Spaces:
Sleeping
Sleeping
File size: 11,145 Bytes
2850cb6 0bf6b56 442107b 2850cb6 86ef0cd 442107b 86ef0cd e700f5c 86ef0cd 442107b 0bf6b56 442107b 0bf6b56 442107b 0bf6b56 86ef0cd 2850cb6 86ef0cd 0bf6b56 86ef0cd 2850cb6 86ef0cd 2850cb6 86ef0cd 2850cb6 86ef0cd 2850cb6 0bf6b56 e700f5c 86ef0cd e700f5c 2850cb6 e700f5c 86ef0cd 0bf6b56 86ef0cd 0bf6b56 2850cb6 86ef0cd e700f5c 86ef0cd 2850cb6 86ef0cd e700f5c 86ef0cd e700f5c 86ef0cd e700f5c 86ef0cd e700f5c 86ef0cd e700f5c 0bf6b56 86ef0cd fb007f1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 | # utils/visualization.py (updated for maximum contrast)
import matplotlib.pyplot as plt
import matplotlib.colors as mcolors
import base64
from io import BytesIO
import numpy as np
def create_visualization(text, explanation, tokenizer, explainer_type):
"""Create HTML visualization of token attributions with maximum contrast"""
try:
# Tokenize the text
tokens = tokenizer.tokenize(text)
# Handle different explanation formats
token_values = {}
if explainer_type == "LIME" and explanation:
# LIME returns list of (feature, weight) tuples
for feature, weight in explanation:
# Extract individual tokens from LIME features
feature_tokens = feature.split()
for token in feature_tokens:
# Clean token (remove punctuation, etc.)
clean_token = token.strip('.,!?;:"()[]{}').lower()
if clean_token:
token_values[clean_token] = weight / len(feature_tokens) if feature_tokens else weight
elif explainer_type in ["SHAP", "Captum"] and explanation:
# SHAP and Captum return list of dicts with 'token' and 'value'
for item in explanation:
if isinstance(item, dict) and 'token' in item and 'value' in item:
token = item['token'].lower()
value = item['value']
token_values[token] = value
# If no explanation data, create a neutral visualization
if not token_values:
html_output = '''
<div style="font-family: monospace; line-height: 2; padding: 15px;
border-radius: 5px; background-color: #f9f9f9;
border: 1px solid #ddd; margin: 10px 0; color: #666;">
<i>Explanation data not available. Showing tokenized text.</i><br>
'''
for token in tokens:
html_output += f'<span style="margin: 2px; padding: 4px 6px; display: inline-block; background-color: #f0f0f0; border: 1px solid #ccc; border-radius: 4px;">{token.replace("##", "")}</span> '
html_output += '</div>'
return html_output
# Normalize scores for coloring
values = list(token_values.values())
max_abs_value = max(abs(min(values)), abs(max(values))) if values else 1
if max_abs_value > 0:
normalized_values = {k: v / max_abs_value for k, v in token_values.items()}
else:
normalized_values = {k: 0 for k in token_values.keys()}
# Create HTML
html_output = '''
<div style="font-family: monospace; line-height: 2; padding: 15px;
border-radius: 5px; background-color: #f9f9f9;
border: 1px solid #ddd; margin: 10px 0;">
'''
# Map tokens to values with high contrast colors
for token in tokens:
clean_token = token.replace('##', '').lower()
if clean_token in normalized_values:
value = token_values[clean_token]
norm_value = normalized_values[clean_token]
# Determine color based on value with maximum contrast
if value < 0:
# Negative values: deep red colors
intensity = min(1.0, 0.6 + 0.4 * abs(norm_value))
if intensity > 0.8:
color = "#cc0000" # Very dark red
elif intensity > 0.6:
color = "#ff4444" # Dark red
elif intensity > 0.4:
color = "#ff8888" # Medium red
else:
color = "#ffcccc" # Light red
text_color = "white" if intensity > 0.5 else "black"
else:
# Positive values: deep blue colors
intensity = min(1.0, 0.6 + 0.4 * norm_value)
if intensity > 0.8:
color = "#0000cc" # Very dark blue
elif intensity > 0.6:
color = "#4444ff" # Dark blue
elif intensity > 0.4:
color = "#8888ff" # Medium blue
else:
color = "#ccccff" # Light blue
text_color = "white" if intensity > 0.5 else "black"
html_output += f'<span style="background-color: {color}; color: {text_color}; border: 2px solid {color}; margin: 2px; padding: 4px 6px; border-radius: 4px; display: inline-block; font-weight: bold;">{token.replace("##", "")}</span> '
else:
html_output += f'<span style="margin: 2px; padding: 4px 6px; display: inline-block; background-color: #f0f0f0; border: 1px solid #ccc; border-radius: 4px;">{token.replace("##", "")}</span> '
html_output += '</div>'
# Add color legend
html_output += '''
<div style="margin-top: 10px; font-size: 12px; color: #666;">
<span style="background-color: #cc0000; color: white; padding: 2px 6px; border-radius: 3px; margin-right: 5px;">Strong negative</span>
<span style="background-color: #ff8888; padding: 2px 6px; border-radius: 3px; margin-right: 5px;">Weak negative</span>
<span style="background-color: #0000cc; color: white; padding: 2px 6px; border-radius: 3px; margin-right: 5px;">Strong positive</span>
<span style="background-color: #8888ff; padding: 2px 6px; border-radius: 3px;">Weak positive</span>
</div>
'''
return html_output
except Exception as e:
print(f"Visualization error: {e}")
return f'<div style="color: red; padding: 10px;">Error creating visualization: {str(e)}</div>'
def create_attribution_plot(explanation, method_name):
"""Create matplotlib visualization of token attributions"""
try:
if not explanation:
return "<p>No explanation data available</p>"
# Handle different explanation formats
if method_name == "LIME":
# LIME: list of (feature, weight) tuples
features = [item[0] for item in explanation][:15] # Show top 15 features
scores = [item[1] for item in explanation][:15]
title = f'Top Feature Attributions ({method_name})'
else:
# SHAP/Captum: list of dicts with 'token' and 'value'
tokens = [item['token'] for item in explanation if isinstance(item, dict) and 'token' in item][:15]
scores = [item['value'] for item in explanation if isinstance(item, dict) and 'value' in item][:15]
features = tokens
title = f'Top Token Attributions ({method_name})'
if not features or not scores:
return "<p>No valid explanation data available for plotting</p>"
# Create plot with better colors
fig, ax = plt.subplots(figsize=(12, 6))
# Create colors based on values - using high contrast colors
colors = ['#ff6b6b' if score < 0 else '#4ecdc4' for score in scores]
# Create horizontal bar chart
y_pos = np.arange(len(features))
bars = ax.barh(y_pos, scores, color=colors, alpha=0.8, edgecolor='black', linewidth=0.5)
# Customize plot
ax.set_yticks(y_pos)
ax.set_yticklabels(features, fontsize=10)
ax.set_xlabel('Attribution Score', fontsize=12, fontweight='bold')
ax.set_title(title, fontsize=14, fontweight='bold')
ax.axvline(x=0, color='black', linestyle='-', alpha=0.5, linewidth=1)
# Add grid for better readability
ax.grid(True, alpha=0.3, axis='x')
# Add value labels on bars
for i, (bar, score) in enumerate(zip(bars, scores)):
width = bar.get_width()
label_x_pos = width + (0.01 * max(scores) if width >= 0 else 0.01 * min(scores))
ax.text(label_x_pos, bar.get_y() + bar.get_height()/2,
f'{score:.4f}', ha='left' if width >= 0 else 'right', va='center',
fontsize=9, fontweight='bold')
# Set background color
ax.set_facecolor('#f8f9fa')
fig.patch.set_facecolor('#f8f9fa')
plt.tight_layout()
# Convert to HTML
buf = BytesIO()
plt.savefig(buf, format='png', dpi=100, bbox_inches='tight', facecolor=fig.get_facecolor())
buf.seek(0)
img_str = base64.b64encode(buf.read()).decode('utf-8')
plt.close(fig)
return f'<img src="data:image/png;base64,{img_str}" style="max-width: 100%; border: 1px solid #ddd; border-radius: 5px;">'
except Exception as e:
print(f"Plot error: {e}")
return f'<div style="color: red; padding: 10px;">Error creating plot: {str(e)}</div>'
# utils/visualization.py (add this function)
def create_confidence_chart(probabilities, class_names=None):
"""Create a bar chart showing class probabilities"""
try:
if class_names is None:
class_names = [f"Class {i}" for i in range(len(probabilities))]
fig, ax = plt.subplots(figsize=(10, 6))
# Create bar chart
bars = ax.bar(range(len(probabilities)), probabilities,
color=['#ff6b6b', '#4ecdc4', '#45b7af', '#556270'][:len(probabilities)],
alpha=0.8, edgecolor='black', linewidth=1)
# Customize chart
ax.set_xlabel('Classes', fontsize=12, fontweight='bold')
ax.set_ylabel('Probability', fontsize=12, fontweight='bold')
ax.set_title('Class Probability Distribution', fontsize=14, fontweight='bold')
ax.set_xticks(range(len(probabilities)))
ax.set_xticklabels(class_names, rotation=45, ha='right')
ax.set_ylim(0, 1)
# Add value labels on bars
for i, (bar, prob) in enumerate(zip(bars, probabilities)):
ax.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 0.01,
f'{prob:.3f}', ha='center', va='bottom', fontweight='bold')
# Add grid
ax.grid(True, alpha=0.3, axis='y')
# Set background color
ax.set_facecolor('#f8f9fa')
fig.patch.set_facecolor('#f8f9fa')
plt.tight_layout()
# Convert to HTML
buf = BytesIO()
plt.savefig(buf, format='png', dpi=100, bbox_inches='tight', facecolor=fig.get_facecolor())
buf.seek(0)
img_str = base64.b64encode(buf.read()).decode('utf-8')
plt.close(fig)
return f'<img src="data:image/png;base64,{img_str}" style="max-width: 100%; border: 1px solid #ddd; border-radius: 5px;">'
except Exception as e:
print(f"Confidence chart error: {e}")
return f'<div style="color: red; padding: 10px;">Error creating confidence chart: {str(e)}</div>' |