Create app.py

app.py

@@ -0,0 +1,80 @@
+import gradio as gr
+import pandas as pd
+import io
+from huggingface_hub import HfApi, HfFolder, Repository
+import os
+
+# Global state
+df = pd.DataFrame()
+
+def upload_csv(file):
+    global df
+    df = pd.read_csv(file.name)
+    
+    if "text" not in df.columns or "label" not in df.columns:
+        return gr.update(visible=False), "CSV must contain 'text' and 'label' columns."
+    
+    # Fill label column if empty
+    df["label"] = df["label"].fillna("")
+    
+    # Return the editable table
+    return gr.Dataframe(
+        value=df,
+        headers=["text", "label"],
+        interactive=True,
+        label="Edit labels below"
+    ), "File uploaded successfully."
+
+def save_edits(updated_table):
+    global df
+    df = pd.DataFrame(updated_table, columns=["text", "label"])
+    return "Changes saved."
+
+def download_csv():
+    # Create a downloadable CSV
+    csv_bytes = df.to_csv(index=False).encode()
+    return gr.File.update(value=io.BytesIO(csv_bytes), filename="annotated_data.csv")
+
+def push_to_hub(repo_name, hf_token):
+    # Authenticate and push to Hugging Face Hub
+    repo_url = f"https://huggingface.co/datasets/{repo_name}"
+    local_path = f"./{repo_name}"
+    
+    if os.path.exists(local_path):
+        os.system(f"rm -rf {local_path}")
+    
+    api = HfApi()
+    api.create_repo(repo_id=repo_name, token=hf_token, repo_type="dataset", exist_ok=True)
+    repo = Repository(local_dir=local_path, clone_from=repo_url, token=hf_token)
+    df.to_csv(f"{local_path}/data.csv", index=False)
+    repo.push_to_hub()
+    
+    return f"Pushed to Hugging Face: {repo_url}"
+
+with gr.Blocks() as demo:
+    gr.Markdown("## 🏷️ CSV Text Labeling Tool")
+    
+    with gr.Row():
+        csv_input = gr.File(label="Upload CSV", file_types=[".csv"])
+        upload_btn = gr.Button("Upload")
+    
+    df_output = gr.Dataframe(headers=["text", "label"], interactive=True, visible=False)
+    upload_status = gr.Textbox(visible=True, interactive=False)
+    
+    with gr.Row():
+        save_btn = gr.Button("Save Changes")
+        download_btn = gr.Button("Download CSV")
+        download_file = gr.File(label="Download", interactive=False)
+    
+    with gr.Row():
+        hf_repo = gr.Textbox(label="HF Dataset Repo (e.g. your-username/my-dataset)")
+        hf_token = gr.Textbox(label="Hugging Face Token", type="password")
+        push_btn = gr.Button("Push to Hugging Face Hub")
+        push_status = gr.Textbox(interactive=False)
+    
+    upload_btn.click(fn=upload_csv, inputs=csv_input, outputs=[df_output, upload_status])
+    save_btn.click(fn=save_edits, inputs=df_output, outputs=upload_status)
+    download_btn.click(fn=download_csv, outputs=download_file)
+    push_btn.click(fn=push_to_hub, inputs=[hf_repo, hf_token], outputs=push_status)
+
+demo.launch()