Modulized subprocess

.gitignore

@@ -0,0 +1,4 @@
+Environment/
+out/
+flagged/
+

app.py

@@ -1,16 +1,25 @@
 import gradio as gr
 import subprocess
+import os
+import sys
 
-dataset = "truthful_qa"
-def greet(model):
-    global dataset
-    command = f"python detect-pretrain-code-contamination/src/run.py --target_model {model} --ref_model roneneldan/TinyStories-Instruct-1M --data {dataset} --output_dir detect-pretrain-code-contamination/out/{dataset} --ratio_gen 0.4"
-    result = subprocess.run(command, shell=True,capture_output=True, text=True)
-    
-    if result.returncode == 0:
-        return result.stdout
-    else:
-        return f"Error: {result.stderr}"
+# Add the path to the "src" directory of detect-pretrain-code-contamination to the sys.path
+project_root = os.path.abspath(os.path.join(os.path.dirname(__file__), "detect-pretrain-code-contamination"))
+src_dir = os.path.join(project_root, "src")
+sys.path.insert(0, src_dir)
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+import run as evaluator  # Import the run module
+
+def evaluate(model):
+    return evaluator.main(
+                target_model="roneneldan/TinyStories-1M",
+                ref_model="roneneldan/TinyStories-Instruct-1M",
+                output_dir="out",
+                data="truthful_qa",
+                length=64,
+                key_name="input",
+                ratio_gen=0.4
+            )  # Call the run_main function directly
+
+iface = gr.Interface(fn=evaluate, inputs="text", outputs="text")
 iface.launch()

requirements.txt

@@ -7,3 +7,4 @@ ipdb
 matplotlib
 scikit-learn
 accelerate
+gradio