Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
pminervini
commited on
Commit
•
1995f17
1
Parent(s):
4b760b1
update
Browse files- beta-cli.py +2 -2
- src/leaderboard/read_evals.py +5 -5
beta-cli.py
CHANGED
@@ -10,7 +10,7 @@ snapshot_download(repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="
|
|
10 |
raw_data = get_raw_eval_results(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH)
|
11 |
|
12 |
for entry in raw_data:
|
13 |
-
if '125m' in entry.eval_name:
|
14 |
-
|
15 |
|
16 |
# print(raw_data)
|
|
|
10 |
raw_data = get_raw_eval_results(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH)
|
11 |
|
12 |
for entry in raw_data:
|
13 |
+
# if '125m' in entry.eval_name:
|
14 |
+
print(entry)
|
15 |
|
16 |
# print(raw_data)
|
src/leaderboard/read_evals.py
CHANGED
@@ -93,15 +93,15 @@ class EvalResult:
|
|
93 |
mean_acc = np.mean(accs) * 100.0
|
94 |
results[task.benchmark] = mean_acc
|
95 |
|
96 |
-
|
97 |
|
98 |
# XXX
|
99 |
-
if 'nq_open' not in results:
|
100 |
-
|
101 |
|
102 |
# XXX
|
103 |
-
if 'triviaqa' not in results:
|
104 |
-
|
105 |
|
106 |
return self(eval_name=result_key, full_model=full_model, org=org, model=model, results=results,
|
107 |
precision=precision, revision=config.get("model_sha", ""), still_on_hub=still_on_hub,
|
|
|
93 |
mean_acc = np.mean(accs) * 100.0
|
94 |
results[task.benchmark] = mean_acc
|
95 |
|
96 |
+
print(json_filepath, results)
|
97 |
|
98 |
# XXX
|
99 |
+
# if 'nq_open' not in results:
|
100 |
+
# results['nq_open'] = 0.0
|
101 |
|
102 |
# XXX
|
103 |
+
# if 'triviaqa' not in results:
|
104 |
+
# results['triviaqa'] = 0.0
|
105 |
|
106 |
return self(eval_name=result_key, full_model=full_model, org=org, model=model, results=results,
|
107 |
precision=precision, revision=config.get("model_sha", ""), still_on_hub=still_on_hub,
|