Jellyfish042 Claude Sonnet 4.5 commited on
Commit
98b668c
·
1 Parent(s): d620a8f

Add progress indicators during evaluation

Browse files

- Add gr.Progress() parameter to run_evaluation function
- Display progress messages: "Evaluating with Qwen3...", "Evaluating with RWKV7...", "Generating visualization..."
- Show progress bar with percentage (10%, 50%, 90%)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Files changed (1) hide show
  1. app.py +4 -4
app.py CHANGED
@@ -182,7 +182,7 @@ def wrap_html_in_iframe(html: str) -> str:
182
  '''
183
 
184
 
185
- def run_evaluation(text: str):
186
  """Run evaluation on both models and generate visualization."""
187
  from core.evaluator import evaluate_hf_single_sample, evaluate_rwkv7_single_sample
188
  from visualization.html_generator import generate_comparison_html
@@ -199,7 +199,7 @@ def run_evaluation(text: str):
199
 
200
  try:
201
  # Step 1: Evaluate Qwen (using cached model)
202
- print("Evaluating with Qwen3...")
203
  result_qwen = evaluate_hf_single_sample(
204
  _qwen_model,
205
  _qwen_tokenizer,
@@ -208,7 +208,7 @@ def run_evaluation(text: str):
208
  )
209
 
210
  # Step 2: Evaluate RWKV7 (using cached model)
211
- print("Evaluating with RWKV7...")
212
  result_rwkv = evaluate_rwkv7_single_sample(
213
  _rwkv_model,
214
  _rwkv_tokenizer,
@@ -216,7 +216,7 @@ def run_evaluation(text: str):
216
  )
217
 
218
  # Step 3: Generate visualization
219
- print("Generating visualization...")
220
  html = generate_comparison_html(
221
  text=text,
222
  byte_losses_a=result_qwen["byte_wise_losses"],
 
182
  '''
183
 
184
 
185
+ def run_evaluation(text: str, progress=gr.Progress()):
186
  """Run evaluation on both models and generate visualization."""
187
  from core.evaluator import evaluate_hf_single_sample, evaluate_rwkv7_single_sample
188
  from visualization.html_generator import generate_comparison_html
 
199
 
200
  try:
201
  # Step 1: Evaluate Qwen (using cached model)
202
+ progress(0.1, desc="Evaluating with Qwen3...")
203
  result_qwen = evaluate_hf_single_sample(
204
  _qwen_model,
205
  _qwen_tokenizer,
 
208
  )
209
 
210
  # Step 2: Evaluate RWKV7 (using cached model)
211
+ progress(0.5, desc="Evaluating with RWKV7...")
212
  result_rwkv = evaluate_rwkv7_single_sample(
213
  _rwkv_model,
214
  _rwkv_tokenizer,
 
216
  )
217
 
218
  # Step 3: Generate visualization
219
+ progress(0.9, desc="Generating visualization...")
220
  html = generate_comparison_html(
221
  text=text,
222
  byte_losses_a=result_qwen["byte_wise_losses"],