add leaderboard arg and inf token

app_text_classification.py

@@ -70,6 +70,7 @@ def get_demo(demo):
         run_local = gr.Checkbox(value=True, label="Run in this Space")
         use_inference = read_inference_type(uid) == "hf_inference_api"
         run_inference = gr.Checkbox(value=use_inference, label="Run with Inference API")
+        inference_token = gr.Textbox(value="", label="HF Token for Inference API", visible=False)
 
     with gr.Accordion(label="Scanner Advance Config (optional)", open=False):
         selected = read_scanners(uid)
@@ -105,7 +106,7 @@ def get_demo(demo):
 
     scanners.change(write_scanners, inputs=[scanners, uid_label])
 
-    run_inference.change(write_inference_type, inputs=[run_inference, uid_label])
+    run_inference.change(write_inference_type, inputs=[run_inference, uid_label], outputs=[inference_token])
 
     gr.on(
         triggers=[label.change for label in column_mappings],

io_utils.py

@@ -1,6 +1,6 @@
 import os
 import subprocess
-
+import gradio as gr
 import yaml
 
 import pipe
@@ -28,6 +28,7 @@ def read_scanners(uid):
     with open(get_yaml_path(uid), "r") as f:
         config = yaml.load(f, Loader=yaml.FullLoader)
         scanners = config.get("detectors", [])
+    f.close()
     return scanners
 
 
@@ -39,6 +40,7 @@ def write_scanners(scanners, uid):
             config["detectors"] = scanners
             # save scanners to detectors in yaml
             yaml.dump(config, f, Dumper=Dumper)
+    f.close()
 
 
 # read model_type from yaml file
@@ -47,6 +49,7 @@ def read_inference_type(uid):
     with open(get_yaml_path(uid), "r") as f:
         config = yaml.load(f, Loader=yaml.FullLoader)
         inference_type = config.get("inference_type", "")
+    f.close()
     return inference_type
 
 
@@ -60,6 +63,9 @@ def write_inference_type(use_inference, uid):
             config["inference_type"] = "hf_pipeline"
         # save inference_type to inference_type in yaml
         yaml.dump(config, f, Dumper=Dumper)
+    f.close()
+    return (gr.update(visible=(use_inference == "hf_inference_api")))
+
 
 
 # read column mapping from yaml file
@@ -69,6 +75,7 @@ def read_column_mapping(uid):
         config = yaml.load(f, Loader=yaml.FullLoader)
         if config:
             column_mapping = config.get("column_mapping", dict())
+    f.close()
     return column_mapping
 
 
@@ -85,6 +92,7 @@ def write_column_mapping(mapping, uid):
     with open(get_yaml_path(uid), "w") as f:
         # save column_mapping to column_mapping in yaml
         yaml.dump(config, f, Dumper=Dumper)
+    f.close()
 
 
 # convert column mapping dataframe to json
@@ -107,6 +115,7 @@ def get_logs_file(uid):
 def write_log_to_user_file(id, log):
     with open(f"./tmp/{id}_log", "a") as f:
         f.write(log)
+    f.close()
 
 
 def save_job_to_pipe(id, job, lock):
@@ -120,8 +129,6 @@ def pop_job_from_pipe():
     job_info = pipe.jobs.pop()
     write_log_to_user_file(job_info[0], f"Running job id {job_info[0]}\n")
     command = job_info[1]
-    print(f"Running job id {job_info[0]}")
-    print(f"Running command {command}")
 
     log_file = open(f"./tmp/{job_info[0]}_log", "a")
     subprocess.Popen(

text_classification_ui_helpers.py

@@ -188,6 +188,10 @@ def try_submit(m_id, d_id, config, split, local, uid):
         return (gr.update(interactive=True), gr.update(visible=False))
     feature_mapping = all_mappings["features"]
 
+    leaderboard_dataset = None
+    if os.environ.get("SPACE_ID") == "giskardai/giskard-evaluator":
+        leaderboard_dataset = "ZeroCommand/test-giskard-report"
+
     # TODO: Set column mapping for some dataset such as `amazon_polarity`
     if local:
         command = [
@@ -216,6 +220,8 @@ def try_submit(m_id, d_id, config, split, local, uid):
             json.dumps(label_mapping),
             "--scan_config",
             get_yaml_path(uid),
+            "--leaderboard_dataset",
+            leaderboard_dataset,
         ]
 
         eval_str = f"[{m_id}]<{d_id}({config}, {split} set)>"