Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Update app.py
Browse files
app.py
CHANGED
@@ -78,23 +78,6 @@ SET = "validation"
|
|
78 |
|
79 |
custom_role_conversions = {"tool-call": "assistant", "tool-response": "user"}
|
80 |
|
81 |
-
### LOAD EVALUATION DATASET
|
82 |
-
|
83 |
-
eval_ds = datasets.load_dataset("gaia-benchmark/GAIA", "2023_all")[SET]
|
84 |
-
eval_ds = eval_ds.rename_columns({"Question": "question", "Final answer": "true_answer", "Level": "task"})
|
85 |
-
|
86 |
-
|
87 |
-
def preprocess_file_paths(row):
|
88 |
-
if len(row["file_name"]) > 0:
|
89 |
-
row["file_name"] = f"data/gaia/{SET}/" + row["file_name"]
|
90 |
-
return row
|
91 |
-
|
92 |
-
|
93 |
-
eval_ds = eval_ds.map(preprocess_file_paths)
|
94 |
-
eval_df = pd.DataFrame(eval_ds)
|
95 |
-
print("Loaded evaluation dataset:")
|
96 |
-
print(eval_df["task"].value_counts())
|
97 |
-
|
98 |
user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36 Edg/119.0.0.0"
|
99 |
|
100 |
BROWSER_CONFIG = {
|
|
|
78 |
|
79 |
custom_role_conversions = {"tool-call": "assistant", "tool-response": "user"}
|
80 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
81 |
user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36 Edg/119.0.0.0"
|
82 |
|
83 |
BROWSER_CONFIG = {
|