GSK-2434-Fix-bugs-mask-token-and-feature-mapping (#19)

- bug fix (b919a1bd62e5b2a50484806461eb95b68b319ec9)


Co-authored-by: zcy <ZeroCommand@users.noreply.huggingface.co>

io_utils.py

@@ -120,7 +120,6 @@ def pop_job_from_pipe():
         job = f.readline().strip()
         remaining = f.readlines()
         f.close()
-    print(job, remaining, ">>>>")
     with open(PIPE_PATH, "w") as f:
         f.write("\n".join(remaining))
         f.close()
@@ -130,10 +129,13 @@ def pop_job_from_pipe():
     if len(job_info) != 2:
         raise ValueError("Invalid job info: ", job_info)
 
-    write_log_to_user_file(job_info[0], f"Running job {job_info}")
+    write_log_to_user_file(job_info[0], f"Running job id {job_info[0]}\n")
     command = job_info[1].split(",")
+    masked_command = command.copy()
+    hf_token_index = masked_command.index("--hf_token")
+    masked_command[hf_token_index + 1] = "hf_********"
+    write_log_to_user_file(job_info[0], f"Running command {masked_command}\n")
 
-    write_log_to_user_file(job_info[0], f"Running command {command}")
     log_file = open(f"./tmp/{job_info[0]}_log", "a")
     subprocess.Popen(
         command,
@@ -141,3 +143,5 @@ def pop_job_from_pipe():
         stdout=log_file,
         stderr=log_file,
     )
+
+

text_classification_ui_helpers.py

@@ -67,19 +67,18 @@ def write_column_mapping_to_config(dataset_id, dataset_config, dataset_split, *l
     for i, label in enumerate(labels[:MAX_LABELS]):
         if label:
             all_mappings["labels"][label] = ds_labels[i]
-
     if "features" not in all_mappings.keys():
         all_mappings["features"] = dict()
     for i, feat in enumerate(labels[MAX_LABELS : (MAX_LABELS + MAX_FEATURES)]):
         if feat:
-            all_mappings["features"][feat] = ds_features[i]
+            # TODO: Substitute 'text' with more features for zero-shot
+            all_mappings["features"]["text"] = feat
     write_column_mapping(all_mappings)
 
 
 def list_labels_and_features_from_dataset(ds_labels, ds_features, model_id2label):
     model_labels = list(model_id2label.values())
     len_model_labels = len(model_labels)
-    print(model_labels, model_id2label, 3 % len_model_labels)
     lables = [
         gr.Dropdown(
             label=f"{label}",