File size: 4,779 Bytes
46246bd 8ec303b 46246bd 8ec303b 46246bd 8ec303b 46246bd 8ec303b edfdcd4 8ec303b 46246bd 8ec303b 46246bd 8ec303b 46246bd 8ec303b 46246bd 8ec303b 46246bd 8ec303b 46246bd | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 | #!/usr/bin/env python3
"""
Gradient.com এ Lychee-GPT-9B চালানোর জন্য script
Gradient Notebook এ এটা run করুন
"""
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
import gradio as gr
import os
print("🚀 Installing dependencies...")
os.system("pip install -q torch transformers gradio accelerate safetensors")
print("📥 Loading Lychee-GPT-9B model...")
print("⏱️ এটা 5-10 মিনিট লাগতে পারে...")
MODEL_ID = "deepreinforce-ai/Ornith-1.0-9B"
try:
# Load tokenizer
tokenizer = AutoTokenizer.from_pretrained(
MODEL_ID,
trust_remote_code=True
)
# Load model
model = AutoModelForCausalLM.from_pretrained(
MODEL_ID,
torch_dtype=torch.float32,
device_map="auto", # Gradient automatically optimizes
trust_remote_code=True,
)
model.eval()
print("✅ Model loaded successfully!")
except Exception as e:
print(f"❌ Error loading model: {e}")
model = None
tokenizer = None
def generate_text(prompt, max_length=256, temperature=0.7, top_p=0.9):
"""Generate text using Lychee-GPT-9B"""
if model is None or tokenizer is None:
return "❌ Model failed to load"
try:
inputs = tokenizer(prompt, return_tensors="pt")
with torch.no_grad():
output = model.generate(
inputs["input_ids"],
max_new_tokens=int(max_length),
temperature=float(temperature),
top_p=float(top_p),
do_sample=True,
pad_token_id=tokenizer.eos_token_id,
)
response = tokenizer.decode(output[0], skip_special_tokens=True)
if prompt in response:
response = response.replace(prompt, "", 1).strip()
return response if response else "No response generated"
except Exception as e:
return f"❌ Error: {str(e)}"
# Create Gradio interface
print("\n🎨 Creating Gradio interface...")
with gr.Blocks(title="Lychee-GPT-9B", theme=gr.themes.Soft()) as demo:
gr.Markdown("""
# 🎉 Lychee-GPT-9B Demo
আপনার নিজস্ব LLM Model - Gradient এ চলছে!
⏱️ **প্রথম response একটু slow হতে পারে (model warming up)**
""")
with gr.Row():
with gr.Column(scale=1):
prompt = gr.Textbox(
label="প্রশ্ন/Prompt",
placeholder="কিছু লিখুন...",
lines=4,
info="আপনার প্রশ্ন বা prompt দিন"
)
with gr.Row():
max_len = gr.Slider(
label="Max Length",
minimum=10,
maximum=512,
value=256,
step=10,
)
with gr.Row():
temp = gr.Slider(
label="Temperature",
minimum=0.0,
maximum=1.0,
value=0.7,
step=0.1,
)
top_p = gr.Slider(
label="Top P",
minimum=0.0,
maximum=1.0,
value=0.9,
step=0.05,
)
submit_btn = gr.Button("🚀 Generate", variant="primary", size="lg")
with gr.Column(scale=1):
output = gr.Textbox(
label="Response",
lines=10,
interactive=False,
)
# Examples
gr.Examples(
examples=[
["বাংলা ভাষা সম্পর্কে বলুন"],
["পাইথন প্রোগ্রামিং কি?"],
["একটি সংক্ষিপ্ত গল্প বলুন"],
["কৃত্রিম বুদ্ধিমত্তা কি?"],
],
inputs=prompt,
label="উদাহরণ প্রশ্ন"
)
# Connect button
submit_btn.click(
fn=generate_text,
inputs=[prompt, max_len, temp, top_p],
outputs=output,
)
# Allow Enter key
prompt.submit(
fn=generate_text,
inputs=[prompt, max_len, temp, top_p],
outputs=output,
)
print("\n🌐 Launching Gradio interface...")
print("=" * 50)
print("Space URL will appear below 👇")
print("=" * 50)
# Launch
demo.launch(
server_name="0.0.0.0",
server_port=7860,
share=True,
show_error=True,
) |