taesiri commited on
Commit
19e99f5
1 Parent(s): 3eb3dd3
Files changed (1) hide show
  1. app.py +18 -3
app.py CHANGED
@@ -19,8 +19,8 @@ import torchvision
19
  from huggingface_hub import HfApi, login, snapshot_download
20
  from PIL import Image
21
 
22
- session_token = os.environ.get("SessionToken")
23
- login(token=session_token)
24
 
25
  csv.field_size_limit(sys.maxsize)
26
 
@@ -83,8 +83,23 @@ def update_snapshot(username):
83
  rows.append(tdf)
84
 
85
  df = pd.DataFrame(rows, columns=columns)
86
- df = df[df["user_id"] == username]
87
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
88
  return df
89
 
90
 
 
19
  from huggingface_hub import HfApi, login, snapshot_download
20
  from PIL import Image
21
 
22
+ # session_token = os.environ.get("SessionToken")
23
+ # login(token=session_token)
24
 
25
  csv.field_size_limit(sys.maxsize)
26
 
 
83
  rows.append(tdf)
84
 
85
  df = pd.DataFrame(rows, columns=columns)
 
86
 
87
+ # download and append all CSV files
88
+ output_dir = snapshot_download(
89
+ repo_id="taesiri/imagenet_hard_review_data_r3",
90
+ allow_patterns="*.csv",
91
+ repo_type="dataset",
92
+ )
93
+ files = glob(f"{output_dir}/*.csv")
94
+
95
+ if len(files) > 0:
96
+ csv_dataframes = [pd.read_csv(file) for file in files]
97
+ csv_dataframes = pd.concat(csv_dataframes, ignore_index=True)
98
+ df = pd.concat([df, csv_dataframes], ignore_index=True)
99
+
100
+ # remove duplicate rows
101
+ df = df.drop_duplicates(subset=["id", "user_id"], keep="last")
102
+ df = df[df["user_id"] == username]
103
  return df
104
 
105