maharashtra,29,2.683,5.038,5.623,Wheat,0,-10.763,0.758,0,1.1161,1.8778,0.2,2.1527,1,0,467,High
+594,Maharashtra,26,0.618,9.35,7.494,Rice,0,-17.54,0.171,0,0.8015,15.1321,1.0,17.5398,1,0,423,High
+595,Punjab,63,1.208,6.526,4.817,Millets,0,-35.692,-0.288,1,0.7382,5.4006,0.0,0.0,0,0,385,High
+596,Maharashtra,42,5.794,15.0,1.446,Rice,0,-24.268,-1.281,0,0.0964,2.589,1.0,24.2684,1,1,458,High
+597,Punjab,37,0.696,4.677,4.863,Sugarcane,0,-6.132,0.44,2,1.0398,6.7184,1.0,6.1319,0,1,412,High
+598,Maharashtra,57,0.5,15.0,6.248,Cotton,1,-3.273,1.38,0,0.4165,30.0,0.5,1.6366,1,1,361,High
+599,Maharashtra,38,0.5,15.0,3.548,Millets,0,-12.243,0.042,0,0.2365,30.0,0.0,0.0,1,0,533,High
+600,Maharashtra,67,8.48,15.0,7.099,Sugarcane,0,5.353,1.617,0,0.4733,1.7689,1.0,0.0,1,0,549,High
+601,Maharashtra,53,2.121,8.778,5.365,Sugarcane,0,-6.039,-1.826,0,0.6112,4.1381,1.0,6.039,1,1,433,High
+602,Maharashtra,39,1.58,15.0,7.244,Rice,0,-23.212,-1.454,1,0.483,9.4942,1.0,23.2118,1,1,357,High
+603,Maharashtra,70,4.996,15.0,5.176,Cotton,0,-16.208,0.555,2,0.3451,3.0023,0.5,8.1038,1,1,403,High
+604,Maharashtra,38,5.42,7.433,1.485,Wheat,1,-33.043,-0.036,1,0.1997,1.3712,0.2,6.6086,1,1,300,High
+605,Maharashtra,54,1.211,3.568,2.823,Sugarcane,0,-13.622,1.014,0,0.7911,2.9456,1.0,13.6221,1,0,453,High
+606,Maharashtra,49,3.549,2.707,2.013,Rice,0,4.682,-0.315,0,0.7437,0.7629,1.0,0.0,1,0,498,High
+607,Maharashtra,61,4.424,15.0,7.316,Millets,0,-1.721,-1.168,0,0.4877,3.3908,0.0,0.0,1,1,478,High
+608,Maharashtra,64,3.19,2.701,6.203,Millets,0,9.03,-0.514,3,2.2969,0.8467,0.0,0.0,1,1,327,High
+609,Maharashtra,45,0.5,9.044,5.573,Cotton,0,-16.543,-1.833,0,0.6162,18.0888,0.5,8.2716,1,1,413,High
+610,Punjab,36,6.26,15.0,2.757,Sugarcane,0,8.191,-1.429,1,0.1838,2.3963,1.0,0.0,0,0,544,High
+611,Maharashtra,51,2.324,14.128,1.88,Wheat,0,-14.9,0.672,0,0.1331,6.0799,0.2,2.9799,1,0,535,High
+612,Punjab,37,0.996,15.0,6.173,Wheat,0,10.074,0.573,0,0.4116,15.0628,0.2,0.0,0,0,589,High
+613,Punjab,69,1.661,14.554,4.056,Cotton,0,-12.328,0.063,0,0.2787,8.7601,0.5,6.1639,0,1,523,High
+614,Maharashtra,47,0.5,15.0,2.193,Wheat,0,-1.392,-2.021,0,0.1462,30.0,0.2,0.2783,1,0,526,High
+615,Punjab,41,0.5,5.687,5.125,Sugarcane,0,10.127,-0.658,1,0.9012,11.3736,1.0,0.0,0,0,479,High
+616,Punjab,47,4.351,7.088,0.803,Cotton,0,-14.901,-1.525,0,0.1132,1.6289,0.5,7.4507,0,0,490,High
+617,Maharashtra,29,0.5,6.566,2.946,Wheat,0,-8.724,0.877,1,0.4487,13.1316,0.2,1.7448,1,0,472,High
+618,Maharashtra,33,1.616,15.0,4.014,Wheat,0,7.333,-0.884,0,0.2676,9.2795,0.2,0.0,1,0,550,High
+619,Punjab,59,2.218,15.0,1.61,Sugarcane,0,-13.995,-1.386,3,0.1073,6.764,1.0,13.995,0,0,390,High
+620,Punjab,37,0.5,15.0,7.885,Rice,0,-31.813,-0.27,0,0.5257,30.0,1.0,31.8134,0,0,416,High
+621,Maharashtra,32,10.844,15.0,2.068,Millets,0,20.0,-0.94,2,0.1379,1.3832,0.0,0.0,1,0,523,High
+622,Maharashtra,51,0.769,4.747,1.478,Wheat,1,-19.046,1.184,2,0.3113,6.173,0.2,3.8093,1,1,300,High
+623,Maharashtra,59,1.088,15.0,2.027,Cotton,0,-9.619,-1.035,3,0.1351,13.7858,0.5,4.8093,1,0,389,High
+624,Maharashtra,54,1.087,15.0,6.268,Rice,0,5.34,-1.699,0,0.4179,13.7998,1.0,0.0,1,0,496,High
+625,Punjab,57,0.786,15.0,3.691,Wheat,0,-21.127,-2.453,1,0.2461,19.0908,0.2,4.2255,0,0,434,High
+626,Maharashtra,36,0.5,15.0,4.741,Millets,0,-27.118,-0.331,0,0.316,30.0,0.0,0.0,1,1,483,High
+627,Maharashtra,28,3.921,3.237,0.658,Cotton,0,-8.194,-0.862,0,0.2031,0.8257,0.5,4.097,1,0,483,High
+628,Punjab,63,10.22,8.008,0.758,Sugarcane,0,-38.896,-1.729,0,0.0947,0.7836,1.0,38.8961,0,1,386,High
+629,Maharashtra,41,0.762,6.765,3.844,Rice,0,-16.319,0.889,0,0.5682,8.8786,1.0,16.3186,1,1,456,High
+630,Maharashtra,42,3.263,15.0,6.625,Wheat,0,-8.324,0.367,3,0.4417,4.5964,0.2,1.6649,1,1,384,High
+631,Punjab,61,4.063,5.259,7.139,Millets,0,-16.049,-0.328,0,1.3576,1.2942,0.0,0.0,0,1,438,High
+632,Punjab,52,5.66,15.0,1.15,Sugarcane,0,-35.152,1.064,1,0.0767,2.6502,1.0,35.1518,0,0,448,High
+633,Maharashtra,50,1.648,15.0,4.538,Rice,0,11.913,0.363,1,0.3025,9.103,1.0,0.0,1,0,528,High
+634,Maharashtra,70,1.083,15.0,7.349,Wheat,0,-10.447,-1.123,1,0.4899,13.8554,0.2,2.0894,1,0,411,High
+635,Maharashtra,37,1.18,15.0,3.41,Rice,0,9.467,-1.339,1,0.2273,12.7071,1.0,0.0,1,0,509,High
+636,Maharashtra,56,3.741,5.527,6.722,Cotton,0,-15.901,0.518,0,1.2161,1.4775,0.5,7.9503,1,0,420,High
+637,Maharashtra,46,5.091,6.187,1.503,Sugarcane,0,-25.2,0.096,1,0.2429,1.2153,1.0,25.2002,1,0,388,High
+638,Maharashtra,48,0.601,15.0,7.515,Sugarcane,1,-7.962,-0.13,0,0.501,24.9475,1.0,7.9623,1,1,301,High
+639,Maharashtra,63,1.391,15.0,6.136,Rice,0,-15.237,-1.146,2,0.409,10.7868,1.0,15.2368,1,1,352,High
+640,Maharashtra,51,6.967,11.808,7.56,Millets,0,-29.666,0.308,1,0.6402,1.695,0.0,0.0,1,1,393,High
+641,Maharashtra,51,4.93,8.72,5.575,Wheat,0,10.047,-1.055,3,0.6393,1.7687,0.2,0.0,1,1,371,High
+642,Maharashtra,58,12.571,2.409,3.226,Cotton,0,-21.334,-0.402,0,1.3388,0.1917,0.5,10.667,1,0,413,High
+643,Maharashtra,41,4.21,5.172,7.536,Wheat,0,-39.604,1.147,0,1.4571,1.2285,0.2,7.9208,1,1,379,High
+644,Maharashtra,35,1.499,15.0,2.567,Wheat,0,7.922,-0.961,0,0.1711,10.009,0.2,0.0,1,0,564,High
+645,Maharashtra,32,2.679,4.374,2.991,Wheat,0,-19.564,-1.873,0,0.6839,1.6323,0.2,3.9129,1,1,417,High
+646,Punjab,25,3.245,14.176,5.756,Wheat,0,-21.117,-1.353,0,0.406,4.3682,0.2,4.2235,0,0,484,High
+647,Maharashtra,70,0.5,15.0,6.24,Wheat,0,-38.481,0.694,1,0.416,30.0,0.2,7.6962,1,1,392,High
+648,Punjab,32,2.359,4.464,7.471,Rice,1,14.114,-0.045,0,1.6735,1.8923,1.0,0.0,0,0,335,High
+649,Punjab,42,3.707,14.122,2.036,Sugarcane,0,-31.342,-0.429,1,0.1442,3.8101,1.0,31.3422,0,0,423,High
+650,Maharashtra,42,0.549,15.0,6.485,Wheat,0,-11.123,-0.809,2,0.4323,27.3064,0.2,2.2247,1,0,401,High
+651,Maharashtra,49,4.551,12.305,6.039,Rice,0,1.689,0.185,1,0.4908,2.7038,1.0,0.0,1,0,474,High
+652,Maharashtra,55,2.63,15.0,0.978,Sugarcane,0,-5.674,-1.149,0,0.0652,5.7044,1.0,5.6742,1,0,527,High
+653,Maharashtra,55,9.124,3.206,3.404,Millets,0,-12.895,0.006,0,1.0617,0.3514,0.0,0.0,1,1,458,High
+654,Punjab,38,0.5,15.0,2.621,Millets,0,-37.525,-1.031,0,0.1747,30.0,0.0,0.0,0,0,498,High
+655,Maharashtra,43,0.925,10.61,2.784,Cotton,0,-24.146,-1.198,0,0.2624,11.47,0.5,12.0729,1,1,440,High
+656,Maharashtra,48,2.761,15.0,7.872,Rice,0,-9.183,-0.937,1,0.5248,5.4337,1.0,9.183,1,1,410,High
+657,Maharashtra,42,1.055,15.0,5.321,Wheat,1,-23.808,-0.675,0,0.3547,14.2137,0.2,4.7615,1,1,300,High
+658,Maharashtra,64,3.76,6.777,5.883,Rice,0,-8.95,-0.349,2,0.8681,1.8024,1.0,8.9496,1,1,347,High
+659,Punjab,51,4.269,14.632,7.825,Millets,0,-1.774,-0.065,1,0.5348,3.4275,0.0,0.0,0,0,485,High
+660,Maharashtra,54,0.5,2.756,3.329,Rice,1,2.438,-3.0,1,1.208,5.5114,1.0,0.0,1,1,300,High
+661,Maharashtra,53,2.164,15.0,6.514,Cotton,0,-30.393,-1.108,0,0.4343,6.9324,0.5,15.1963,1,1,403,High
+662,Punjab,59,2.133,15.0,3.759,Cotton,0,20.0,-0.692,0,0.2506,7.0315,0.5,0.0,0,0,599,High
+663,Punjab,47,1.383,15.0,7.022,Cotton,0,-0.488,1.205,1,0.4681,10.8485,0.5,0.2439,0,0,523,High
+664,Maharashtra,29,8.106,15.0,1.855,Sugarcane,0,-12.026,-1.38,0,0.1237,1.8505,1.0,12.0257,1,0,508,High
+665,Maharashtra,50,5.272,15.0,7.608,Wheat,0,4.343,-0.137,0,0.5072,2.8451,0.2,0.0,1,0,514,High
+666,Maharashtra,37,3.584,13.715,2.141,Rice,0,20.0,1.195,0,0.1561,3.8265,1.0,0.0,1,0,627,Medium
+667,Maharashtra,39,3.759,6.02,2.947,Cotton,0,-33.581,-1.613,0,0.4894,1.6017,0.5,16.7904,1,1,378,High
+668,Maharashtra,52,1.858,6.891,6.171,Wheat,0,-15.379,0.094,1,0.8956,3.7086,0.2,3.0758,1,0,395,High
+669,Maharashtra,60,7.615,15.0,3.453,Rice,0,-47.676,-1.102,0,0.2302,1.9699,1.0,47.6765,1,1,345,High
+670,Maharashtra,45,3.558,15.0,5.062,Wheat,0,-9.11,-0.805,1,0.3374,4.2154,0.2,1.822,1,0,456,High
+671,Maharashtra,49,3.917,10.07,3.834,Wheat,0,-6.519,-1.103,0,0.3808,2.5709,0.2,1.3038,1,0,481,High
+672,Punjab,65,5.935,12.639,6.068,Sugarcane,0,-12.587,-1.032,1,0.4801,2.1297,1.0,12.5874,0,1,423,High
+673,Maharashtra,48,0.961,5.908,2.212,Millets,0,-21.818,2.0,2,0.3745,6.1483,0.0,0.0,1,1,420,High
+674,Maharashtra,70,4.936,15.0,0.938,Millets,0,-42.743,0.544,2,0.0626,3.0387,0.0,0.0,1,1,404,High
+675,Maharashtra,52,0.652,15.0,2.749,Millets,0,-6.397,0.802,0,0.1833,23.019,0.0,0.0,1,0,559,High
+676,Punjab,25,1.271,15.0,4.057,Rice,0,9.832,0.192,0,0.2705,11.8024,1.0,0.0,0,0,609,Medium
+677,Maharashtra,54,1.227,6.034,1.76,Rice,0,-20.865,-1.906,0,0.2917,4.9187,1.0,20.8653,1,0,396,High
+678,Maharashtra,50,11.523,9.003,3.161,Cotton,0,-5.399,0.095,1,0.3511,0.7814,0.5,2.6996,1,0,471,High
+679,Punjab,60,5.825,14.66,3.5,Millets,0,-35.522,0.773,1,0.2387,2.5168,0.0,0.0,0,0,471,High
+680,Punjab,53,3.624,15.0,0.93,Millets,0,-5.142,-1.153,1,0.062,4.1395,0.0,0.0,0,0,529,High
+681,Punjab,43,3.895,10.252,4.875,Millets,0,-17.901,0.665,0,0.4755,2.6318,0.0,0.0,0,0,515,High
+682,Maharashtra,30,2.477,6.06,7.133,Rice,1,-23.436,0.404,1,1.1771,2.4462,1.0,23.4363,1,1,300,High
+683,Maharashtra,42,8.818,15.0,1.638,Millets,0,-23.405,1.552,0,0.1092,1.7012,0.0,0.0,1,0,557,High
+684,Punjab,35,2.057,15.0,4.984,Cotton,0,-17.904,-0.718,1,0.3322,7.2919,0.5,8.9521,0,0,461,High
+685,Punjab,38,1.437,5.853,5.486,Rice,0,-6.264,-1.398,0,0.9372,4.0743,1.0,6.2635,0,0,458,High
+686,Maharashtra,52,0.534,15.0,3.645,Cotton,1,-31.682,-0.662,1,0.243,28.0941,0.5,15.8412,1,1,300,High
+687,Maharashtra,65,4.628,15.0,5.758,Rice,0,-17.957,1.136,0,0.3839,3.241,1.0,17.9567,1,0,471,High
+688,Maharashtra,50,3.978,13.564,3.58,Wheat,0,-12.622,-2.128,0,0.264,3.41,0.2,2.5244,1,1,468,High
+689,Punjab,31,1.455,12.983,4.285,Cotton,0,-28.486,-2.368,1,0.33,8.9218,0.5,14.2428,0,1,396,High
+690,Maharashtra,50,0.5,15.0,0.555,Wheat,0,14.95,-0.211,0,0.037,30.0,0.2,0.0,1,0,611,Medium
+691,Maharashtra,59,5.479,6.114,5.688,Wheat,0,-24.802,-0.743,1,0.9303,1.116,0.2,4.9605,1,0,356,High
+692,Maharashtra,38,0.5,12.479,5.46,Millets,0,-22.042,-0.292,1,0.4376,24.9574,0.0,0.0,1,0,430,High
+693,Maharashtra,43,3.649,2.177,0.788,Wheat,1,-26.675,0.645,1,0.3619,0.5967,0.2,5.3349,1,1,300,High
+694,Maharashtra,45,10.545,15.0,3.262,Cotton,1,14.977,-0.853,0,0.2175,1.4224,0.5,0.0,1,1,399,High
+695,Maharashtra,36,1.727,3.493,7.843,Sugarcane,1,-3.706,-2.173,0,2.2452,2.0227,1.0,3.7064,1,1,300,High
+696,Punjab,53,0.804,15.0,3.656,Millets,0,-26.049,1.744,1,0.2437,18.6675,0.0,0.0,0,1,513,High
+697,Maharashtra,44,3.661,9.355,4.265,Rice,0,4.593,-1.297,0,0.4559,2.5553,1.0,0.0,1,1,496,High
+698,Maharashtra,50,5.097,10.285,7.324,Wheat,0,-8.99,-1.359,1,0.7122,2.0178,0.2,1.7979,1,1,395,High
+699,Maharashtra,34,2.372,7.373,5.879,Cotton,0,-8.141,0.804,0,0.7974,3.1085,0.5,4.0706,1,0,478,High
+700,Maharashtra,40,10.237,9.675,4.866,Rice,0,-17.167,-0.684,0,0.5029,0.9451,1.0,17.1672,1,1,437,High
+701,Maharashtra,54,0.5,15.0,6.484,Wheat,0,-23.241,-0.363,0,0.4322,30.0,0.2,4.6482,1,0,456,High
+702,Maharashtra,41,2.108,1.665,6.983,Millets,0,-6.934,0.183,0,4.1941,0.7897,0.0,0.0,1,0,420,High
+703,Maharashtra,66,3.802,15.0,3.927,Rice,0,-20.176,0.784,0,0.2618,3.9449,1.0,20.1763,1,1,474,High
+704,Maharashtra,40,5.955,15.0,4.063,Sugarcane,0,15.251,-0.633,1,0.2709,2.5187,1.0,0.0,1,0,529,High
+705,Maharashtra,48,0.593,4.638,3.395,Millets,0,-31.039,1.22,1,0.732,7.8269,0.0,0.0,1,0,407,High
+706,Punjab,56,4.828,5.154,7.75,Millets,0,11.042,-2.436,0,1.5038,1.0674,0.0,0.0,0,0,452,High
+707,Maharashtra,28,2.414,15.0,5.729,Cotton,0,-13.67,0.46,3,0.3819,6.2147,0.5,6.8349,1,1,382,High
+708,Punjab,25,1.515,5.718,1.124,Millets,0,-7.514,-1.14,1,0.1966,3.7734,0.0,0.0,0,0,483,High
+709,Maharashtra,27,0.5,4.842,6.972,Cotton,0,0.497,1.072,2,1.44,9.6839,0.5,0.0,1,0,404,High
+710,Punjab,60,6.014,15.0,4.104,Wheat,0,-21.454,-0.158,1,0.2736,2.4942,0.2,4.2907,0,0,472,High
+711,Punjab,38,0.513,15.0,1.016,Rice,0,0.964,-0.973,1,0.0677,29.2205,1.0,0.0,0,0,556,High
+712,Maharashtra,70,1.112,5.753,4.623,Rice,0,-52.613,0.363,0,0.8036,5.172,1.0,52.6132,1,1,300,High
+713,Maharashtra,38,1.028,14.488,3.629,Sugarcane,0,-22.8,0.593,0,0.2505,14.0977,1.0,22.7998,1,0,475,High
+714,Maharashtra,47,6.153,5.362,7.083,Cotton,0,-25.861,0.342,0,1.3209,0.8714,0.5,12.9305,1,0,389,High
+715,Maharashtra,64,3.351,15.0,2.03,Cotton,0,-30.803,-1.33,1,0.1353,4.4767,0.5,15.4013,1,0,399,High
+716,Punjab,69,4.779,15.0,6.698,Rice,1,-15.224,-0.784,0,0.4466,3.1388,1.0,15.2243,0,1,300,High
+717,Punjab,70,0.865,15.0,4.683,Wheat,0,-6.214,-0.304,1,0.3122,17.3427,0.2,1.2428,0,0,494,High
+718,Punjab,60,2.941,15.0,0.91,Rice,1,-1.46,-0.341,0,0.0606,5.0995,1.0,1.4601,0,1,410,High
+719,Maharashtra,57,2.544,15.0,7.789,Sugarcane,0,-8.387,0.45,2,0.5193,5.896,1.0,8.3868,1,0,395,High
+720,Punjab,52,5.366,12.179,4.345,Rice,0,-21.985,-2.315,0,0.3567,2.2696,1.0,21.9854,0,1,425,High
+721,Punjab,54,7.096,4.733,2.682,Rice,1,-6.465,0.034,0,0.5666,0.6669,1.0,6.4648,0,1,329,High
+722,Maharashtra,38,0.5,15.0,4.021,Cotton,0,-10.136,-0.59,0,0.268,30.0,0.5,5.0679,1,0,512,High
+723,Maharashtra,35,3.363,15.0,6.934,Wheat,0,2.041,-1.139,0,0.4623,4.4606,0.2,0.0,1,0,503,High
+724,Punjab,45,0.5,5.41,2.787,Cotton,0,-0.346,-1.042,0,0.5151,10.8194,0.5,0.173,0,0,512,High
+725,Maharashtra,43,2.388,9.123,1.945,Cotton,0,-11.543,-0.515,1,0.2132,3.8206,0.5,5.7714,1,1,449,High
+726,Punjab,40,1.017,15.0,1.468,Millets,1,-25.459,-3.0,0,0.0979,14.7422,0.0,0.0,0,1,316,High
+727,Maharashtra,53,1.099,11.668,2.734,Wheat,0,-21.794,-1.001,0,0.2343,10.6152,0.2,4.3589,1,1,463,High
+728,Punjab,56,1.306,5.552,6.35,Rice,0,-20.247,-1.176,2,1.1438,4.251,1.0,20.2468,0,1,307,High
+729,Maharashtra,46,2.913,15.0,4.037,Cotton,0,-16.729,0.233,0,0.2691,5.1494,0.5,8.3645,1,0,499,High
+730,Maharashtra,63,1.22,15.0,2.206,Millets,0,-7.979,-1.106,2,0.1471,12.2989,0.0,0.0,1,0,436,High
+731,Maharashtra,31,3.958,14.252,1.746,Cotton,0,-1.272,-0.515,3,0.1225,3.6007,0.5,0.6362,1,0,437,High
+732,Punjab,43,1.555,5.911,2.995,Rice,0,-12.767,1.242,0,0.5068,3.8005,1.0,12.7668,0,1,508,High
+733,Punjab,36,0.5,3.528,7.549,Sugarcane,0,-11.758,0.694,0,2.1396,7.0567,1.0,11.7581,0,0,441,High
+734,Maharashtra,25,4.594,14.975,3.073,Wheat,0,-9.85,0.602,0,0.2052,3.2596,0.2,1.9701,1,1,554,High
+735,Maharashtra,44,1.01,15.0,7.911,Wheat,0,-19.739,-0.8,1,0.5274,14.8573,0.2,3.9477,1,1,401,High
+736,Punjab,44,1.701,15.0,4.139,Rice,0,-25.881,0.243,1,0.2759,8.8186,1.0,25.8809,0,1,438,High
+737,Maharashtra,63,3.47,12.159,1.844,Rice,0,-38.332,-0.534,1,0.1516,3.5043,1.0,38.3316,1,0,346,High
+738,Punjab,53,1.212,7.603,7.184,Wheat,0,-22.427,-0.496,1,0.9448,6.271,0.2,4.4854,0,0,389,High
+739,Maharashtra,33,0.5,5.085,3.953,Rice,0,-4.903,0.11,0,0.7773,10.1697,1.0,4.9028,1,1,478,High
+740,Maharashtra,67,1.405,15.0,5.823,Wheat,0,-22.544,-0.897,2,0.3882,10.6774,0.2,4.5087,1,1,361,High
+741,Maharashtra,60,8.671,15.0,7.878,Wheat,1,-15.567,1.193,1,0.5252,1.7298,0.2,3.1134,1,1,300,High
+742,Maharashtra,52,3.079,3.537,0.876,Cotton,0,-5.004,-0.46,1,0.2477,1.1487,0.5,2.5018,1,0,445,High
+743,Punjab,42,3.339,15.0,1.225,Millets,1,-4.19,0.318,2,0.0817,4.4917,0.0,0.0,0,1,345,High
+744,Maharashtra,33,3.001,15.0,5.903,Sugarcane,0,-28.193,1.156,1,0.3936,4.9978,1.0,28.193,1,1,405,High
+745,Maharashtra,41,0.666,15.0,2.614,Sugarcane,0,2.956,0.732,0,0.1742,22.5342,1.0,0.0,1,0,584,High
+746,Punjab,58,1.626,4.163,1.882,Rice,0,4.91,-1.709,0,0.4521,2.5611,1.0,0.0,0,0,505,High
+747,Maharashtra,68,4.17,4.085,0.795,Rice,1,-28.948,0.145,0,0.1947,0.9796,1.0,28.9477,1,1,300,High
+748,Maharashtra,64,0.5,10.814,0.664,Wheat,0,-24.358,0.551,0,0.0614,21.6276,0.2,4.8715,1,0,500,High
+749,Punjab,39,0.977,15.0,6.309,Rice,1,5.866,0.592,3,0.4206,15.3597,1.0,0.0,0,1,300,High
+750,Punjab,32,0.97,15.0,1.508,Millets,1,-14.203,-1.413,0,0.1006,15.4605,0.0,0.0,0,0,374,High
+751,Maharashtra,47,1.697,15.0,4.008,Rice,0,-45.297,-0.957,1,0.2672,8.8391,1.0,45.2967,1,0,312,High
+752,Punjab,25,11.792,10.033,5.044,Wheat,0,-32.995,0.695,1,0.5028,0.8508,0.2,6.599,0,1,446,High
+753,Punjab,49,0.5,5.36,3.713,Cotton,0,-15.741,-1.062,0,0.6927,10.7194,0.5,7.8703,0,0,453,High
+754,Maharashtra,43,2.194,15.0,0.703,Wheat,0,-29.446,-0.6,2,0.0469,6.8375,0.2,5.8891,1,0,415,High
+755,Maharashtra,39,0.593,5.631,6.376,Rice,0,-18.201,1.276,2,1.1324,9.4948,1.0,18.2014,1,1,342,High
+756,Punjab,26,10.588,15.0,0.579,Cotton,0,-11.514,-0.348,1,0.0386,1.4167,0.5,5.757,0,1,545,High
+757,Punjab,51,0.5,9.704,6.214,Cotton,0,9.934,-1.027,0,0.6404,19.4071,0.5,0.0,0,0,522,High
+758,Maharashtra,39,1.551,15.0,7.2,Millets,0,-16.357,-1.178,0,0.48,9.6741,0.0,0.0,1,0,458,High
+759,Punjab,31,4.017,15.0,3.746,Rice,0,4.088,0.033,0,0.2498,3.7342,1.0,0.0,0,0,594,High
+760,Punjab,25,3.658,10.918,4.517,Millets,0,9.431,-1.633,0,0.4137,2.9848,0.0,0.0,0,0,546,High
+761,Maharashtra,45,1.648,8.04,4.077,Sugarcane,0,9.515,-0.547,1,0.5071,4.8779,1.0,0.0,1,0,473,High
+762,Maharashtra,66,1.276,9.275,3.419,Cotton,0,-7.729,0.033,3,0.3687,7.2665,0.5,3.8643,1,1,363,High
+763,Maharashtra,65,1.521,15.0,7.156,Sugarcane,0,-41.449,0.999,1,0.477,9.8648,1.0,41.4493,1,1,322,High
+764,Punjab,40,0.921,12.958,7.205,Rice,0,-5.088,0.368,2,0.556,14.0742,1.0,5.0876,0,0,439,High
+765,Maharashtra,38,0.689,7.576,3.019,Millets,0,3.726,-1.014,1,0.3985,10.9874,0.0,0.0,1,1,465,High
+766,Maharashtra,51,1.979,10.671,5.113,Wheat,0,-13.216,0.878,0,0.4792,5.3909,0.2,2.6432,1,1,490,High
+767,Punjab,33,0.938,12.973,7.452,Cotton,0,14.169,-0.6,2,0.5744,13.8287,0.5,0.0,0,1,471,High
+768,Punjab,49,1.017,2.219,7.715,Cotton,0,0.898,-1.536,0,3.4771,2.1823,0.5,0.0,0,0,425,High
+769,Maharashtra,54,3.208,15.0,4.317,Rice,0,-3.476,-0.472,2,0.2878,4.6759,1.0,3.4759,1,0,434,High
+770,Punjab,60,10.377,5.563,3.896,Millets,1,4.296,1.138,0,0.7003,0.5361,0.0,0.0,0,0,377,High
+771,Maharashtra,44,2.776,4.842,2.59,Rice,0,-6.999,-0.135,0,0.5349,1.7441,1.0,6.9995,1,0,473,High
+772,Punjab,43,0.5,3.159,7.692,Wheat,0,-49.392,-0.287,2,2.4351,6.3172,0.2,9.8784,0,1,300,High
+773,Maharashtra,34,0.5,2.901,1.571,Cotton,0,-19.487,1.337,1,0.5417,5.801,0.5,9.7435,1,0,433,High
+774,Maharashtra,35,9.074,4.119,4.849,Millets,0,5.958,-1.731,2,1.1771,0.454,0.0,0.0,1,1,383,High
+775,Punjab,42,1.058,6.909,2.619,Rice,0,-24.114,-1.586,2,0.3791,6.5273,1.0,24.1139,0,1,340,High
+776,Maharashtra,49,0.5,15.0,2.593,Wheat,0,-15.352,-1.916,0,0.1728,30.0,0.2,3.0704,1,0,488,High
+777,Maharashtra,56,2.742,7.999,3.537,Wheat,0,-27.673,0.073,1,0.4422,2.9175,0.2,5.5345,1,1,397,High
+778,Punjab,35,2.932,14.044,7.991,Sugarcane,0,14.759,-0.134,0,0.569,4.7904,1.0,0.0,0,0,561,High
+779,Punjab,63,3.137,15.0,0.912,Rice,0,13.222,1.243,1,0.0608,4.7821,1.0,0.0,0,0,608,Medium
+780,Punjab,42,0.933,13.529,4.839,Rice,0,-25.258,-0.057,1,0.3577,14.4934,1.0,25.258,0,0,421,High
+781,Punjab,45,0.5,11.451,1.119,Rice,1,-9.791,-0.704,0,0.0977,22.9013,1.0,9.7905,0,0,360,High
+782,Maharashtra,36,10.059,4.682,1.659,Sugarcane,0,-22.207,-0.298,1,0.3543,0.4655,1.0,22.2074,1,1,392,High
+783,Maharashtra,25,0.867,15.0,3.797,Rice,0,-2.776,-0.182,1,0.2531,17.3086,1.0,2.7758,1,0,504,High
+784,Maharashtra,56,3.381,15.0,4.328,Wheat,0,-4.355,-1.972,0,0.2886,4.4362,0.2,0.871,1,0,488,High
+785,Maharashtra,54,3.679,3.17,2.441,Rice,0,0.176,2.0,2,0.7699,0.8618,1.0,0.0,1,0,447,High
+786,Punjab,42,2.825,5.37,2.873,Millets,0,-7.021,0.22,1,0.5351,1.9009,0.0,0.0,0,0,482,High
+787,Maharashtra,46,1.126,9.739,1.152,Millets,0,20.0,-0.09,2,0.1183,8.6504,0.0,0.0,1,0,504,High
+788,Maharashtra,51,0.949,5.631,2.774,Wheat,0,-27.419,0.315,0,0.4926,5.9364,0.2,5.4838,1,1,439,High
+789,Maharashtra,26,2.731,15.0,7.749,Cotton,0,-37.659,0.409,0,0.5166,5.4916,0.5,18.8295,1,0,412,High
+790,Punjab,39,6.032,6.803,7.585,Rice,0,-17.896,-0.441,1,1.115,1.1279,1.0,17.8964,0,0,378,High
+791,Maharashtra,55,8.797,15.0,3.012,Rice,0,-35.52,-0.931,0,0.2008,1.7051,1.0,35.5198,1,0,402,High
+792,Maharashtra,30,0.5,15.0,4.781,Cotton,0,-34.387,0.299,3,0.3188,30.0,0.5,17.1935,1,1,331,High
+793,Maharashtra,49,0.505,14.441,7.322,Wheat,0,-27.02,0.086,0,0.507,28.6157,0.2,5.4039,1,1,445,High
+794,Punjab,29,8.687,15.0,0.546,Wheat,0,-25.833,-1.11,1,0.0364,1.7267,0.2,5.1666,0,1,498,High
+795,Maharashtra,51,4.002,15.0,6.58,Millets,0,-26.319,1.119,1,0.4387,3.7486,0.0,0.0,1,0,441,High
+796,Punjab,45,6.437,7.759,7.613,Cotton,0,-4.227,-0.007,0,0.9812,1.2054,0.5,2.1134,0,0,486,High
+797,Punjab,26,0.679,6.341,1.185,Wheat,0,-19.593,0.43,0,0.1869,9.3342,0.2,3.9186,0,0,524,High
+798,Maharashtra,59,2.657,12.968,1.711,Wheat,0,-14.693,1.092,1,0.1319,4.8805,0.2,2.9385,1,0,494,High
+799,Punjab,36,3.623,15.0,7.706,Wheat,0,1.413,-0.278,1,0.5137,4.1398,0.2,0.0,0,0,498,High
+800,Punjab,35,3.419,13.616,4.532,Cotton,0,-23.495,0.017,0,0.3328,3.9821,0.5,11.7474,0,0,497,High
+801,Punjab,47,1.574,3.45,0.606,Wheat,0,-15.968,0.51,2,0.1758,2.1926,0.2,3.1936,0,0,432,High
+802,Maharashtra,59,0.5,6.409,2.432,Wheat,1,-6.514,-1.025,0,0.3795,12.8185,0.2,1.3027,1,0,300,High
+803,Maharashtra,33,2.623,11.482,3.875,Sugarcane,0,-41.124,-0.246,0,0.3374,4.3768,1.0,41.1235,1,0,371,High
+804,Punjab,46,0.875,11.467,5.697,Millets,0,-47.774,-0.174,1,0.4968,13.0979,0.0,0.0,0,0,391,High
+805,Maharashtra,50,1.795,15.0,5.999,Cotton,0,-6.808,-1.256,0,0.3999,8.357,0.5,3.4041,1,0,477,High
+806,Punjab,53,9.48,4.02,7.966,Rice,0,-4.62,2.0,2,1.9817,0.424,1.0,4.6198,0,0,410,High
+807,Punjab,47,1.528,15.0,2.656,Wheat,0,18.963,-1.707,0,0.1771,9.819,0.2,0.0,0,0,595,High
+808,Punjab,41,5.494,15.0,4.537,Wheat,0,-9.629,-1.598,0,0.3024,2.7302,0.2,1.9257,0,1,519,High
+809,Maharashtra,35,0.627,15.0,4.5,Rice,1,0.296,1.014,0,0.3,23.9325,1.0,0.0,1,1,391,High
+810,Maharashtra,46,3.348,8.815,3.542,Sugarcane,1,-14.932,0.413,1,0.4018,2.6326,1.0,14.9321,1,1,300,High
+811,Punjab,32,11.318,15.0,1.769,Millets,0,-7.542,-1.326,1,0.1179,1.3254,0.0,0.0,0,0,531,High
+812,Punjab,25,0.5,15.0,7.831,Millets,0,13.009,0.307,1,0.5221,30.0,0.0,0.0,0,1,543,High
+813,Maharashtra,50,0.5,14.245,6.9,Rice,1,-43.86,1.168,0,0.4844,28.4894,1.0,43.8599,1,1,300,High
+814,Punjab,56,1.705,11.256,4.237,Millets,0,-5.796,-0.33,0,0.3765,6.6009,0.0,0.0,0,0,526,High
+815,Punjab,25,0.5,15.0,5.024,Sugarcane,0,20.0,2.0,2,0.335,30.0,1.0,0.0,0,0,578,High
+816,Maharashtra,33,4.14,1.607,2.204,Wheat,0,-7.769,-2.024,0,1.372,0.3881,0.2,1.5539,1,1,435,High
+817,Maharashtra,52,7.379,15.0,4.578,Wheat,0,12.759,0.394,2,0.3052,2.0327,0.2,0.0,1,0,493,High
+818,Maharashtra,70,1.708,13.476,5.712,Sugarcane,0,-21.438,0.867,0,0.4239,7.8921,1.0,21.4382,1,1,440,High
+819,Maharashtra,45,0.898,10.196,4.364,Wheat,0,-12.481,0.591,2,0.428,11.3532,0.2,2.4961,1,1,414,High
+820,Maharashtra,36,1.71,15.0,7.927,Cotton,0,-11.497,-0.504,0,0.5285,8.7739,0.5,5.7486,1,0,465,High
+821,Punjab,43,3.856,7.1,5.085,Wheat,0,-45.264,-0.59,1,0.7162,1.8412,0.2,9.0528,0,1,356,High
+822,Maharashtra,61,0.5,8.825,3.282,Wheat,0,-29.523,0.144,0,0.372,17.6491,0.2,5.9045,1,1,441,High
+823,Maharashtra,37,2.671,15.0,6.291,Rice,0,-28.453,0.516,0,0.4194,5.6163,1.0,28.4531,1,1,427,High
+824,Maharashtra,35,2.85,6.261,1.778,Millets,0,0.331,0.302,0,0.2841,2.1967,0.0,0.0,1,0,531,High
+825,Maharashtra,39,3.046,7.845,3.384,Wheat,0,1.082,0.74,0,0.4313,2.5758,0.2,0.0,1,0,531,High
+826,Punjab,34,0.831,10.831,3.102,Wheat,0,-7.139,-1.319,0,0.2864,13.0285,0.2,1.4279,0,0,524,High
+827,Maharashtra,46,3.756,15.0,3.739,Rice,1,0.362,-0.528,1,0.2492,3.9938,1.0,0.0,1,1,321,High
+828,Maharashtra,64,0.5,14.386,6.363,Wheat,1,-14.462,-1.24,1,0.4423,28.7716,0.2,2.8924,1,0,300,High
+829,Punjab,49,0.667,4.331,4.131,Rice,0,6.029,-0.686,1,0.9538,6.4969,1.0,0.0,0,0,467,High
+830,Maharashtra,42,6.298,7.959,6.455,Rice,0,-11.868,0.715,1,0.8111,1.2636,1.0,11.8682,1,0,412,High
+831,Maharashtra,41,4.027,7.494,4.824,Wheat,0,-11.132,-0.192,1,0.6437,1.8612,0.2,2.2264,1,0,425,High
+832,Maharashtra,26,0.5,11.453,3.066,Cotton,0,-15.984,2.0,0,0.2677,22.9053,0.5,7.9919,1,0,538,High
+833,Maharashtra,56,0.857,15.0,6.575,Wheat,1,-1.733,-0.107,0,0.4383,17.5119,0.2,0.3466,1,0,332,High
+834,Maharashtra,44,0.724,15.0,2.287,Rice,1,-18.318,-0.24,0,0.1525,20.7197,1.0,18.318,1,1,314,High
+835,Maharashtra,43,0.947,2.942,2.183,Cotton,0,2.646,-0.194,2,0.7419,3.1055,0.5,0.0,1,1,415,High
+836,Maharashtra,70,0.853,15.0,4.284,Rice,0,-28.982,-0.938,0,0.2856,17.5788,1.0,28.9822,1,1,402,High
+837,Punjab,49,0.5,14.91,2.456,Millets,0,-28.484,-0.358,0,0.1647,29.8207,0.0,0.0,0,0,525,High
+838,Maharashtra,36,1.843,15.0,0.568,Wheat,0,-1.689,0.828,1,0.0379,8.1403,0.2,0.3377,1,0,556,High
+839,Maharashtra,40,3.959,15.0,5.362,Wheat,0,-13.328,-0.186,0,0.3575,3.7893,0.2,2.6656,1,1,498,High
+840,Punjab,59,2.8,15.0,4.333,Cotton,0,-18.808,-1.093,2,0.2889,5.3575,0.5,9.4042,0,0,406,High
+841,Maharashtra,46,3.352,15.0,3.841,Rice,0,1.351,-1.225,2,0.2561,4.4747,1.0,0.0,1,0,444,High
+842,Punjab,28,0.5,15.0,7.752,Millets,0,-21.109,-0.75,0,0.5168,30.0,0.0,0.0,0,0,490,High
+843,Maharashtra,56,9.078,15.0,4.784,Cotton,0,-15.003,-0.226,0,0.3189,1.6523,0.5,7.5015,1,0,488,High
+844,Punjab,37,5.47,11.265,6.976,Cotton,0,8.988,0.894,0,0.6193,2.0595,0.5,0.0,0,0,565,High
+845,Maharashtra,67,4.906,15.0,1.546,Wheat,0,11.087,-1.64,1,0.1031,3.0577,0.2,0.0,1,1,513,High
+846,Maharashtra,58,5.195,15.0,4.769,Rice,0,11.586,-0.62,0,0.3179,2.8876,1.0,0.0,1,0,545,High
+847,Maharashtra,40,8.096,15.0,7.01,Sugarcane,0,-16.232,-0.156,1,0.4673,1.8527,1.0,16.2319,1,0,415,High
+848,Punjab,60,2.357,4.388,4.873,Wheat,0,6.461,-2.134,1,1.1105,1.8615,0.2,0.0,0,1,428,High
+849,Maharashtra,46,0.67,4.346,5.885,Cotton,1,-26.234,-1.134,0,1.354,6.4825,0.5,13.1171,1,1,300,High
+850,Maharashtra,55,2.877,15.0,3.29,Rice,0,-25.604,-0.98,0,0.2194,5.2129,1.0,25.6035,1,0,431,High
+851,Maharashtra,51,4.075,13.687,1.193,Rice,0,-47.257,-1.096,1,0.0872,3.359,1.0,47.2568,1,1,323,High
+852,Punjab,35,4.017,7.454,3.934,Millets,0,13.093,-0.108,0,0.5278,1.8554,0.0,0.0,0,0,565,High
+853,Maharashtra,37,2.211,13.637,3.459,Wheat,0,-24.96,-0.461,2,0.2536,6.1679,0.2,4.9919,1,1,395,High
+854,Punjab,57,0.5,15.0,7.003,Rice,1,-13.898,-0.855,0,0.4669,30.0,1.0,13.898,0,1,300,High
+855,Maharashtra,41,1.391,15.0,2.33,Rice,0,-25.655,0.128,1,0.1553,10.7804,1.0,25.6549,1,0,428,High
+856,Maharashtra,40,7.606,4.89,7.751,Wheat,1,-1.573,1.423,2,1.585,0.6429,0.2,0.3145,1,1,300,High
+857,Maharashtra,34,2.634,11.143,7.686,Wheat,0,-17.635,1.339,2,0.6898,4.2296,0.2,3.5269,1,1,391,High
+858,Maharashtra,50,2.319,6.398,2.357,Cotton,1,-31.572,-2.223,0,0.3684,2.7594,0.5,15.7861,1,1,300,High
+859,Maharashtra,70,0.939,8.934,3.11,Cotton,0,-5.375,0.277,0,0.3482,9.5098,0.5,2.6875,1,1,496,High
+860,Maharashtra,32,1.379,8.75,7.217,Rice,1,-33.506,-0.478,1,0.8247,6.3468,1.0,33.506,1,1,300,High
+861,Punjab,45,6.771,8.314,7.75,Wheat,0,-24.561,-0.145,0,0.9322,1.2278,0.2,4.9122,0,0,438,High
+862,Maharashtra,62,3.294,4.463,3.826,Millets,0,-43.603,-0.698,0,0.8572,1.355,0.0,0.0,1,1,372,High
+863,Maharashtra,44,4.645,11.0,2.907,Rice,0,-33.359,-0.182,2,0.2643,2.3682,1.0,33.3594,1,1,324,High
+864,Maharashtra,50,1.817,7.78,4.446,Wheat,1,-9.322,-0.149,0,0.5715,4.2809,0.2,1.8644,1,1,300,High
+865,Maharashtra,32,2.985,1.169,5.658,Cotton,0,-41.468,-0.879,0,4.8388,0.3918,0.5,20.734,1,1,305,High
+866,Maharashtra,50,0.857,15.0,2.551,Cotton,0,-34.217,-1.196,0,0.1701,17.5113,0.5,17.1086,1,0,435,High
+867,Maharashtra,43,3.667,15.0,6.024,Cotton,1,-12.036,-0.749,1,0.4016,4.0905,0.5,6.0178,1,1,300,High
+868,Maharashtra,57,1.674,9.016,0.887,Sugarcane,0,-20.154,-0.066,1,0.0984,5.3865,1.0,20.1541,1,1,418,High
+869,Punjab,59,1.753,12.367,0.87,Wheat,0,17.881,0.299,0,0.0703,7.0544,0.2,0.0,0,0,629,Medium
+870,Punjab,70,3.144,4.772,3.397,Wheat,0,15.508,0.553,0,0.712,1.5178,0.2,0.0,0,0,554,High
+871,Maharashtra,52,8.26,15.0,3.151,Cotton,0,-25.019,-0.794,0,0.2101,1.816,0.5,12.5094,1,0,466,High
+872,Maharashtra,49,0.5,15.0,2.867,Rice,1,-3.643,-0.016,1,0.1911,30.0,1.0,3.6429,1,0,328,High
+873,Punjab,47,5.225,13.809,4.183,Rice,0,-0.761,0.663,0,0.3029,2.6432,1.0,0.7608,0,0,576,High
+874,Maharashtra,41,1.038,15.0,2.906,Sugarcane,0,-20.962,0.866,2,0.1937,14.457,1.0,20.9622,1,1,414,High
+875,Maharashtra,49,1.761,14.016,1.942,Rice,1,-34.433,-0.359,0,0.1385,7.9615,1.0,34.4331,1,1,300,High
+876,Maharashtra,41,0.5,4.181,2.821,Wheat,0,-12.184,-0.521,3,0.6747,8.3626,0.2,2.4369,1,1,335,High
+877,Punjab,47,1.075,8.763,7.812,Cotton,0,-14.593,-0.109,0,0.8914,8.1554,0.5,7.2963,0,0,452,High
+878,Punjab,61,2.095,13.539,3.583,Cotton,0,-18.536,-1.45,1,0.2647,6.4615,0.5,9.2682,0,1,439,High
+879,Maharashtra,33,0.5,8.469,3.835,Wheat,0,-5.365,-0.494,0,0.4529,16.9371,0.2,1.073,1,1,495,High
+880,Maharashtra,59,2.026,10.227,4.824,Cotton,0,-23.261,0.445,0,0.4717,5.0469,0.5,11.6305,1,0,441,High
+881,Maharashtra,61,7.856,11.324,7.875,Rice,1,-29.28,0.342,2,0.6955,1.4413,1.0,29.28,1,1,300,High
+882,Punjab,44,0.5,11.419,2.312,Cotton,0,-10.364,-0.57,2,0.2025,22.8384,0.5,5.1819,0,0,452,High
+883,Maharashtra,25,4.337,12.703,5.61,Wheat,0,-6.21,-0.082,0,0.4416,2.9286,0.2,1.2421,1,0,510,High
+884,Maharashtra,38,1.581,5.215,5.87,Wheat,0,-29.65,-1.125,1,1.1256,3.298,0.2,5.9299,1,0,336,High
+885,Punjab,61,3.194,6.865,5.055,Cotton,0,1.921,0.403,1,0.7364,2.1491,0.5,0.0,0,0,480,High
+886,Maharashtra,45,0.904,4.359,1.164,Wheat,1,-38.557,-0.378,2,0.2671,4.8209,0.2,7.7114,1,1,300,High
+887,Punjab,33,0.521,15.0,5.577,Rice,1,-20.672,-0.576,0,0.3718,28.7771,1.0,20.6717,0,0,302,High
+888,Maharashtra,39,0.525,2.26,0.981,Rice,0,-15.87,-0.662,1,0.4339,4.3076,1.0,15.8701,1,0,392,High
+889,Maharashtra,55,0.5,15.0,4.971,Rice,0,-21.543,-0.363,0,0.3314,30.0,1.0,21.5434,1,1,446,High
+890,Maharashtra,52,0.616,15.0,6.047,Millets,0,-20.476,-0.748,0,0.4031,24.3474,0.0,0.0,1,1,466,High
+891,Punjab,42,3.274,9.566,2.082,Rice,0,16.066,0.358,2,0.2177,2.922,1.0,0.0,0,0,525,High
+892,Punjab,41,6.403,1.518,6.812,Rice,0,-24.763,-0.695,0,4.4872,0.2371,1.0,24.7633,0,0,357,High
+893,Punjab,40,5.048,9.124,0.728,Cotton,0,20.0,1.293,0,0.0797,1.8075,0.5,0.0,0,0,647,Medium
+894,Maharashtra,34,3.467,11.573,6.829,Rice,0,-5.443,1.165,1,0.5901,3.3381,1.0,5.4431,1,1,463,High
+895,Punjab,64,0.5,5.428,3.952,Wheat,0,-10.619,-1.192,1,0.7282,10.8555,0.2,2.1238,0,1,421,High
+896,Maharashtra,41,1.024,8.067,2.132,Rice,0,-4.856,-0.23,0,0.2643,7.878,1.0,4.8564,1,0,504,High
+897,Maharashtra,60,1.112,15.0,1.503,Rice,0,20.0,-0.951,0,0.1002,13.4849,1.0,0.0,1,1,588,High
+898,Punjab,63,0.863,15.0,7.65,Wheat,0,-26.904,-0.627,2,0.51,17.3838,0.2,5.3808,0,1,369,High
+899,Maharashtra,57,2.171,5.814,1.045,Cotton,0,-28.806,1.702,0,0.1797,2.6784,0.5,14.4031,1,0,464,High
+900,Maharashtra,40,2.301,3.394,1.993,Sugarcane,0,-23.323,-0.608,1,0.5873,1.4751,1.0,23.323,1,0,360,High
+901,Maharashtra,50,1.324,9.378,3.109,Wheat,1,-11.518,0.06,1,0.3315,7.0816,0.2,2.3036,1,1,300,High
+902,Maharashtra,45,1.31,10.27,5.152,Cotton,0,-37.887,0.241,0,0.5016,7.8404,0.5,18.9433,1,0,398,High
+903,Maharashtra,34,5.29,15.0,4.586,Wheat,0,-11.665,-1.301,2,0.3057,2.8353,0.2,2.333,1,1,412,High
+904,Maharashtra,49,4.672,7.939,1.442,Rice,0,13.403,-1.155,1,0.1817,1.6993,1.0,0.0,1,0,497,High
+905,Maharashtra,31,1.104,15.0,2.673,Wheat,0,-9.729,-1.506,0,0.1782,13.5916,0.2,1.9458,1,0,514,High
+906,Punjab,59,7.353,15.0,2.291,Cotton,1,-16.39,-1.186,1,0.1527,2.04,0.5,8.1951,0,1,300,High
+907,Punjab,39,9.161,15.0,2.102,Cotton,0,-21.272,-3.0,1,0.1401,1.6374,0.5,10.6362,0,1,442,High
+908,Maharashtra,47,1.187,11.723,3.584,Wheat,0,1.478,0.245,1,0.3057,9.8769,0.2,0.0,1,1,497,High
+909,Maharashtra,48,1.313,10.522,5.78,Millets,1,-24.387,1.754,2,0.5493,8.0148,0.0,0.0,1,1,300,High
+910,Maharashtra,42,2.113,9.158,2.131,Millets,0,15.696,-0.152,0,0.2327,4.3334,0.0,0.0,1,0,564,High
+911,Maharashtra,52,8.497,15.0,3.25,Wheat,0,17.0,-1.716,0,0.2167,1.7654,0.2,0.0,1,0,558,High
+912,Maharashtra,39,6.27,7.375,3.432,Cotton,0,-8.149,-0.943,0,0.4653,1.1763,0.5,4.0744,1,0,472,High
+913,Punjab,55,0.5,12.631,7.823,Rice,0,-12.173,-0.759,0,0.6194,25.2626,1.0,12.1727,0,0,459,High
+914,Maharashtra,29,1.496,12.692,7.66,Cotton,0,3.46,-1.318,0,0.6035,8.4819,0.5,0.0,1,1,484,High
+915,Maharashtra,47,2.418,8.155,6.501,Wheat,0,4.422,0.56,2,0.7972,3.3723,0.2,0.0,1,0,418,High
+916,Maharashtra,68,2.098,4.063,2.312,Cotton,0,-24.09,0.51,1,0.5691,1.9365,0.5,12.0449,1,0,385,High
+917,Maharashtra,33,0.647,15.0,1.01,Millets,1,-19.898,-1.455,1,0.0674,23.2017,0.0,0.0,1,1,300,High
+918,Punjab,37,5.873,11.602,2.877,Cotton,0,-31.835,0.425,1,0.248,1.9755,0.5,15.9176,0,1,447,High
+919,Punjab,51,3.39,15.0,2.501,Cotton,0,-5.864,1.227,2,0.1667,4.4251,0.5,2.932,0,0,512,High
+920,Maharashtra,47,5.459,7.95,4.294,Cotton,0,-2.001,1.603,0,0.5402,1.4563,0.5,1.0003,1,1,528,High
+921,Punjab,49,5.872,15.0,6.333,Sugarcane,0,-32.249,2.0,2,0.4222,2.5545,1.0,32.2489,0,0,384,High
+922,Maharashtra,51,4.139,6.041,6.937,Wheat,0,-18.828,-1.202,0,1.1484,1.4595,0.2,3.7656,1,1,392,High
+923,Punjab,53,1.735,15.0,6.836,Wheat,1,-32.209,2.0,2,0.4558,8.6444,0.2,6.4418,0,1,300,High
+924,Punjab,58,2.829,3.01,3.198,Cotton,0,-25.482,-0.654,1,1.0624,1.0639,0.5,12.741,0,1,378,High
+925,Maharashtra,50,0.524,15.0,5.654,Sugarcane,0,-3.735,-0.892,0,0.377,28.606,1.0,3.7347,1,0,498,High
+926,Maharashtra,42,3.359,15.0,2.846,Rice,0,-23.002,0.027,1,0.1898,4.4652,1.0,23.0022,1,0,431,High
+927,Maharashtra,53,0.593,6.447,0.957,Millets,0,-16.512,-0.882,1,0.1484,10.8781,0.0,0.0,1,0,432,High
+928,Maharashtra,68,3.552,15.0,5.762,Wheat,1,-36.41,0.099,0,0.3841,4.223,0.2,7.2819,1,1,300,High
+929,Punjab,43,0.783,15.0,0.574,Cotton,0,-9.195,-1.62,1,0.0382,19.1686,0.5,4.5976,0,0,514,High
+930,Maharashtra,33,0.5,2.068,3.569,Wheat,1,-12.664,-0.539,0,1.7258,4.1358,0.2,2.5328,1,1,300,High
+931,Maharashtra,58,0.542,15.0,2.97,Wheat,0,-22.934,0.986,0,0.198,27.6756,0.2,4.5869,1,0,516,High
+932,Punjab,44,0.5,1.658,5.7,Rice,0,-14.835,-0.357,2,3.4389,3.3152,1.0,14.8352,0,0,327,High
+933,Punjab,25,3.805,5.626,6.247,Millets,0,-9.296,-0.158,1,1.1103,1.4785,0.0,0.0,0,0,445,High
+934,Maharashtra,55,3.955,15.0,2.781,Cotton,0,-24.223,0.175,2,0.1854,3.7929,0.5,12.1115,1,0,403,High
+935,Maharashtra,39,2.17,7.649,6.587,Cotton,0,-34.144,0.871,1,0.8612,3.5242,0.5,17.072,1,1,353,High
+936,Maharashtra,44,0.528,4.763,2.704,Rice,0,15.084,-1.955,0,0.5678,9.0222,1.0,0.0,1,0,495,High
+937,Maharashtra,49,0.5,10.398,3.847,Cotton,0,-21.499,0.285,1,0.37,20.7968,0.5,10.7495,1,0,422,High
+938,Punjab,47,0.5,15.0,5.953,Wheat,0,-13.451,-0.568,1,0.3968,30.0,0.2,2.6903,0,0,473,High
+939,Maharashtra,54,0.544,2.814,5.697,Cotton,0,3.163,-0.133,3,2.0244,5.1763,0.5,0.0,1,0,331,High
+940,Maharashtra,40,6.645,5.254,7.905,Wheat,0,20.0,0.76,0,1.5047,0.7906,0.2,0.0,1,0,511,High
+941,Maharashtra,51,0.829,15.0,2.189,Cotton,1,-30.667,-1.494,2,0.1459,18.0857,0.5,15.3336,1,1,300,High
+942,Maharashtra,42,1.312,15.0,5.467,Millets,0,-5.253,-1.313,2,0.3645,11.4318,0.0,0.0,1,0,414,High
+943,Punjab,54,0.5,11.215,5.073,Wheat,0,-22.613,-2.235,1,0.4523,22.4301,0.2,4.5227,0,1,401,High
+944,Punjab,52,0.754,4.521,7.363,Sugarcane,0,-20.082,-0.849,2,1.6286,5.9924,1.0,20.0819,0,1,300,High
+945,Maharashtra,34,2.197,15.0,1.673,Rice,0,-22.803,1.03,0,0.1115,6.8273,1.0,22.8032,1,1,507,High
+946,Maharashtra,58,2.808,15.0,6.138,Cotton,0,-28.885,-0.421,3,0.4092,5.3427,0.5,14.4424,1,1,300,High
+947,Maharashtra,39,0.845,15.0,4.762,Cotton,0,-12.619,-1.64,1,0.3175,17.7617,0.5,6.3097,1,1,432,High
+948,Punjab,55,0.5,6.36,2.56,Wheat,0,-34.118,-0.786,0,0.4025,12.7206,0.2,6.8235,0,0,437,High
+949,Punjab,49,1.488,11.714,4.058,Millets,1,2.202,-1.788,2,0.3464,7.8724,0.0,0.0,0,1,300,High
+950,Maharashtra,67,0.801,10.716,3.003,Rice,0,-11.175,-1.425,0,0.2803,13.386,1.0,11.175,1,0,451,High
+951,Punjab,42,0.744,15.0,4.845,Rice,0,-5.663,0.089,0,0.323,20.1703,1.0,5.6635,0,0,549,High
+952,Punjab,41,9.646,15.0,7.168,Millets,0,-18.993,-1.474,1,0.4778,1.555,0.0,0.0,0,1,441,High
+953,Maharashtra,45,2.874,15.0,4.77,Rice,0,-17.508,0.813,0,0.318,5.2198,1.0,17.5081,1,0,486,High
+954,Punjab,41,2.442,5.145,5.238,Wheat,0,-7.687,-0.371,0,1.018,2.1073,0.2,1.5373,0,0,480,High
+955,Maharashtra,55,1.612,15.0,4.391,Wheat,0,-29.461,-0.214,2,0.2927,9.3029,0.2,5.8922,1,1,378,High
+956,Maharashtra,26,1.681,15.0,0.884,Sugarcane,0,-17.37,-0.578,0,0.0589,8.9243,1.0,17.3699,1,0,509,High
+957,Maharashtra,32,2.335,15.0,2.246,Rice,0,-24.894,0.473,0,0.1498,6.4226,1.0,24.8942,1,0,484,High
+958,Maharashtra,32,3.576,15.0,4.19,Rice,0,-25.359,0.483,0,0.2793,4.1941,1.0,25.3585,1,1,463,High
+959,Maharashtra,56,3.633,5.006,2.827,Wheat,0,-43.648,0.035,0,0.5646,1.3779,0.2,8.7295,1,0,389,High
+960,Punjab,66,0.565,11.838,1.863,Millets,0,8.09,0.265,1,0.1574,20.9484,0.0,0.0,0,0,553,High
+961,Maharashtra,44,2.08,7.963,4.281,Cotton,0,-22.298,-0.512,1,0.5376,3.8283,0.5,11.149,1,0,385,High
+962,Punjab,43,1.593,3.475,2.921,Cotton,1,5.451,0.242,2,0.8405,2.1819,0.5,0.0,0,1,300,High
+963,Maharashtra,46,6.133,15.0,2.436,Wheat,1,4.032,-0.638,1,0.1624,2.4458,0.2,0.0,1,1,343,High
+964,Maharashtra,59,2.994,15.0,3.98,Millets,0,-31.29,-0.298,0,0.2653,5.0095,0.0,0.0,1,0,466,High
+965,Maharashtra,34,2.281,2.795,1.5,Sugarcane,0,-28.326,-0.575,2,0.5365,1.2254,1.0,28.3262,1,1,306,High
+966,Maharashtra,48,0.5,11.872,7.458,Millets,0,20.0,-0.757,1,0.6282,23.7438,0.0,0.0,1,1,481,High
+967,Punjab,57,2.791,15.0,7.008,Rice,0,14.264,-0.989,1,0.4672,5.3735,1.0,0.0,0,0,508,High
+968,Maharashtra,51,0.5,13.443,5.807,Rice,0,-12.473,0.168,1,0.432,26.8861,1.0,12.4728,1,0,433,High
+969,Maharashtra,47,1.234,11.128,1.297,Wheat,0,-8.139,1.011,2,0.1165,9.019,0.2,1.6279,1,0,468,High
+970,Maharashtra,57,0.5,4.309,6.319,Sugarcane,0,-21.463,0.871,1,1.4666,8.6173,1.0,21.4627,1,1,346,High
+971,Punjab,25,3.534,14.676,6.56,Cotton,0,-5.515,-1.557,0,0.447,4.1532,0.5,2.7574,0,0,511,High
+972,Punjab,53,0.693,6.65,6.258,Rice,0,-22.758,0.989,1,0.9411,9.5961,1.0,22.7577,0,1,390,High
+973,Punjab,37,1.492,2.113,1.355,Sugarcane,1,-42.485,-0.618,0,0.6412,1.4161,1.0,42.4853,0,1,300,High
+974,Maharashtra,25,6.781,15.0,5.329,Rice,0,-31.823,0.301,0,0.3553,2.2122,1.0,31.8233,1,0,429,High
+975,Maharashtra,51,0.686,15.0,6.757,Cotton,0,-11.684,-1.206,0,0.4505,21.8767,0.5,5.8418,1,0,458,High
+976,Maharashtra,61,2.127,12.697,4.83,Wheat,0,-28.966,-0.158,0,0.3804,5.9702,0.2,5.7932,1,1,441,High
+977,Maharashtra,43,1.63,14.633,5.348,Wheat,0,-13.009,0.014,0,0.3655,8.9748,0.2,2.6018,1,0,498,High
+978,Maharashtra,56,0.5,15.0,3.938,Sugarcane,0,-28.12,-1.785,0,0.2625,30.0,1.0,28.1199,1,1,404,High
+979,Maharashtra,64,4.722,14.577,4.465,Cotton,0,-23.735,0.308,1,0.3063,3.0874,0.5,11.8677,1,0,424,High
+980,Maharashtra,61,0.5,15.0,0.758,Wheat,1,0.797,1.046,2,0.0505,30.0,0.2,0.0,1,1,339,High
+981,Maharashtra,65,1.929,9.602,5.292,Wheat,0,-8.288,-0.475,3,0.5511,4.9784,0.2,1.6576,1,0,339,High
+982,Punjab,54,2.468,15.0,1.111,Wheat,0,-12.145,1.06,0,0.0741,6.0789,0.2,2.429,0,0,590,High
+983,Maharashtra,46,2.97,15.0,3.763,Wheat,0,1.184,-0.49,1,0.2508,5.0505,0.2,0.0,1,0,500,High
+984,Maharashtra,26,3.5,14.228,2.695,Cotton,0,-4.791,0.55,2,0.1894,4.0649,0.5,2.3954,1,0,479,High
+985,Punjab,42,0.876,4.41,4.591,Rice,0,9.698,-2.519,0,1.0409,5.0337,1.0,0.0,0,1,480,High
+986,Maharashtra,35,0.5,15.0,1.501,Rice,1,-18.898,2.0,1,0.1,30.0,1.0,18.8979,1,1,330,High
+987,Punjab,70,3.856,6.262,5.687,Rice,1,-18.445,1.195,1,0.9083,1.6239,1.0,18.4445,0,0,300,High
+988,Maharashtra,43,2.301,7.311,6.463,Wheat,0,-31.517,-0.585,0,0.884,3.1768,0.2,6.3033,1,0,387,High
+989,Maharashtra,46,5.426,6.345,1.725,Sugarcane,0,-18.435,-2.272,2,0.2719,1.1694,1.0,18.4347,1,1,327,High
+990,Punjab,52,5.168,5.191,1.282,Sugarcane,0,2.472,-0.012,2,0.247,1.0046,1.0,0.0,0,0,469,High
+991,Punjab,46,5.561,15.0,1.591,Sugarcane,0,7.5,-1.744,0,0.1061,2.6975,1.0,0.0,0,0,583,High
+992,Punjab,65,1.991,15.0,7.078,Cotton,1,2.405,0.145,0,0.4719,7.5327,0.5,0.0,0,1,363,High
+993,Maharashtra,38,1.218,15.0,4.461,Cotton,0,-38.913,0.112,0,0.2974,12.3181,0.5,19.4564,1,0,431,High
+994,Maharashtra,47,4.704,4.58,4.004,Cotton,0,4.247,-1.241,0,0.8743,0.9736,0.5,0.0,1,0,471,High
+995,Maharashtra,36,1.801,12.897,1.3,Rice,0,-39.578,0.821,1,0.1008,7.1628,1.0,39.5778,1,0,390,High
+996,Maharashtra,29,0.608,4.328,5.372,Rice,0,1.472,0.506,1,1.241,7.1194,1.0,0.0,1,0,448,High
+997,Punjab,38,5.793,6.983,7.01,Wheat,1,-17.517,0.676,0,1.0039,1.2055,0.2,3.5034,0,1,300,High
+998,Maharashtra,45,6.425,5.844,6.711,Millets,0,-1.462,-0.291,2,1.1483,0.9096,0.0,0.0,1,1,378,High
+999,Punjab,40,1.884,15.0,5.195,Rice,0,-31.987,-1.716,0,0.3464,7.9625,1.0,31.9867,0,1,409,High
+1000,Maharashtra,37,0.5,8.873,1.065,Rice,0,5.568,0.678,0,0.12,17.7461,1.0,0.0,1,0,572,High

requirements.txt

@@ -1,3 +1,9 @@
-altair
+streamlit
+google-genai
+scikit-learn
 pandas
-streamlit
+numpy
+joblib
+matplotlib
+plotly
+fpdf2>=2.7.0

src/streamlit_app.py

@@ -1,40 +0,0 @@
-import altair as alt
-import numpy as np
-import pandas as pd
-import streamlit as st
-
-"""
-# Welcome to Streamlit!
-
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
-
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
-
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))