Updated files

main.py

@@ -7,17 +7,17 @@ import os
 import pickle
 
 # setup
-FASTAPI_INCOME_CLASSIFICATION = os.path.abspath('.')
+SRC = os.path.abspath('./SRC/Assets')
 
 # Load the pipeline using pickle
-pipeline_path = os.path.join(FASTAPI_INCOME_CLASSIFICATION, 'rfc_pipeline.pkl')
+pipeline_path = os.path.join(SRC, 'pipeline.pkl')
 with open(pipeline_path, 'rb') as file:
-    rfc_pipeline = pickle.load(file)
+    pipeline = pickle.load(file)
 
 # Load the encoder using pickle
-encoder_path = os.path.join(FASTAPI_INCOME_CLASSIFICATION, 'encoder.pkl')
-with open(encoder_path, 'rb') as file:
-    encoder = pickle.load(file)
+model_path = os.path.join(SRC, 'rfc_model.pkl')
+with open(model_path, 'rb') as file:
+    model = pickle.load(file)
 
 app = FastAPI(
     title= 'Income Classification FastAPI',
@@ -71,21 +71,24 @@ def home():
 @app.post('/classify', response_model=IncomePredictionOutput)
 def income_classification(income: IncomePredictionInput):
     try:
-        df = pd.DataFrame([income.model_dump()])
-           
-        # Make predictions
-        prediction = rfc_pipeline.predict(df)
-        output = rfc_pipeline.predict_proba(df)
+        # Convert input data to DataFrame
+        input_df = pd.DataFrame([dict(income)])
+
+        # Preprocess the input data through the pipeline
+        input_df_transformed = pipeline.transform(input_df)
 
-        prediction_result = "Income over $50K" if prediction[0] == 1 else "Income under $50K"
-        return {"income_prediction": prediction_result, "prediction_probability": output[0][1]}
+        # Make predictions
+        prediction = model.predict(input_df_transformed)
+        probability = model.predict_proba(input_df_transformed).max(axis=1)[0]
 
+        prediction_result = "Above Limit" if prediction[0] == 1 else "Below Limit"
+        return {"income_prediction": prediction_result, "prediction_probability": probability}
 
     except Exception as e:
-        # Return error message and details if an exception occurs
         error_detail = str(e)
         raise HTTPException(status_code=500, detail=f"Error during classification: {error_detail}")
 
 
 if __name__ == '__main__':
-    uvicorn.run('main:app', reload=True)
+    uvicorn.run('main:app', reload=True)
+    

encoder.pkl → pipeline.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98baac762b1becb7d5b699f21576a27ebe7555a83826444f23d540dcfc7d01d1
-size 270
+oid sha256:dd28c23cc70beba35906a28c1b6937630ffb3bb4a7c5e8cb8276f66a33eb60e4
+size 4811

requirements.txt

@@ -1,5 +1,6 @@
-fastapi
-uvicorn
-pandas
-pydantic
-scikit-learn==1.3.0 
+fastapi==0.108.0
+uvicorn==0.25.0
+pandas==2.1.4
+pydantic==2.5.3
+pydantic_core==2.14.6
+scikit-learn==1.3.2 

rfc_pipeline.pkl → rfc_model.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de6be7713061b3b93f04b7e4c3239216dc6f51b4b18ead0b60fce19c0cbe594e
-size 223631035
+oid sha256:cd6ad029aa941d54353a585b399b77c15380a06cfa80a2c81faf9697effe0aac
+size 267723561