Update README.md

README.md

@@ -1,5 +1,6 @@
 ---
 license: gemma
+pipeline_tag: text-generation
 language:
 - en
 - km
@@ -12,8 +13,7 @@ tags:
 - unsloth
 - qlora
 base_model:
-- google/gemma-4-12B-it
-pipeline_tag: text-classification
+- unsloth/gemma-4-E4B-it
 ---
 
 # Gemma‑4 HS Code Classifier (Cambodia Customs)
@@ -29,7 +29,7 @@ Built with **[Unsloth](https://github.com/unslothai/unsloth)** for fast, memory
 Given a plain‑English product description, the model generates:
 
 ```text
-HS Code: 6105.10.00
+HS Code: 61091000
 Unit: PIECE
 Customs Duty: 25%
 Special Tax: 0%
@@ -44,12 +44,16 @@ For production, always use the included **lookup table** (`hs_code_lookup.json`)
 
 ## 🚀 Quick start (in Colab or locally)
 
+This repository contains **only the LoRA adapter**, not the full model.  
+Loading it will automatically download the base model (`unsloth/gemma-4-E4B-it`) and apply the adapter in 4-bit.
+
 ```python
 from unsloth import FastModel
 
 model, tokenizer = FastModel.from_pretrained(
-    "Sothay/gemma4-hscode-classifier",
-    load_in_4bit=True,
+    "Sothay/gemma4-hscode-classifier",   # LoRA adapter on Hugging Face
+    load_in_4bit = True,                 # required – the adapter was trained in 4-bit
+    max_seq_length = 1024,
 )
 
 # ---------- Inference with the authoritative lookup table (recommended) ----------
@@ -59,8 +63,13 @@ with open("hs_code_lookup.json") as f:
     rate_lookup = json.load(f)
 
 def predict_hs_code(description: str) -> dict:
+    system_prompt = (
+        "You are a customs compliance AI. Classify the product description to its "
+        "correct 8-digit HS code and output the corresponding trade rates (Customs Duty, "
+        "Special Tax, VAT, Excise Tax) and unit."
+    )
     messages = [
-        {"role": "system", "content": [{"type": "text", "text": "You are a customs compliance AI..."}]},
+        {"role": "system", "content": [{"type": "text", "text": system_prompt}]},
         {"role": "user",   "content": [{"type": "text", "text": f"Description: {description}"}]},
     ]
     inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to("cuda")
@@ -86,7 +95,7 @@ print(predict_hs_code("Men's cotton knitted T-shirt"))
 - **Dataset**: Custom Cambodian HS‑code dataset (`hs_code.csv`) with descriptions, codes, and official rates
   - Cleaned, deduplicated, split into 90/10 train/validation
   - Chat roles fixed to system/user/assistant (Gemma‑4 standard)
-- **Training config**: 2 epochs, effective batch size 8, learning rate 2e‑4, linear schedule, eval & save every epoch, best model loaded
+- **Training config**: 3 epochs, effective batch size 8, learning rate 2e‑4, linear schedule, eval & save every epoch, best model loaded
 - **Hardware**: Google Colab T4 (16 GB) – peak memory ~10 GB thanks to QLoRA
 - **Accuracy**: Evaluated on held‑out examples (exact HS‑code match) – see model card for current numbers
 
@@ -117,10 +126,12 @@ hf_hub_download("Sothay/gemma4-hscode-classifier", "hs_code_lookup.json")
 | File | Description |
 |------|-------------|
 | `adapter_model.safetensors` | LoRA adapter weights (few MB) |
+| `adapter_config.json` | Adapter configuration (references base model) |
 | `tokenizer.json`, `tokenizer_config.json` | Tokenizer files |
 | `hs_code_lookup.json` | Authoritative rate table for production inference |
 | `README.md` | This file |
 
+> **Note**: Only the adapter is stored here – the full Gemma‑4 base model is automatically fetched from Unsloth when you call `FastModel.from_pretrained`.  
 > If you need a **merged, full‑precision model** (for vLLM, TGI, etc.), generate it locally with Unsloth:
 > ```python
 > model.save_pretrained_merged("merged_fp16", tokenizer, save_method="merged_16bit")
@@ -130,7 +141,7 @@ hf_hub_download("Sothay/gemma4-hscode-classifier", "hs_code_lookup.json")
 
 ## 🦙 Ollama / llama.cpp (GGUF)
 
-Export a quantized GGUF directly from the LoRA adapter:
+Export a quantized GGUF directly from the loaded adapter:
 
 ```python
 model.save_pretrained_gguf("gguf_model", tokenizer, quantization_method="q4_k_m")
@@ -167,4 +178,4 @@ The HS‑code dataset and lookup table are the property of their respective owne
 ---
 
 **Author**: [Sothay](https://huggingface.co/Sothay)  
-**Model card version**: 1.0
+**Model card version**: 1.1