Upload folder using huggingface_hub

CMD.bash

@@ -0,0 +1,14 @@
+
+# 2. Virtual environment (recommended)
+python -m venv venv
+source venv/bin/activate  # Linux/Mac
+# venv\Scripts\activate   # Windows
+
+# 3. Install dependencies
+pip install -r requirements.txt
+
+# 4. Model load karo (pehli baar thoda time lagega)
+python -c "from model import model_instance; print('Model ready')"
+
+# 5. Server start karo
+uvicorn app:app --reload --host 0.0.0.0 --port 8000

app.py

@@ -0,0 +1,69 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel, Field
+from typing import List
+from model import model_instance
+import time
+import logging
+
+# Logging setup
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+app = FastAPI(
+    title="Sentence Embedding API",
+    description="Aapke trained model se text embedding nikaalne ka API",
+    version="1.0.0"
+)
+
+# Request body ka structure
+class TextInput(BaseModel):
+    text: str = Field(..., min_length=1, max_length=512, example="Mera naam Bahadur hai")
+    
+class EmbeddingResponse(BaseModel):
+    embedding: List[float]
+    input_text: str
+    inference_time_ms: float
+
+# Health check endpoint
+@app.get("/")
+def root():
+    return {"message": "API is running! Go to /docs for Swagger UI"}
+
+@app.get("/health")
+def health_check():
+    return {"status": "healthy", "model_loaded": True}
+
+# Main prediction endpoint
+@app.post("/embed", response_model=EmbeddingResponse)
+async def get_embedding(input_data: TextInput):
+    try:
+        logger.info(f"Processing text: {input_data.text[:50]}...")
+        
+        start_time = time.time()
+        embedding = model_instance.get_embedding(input_data.text)
+        inference_time = (time.time() - start_time) * 1000  # milliseconds
+        
+        return EmbeddingResponse(
+            embedding=embedding,
+            input_text=input_data.text,
+            inference_time_ms=round(inference_time, 2)
+        )
+    
+    except Exception as e:
+        logger.error(f"Error: {str(e)}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+# Batch processing (optional)
+class BatchTextInput(BaseModel):
+    texts: List[str]
+
+@app.post("/embed/batch")
+async def get_batch_embeddings(input_data: BatchTextInput):
+    results = []
+    for text in input_data.texts:
+        embedding = model_instance.get_embedding(text)
+        results.append({
+            "text": text,
+            "embedding": embedding
+        })
+    return {"results": results, "count": len(results)}

model.py

@@ -0,0 +1,18 @@
+from sentence_transformers import SentenceTransformer
+import torch
+
+class EmbeddingModel:
+    def __init__(self, model_name="embedingHF/Sentence_Transformer"):
+        # Aapka apna HF model ya koi bhi pre-trained
+        self.device = "cuda" if torch.cuda.is_available() else "cpu"
+        print(f"Loading model on {self.device}...")
+        self.model = SentenceTransformer(model_name, device=self.device)
+        print("Model loaded successfully!")
+    
+    def get_embedding(self, text: str):
+        """Convert text to vector embedding"""
+        embedding = self.model.encode(text, convert_to_tensor=True)
+        return embedding.cpu().numpy().tolist()
+
+# Global instance (ek baar load hoga, baar baar nahi)
+model_instance = EmbeddingModel()

requirements.txt

@@ -0,0 +1,6 @@
+fastapi>=0.104.1
+uvicorn>=0.24.0
+torch>=2.1.0
+transformers>=4.35.0
+sentence-transformers>=2.2.2
+pydantic>=2.4.2

test_request.py

@@ -0,0 +1,21 @@
+import requests
+import json
+
+# API ko call karo
+url = "http://localhost:8000/embed"
+
+payload = {
+    "text": "Mujhe professional AI developer banna hai!"
+}
+
+response = requests.post(url, json=payload)
+
+if response.status_code == 200:
+    result = response.json()
+    print(f"✅ Input: {result['input_text']}")
+    print(f"📊 Embedding dimension: {len(result['embedding'])}")
+    print(f"⚡ Time taken: {result['inference_time_ms']} ms")
+    print(f"🔢 First 5 values: {result['embedding'][:5]}")
+else:
+    print(f"❌ Error: {response.status_code}")
+    print(response.text)