Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Bug Fix: KeyError and Repeat Counts in evaluation #178

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Bug Fix: KeyError and Repeat Counts in evaluation #178

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
82473c5
Bug Fix: Resolved KeyError Occurring During the Round Loop
realgump Oct 13, 2023
92f8a33
Bug Fix: Duplicate Preference Counts in Some Cases.
realgump Oct 13, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion toolbench/tooleval/eval_preference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -175,22 +175,28 @@ def get_preference(query_id, task_status, answer_statuss, ref_example, output_ex
for qid in test_ids:
if qid not in prefer_dict:
prefer_dict[qid] = {reference_model: 0, output_model: 0, f"round_{i}": "incomplete"}
elif f"round_{i}" not in prefer_dict[qid]:
prefer_dict[qid][f"round_{i}"] = "incomplete"
elif prefer_dict[qid][f"round_{i}"] == "complete":
continue
if qid in ref_pass_result_dict and qid in output_pass_result_dict:
if ref_pass_result_dict[qid]["machine_label"] == "passed" and output_pass_result_dict[qid]["machine_label"] == "failed":
prefer_dict[qid][reference_model] += 1
prefer_dict[qid][f"round_{i}"] = "complete"
continue
elif ref_pass_result_dict[qid]["machine_label"] == "failed" and output_pass_result_dict[qid]["machine_label"] == "passed":
prefer_dict[qid][output_model] += 1
prefer_dict[qid][f"round_{i}"] = "complete"
continue

if qid not in reference_examples:
prefer_dict[qid][output_model] += 1
prefer_dict[qid][f"round_{i}"] = "complete"
continue
if qid not in output_examples:
print(f"Query {qid} not in output model converted answers!")
prefer_dict[qid][reference_model] += 1
prefer_dict[qid][f"round_{i}"] = "complete"
continue

ref_example = reference_examples[qid]
Expand Down Expand Up @@ -262,4 +268,4 @@ def get_preference(query_id, task_status, answer_statuss, ref_example, output_ex
lose_rate /= len(prefer_dict)
tie_rate /= len(prefer_dict)
print(f"Test set: {test_set}. Reference model: {reference_model}, Candidate model: {output_model}. Win rate: {str(win_rate)}, Tie rate: {str(tie_rate)}")