Spaces:
Sleeping
Sleeping
meg-huggingface
commited on
Commit
•
39456a9
1
Parent(s):
313cc30
Adding print debug statements
Browse files- src/leaderboard/read_evals.py +6 -0
- src/populate.py +4 -0
src/leaderboard/read_evals.py
CHANGED
@@ -158,6 +158,8 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
|
|
158 |
"""From the path of the results folder root, extract all needed info for results"""
|
159 |
model_result_filepaths = []
|
160 |
|
|
|
|
|
161 |
for root, _, files in os.walk(results_path):
|
162 |
# We should only have json files in model results
|
163 |
if len(files) == 0 or any([not f.endswith(".json") for f in files]):
|
@@ -172,6 +174,8 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
|
|
172 |
for file in files:
|
173 |
model_result_filepaths.append(os.path.join(root, file))
|
174 |
|
|
|
|
|
175 |
eval_results = {}
|
176 |
for model_result_filepath in model_result_filepaths:
|
177 |
# Creation of result
|
@@ -185,6 +189,8 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
|
|
185 |
else:
|
186 |
eval_results[eval_name] = eval_result
|
187 |
|
|
|
|
|
188 |
results = []
|
189 |
for v in eval_results.values():
|
190 |
try:
|
|
|
158 |
"""From the path of the results folder root, extract all needed info for results"""
|
159 |
model_result_filepaths = []
|
160 |
|
161 |
+
print("Getting raw eval results from:")
|
162 |
+
print(os.walk(results_path))
|
163 |
for root, _, files in os.walk(results_path):
|
164 |
# We should only have json files in model results
|
165 |
if len(files) == 0 or any([not f.endswith(".json") for f in files]):
|
|
|
174 |
for file in files:
|
175 |
model_result_filepaths.append(os.path.join(root, file))
|
176 |
|
177 |
+
print("model results filepaths are")
|
178 |
+
print(model_result_filepaths)
|
179 |
eval_results = {}
|
180 |
for model_result_filepath in model_result_filepaths:
|
181 |
# Creation of result
|
|
|
189 |
else:
|
190 |
eval_results[eval_name] = eval_result
|
191 |
|
192 |
+
print("eval results are")
|
193 |
+
print(eval_results)
|
194 |
results = []
|
195 |
for v in eval_results.values():
|
196 |
try:
|
src/populate.py
CHANGED
@@ -10,6 +10,8 @@ from src.leaderboard.read_evals import get_raw_eval_results
|
|
10 |
|
11 |
def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:
|
12 |
raw_data = get_raw_eval_results(results_path, requests_path)
|
|
|
|
|
13 |
all_data_json = [v.to_dict() for v in raw_data]
|
14 |
|
15 |
df = pd.DataFrame.from_records(all_data_json)
|
@@ -23,6 +25,8 @@ def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchm
|
|
23 |
|
24 |
def get_evaluation_queue_df(save_path: str, cols: list) -> list[pd.DataFrame]:
|
25 |
entries = [entry for entry in os.listdir(save_path) if not entry.startswith(".")]
|
|
|
|
|
26 |
all_evals = []
|
27 |
|
28 |
for entry in entries:
|
|
|
10 |
|
11 |
def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:
|
12 |
raw_data = get_raw_eval_results(results_path, requests_path)
|
13 |
+
print("To get the leaderboard df, the raw_data is:")
|
14 |
+
print(raw_data)
|
15 |
all_data_json = [v.to_dict() for v in raw_data]
|
16 |
|
17 |
df = pd.DataFrame.from_records(all_data_json)
|
|
|
25 |
|
26 |
def get_evaluation_queue_df(save_path: str, cols: list) -> list[pd.DataFrame]:
|
27 |
entries = [entry for entry in os.listdir(save_path) if not entry.startswith(".")]
|
28 |
+
print("To get the evaluation queue df, the entries are:")
|
29 |
+
print(entries)
|
30 |
all_evals = []
|
31 |
|
32 |
for entry in entries:
|