Initial commit for Flask Docker Space

Dockerfile

@@ -0,0 +1,17 @@
+# Use a base image with Python and CUDA if you use torch GPU (or just python:3.10)
+FROM python:3.10-slim
+
+# Set work directory
+WORKDIR /app
+
+# Copy your code
+COPY . /app
+
+# Install dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Expose the Flask port
+EXPOSE 7860
+
+# Start the Flask app
+CMD ["python", "app.py"]

app.py

@@ -0,0 +1,159 @@
+from flask import Flask, request, jsonify
+from flask_cors import CORS
+import os
+import pandas as pd
+# from utils.analytics import get_summary_stats
+from model.mood_predict import mood_predict
+from utils.mapping import map_to_mood
+import datetime
+
+app = Flask(__name__)
+
+CORS(app) 
+
+# --- Data File ---
+DATA_DIR = 'data'
+LOGS_CSV = os.path.join(DATA_DIR, 'logs.csv')
+LOG_COLUMNS = ['date', 'exercise', 'water', 'reading', 'meditation', 'mood', 'journal_text']
+
+# --- Helper Function ---
+def init_log_file():
+    """Creates the log file with headers if it doesn't exist."""
+    if not os.path.exists("data"):
+        os.makedirs("data")
+    if not os.path.exists(LOGS_CSV):
+        df = pd.DataFrame(columns=["date", "exercise", "water", "reading", "meditation", "mood", "journal_text"])
+        df.to_csv(LOGS_CSV, index=False)
+
+
+# --- API Endpoints ---
+
+@app.route("/")
+def home():
+    return jsonify({'message': "MindTrack Backend is running!"}), 201
+
+@app.route("/log", methods=["POST"])
+def log_habit():
+    """
+    Saves a new log entry.
+    This is now a "read-modify-write" operation to handle
+    duplicate dates (overwrite) and file creation.
+    """
+    new_log_data = request.json
+    
+    if not new_log_data or 'date' not in new_log_data:
+        return jsonify({"error": "No data or date provided"}), 400
+        
+    try:
+        # 1. Ensure the data directory exists
+        os.makedirs(DATA_DIR, exist_ok=True)
+        
+        # 2. Load existing data if file exists
+        if os.path.exists(LOGS_CSV):
+            df = pd.read_csv(LOGS_CSV)
+        else:
+            # Create an empty DataFrame WITH the correct columns
+            df = pd.DataFrame(columns=LOG_COLUMNS)
+
+        # 3. Check for and remove duplicate date (for overwrite)
+        new_date = new_log_data['date']
+        if not df.empty and new_date in df['date'].values:
+            print(f"Duplicate date found: {new_date}. Overwriting old entry.")
+            df = df[df['date'] != new_date] # Keep all rows *except* the one with the duplicate date
+        
+        # 4. Create a DataFrame for the new entry, ensuring it also has all columns
+        df_entry = pd.DataFrame([new_log_data], columns=LOG_COLUMNS)
+        
+        # 5. Append new entry to the (potentially filtered) DataFrame
+        df_updated = pd.concat([df, df_entry], ignore_index=True)
+        
+        # 6. Sort by date for consistency (optional but good practice)
+        df_updated = df_updated.sort_values(by='date')
+        
+        # 7. Save the entire updated DataFrame back to the CSV
+        # header=True is the default and correct
+        df_updated.to_csv(LOGS_CSV, index=False)
+        
+        return jsonify({"message": "Log saved successfully"}), 201
+        
+    except Exception as e:
+        print(f"Error saving log: {e}")
+        return jsonify({"error": "Failed to save log"}), 500
+
+
+@app.route('/get_all_logs', methods=['GET'])
+def get_all_logs():
+    """
+    Reads all log entries from the CSV and returns them as JSON.
+    This is the "Single Source of Truth" endpoint for the dashboard.
+    """
+    if not os.path.exists(LOGS_CSV):
+        # If the file doesn't exist yet, just return an empty list
+        return jsonify([]) 
+        
+    try:
+        df = pd.read_csv(LOGS_CSV)
+        
+        # Convert DataFrame to JSON (orient='records' gives a list of dicts)
+        logs_json = df.to_dict(orient='records')
+        
+        return jsonify(logs_json)
+        
+    except Exception as e:
+        print(f"Error reading logs: {e}")
+        return jsonify({"error": "Failed to retrieve logs"}), 500
+
+
+@app.route("/predict_mood", methods=["POST"])
+def predict_mood():
+    """
+    Predicts the sentiment of a given journal text.
+    """
+    try:
+        data = request.json
+        text = data.get("text")
+        
+        if not text or text.strip() == "":
+            return jsonify({"error": "No text provided"}), 400
+        
+        pred_moods = mood_predict(text)
+        mood = pred_moods.get('label', 'Neutral')
+        score = pred_moods.get('score', 0.0)
+
+        mapped_mood = map_to_mood(mood)
+        
+        return jsonify({
+            "mood": mapped_mood,
+            "score": score
+        }), 200
+
+    except Exception as e:
+        app.logger.error(f"Error in /predict_mood: {e}")
+        return jsonify({"error": str(e)}), 500
+
+
+@app.route('/reset_logs', methods=['POST'])
+def reset_logs():
+    """
+    Deletes the logs.csv file to reset the dashboard to sample data.
+    This is for demo purposes.
+    """
+    try:
+        if os.path.exists(LOGS_CSV):
+            os.remove(LOGS_CSV)
+            print("logs.csv has been deleted.")
+            return jsonify({"message": "Log file deleted successfully. Dashboard will reset to sample data."}), 200
+        else:
+            print("logs.csv not found, no action needed.")
+            return jsonify({"message": "No log file to delete."}), 200
+    except Exception as e:
+        print(f"Error deleting log file: {e}")
+        return jsonify({"error": f"Failed to delete log file: {e}"}), 500
+
+
+# --- Main execution ---
+if __name__ == "__main__":
+    init_log_file() # Ensure log file exists on startup
+    # For Render deployment, Render sets the PORT env variable.
+    # port = int(os.environ.get("PORT", 5000))
+    app.run(host="0.0.0.0", port=7860, debug=True) # Debug=True is fine for hackathon

model/mood_predict.py

@@ -0,0 +1,15 @@
+from transformers import pipeline
+
+# Load a BERT emotion classifier
+emotion_classifier = pipeline(
+    "text-classification",
+    model="j-hartmann/emotion-english-distilroberta-base",
+    return_all_scores=False,
+    top_k = 1 
+)
+
+def mood_predict(text: str):
+    preds = emotion_classifier(text)[0]
+    print(preds)
+    return preds[0]  # return dict like content 
+

requirements.txt

@@ -0,0 +1,7 @@
+flask
+flask-cors
+transformers
+torch
+pandas
+huggingface_hub[hf_xet]
+gunicorn

utils/mapping.py

@@ -0,0 +1,11 @@
+def map_to_mood(label):
+    mapping = {
+        'joy': 'Happy',
+        'neutral': 'Neutral',
+        'sadness': 'Sad',
+        'anger': 'Angry',
+        'disgust': 'Angry',
+        'fear': 'Sad',
+        'surprise': 'Neutral'
+    }
+    return mapping.get(label, 'Neutral')