Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Update src/leaderboard/read_evals.py
Browse files
src/leaderboard/read_evals.py
CHANGED
@@ -109,7 +109,7 @@ class EvalResult:
|
|
109 |
|
110 |
def to_dict(self):
|
111 |
"""Converts the Eval Result to a dict compatible with our dataframe display"""
|
112 |
-
average = sum([v for v in self.results.values() if v is not None]) / len(
|
113 |
data_dict = {
|
114 |
"eval_name": self.eval_name, # not a column, just a save name,
|
115 |
AutoEvalColumn.precision.name: self.precision.value.name,
|
@@ -127,11 +127,11 @@ class EvalResult:
|
|
127 |
}
|
128 |
|
129 |
for task in Tasks:
|
130 |
-
|
131 |
-
data_dict[task_value.col_name] = self.results[task_value.metric]
|
132 |
|
133 |
return data_dict
|
134 |
|
|
|
135 |
def get_request_file_for_model(requests_path, model_name, precision):
|
136 |
"""Selects the correct request file for a given model. Only keeps runs tagged as FINISHED"""
|
137 |
request_files = os.path.join(
|
|
|
109 |
|
110 |
def to_dict(self):
|
111 |
"""Converts the Eval Result to a dict compatible with our dataframe display"""
|
112 |
+
average = sum([v for v in self.results.values() if v is not None]) / len(Tasks)
|
113 |
data_dict = {
|
114 |
"eval_name": self.eval_name, # not a column, just a save name,
|
115 |
AutoEvalColumn.precision.name: self.precision.value.name,
|
|
|
127 |
}
|
128 |
|
129 |
for task in Tasks:
|
130 |
+
data_dict[task.value.col_name] = self.results[task.value.benchmark]
|
|
|
131 |
|
132 |
return data_dict
|
133 |
|
134 |
+
|
135 |
def get_request_file_for_model(requests_path, model_name, precision):
|
136 |
"""Selects the correct request file for a given model. Only keeps runs tagged as FINISHED"""
|
137 |
request_files = os.path.join(
|