Add backend and frontend project files

app.py

@@ -1,101 +1,92 @@
-# -*- coding: utf-8 -*-
-"""app.py
-
-Automatically generated by Colab.
-
-Original file is located at
-    https://colab.research.google.com/drive/1E5d9dWFZwd3QoYrwG2SkR-slXGm2aMfL
-"""
-
 import pickle
 import json
 import torch
-from fastapi import FastAPI
-from pydantic import BaseModel
+from flask import Flask, request, jsonify
+# CORRECTED: Changed back to BertTokenizer to match your model
 from transformers import BertTokenizer
+import io
+from waitress import serve
 
 # 1. Initialize App and Load Assets
 # ==================================
-app = FastAPI(
-    title="TEXI'OR Sentiment Analysis API",
-    description="An API to classify text into different sentiments using a fine-tuned BERT model.",
-    version="1.0"
-)
+app = Flask(__name__)
 
-# Pydantic model for input data validation
-class TextInput(BaseModel):
-    text: str
+device = torch.device('cpu')
+print("✅ Using device: cpu")
 
-# Use CUDA if available, otherwise CPU
-device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
-print(f"✅ Using device: {device}")
+# Custom Unpickler to load a GPU-trained model onto a CPU
+class CPU_Unpickler(pickle.Unpickler):
+    def find_class(self, module, name):
+        if module == 'torch.storage' and name == '_load_from_bytes':
+            return lambda b: torch.load(io.BytesIO(b), map_location='cpu')
+        else:
+            return super().find_class(module, name)
 
-# Load the fine-tuned model and tokenizer
-try:
-    with open('bert_sentiment_model (1).pkl', 'rb') as f:
-        model = pickle.load(f)
-    model.to(device)
-    model.eval()  # Set model to evaluation mode
-    print("✅ Model 'bert_sentiment_model (1).pkl' loaded successfully.")
-except FileNotFoundError:
-    print("❌ Model file not found. Make sure 'bert_sentiment_model (1).pkl' is uploaded.")
-    model = None
+# Load model, dictionary, and tokenizer
+model = None
+idx2label = None
+tokenizer = None
 
 try:
-    with open('label_dict.json', 'r') as f:
+    with open('bert_sentiment_model.pkl', 'rb') as f:
+        model = CPU_Unpickler(f).load()
+    model.to(device)
+    model.eval()
+    print("✅ Model 'bert_sentiment_model.pkl' loaded successfully.")
+    
+    with open('Label_Dict.json', 'r') as f:
         label_dict = json.load(f)
-    # Create an inverse mapping from index to label string (e.g., {0: "nocode"})
     idx2label = {int(v): k for k, v in label_dict.items()}
-    print("✅ Label dictionary 'label_dict.json' loaded successfully.")
-except FileNotFoundError:
-    print("❌ Label dictionary 'label_dict.json' not found.")
-    idx2label = None
+    print("✅ Label dictionary 'Label_Dict.json' loaded successfully.")
+
+    # CORRECTED: Use the original BERT tokenizer
+    tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+    print("✅ Tokenizer loaded successfully.")
 
-tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
-print("✅ Tokenizer loaded successfully.")
+except Exception as e:
+    print(f"❌ Error loading assets: {e}")
 
 
 # 2. Define API Endpoints
 # ========================
-@app.get("/")
-def read_root():
-    return {"message": "Welcome to the TEXI'OR API. Use the /predict endpoint to get a sentiment."}
-
-@app.post("/predict")
-def predict_sentiment(text_input: TextInput):
-    """
-    Predicts the sentiment of a given text.
-    - Input: A JSON with a "text" field.
-    - Output: A JSON with the predicted "sentiment" and "confidence" score.
-    """
+@app.route("/predict", methods=['POST'])
+def predict_sentiment():
     if not all([model, idx2label, tokenizer]):
-        return {"error": "API is not ready. A model, label, or tokenizer component failed to load."}
+        return jsonify({"error": "API is not ready. Check server logs."}), 500
+
+    if not request.is_json:
+        return jsonify({"error": "Request must be JSON"}), 400
 
-    text = text_input.text
+    data = request.get_json()
+    text = data.get("text")
 
-    # Tokenize the input text
-    inputs = tokenizer(
-        text, padding=True, truncation=True, max_length=150, return_tensors='pt'
-    )
+    if not text:
+        return jsonify({"error": "Missing 'text' field in request"}), 400
 
-    # Move tensors to the correct device
+    # Tokenize and predict
+    inputs = tokenizer(text, padding=True, truncation=True, max_length=150, return_tensors='pt')
     input_ids = inputs['input_ids'].to(device)
     attention_mask = inputs['attention_mask'].to(device)
-
-    # Get model predictions without calculating gradients
+    
     with torch.no_grad():
         outputs = model(input_ids=input_ids, attention_mask=attention_mask)
 
-    # Process the model output to get probabilities and the predicted class
+    # Process output
     logits = outputs.logits
     probabilities = torch.nn.functional.softmax(logits, dim=-1)
     confidence, predicted_class_idx = torch.max(probabilities, dim=1)
-
-    # Map the predicted index back to its string label
-    predicted_label = idx2label.get(predicted_class_idx.item(), "Unknown Label")
+    
+    predicted_label = idx2label.get(predicted_class_idx.item(), "Unknown")
     confidence_score = confidence.item()
 
-    return {
+    return jsonify({
         "sentiment": predicted_label,
         "confidence": round(confidence_score, 4)
-    }
+    })
+
+# 3. Run the Server
+# ===================
+if __name__ == "__main__":
+    print("🚀 Starting Flask server with Waitress...")
+    # Using 0.0.0.0 makes the server accessible on your local network
+    serve(app, host='0.0.0.0', port=5000)

requirements.txt

@@ -1,14 +1,5 @@
-{
-    "nocode": 0,
-    "happy": 1,
-    "not-relevant": 2,
-    "angry": 3,
-    "surprise": 4,
-    "sad": 5,
-    "happy|surprise": 6,
-    "happy|sad": 7,
-    "disgust|angry": 8,
-    "disgust": 9,
-    "sad|disgust": 10,
-    "sad|angry": 11
-}
+Flask
+torch
+transformers
+scikit-learn
+waitress

texior-app/.gitignore

@@ -0,0 +1,72 @@
+# Learn more https://docs.github.com/en/get-started/getting-started-with-git/ignoring-files
+
+# dependencies
+node_modules/
+
+# Expo
+.expo/
+dist/
+web-build/
+expo-env.d.ts
+
+# Native
+.kotlin/
+*.orig.*
+*.jks
+*.p8
+*.p12
+*.key
+*.mobileprovision
+
+# Metro
+.metro-health-check*
+
+# debug
+npm-debug.*
+yarn-debug.*
+yarn-error.*
+
+# macOS
+.DS_Store
+*.pem
+
+# local env files
+.env*.local
+
+# typescript
+*.tsbuildinfo
+
+# generated native folders
+/ios
+/android
+==================================
+Python & Flask Backend Specific
+==================================
+Ignore the virtual environment folder
+/texior-api/venv/
+
+Ignore Python's compiled file cache
+/texior-api/pycache/
+/texior-api/*.pyc
+
+==================================
+React Native & Expo Frontend Specific
+==================================
+Ignore the massive node modules folder
+/texior-app/node_modules/
+
+Ignore local Expo configuration and cache
+/texior-app/.expo/
+
+Ignore log files
+/texior-app/npm-debug.log
+/texior-app/yarn-error.log
+
+==================================
+General - IDE and OS Files
+==================================
+Ignore VS Code editor settings
+.vscode/
+
+Ignore macOS generated files
+.DS_Store

texior-app/App.js

@@ -0,0 +1,297 @@
+import React, { useState } from 'react';
+import {
+  StyleSheet,
+  Text,
+  View,
+  TextInput,
+  TouchableOpacity,
+  ActivityIndicator,
+  Keyboard,
+  ScrollView,
+  SafeAreaView,
+  StatusBar,
+  Platform,
+} from 'react-native';
+
+const API_URL = "http://192.168.1.8:5000/predict"; // update if needed
+
+export default function App() {
+  const [text, setText] = useState('');
+  const [sentiment, setSentiment] = useState(null);
+  const [confidence, setConfidence] = useState(null);
+  const [isLoading, setIsLoading] = useState(false);
+  const [error, setError] = useState(null);
+
+  const handlePredict = async () => {
+    if (text.trim() === '') {
+      setError('Please enter some text to analyze.');
+      return;
+    }
+
+    if (API_URL.includes("YOUR_COMPUTER_IP_ADDRESS")) {
+      setError("Please update the API_URL with your actual IP address.");
+      return;
+    }
+
+    Keyboard.dismiss();
+    setIsLoading(true);
+    setError(null);
+
+    try {
+      const response = await fetch(API_URL, {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({ text }),
+      });
+
+      let data;
+      try {
+        data = await response.json();
+      } catch {
+        throw new Error('Invalid response from server. Expected JSON.');
+      }
+
+      if (response.ok) {
+        setSentiment(data.sentiment);
+        setConfidence(data.confidence);
+        // NOTE: we intentionally do NOT clear the input so text is retained
+      } else {
+        throw new Error(data.error || 'API returned an error.');
+      }
+    } catch (err) {
+      console.error(err);
+      setError(
+        'Failed to connect to the API. Ensure your Flask server is running and your device is on the same Wi-Fi network.'
+      );
+    } finally {
+      setIsLoading(false);
+    }
+  };
+
+  const getSentimentStyle = (label) => {
+    switch (label?.toLowerCase()) {
+      case 'happy':
+        return { color: '#27ae60' };
+      case 'sad':
+        return { color: '#2980b9' };
+      case 'angry':
+        return { color: '#e74c3c' };
+      case 'surprise':
+        return { color: '#f1c40f' };
+      case 'disgust':
+        return { color: '#8e44ad' };
+      default:
+        return { color: '#7f8c8d' };
+    }
+  };
+
+  return (
+    <SafeAreaView style={styles.safeArea}>
+      <StatusBar
+        barStyle="dark-content"
+        backgroundColor={styles.safeArea.backgroundColor}
+      />
+      <ScrollView contentContainerStyle={styles.container}>
+        <View style={styles.header}>
+          <Text style={styles.title}>TEXI'OR</Text>
+          <Text style={styles.subtitle}>Mood & Sentiment Analyzer</Text>
+        </View>
+
+        <View style={styles.card}>
+          <TextInput
+            style={styles.input}
+            placeholder="How are you feeling today?"
+            placeholderTextColor="#9aa0a6"
+            value={text}
+            onChangeText={setText}
+            multiline
+            textAlignVertical="top"
+          />
+
+          <TouchableOpacity
+            style={[styles.button, isLoading && styles.buttonDisabled]}
+            onPress={handlePredict}
+            disabled={isLoading}
+            activeOpacity={0.85}
+          >
+            {isLoading ? (
+              <ActivityIndicator size="small" color="#fff" />
+            ) : (
+              <Text style={styles.buttonText}>Analyze Mood</Text>
+            )}
+          </TouchableOpacity>
+
+          {error ? <Text style={styles.errorText}>{error}</Text> : null}
+
+          {sentiment && Number.isFinite(confidence) && (
+            <View style={styles.resultContainer}>
+              <Text style={styles.resultLabel}>PREDICTED MOOD</Text>
+              <Text style={[styles.resultSentiment, getSentimentStyle(sentiment)]}>
+                {String(sentiment).toUpperCase()}
+              </Text>
+
+              <View style={styles.confidenceBarBackground}>
+                <View
+                  style={[
+                    styles.confidenceBarFill,
+                    { width: `${Math.max(0, Math.min(100, (confidence * 100).toFixed(0)))}%` },
+                  ]}
+                />
+              </View>
+
+              <Text style={styles.resultConfidence}>
+                {(Number.isFinite(confidence) ? (confidence * 100).toFixed(1) : '0.0')}% Confidence
+              </Text>
+            </View>
+          )}
+        </View>
+
+        
+      </ScrollView>
+    </SafeAreaView>
+  );
+}
+
+const styles = StyleSheet.create({
+  safeArea: {
+    flex: 1,
+    backgroundColor: '#f4f6f8',
+  },
+  container: {
+    flexGrow: 1,
+    padding: 22,
+    alignItems: 'center',
+    justifyContent: 'center',
+  },
+  header: {
+    alignItems: 'center',
+    marginBottom: 18,
+  },
+  title: {
+    fontSize: 44,
+    fontWeight: '800',
+    color: '#243447',
+    letterSpacing: 1,
+  },
+  subtitle: {
+    fontSize: 16,
+    color: '#61707a',
+    marginTop: 6,
+  },
+
+  card: {
+    width: '100%',
+    backgroundColor: '#ffffff',
+    borderRadius: 16,
+    padding: 18,
+    // subtle shadow
+    ...Platform.select({
+      ios: {
+        shadowColor: '#000',
+        shadowOpacity: 0.06,
+        shadowOffset: { width: 0, height: 6 },
+        shadowRadius: 12,
+      },
+      android: {
+        elevation: 4,
+      },
+    }),
+  },
+
+  input: {
+    width: '100%',
+    minHeight: 120,
+    maxHeight: 220,
+    borderRadius: 12,
+    borderWidth: 1,
+    borderColor: '#e6ecef',
+    padding: 14,
+    fontSize: 16,
+    backgroundColor: '#fbfdff',
+    color: '#243447',
+  },
+
+  button: {
+    marginTop: 14,
+    backgroundColor: '#3276d6',
+    paddingVertical: 13,
+    borderRadius: 12,
+    alignItems: 'center',
+    justifyContent: 'center',
+  },
+  buttonDisabled: {
+    opacity: 0.78,
+  },
+  buttonText: {
+    color: '#fff',
+    fontWeight: '700',
+    fontSize: 16,
+    letterSpacing: 0.3,
+  },
+
+  errorText: {
+    marginTop: 12,
+    color: '#e04444',
+    fontSize: 14,
+    textAlign: 'center',
+  },
+
+  resultContainer: {
+    marginTop: 18,
+    alignItems: 'center',
+    width: '100%',
+    paddingVertical: 14,
+    paddingHorizontal: 10,
+    borderRadius: 12,
+    backgroundColor: '#fcfeff',
+    borderWidth: 1,
+    borderColor: '#e9f0f6',
+  },
+  resultLabel: {
+    fontSize: 12,
+    color: '#6b7780',
+    fontWeight: '600',
+    letterSpacing: 1,
+  },
+  resultSentiment: {
+    fontSize: 30,
+    marginTop: 8,
+    fontWeight: '800',
+  },
+
+  confidenceBarBackground: {
+    width: '92%',
+    height: 10,
+    borderRadius: 6,
+    backgroundColor: '#eef3f7',
+    marginTop: 12,
+    overflow: 'hidden',
+  },
+  confidenceBarFill: {
+    height: '100%',
+    backgroundColor: '#46a0ff',
+  },
+  resultConfidence: {
+    marginTop: 10,
+    fontSize: 15,
+    color: '#34495e',
+    fontWeight: '600',
+  },
+
+  footer: {
+    marginTop: 18,
+    alignItems: 'center',
+  },
+  footerText: {
+    color: '#8b99a2',
+    fontSize: 13,
+  },
+  code: {
+    fontFamily: Platform.select({ ios: 'Menlo', android: 'monospace' }),
+    backgroundColor: '#eef6ff',
+    paddingHorizontal: 6,
+    paddingVertical: 2,
+    borderRadius: 4,
+    color: '#2a6fdb',
+  },
+});

texior-app/app.json

@@ -0,0 +1,29 @@
+{
+  "expo": {
+    "name": "texior-app",
+    "slug": "texior-app",
+    "version": "1.0.0",
+    "orientation": "portrait",
+    "icon": "./assets/icon.png",
+    "userInterfaceStyle": "light",
+    "newArchEnabled": true,
+    "splash": {
+      "image": "./assets/splash-icon.png",
+      "resizeMode": "contain",
+      "backgroundColor": "#ffffff"
+    },
+    "ios": {
+      "supportsTablet": true
+    },
+    "android": {
+      "adaptiveIcon": {
+        "foregroundImage": "./assets/adaptive-icon.png",
+        "backgroundColor": "#ffffff"
+      },
+      "edgeToEdgeEnabled": true
+    },
+    "web": {
+      "favicon": "./assets/favicon.png"
+    }
+  }
+}

texior-app/index.js

@@ -0,0 +1,8 @@
+import { registerRootComponent } from 'expo';
+
+import App from './App';
+
+// registerRootComponent calls AppRegistry.registerComponent('main', () => App);
+// It also ensures that whether you load the app in Expo Go or in a native build,
+// the environment is set up appropriately
+registerRootComponent(App);

texior-app/package.json

@@ -0,0 +1,22 @@
+{
+  "name": "texior-app",
+  "license": "0BSD",
+  "version": "1.0.0",
+  "main": "index.js",
+  "scripts": {
+    "start": "expo start",
+    "android": "expo start --android",
+    "ios": "expo start --ios",
+    "web": "expo start --web"
+  },
+  "dependencies": {
+    "expo": "~54.0.12",
+    "expo-status-bar": "~3.0.8",
+    "react": "19.1.0",
+    "react-dom": "19.1.0",
+    "react-native": "0.81.4",
+    "react-native-linear-gradient": "^2.8.3",
+    "react-native-web": "^0.21.0"
+  },
+  "private": true
+}