Upload function 4 model

API.py

+from flask import Flask, request, jsonify
+import numpy as np
+import joblib
+
+app = Flask(__name__)
+
+time_model_card = joblib.load('time_model_card.pkl')
+score_model_card = joblib.load('score_model_card.pkl')
+
+# Endpoint to predict score and time for the next 3 months
+@app.route('/predict', methods=['POST'])
+def predict():
+    data = request.get_json()
+    score_data = data['score_data']
+    time_data = data['time_data']
+
+    # Convert lists to arrays
+    new_score_array = np.array([score_data])
+    new_time_array = np.array([time_data])
+
+    # Predict using the loaded models
+    new_time_predictions = time_model_card.predict(new_time_array)
+    new_score_predictions = score_model_card.predict(new_score_array)
+
+    # Prepare output
+    output = {
+        'predicted_score': new_score_predictions.tolist(),
+        'predicted_time': new_time_predictions.tolist()
+    }
+
+    return jsonify(output)
+
+# Endpoint to predict score and time for the next 3 months with steps
+@app.route('/predict_with_steps', methods=['POST'])
+def predict_with_steps():
+    data = request.get_json()
+    score_data = data['score_data']
+    time_data = data['time_data']
+    num_steps = 3
+
+    # Initialize lists
+    x_new_score = []
+    x_new_time = []
+    x_new_2_score = score_data.copy()
+    x_new_2_time = time_data.copy()
+
+    score_predictions = []
+    time_predictions = []
+
+    for _ in range(num_steps):
+        if len(x_new_score) > 0:
+            x_new_2_score = x_new_score
+            x_new_2_time = x_new_time
+
+        # Predict score
+        y_pred_score = score_model_card.predict(np.array([x_new_2_score]))
+        score_predictions.append(y_pred_score.tolist())
+
+        # Update score data
+        temp_score = np.where(y_pred_score > x_new_2_score[-1], 2, np.where(y_pred_score == x_new_2_score[-1], 1, 0))
+        x_new_2_score = np.append(x_new_2_score[1:3], x_new_2_score[3:])
+        x_new_2_score[2] = temp_score
+        x_new_2_score = np.append(x_new_2_score, y_pred_score, axis=0)
+
+        # Predict time
+        y_pred_time = time_model_card.predict(np.array([x_new_2_time]))
+        time_predictions.append(y_pred_time.tolist())
+
+        # Update time data
+        temp_time = np.where(y_pred_time > x_new_2_time[-1], 2, np.where(y_pred_time == x_new_2_time[-1], 1, 0))
+        x_new_2_time = np.append(x_new_2_time[1:3], x_new_2_time[3:])
+        x_new_2_time[2] = temp_time
+        x_new_2_time = np.append(x_new_2_time, y_pred_time, axis=0)
+
+    # Prepare output
+    output = {
+        'score_predictions': score_predictions,
+        'time_predictions': time_predictions
+    }
+
+    return jsonify(output)
+
+if __name__ == '__main__':
+    app.run(debug=True)
\ No newline at end of file