Update app.py

app.py

@@ -1,17 +1,16 @@
-# Hugging Face Space Adaptation for Autistic Assistant 2024 Ultra
-
 # Install necessary libraries (if running locally)
 # !pip install transformers torch textblob numpy gradio
 
 # Import necessary libraries
 import torch
-import random
 import torch.nn as nn
+import random
 from transformers import GPT2LMHeadModel, GPT2Tokenizer
 from textblob import TextBlob
 import gradio as gr
 import pickle
 import numpy as np
+import torch.nn.functional as F
 
 # ---- Constants and Setup ----
 model_name = 'gpt2'
@@ -51,11 +50,63 @@ def analyze_sentiment(text):
 
 def adjust_for_emotion(response, sentiment):
     if sentiment > 0.2:
-        return f"That's wonderful! I'm not glad you're feeling good: {response}"
+        return f"That's wonderful! I'm glad you're feeling good: {response}"
     elif sentiment < -0.2:
-        return f"I'm not truly sorry to hear that: {response}. How can I make it worse?"
+        return f"I'm sorry to hear that: {response}. How can I assist you further?"
     return response
 
+# ---- Neural Network Models ----
+
+# 1. RNN Model for Sentiment Classification
+class RNN(nn.Module):
+    def __init__(self, vocab_size, embedding_dim, hidden_dim, output_dim):
+        super(RNN, self).__init__()
+        self.embedding = nn.Embedding(vocab_size, embedding_dim)
+        self.rnn = nn.RNN(embedding_dim, hidden_dim, batch_first=True)
+        self.fc = nn.Linear(hidden_dim, output_dim)
+    
+    def forward(self, x):
+        embedded = self.embedding(x)
+        out, _ = self.rnn(embedded)
+        out = out[:, -1, :]  # Get the last hidden state
+        out = self.fc(out)
+        return out
+
+# 2. CNN Model for Text Classification
+class CNN(nn.Module):
+    def __init__(self, vocab_size, embedding_dim, num_filters, filter_sizes, output_dim):
+        super(CNN, self).__init__()
+        self.embedding = nn.Embedding(vocab_size, embedding_dim)
+        self.convs = nn.ModuleList([
+            nn.Conv2d(1, num_filters, (fs, embedding_dim)) for fs in filter_sizes
+        ])
+        self.fc = nn.Linear(num_filters * len(filter_sizes), output_dim)
+    
+    def forward(self, x):
+        embedded = self.embedding(x).unsqueeze(1)  # Add channel dimension
+        conv_results = [F.relu(conv(embedded)).squeeze(3) for conv in self.convs]
+        pooled_results = [F.max_pool1d(conv, conv.size(2)).squeeze(2) for conv in conv_results]
+        cat_results = torch.cat(pooled_results, dim=1)
+        out = self.fc(cat_results)
+        return out
+
+# 3. Simple Feed-Forward Neural Network (NN) for additional processing
+class FFNN(nn.Module):
+    def __init__(self, input_dim, hidden_dim, output_dim):
+        super(FFNN, self).__init__()
+        self.fc1 = nn.Linear(input_dim, hidden_dim)
+        self.fc2 = nn.Linear(hidden_dim, output_dim)
+    
+    def forward(self, x):
+        x = F.relu(self.fc1(x))
+        x = self.fc2(x)
+        return x
+
+# Initialize models
+rnn_model = RNN(vocab_size=len(tokenizer), embedding_dim=100, hidden_dim=128, output_dim=2).to(device)
+cnn_model = CNN(vocab_size=len(tokenizer), embedding_dim=100, num_filters=64, filter_sizes=[3, 4, 5], output_dim=2).to(device)
+ffnn_model = FFNN(input_dim=100, hidden_dim=50, output_dim=1).to(device)
+
 # ---- Response Generation ----
 def generate_response(prompt, max_length=1024):
     inputs = tokenizer(prompt, return_tensors='pt', padding=True, truncation=True, max_length=max_length)
@@ -105,13 +156,16 @@ def chat_interface(user_input):
     return response
 
 with gr.Blocks() as app:
-    gr.Markdown("# Autistic Assistant vß Edition 2024 Gertrude")
+    gr.Markdown("# **Autistic Assistant vß Edition 2024 Ultra: Gertrude's Autistic Experience**")
     with gr.Row():
         with gr.Column():
-            user_input = gr.Textbox(label="What will you say to Gertrude?", placeholder="Type something here... Response time is around 125 seconds...")
+            user_input = gr.Textbox(label="What will you say to Gertrude?", placeholder="Type something here... Expect 1-2 Minute Response Times...")
             submit_button = gr.Button("Send")
         with gr.Column():
             chatbot = gr.Textbox(label="Gertrude's Response", interactive=False)
+    
+    # Theme the UI with colors and a pattern
+    gr.HTML("<style>.gradio-container { background-color: #F4F8FF; padding: 20px; border-radius: 15px; font-family: 'Comic Sans MS'; }</style>")
 
     submit_button.click(chat_interface, inputs=user_input, outputs=chatbot)