Update app.py

app.py

@@ -1,93 +1,47 @@
-import os
 import gradio as gr
-from fastapi import FastAPI, HTTPException
-from fastapi.middleware.cors import CORSMiddleware
-from pydantic import BaseModel
-from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+from transformers import AutoTokenizer, AutoModelForSeq2SeqGeneration
 import torch
 
-os.environ['TRANSFORMERS_CACHE'] = '/tmp/transformers_cache'
-os.makedirs('/tmp/transformers_cache', exist_ok=True)
-
-app = FastAPI(title="DeepSeek LLM Interface")
-
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-model_name = "deepseek-ai/deepseek-llm-7b-base"
-tokenizer = AutoTokenizer.from_pretrained(model_name, cache_dir='/tmp/transformers_cache')
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    cache_dir='/tmp/transformers_cache',
-    torch_dtype=torch.float16,
-    device_map="auto"
-)
-
-def generate_response(prompt, max_length=500, temperature=0.7):
-    """Generate response using the DeepSeek model"""
-    try:
-        inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
-        outputs = model.generate(
-            **inputs,
-            max_length=max_length,
-            temperature=temperature,
-            do_sample=True,
-            pad_token_id=tokenizer.eos_token_id
-        )
-        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-        return response
-    except Exception as e:
-        print(f"Error in generate_response: {str(e)}")
-        return f"Error generating response: {str(e)}"
-
-class GenerationRequest(BaseModel):
-    prompt: str
-    max_length: int = 500
-    temperature: float = 0.7
-
-class GenerationResponse(BaseModel):
-    response: str
-
-@app.post("/generate", response_model=GenerationResponse)
-async def generate_text(request: GenerationRequest):
-    try:
-        response = generate_response(
-            request.prompt,
-            max_length=request.max_length,
-            temperature=request.temperature
-        )
-        return GenerationResponse(response=response)
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-@app.get("/health")
-async def health_check():
-    return {"status": "healthy"}
-
-def gradio_generate(prompt, max_length, temperature):
-    return generate_response(prompt, int(max_length), float(temperature))
-
-interface = gr.Interface(
-    fn=gradio_generate,
+# Initialize model and tokenizer
+model_name = "deepseek-ai/deepseek-model"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSeq2SeqGeneration.from_pretrained(model_name)
+
+def summarize_text(text, max_length=150, min_length=50):
+    """
+    Summarize the input text using the DeepSeek model
+    """
+    # Prepare the input
+    inputs = tokenizer(text, return_tensors="pt", max_length=1024, truncation=True)
+    
+    # Generate summary
+    summary_ids = model.generate(
+        inputs["input_ids"],
+        max_length=max_length,
+        min_length=min_length,
+        length_penalty=2.0,
+        num_beams=4,
+        early_stopping=True
+    )
+    
+    # Decode and return the summary
+    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
+    return summary
+
+# Create the Gradio interface
+iface = gr.Interface(
+    fn=summarize_text,
     inputs=[
-        gr.Textbox(label="Prompt", placeholder="Enter your prompt here..."),
-        gr.Slider(minimum=50, maximum=1000, value=500, step=50, label="Max Length"),
-        gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature")
+        gr.Textbox(label="Input Text", placeholder="Enter the text you want to summarize...", lines=10),
+        gr.Slider(minimum=50, maximum=300, value=150, label="Maximum Summary Length"),
+        gr.Slider(minimum=30, maximum=150, value=50, label="Minimum Summary Length")
     ],
-    outputs=gr.Textbox(label="Generated Response"),
-    title="DeepSeek LLM Interface",
-    description="Enter a prompt to generate text using DeepSeek LLM",
+    outputs=gr.Textbox(label="Summary"),
+    title="Text Summarization with DeepSeek",
+    description="Enter your text and get an AI-generated summary using the DeepSeek model.",
     examples=[
-        ["Write a short story about a mysterious garden"],
-        ["Explain quantum computing in simple terms"],
-        ["Create a recipe for chocolate chip cookies"]
+        ["The artificial intelligence revolution has transformed various sectors of the economy, from healthcare to finance. Machine learning algorithms are now capable of detecting diseases, predicting market trends, and automating complex tasks. This technological advancement has raised both excitement about the potential benefits and concerns about job displacement and ethical implications."],
     ]
 )
 
-app = gr.mount_gradio_app(app, interface, path="/")
-
+iface.launch()