Create app.py

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+import pandas as pd
+from datasets import Dataset, DatasetDict
+from huggingface_hub import notebook_login, HfApi
+
+def process_and_upload(json_file_path, hf_username, dataset_name):
+    # Load your combined JSON file
+    df = pd.read_json(json_file_path.name)
+    print(f"DataFrame shape: {df.shape}")
+    print(f"DataFrame columns: {df.columns.tolist()}")
+
+    # Create HuggingFace Dataset
+    dataset = Dataset.from_pandas(df)
+    print("Dataset created successfully!")
+
+    # Create a DatasetDict with a 'train' split
+    dataset_dict = DatasetDict({"train": dataset})
+
+    # Log in to Hugging Face (or use existing token if available)
+    try:
+        notebook_login()
+    except:
+        # Assume user has already logged in or token is set
+        pass
+
+    # Push to the Hugging Face Hub
+    api = HfApi()
+    api.create_repo(repo_id=f"{hf_username}/{dataset_name}", repo_type="dataset", private=False, exist_ok=True)
+    dataset_dict.push_to_hub(f"{hf_username}/{dataset_name}", private=False)
+
+    return f"Dataset '{dataset_name}' uploaded to Hugging Face Hub under user '{hf_username}'"
+
+with gr.Blocks() as demo:
+    gr.Markdown("## Upload and Process JSON to Hugging Face Dataset")
+    with gr.Row():
+        json_file_input = gr.File(label="Select JSON file")
+        hf_username_input = gr.Textbox(label="Hugging Face Username", placeholder="Your HF username")
+        dataset_name_input = gr.Textbox(label="Dataset Name", placeholder="Name for your dataset")
+    submit_button = gr.Button("Upload to Hugging Face")
+    output_label = gr.Label(label="Output")
+
+    submit_button.click(
+        process_and_upload,
+        inputs=[json_file_input, hf_username_input, dataset_name_input],
+        outputs=output_label
+    )
+
+demo.launch()