Spaces:
Running
Running
lixuejing
commited on
Commit
·
191b379
1
Parent(s):
8163d34
updat
Browse files- src/display/utils.py +2 -2
- src/leaderboard/read_evals.py +5 -4
src/display/utils.py
CHANGED
|
@@ -27,7 +27,7 @@ auto_eval_column_dict = []
|
|
| 27 |
auto_eval_column_dict.append(["model_type_symbol", ColumnContent, ColumnContent("T", "str", True, never_hidden=True)])
|
| 28 |
auto_eval_column_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
|
| 29 |
#Scores
|
| 30 |
-
auto_eval_column_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
|
| 31 |
for task in Tasks:
|
| 32 |
auto_eval_column_dict.append([task.name, ColumnContent, ColumnContent(task.value.col_name, "number", True)])
|
| 33 |
# Model information
|
|
@@ -51,7 +51,7 @@ auto_eval_column_quota_dict = []
|
|
| 51 |
auto_eval_column_quota_dict.append(["model_type_symbol", ColumnContent, ColumnContent("T", "str", True, never_hidden=True)])
|
| 52 |
auto_eval_column_quota_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
|
| 53 |
#Scores
|
| 54 |
-
auto_eval_column_quota_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
|
| 55 |
for task in Quotas:
|
| 56 |
auto_eval_column_quota_dict.append([task.name, ColumnContent, ColumnContent(task.value.col_name, "number", True)])
|
| 57 |
# Model information
|
|
|
|
| 27 |
auto_eval_column_dict.append(["model_type_symbol", ColumnContent, ColumnContent("T", "str", True, never_hidden=True)])
|
| 28 |
auto_eval_column_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
|
| 29 |
#Scores
|
| 30 |
+
#auto_eval_column_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
|
| 31 |
for task in Tasks:
|
| 32 |
auto_eval_column_dict.append([task.name, ColumnContent, ColumnContent(task.value.col_name, "number", True)])
|
| 33 |
# Model information
|
|
|
|
| 51 |
auto_eval_column_quota_dict.append(["model_type_symbol", ColumnContent, ColumnContent("T", "str", True, never_hidden=True)])
|
| 52 |
auto_eval_column_quota_dict.append(["model", ColumnContent, ColumnContent("Model", "markdown", True, never_hidden=True)])
|
| 53 |
#Scores
|
| 54 |
+
#auto_eval_column_quota_dict.append(["average", ColumnContent, ColumnContent("Average ⬆️", "number", True)])
|
| 55 |
for task in Quotas:
|
| 56 |
auto_eval_column_quota_dict.append([task.name, ColumnContent, ColumnContent(task.value.col_name, "number", True)])
|
| 57 |
# Model information
|
src/leaderboard/read_evals.py
CHANGED
|
@@ -143,10 +143,11 @@ class EvalResult:
|
|
| 143 |
"""Converts the Eval Result to a dict compatible with our dataframe display"""
|
| 144 |
average = 0
|
| 145 |
nums = 0
|
| 146 |
-
for v in self.results.values():
|
| 147 |
-
if
|
| 148 |
-
|
| 149 |
-
|
|
|
|
| 150 |
if nums ==0:
|
| 151 |
average = 0
|
| 152 |
else:
|
|
|
|
| 143 |
"""Converts the Eval Result to a dict compatible with our dataframe display"""
|
| 144 |
average = 0
|
| 145 |
nums = 0
|
| 146 |
+
for k,v in self.results.values():
|
| 147 |
+
if k not in ["Visual Grounding","Counting","State & Activity Understanding","Dynamic","Relative direction","Multi-view matching","Relative distance","Depth estimation","Relative shape","Size estimation","Trajectory","Future prediction","Goal Decomposition","Navigation"]:
|
| 148 |
+
if v is not None and v != 0:
|
| 149 |
+
average += v
|
| 150 |
+
nums += 1
|
| 151 |
if nums ==0:
|
| 152 |
average = 0
|
| 153 |
else:
|