Spaces:
Sleeping
Sleeping
TTsamurai
commited on
Commit
·
64289c0
1
Parent(s):
fe473a7
change evaluation
Browse files
app.py
CHANGED
|
@@ -307,7 +307,7 @@ def create_demo():
|
|
| 307 |
)
|
| 308 |
reason = gr.Textbox(
|
| 309 |
scale=1,
|
| 310 |
-
label="Reason for Your Choice (Explain Your Reasoning & Highlight Useful Parts of Conversation)",
|
| 311 |
lines=5,
|
| 312 |
)
|
| 313 |
with gr.Row():
|
|
@@ -319,7 +319,13 @@ def create_demo():
|
|
| 319 |
label="Confidence in Your Decision (1 = Very Low, 7 = Very High)",
|
| 320 |
show_label=True,
|
| 321 |
)
|
| 322 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 323 |
return {
|
| 324 |
"comp": comp,
|
| 325 |
"system_instruction_context": context,
|
|
@@ -335,6 +341,7 @@ def create_demo():
|
|
| 335 |
"evaluation_send_button": evaluation_send_button,
|
| 336 |
"general_instruction": general_instruction,
|
| 337 |
"round_instruction": round_instruction,
|
|
|
|
| 338 |
}
|
| 339 |
|
| 340 |
def tab_creation_preference_stage():
|
|
@@ -460,6 +467,7 @@ def create_demo():
|
|
| 460 |
likelihood,
|
| 461 |
confidence,
|
| 462 |
evaluation_send_button,
|
|
|
|
| 463 |
) = (
|
| 464 |
tabs["comp"],
|
| 465 |
tabs["system_instruction_context"],
|
|
@@ -473,6 +481,7 @@ def create_demo():
|
|
| 473 |
tabs["likelihood"],
|
| 474 |
tabs["confidence"],
|
| 475 |
tabs["evaluation_send_button"],
|
|
|
|
| 476 |
)
|
| 477 |
system_instruction = ""
|
| 478 |
start_conversation.click(
|
|
@@ -544,7 +553,7 @@ def create_demo():
|
|
| 544 |
[tab_session, chatbot],
|
| 545 |
)
|
| 546 |
evaluation_send_button.click(
|
| 547 |
-
lambda user_id, comp, tab_session, dropdown, reason, likelihood, confidence, evaluation_send_button: respond_evaluation(
|
| 548 |
user_id,
|
| 549 |
tab_session,
|
| 550 |
{
|
|
@@ -555,6 +564,7 @@ def create_demo():
|
|
| 555 |
},
|
| 556 |
comp,
|
| 557 |
evaluation_send_button,
|
|
|
|
| 558 |
),
|
| 559 |
[
|
| 560 |
user_id,
|
|
@@ -565,8 +575,9 @@ def create_demo():
|
|
| 565 |
likelihood,
|
| 566 |
confidence,
|
| 567 |
evaluation_send_button,
|
|
|
|
| 568 |
],
|
| 569 |
-
[tab_session, dropdown, reason, likelihood, confidence, evaluation_send_button],
|
| 570 |
)
|
| 571 |
|
| 572 |
def click_control_preference_stage(tabs, user_id, user_preference_elicitation_session):
|
|
@@ -997,14 +1008,22 @@ def create_demo():
|
|
| 997 |
# log_action(user_id, "Session", "History", history)
|
| 998 |
return tab_data, history
|
| 999 |
|
| 1000 |
-
def respond_evaluation(user_id, tab_data, evals, tab_name, evaluation_send_button):
|
| 1001 |
|
| 1002 |
# dropdown, readon_button, multi-evaluator
|
| 1003 |
print(f"Tab: {tab_name}\nEvaluation: {evals}")
|
| 1004 |
if evals["selection"] is None or evals["likelihood"] is None or evals["confidence"] is None:
|
| 1005 |
-
return
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1006 |
<strong>Please make sure that you answer all the questions.</strong>
|
| 1007 |
-
</div>"""
|
|
|
|
| 1008 |
else:
|
| 1009 |
save_feedback(
|
| 1010 |
user_id,
|
|
@@ -1028,6 +1047,9 @@ def create_demo():
|
|
| 1028 |
evals["likelihood"],
|
| 1029 |
evals["confidence"],
|
| 1030 |
evaluation_send_button,
|
|
|
|
|
|
|
|
|
|
| 1031 |
)
|
| 1032 |
|
| 1033 |
def respond_final_ranking(
|
|
|
|
| 307 |
)
|
| 308 |
reason = gr.Textbox(
|
| 309 |
scale=1,
|
| 310 |
+
label="Reason for Your Choice (Explain Your Reasoning & Highlight Useful Parts of Conversation)[Optional]",
|
| 311 |
lines=5,
|
| 312 |
)
|
| 313 |
with gr.Row():
|
|
|
|
| 319 |
label="Confidence in Your Decision (1 = Very Low, 7 = Very High)",
|
| 320 |
show_label=True,
|
| 321 |
)
|
| 322 |
+
with gr.Row():
|
| 323 |
+
textbox = gr.HTML(
|
| 324 |
+
"""<div style="background-color: #f8d7da; color: #721c24; padding: 15px; border: 1px solid #f5c6cb; border-radius: 5px; margin-bottom: 20px;">
|
| 325 |
+
Make sure to answer all the questions.
|
| 326 |
+
</div>"""
|
| 327 |
+
)
|
| 328 |
+
evaluation_send_button = gr.Button(value="Send: Evaluation")
|
| 329 |
return {
|
| 330 |
"comp": comp,
|
| 331 |
"system_instruction_context": context,
|
|
|
|
| 341 |
"evaluation_send_button": evaluation_send_button,
|
| 342 |
"general_instruction": general_instruction,
|
| 343 |
"round_instruction": round_instruction,
|
| 344 |
+
"textbox": textbox,
|
| 345 |
}
|
| 346 |
|
| 347 |
def tab_creation_preference_stage():
|
|
|
|
| 467 |
likelihood,
|
| 468 |
confidence,
|
| 469 |
evaluation_send_button,
|
| 470 |
+
textbox,
|
| 471 |
) = (
|
| 472 |
tabs["comp"],
|
| 473 |
tabs["system_instruction_context"],
|
|
|
|
| 481 |
tabs["likelihood"],
|
| 482 |
tabs["confidence"],
|
| 483 |
tabs["evaluation_send_button"],
|
| 484 |
+
tabs["textbox"],
|
| 485 |
)
|
| 486 |
system_instruction = ""
|
| 487 |
start_conversation.click(
|
|
|
|
| 553 |
[tab_session, chatbot],
|
| 554 |
)
|
| 555 |
evaluation_send_button.click(
|
| 556 |
+
lambda user_id, comp, tab_session, dropdown, reason, likelihood, confidence, evaluation_send_button, textbox: respond_evaluation(
|
| 557 |
user_id,
|
| 558 |
tab_session,
|
| 559 |
{
|
|
|
|
| 564 |
},
|
| 565 |
comp,
|
| 566 |
evaluation_send_button,
|
| 567 |
+
textbox,
|
| 568 |
),
|
| 569 |
[
|
| 570 |
user_id,
|
|
|
|
| 575 |
likelihood,
|
| 576 |
confidence,
|
| 577 |
evaluation_send_button,
|
| 578 |
+
textbox,
|
| 579 |
],
|
| 580 |
+
[tab_session, dropdown, reason, likelihood, confidence, evaluation_send_button, textbox],
|
| 581 |
)
|
| 582 |
|
| 583 |
def click_control_preference_stage(tabs, user_id, user_preference_elicitation_session):
|
|
|
|
| 1008 |
# log_action(user_id, "Session", "History", history)
|
| 1009 |
return tab_data, history
|
| 1010 |
|
| 1011 |
+
def respond_evaluation(user_id, tab_data, evals, tab_name, evaluation_send_button, textbox):
|
| 1012 |
|
| 1013 |
# dropdown, readon_button, multi-evaluator
|
| 1014 |
print(f"Tab: {tab_name}\nEvaluation: {evals}")
|
| 1015 |
if evals["selection"] is None or evals["likelihood"] is None or evals["confidence"] is None:
|
| 1016 |
+
return (
|
| 1017 |
+
tab_data,
|
| 1018 |
+
evals["selection"],
|
| 1019 |
+
evals["reason"],
|
| 1020 |
+
evals["likelihood"],
|
| 1021 |
+
evals["confidence"],
|
| 1022 |
+
evaluation_send_button,
|
| 1023 |
+
"""<div style="background-color: #f8d7da; color: #721c24; padding: 15px; border: 1px solid #f5c6cb; border-radius: 5px; margin-bottom: 20px;">
|
| 1024 |
<strong>Please make sure that you answer all the questions.</strong>
|
| 1025 |
+
</div>""",
|
| 1026 |
+
)
|
| 1027 |
else:
|
| 1028 |
save_feedback(
|
| 1029 |
user_id,
|
|
|
|
| 1047 |
evals["likelihood"],
|
| 1048 |
evals["confidence"],
|
| 1049 |
evaluation_send_button,
|
| 1050 |
+
"""<div style="background-color: #d4edda; color: #155724; padding: 15px; border: 1px solid #c3e6cb; border-radius: 5px; margin-bottom: 20px;">
|
| 1051 |
+
<strong>Evaluation received. You may now go to the next tab.</strong>
|
| 1052 |
+
</div>""",
|
| 1053 |
)
|
| 1054 |
|
| 1055 |
def respond_final_ranking(
|