text input as json

app.py

@@ -8,7 +8,7 @@ from qwen_classifier.predict import predict_single  # Your existing function
 import torch
 from huggingface_hub import login
 from qwen_classifier.model import QwenClassifier
-import os
+from pydantic import BaseModel
 
 app = FastAPI(title="Qwen Classifier")
 
@@ -30,6 +30,11 @@ async def load_model():
     )
     print("Model loaded successfully!")
 
+
+
+class PredictionRequest(BaseModel):
+    text: str  # ← Enforces that 'text' must be a non-empty string
+
 @app.post("/predict")
-async def predict(text: str):
-    return predict_single(text, backend="local")
+async def predict(request: PredictionRequest):  # ← Validates input automatically
+    return predict_single(request.text, backend="local")

qwen_classifier/evaluate.py

@@ -43,7 +43,6 @@ def preprocessing(df):
 
 
 def evaluate_model(test_data_path):
-    # Load your test data
-    # Implement evaluation logic
-    # Return metrics like precision, recall, f1-score
+    df = load_data(test_data_path)
+    df = preprocessing(df)
     return metrics

qwen_classifier/predict.py

@@ -1,6 +1,6 @@
 import torch
 import requests
-from .config import TAG_NAMES
+from .config import TAG_NAMES, SPACE_URL
 
 # Local model setup (only load if needed)
 local_model = None
@@ -31,18 +31,22 @@ def _predict_local(text, hf_repo):
     return _process_output(logits)
 
 def _predict_hf_api(text, hf_token=None):
-    # Use your Space endpoint instead of direct model API
-    SPACE_URL = "https://keivanr-qwen-classifier-demo.hf.space"
+    
     
     try:
         response = requests.post(
             f"{SPACE_URL}/predict",
-            json={"text": text},
-            headers={"Authorization": f"Bearer {hf_token}"} if hf_token else {}
+            json={"text": text},  # This matches the Pydantic model
+            headers={
+                "Authorization": f"Bearer {hf_token}",
+                "Content-Type": "application/json"
+            } if hf_token else {"Content-Type": "application/json"},
+            timeout=10
         )
+        response.raise_for_status()  # Raise HTTP errors
         return response.json()
-    except Exception as e:
-        raise ValueError(f"Space API Error: {str(e)}")
+    except requests.exceptions.RequestException as e:
+        raise ValueError(f"API Error: {str(e)}\nResponse: {e.response.text if hasattr(e, 'response') else ''}")
 
 def _process_output(logits):
     probs = torch.sigmoid(logits)

setup.py

@@ -11,7 +11,8 @@ setup(
         'scikit-learn',
         'huggingface_hub',
         'requests',
-        'pandas'
+        'pandas',
+        'pydantic'
     ],
     entry_points={
         'console_scripts': [