Delete app.py

app.py

@@ -1,128 +0,0 @@
-
-import gradio as gr
-import torch
-import numpy as np
-import pandas as pd
-from sklearn.manifold import TSNE
-from sklearn.decomposition import PCA
-import plotly.express as px
-
-# Load data
-def load_data():
-    print("Loading data...")
-    data = torch.load("demo_data.pt")
-    records = []
-    for i, d in enumerate(data):
-        records.append({
-            "index": i,
-            "tech": d['tech'],
-            "snr": d['snr'],
-            "mod": d['mod'],
-            "mob": d['mob'],
-            "embedding": d['embedding'].numpy(),
-            "spectrogram": d['data'].numpy().flatten()
-        })
-    df = pd.DataFrame(records)
-    print(f"Loaded {len(df)} samples.")
-    return df
-
-df = load_data()
-
-# Get unique values for filters
-tech_choices = sorted(list(df['tech'].unique()))
-snr_choices = sorted(list(df['snr'].unique()))
-mod_choices = sorted(list(df['mod'].unique()))
-mob_choices = sorted(list(df['mob'].unique()))
-
-def plot_tsne(tech_filter, snr_filter, mod_filter, mob_filter, representation, color_by, perplexity, n_iter):
-    # Filter data
-    filtered_df = df.copy()
-    
-    if tech_filter and len(tech_filter) > 0:
-        filtered_df = filtered_df[filtered_df['tech'].isin(tech_filter)]
-        
-    if snr_filter and len(snr_filter) > 0:
-        filtered_df = filtered_df[filtered_df['snr'].isin(snr_filter)]
-        
-    if mod_filter and len(mod_filter) > 0:
-        filtered_df = filtered_df[filtered_df['mod'].isin(mod_filter)]
-        
-    if mob_filter and len(mob_filter) > 0:
-        filtered_df = filtered_df[filtered_df['mob'].isin(mob_filter)]
-        
-    if len(filtered_df) < 5:
-        return None, f"Not enough data points ({len(filtered_df)}). Need at least 5."
-        
-    # Select features
-    if representation == "LWM Embedding":
-        features = np.stack(filtered_df['embedding'].values)
-    else:
-        features = np.stack(filtered_df['spectrogram'].values)
-        # PCA for raw spectrograms to speed up t-SNE
-        if features.shape[1] > 50:
-            pca = PCA(n_components=50, random_state=42)
-            features = pca.fit_transform(features)
-            
-    # Run t-SNE
-    # Adjust perplexity if N is small
-    eff_perplexity = min(perplexity, len(filtered_df) - 1)
-    
-    tsne = TSNE(n_components=2, perplexity=eff_perplexity, n_iter=n_iter, random_state=42, init='pca', learning_rate='auto')
-    projections = tsne.fit_transform(features)
-    
-    filtered_df['x'] = projections[:, 0]
-    filtered_df['y'] = projections[:, 1]
-    
-    # Plot
-    fig = px.scatter(
-        filtered_df, 
-        x='x', 
-        y='y', 
-        color=color_by,
-        hover_data=['tech', 'snr', 'mod', 'mob'],
-        title=f"t-SNE of {representation} ({len(filtered_df)} samples)",
-        template="plotly_white"
-    )
-    fig.update_layout(legend_title_text=color_by.capitalize())
-    
-    return fig, f"Displayed {len(filtered_df)} samples."
-
-# UI
-with gr.Blocks(title="LWM-Spectro Demo") as demo:
-    gr.Markdown("# 🔬 LWM-Spectro Interactive t-SNE Demo")
-    gr.Markdown("""
-    Compare **LWM embeddings** vs **Raw Spectrograms** for wireless signal classification.
-    
-    *   **LWM Embedding**: 128-dim vector from the pre-trained Large Wireless Model.
-    *   **Raw Spectrogram**: Flattened 128x128 spectrogram (reduced via PCA before t-SNE).
-    """)
-    
-    with gr.Row():
-        with gr.Column(scale=1, min_width=300):
-            gr.Markdown("### Filters")
-            tech_filter = gr.CheckboxGroup(choices=tech_choices, value=tech_choices, label="Technology")
-            snr_filter = gr.Dropdown(choices=snr_choices, value=None, multiselect=True, label="SNR (Empty = All)")
-            mod_filter = gr.Dropdown(choices=mod_choices, value=None, multiselect=True, label="Modulation (Empty = All)")
-            mob_filter = gr.Dropdown(choices=mob_choices, value=None, multiselect=True, label="Mobility (Empty = All)")
-            
-            gr.Markdown("### Visualization Settings")
-            representation = gr.Radio(choices=["LWM Embedding", "Raw Spectrogram"], value="LWM Embedding", label="Representation")
-            color_by = gr.Dropdown(choices=["tech", "snr", "mod", "mob"], value="tech", label="Color By")
-            
-            with gr.Accordion("Advanced t-SNE Settings", open=False):
-                perplexity = gr.Slider(minimum=5, maximum=50, value=30, step=1, label="Perplexity")
-                n_iter = gr.Slider(minimum=250, maximum=2000, value=1000, step=50, label="Iterations")
-                
-            btn = gr.Button("Update Plot", variant="primary")
-            status = gr.Textbox(label="Status", interactive=False)
-            
-        with gr.Column(scale=3):
-            plot = gr.Plot(label="t-SNE Visualization")
-            
-    btn.click(plot_tsne, inputs=[tech_filter, snr_filter, mod_filter, mob_filter, representation, color_by, perplexity, n_iter], outputs=[plot, status])
-    
-    # Initial load
-    demo.load(plot_tsne, inputs=[tech_filter, snr_filter, mod_filter, mob_filter, representation, color_by, perplexity, n_iter], outputs=[plot, status])
-
-if __name__ == "__main__":
-    demo.launch()