File size: 16,339 Bytes
080d211 4e580b0 080d211 2dad6a6 4e580b0 2dad6a6 4e580b0 2dad6a6 4e580b0 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 2dad6a6 080d211 4e580b0 080d211 2dad6a6 4e580b0 080d211 2dad6a6 080d211 2dad6a6 080d211 4e580b0 080d211 2dad6a6 080d211 4e580b0 2dad6a6 4e580b0 2dad6a6 4e580b0 2dad6a6 4e580b0 080d211 4e580b0 2dad6a6 080d211 80d0dfc |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 |
import json
import gradio as gr
import plotly.graph_objects as go
import networkx as nx
from typing import List, Dict, Optional
from langchain_openai.chat_models import ChatOpenAI
from dialog2graph.pipelines.model_storage import ModelStorage
from dialog2graph.pipelines.d2g_llm.pipeline import D2GLLMPipeline
from dialog2graph.pipelines.helpers.parse_data import PipelineRawDataType
# Initialize the pipeline
def initialize_pipeline():
ms = ModelStorage()
ms.add(
"my_filling_model",
config={"model_name": "gpt-3.5-turbo"},
model_type=ChatOpenAI,
)
return D2GLLMPipeline("d2g_pipeline", model_storage=ms, filling_llm="my_filling_model")
def load_dialog_data(json_file: str, custom_dialog_json: Optional[str] = None) -> List[List[Dict[str, str]]]:
"""Load dialog data from JSON file or custom JSON string and ensure it's in list[list[dict]] format"""
if json_file == "custom" and custom_dialog_json:
try:
data = json.loads(custom_dialog_json)
except json.JSONDecodeError as e:
gr.Error(f"Invalid JSON format in custom dialog: {str(e)}")
return []
else:
file_path = f"{json_file}.json"
try:
with open(file_path, 'r') as f:
data = json.load(f)
except FileNotFoundError:
gr.Error(f"File {file_path} not found!")
return []
except json.JSONDecodeError:
gr.Error(f"Invalid JSON format in {file_path}!")
return []
# Convert to list[list[dict]] format if needed
if not data:
return []
# Check if it's already list[list[dict]]
if isinstance(data, list) and len(data) > 0:
if isinstance(data[0], list):
# Already in list[list[dict]] format
return data
elif isinstance(data[0], dict):
# Convert list[dict] to list[list[dict]]
return [data]
# If it's something else, wrap it in double list
return [data] if isinstance(data, list) else [[data]]
def create_network_visualization(graph: nx.Graph) -> go.Figure:
"""Create a Plotly network visualization from NetworkX graph"""
# Get node positions using spring layout
pos = nx.spring_layout(graph, k=1, iterations=50)
# Extract node and edge information
node_x = []
node_y = []
node_text = []
node_ids = []
for node in graph.nodes():
x, y = pos[node]
node_x.append(x)
node_y.append(y)
# Get node attributes if available
node_attrs = graph.nodes[node]
node_label = node_attrs.get('label', str(node))
node_text.append(f"Node {node}<br>{node_label}")
node_ids.append(node)
# Create edge traces
edge_x = []
edge_y = []
edge_info = []
for edge in graph.edges():
x0, y0 = pos[edge[0]]
x1, y1 = pos[edge[1]]
edge_x.extend([x0, x1, None])
edge_y.extend([y0, y1, None])
# Get edge attributes if available
edge_attrs = graph.edges[edge]
edge_label = edge_attrs.get('label', f"{edge[0]}-{edge[1]}")
edge_info.append(edge_label)
# Create the edge trace
edge_trace = go.Scatter(
x=edge_x, y=edge_y,
line=dict(width=2, color='#888'),
hoverinfo='none',
mode='lines'
)
# Create the node trace
node_trace = go.Scatter(
x=node_x, y=node_y,
mode='markers+text',
hoverinfo='text',
hovertext=node_text,
text=[str(node) for node in node_ids],
textposition="middle center",
marker=dict(
size=20,
line=dict(width=2)
)
)
# Color nodes by number of connections
node_adjacencies = []
for node in graph.nodes():
node_adjacencies.append(len(list(graph.neighbors(node))))
# Update marker color
node_trace.marker = dict(
showscale=True,
colorscale='YlGnBu',
reversescale=True,
color=node_adjacencies,
size=20,
colorbar=dict(
thickness=15,
len=0.5,
x=1.02,
title="Node Connections",
xanchor="left"
),
line=dict(width=2)
)
# Create the figure
fig = go.Figure(data=[edge_trace, node_trace],
layout=go.Layout(
title=dict(
text='Dialog Graph Visualization',
font=dict(
size=16,
),
),
showlegend=False,
hovermode='closest',
margin=dict(b=20,l=5,r=5,t=40),
annotations=[ dict(
text="Hover over nodes for more information",
showarrow=False,
xref="paper", yref="paper",
x=0.005, y=-0.002,
xanchor='left', yanchor='bottom',
font=dict(color="#888", size=12)
)],
xaxis=dict(showgrid=False, zeroline=False, showticklabels=False),
yaxis=dict(showgrid=False, zeroline=False, showticklabels=False),
plot_bgcolor='white'
))
return fig
def create_chat_visualization(dialog_data: List[Dict[str, str]], dialog_index: int = 0, total_dialogs: int = 1) -> str:
"""Create a chat-like visualization of the dialog with navigation info"""
chat_html = f"""
<div style="margin-bottom: 10px; text-align: center; font-weight: bold; color: #666;">
Dialog {dialog_index + 1} of {total_dialogs}
</div>
<div style="max-height: 500px; overflow-y: auto; border: 1px solid #ddd; border-radius: 10px; padding: 20px; background-color: #f9f9f9;">
"""
for i, turn in enumerate(dialog_data):
participant = turn['participant']
text = turn['text']
if participant == 'assistant':
# Assistant messages on the left with blue background
chat_html += f"""
<div style="display: flex; justify-content: flex-start; margin-bottom: 15px;">
<div style="max-width: 70%; background-color: #e3f2fd; padding: 12px 16px; border-radius: 18px; border-bottom-left-radius: 4px; box-shadow: 0 1px 2px rgba(0,0,0,0.1);">
<div style="font-weight: bold; color: #1976d2; font-size: 12px; margin-bottom: 4px;">Assistant</div>
<div style="color: #333; line-height: 1.4;">{text}</div>
</div>
</div>
"""
else:
# User messages on the right with green background
chat_html += f"""
<div style="display: flex; justify-content: flex-end; margin-bottom: 15px;">
<div style="max-width: 70%; background-color: #e8f5e8; padding: 12px 16px; border-radius: 18px; border-bottom-right-radius: 4px; box-shadow: 0 1px 2px rgba(0,0,0,0.1);">
<div style="font-weight: bold; color: #388e3c; font-size: 12px; margin-bottom: 4px;">User</div>
<div style="color: #333; line-height: 1.4;">{text}</div>
</div>
</div>
"""
chat_html += "</div>"
return chat_html
def process_dialog_and_visualize(dialog_choice: str, custom_dialog: str = "", current_dialog_index: int = 0) -> tuple:
"""Process the selected dialog and create visualization"""
try:
# Load the selected dialog data
dialog_data_list = load_dialog_data(dialog_choice, custom_dialog if dialog_choice == "custom" else None)
if not dialog_data_list:
return None, "Failed to load dialog data", "", 0, len(dialog_data_list), gr.update(visible=False), gr.update(visible=False)
# Ensure current_dialog_index is within bounds
current_dialog_index = max(0, min(current_dialog_index, len(dialog_data_list) - 1))
# Initialize pipeline
pipe = initialize_pipeline()
# Process the data (use all dialogs for graph generation)
data = PipelineRawDataType(dialogs=dialog_data_list)
graph, report = pipe.invoke(data)
# Create visualization
fig = create_network_visualization(graph.graph)
# Create chat visualization for the current dialog
current_dialog = dialog_data_list[current_dialog_index]
chat_viz = create_chat_visualization(current_dialog, current_dialog_index, len(dialog_data_list))
# Create summary information
num_nodes = graph.graph.number_of_nodes()
num_edges = graph.graph.number_of_edges()
total_turns = sum(len(dialog) for dialog in dialog_data_list)
summary = f"""
## Graph Summary
- **Number of nodes**: {num_nodes}
- **Number of edges**: {num_edges}
- **Total dialogs**: {len(dialog_data_list)}
- **Total dialog turns**: {total_turns}
- **Currently viewing**: Dialog {current_dialog_index + 1} ({len(current_dialog)} turns)
## Processing Report
Generated graph from {len(dialog_data_list)} dialog(s) with {total_turns} total turns resulting in {num_nodes} nodes and {num_edges} edges.
"""
# Show navigation buttons only if there are multiple dialogs
show_nav = len(dialog_data_list) > 1
return (fig, summary, chat_viz, current_dialog_index, len(dialog_data_list),
gr.update(visible=show_nav), gr.update(visible=show_nav))
except Exception as e:
return None, f"Error processing dialog: {str(e)}", "", 0, 0, gr.update(visible=False), gr.update(visible=False)
# Create the Gradio interface
def create_gradio_app():
with gr.Blocks(title="Dialog2Graph Visualizer") as app:
gr.Markdown("# Dialog2Graph Interactive Visualizer")
gr.Markdown("Select a dialog dataset to process and visualize as a graph network using Plotly.")
# State variables for dialog navigation
current_dialog_index_state = gr.State(0)
total_dialogs_state = gr.State(0)
with gr.Row():
with gr.Column(scale=1):
dialog_selector = gr.Radio(
choices=["dialog1", "dialog2", "dialog3", "dialog4", "custom"],
label="Select Dialog Dataset",
value="dialog1",
info="Choose one of the available dialog datasets or use custom JSON"
)
custom_dialog_input = gr.Textbox(
label="Custom Dialog JSON",
placeholder='[{"text": "Hello! How can I help?", "participant": "assistant"}, {"text": "I need assistance", "participant": "user"}]',
lines=8,
visible=False,
info="Enter dialog data as JSON array with 'text' and 'participant' fields"
)
process_btn = gr.Button(
"Process Dialog & Generate Graph",
variant="primary",
size="lg"
)
with gr.Accordion("Dialog Datasets Info", open=False):
gr.Markdown("""
- **dialog1**: Hotel booking conversation
- **dialog2**: Food delivery conversation
- **dialog3**: Technical support conversation
- **dialog4**: Multiple dialogs (calendar, support, subscription)
- **custom**: Provide your own dialog as JSON
""")
with gr.Column(scale=3):
plot_output = gr.Plot(label="Graph Visualization")
with gr.Row():
with gr.Column(scale=1):
summary_output = gr.Markdown(label="Analysis Summary")
with gr.Column(scale=1):
gr.Markdown("### Dialog Conversation")
# Navigation controls for multiple dialogs
with gr.Row(visible=False) as nav_row:
prev_btn = gr.Button("β Previous Dialog", size="sm")
next_btn = gr.Button("Next Dialog β", size="sm")
chat_output = gr.HTML(label="Chat Visualization")
# Navigation functions
def navigate_dialog(direction: int, current_index: int, total_dialogs: int, dialog_choice: str, custom_dialog: str):
if total_dialogs <= 1:
return current_index, "", ""
new_index = current_index + direction
new_index = max(0, min(new_index, total_dialogs - 1))
try:
dialog_data_list = load_dialog_data(dialog_choice, custom_dialog if dialog_choice == "custom" else None)
if dialog_data_list and new_index < len(dialog_data_list):
current_dialog = dialog_data_list[new_index]
chat_viz = create_chat_visualization(current_dialog, new_index, len(dialog_data_list))
total_turns = sum(len(dialog) for dialog in dialog_data_list)
summary = f"""
## Graph Summary
- **Total dialogs**: {len(dialog_data_list)}
- **Total dialog turns**: {total_turns}
- **Currently viewing**: Dialog {new_index + 1} ({len(current_dialog)} turns)
## Processing Report
Navigate between dialogs to view different conversations.
"""
return new_index, summary, chat_viz
except Exception:
pass
return current_index, "", ""
# Event handlers
def toggle_custom_input(choice):
return gr.update(visible=(choice == "custom"))
dialog_selector.change(
fn=toggle_custom_input,
inputs=[dialog_selector],
outputs=[custom_dialog_input]
)
process_btn.click(
fn=process_dialog_and_visualize,
inputs=[dialog_selector, custom_dialog_input, current_dialog_index_state],
outputs=[plot_output, summary_output, chat_output, current_dialog_index_state, total_dialogs_state, nav_row, nav_row]
)
# Navigation button handlers
prev_btn.click(
fn=lambda curr_idx, total, choice, custom: navigate_dialog(-1, curr_idx, total, choice, custom),
inputs=[current_dialog_index_state, total_dialogs_state, dialog_selector, custom_dialog_input],
outputs=[current_dialog_index_state, summary_output, chat_output]
)
next_btn.click(
fn=lambda curr_idx, total, choice, custom: navigate_dialog(1, curr_idx, total, choice, custom),
inputs=[current_dialog_index_state, total_dialogs_state, dialog_selector, custom_dialog_input],
outputs=[current_dialog_index_state, summary_output, chat_output]
)
# Auto-process on selection change (but not for custom to avoid premature processing)
def auto_process(choice, custom_text, curr_idx):
if choice != "custom":
return process_dialog_and_visualize(choice, custom_text, curr_idx)
else:
return None, "Select 'Process Dialog & Generate Graph' to process custom dialog", "", 0, 0, gr.update(visible=False), gr.update(visible=False)
dialog_selector.change(
fn=auto_process,
inputs=[dialog_selector, custom_dialog_input, current_dialog_index_state],
outputs=[plot_output, summary_output, chat_output, current_dialog_index_state, total_dialogs_state, nav_row, nav_row]
)
return app
if __name__ == "__main__":
app = create_gradio_app()
app.launch()
|