Add Hugging Face Dataset integration for storing submissions

app.py

@@ -4,11 +4,36 @@ import json
 import uuid
 from datetime import datetime
 import shutil
+from huggingface_hub import HfApi, create_repo, upload_file, upload_folder
 
 # Create directories for data storage
 os.makedirs("uploaded_images", exist_ok=True)
 os.makedirs("submissions", exist_ok=True)
 
+# Hugging Face Dataset configuration
+HF_TOKEN = os.environ.get("HF_TOKEN")  # You'll need to set this as a secret in your Space
+DATASET_NAME = "srishtiy/se-culture-dataset-results"  # Change to your username/dataset-name
+DATASET_CREATED = False
+
+def setup_hf_dataset():
+    """Initialize the Hugging Face dataset if it doesn't exist"""
+    global DATASET_CREATED
+    if not DATASET_CREATED and HF_TOKEN:
+        try:
+            api = HfApi()
+            create_repo(
+                DATASET_NAME,
+                repo_type="dataset",
+                token=HF_TOKEN,
+                exist_ok=True
+            )
+            DATASET_CREATED = True
+            print(f"Dataset {DATASET_NAME} is ready")
+        except Exception as e:
+            print(f"Error setting up dataset: {e}")
+    elif not HF_TOKEN:
+        print("Warning: HF_TOKEN not set. Data will be stored locally only.")
+
 def process_submission(input_img, text_answer, multiple_choice, city, country, se_asia_relevance, culture_knowledge, native_caption, english_caption):
     # Generate unique ID for this submission
     submission_id = str(uuid.uuid4())
@@ -31,7 +56,7 @@ def process_submission(input_img, text_answer, multiple_choice, city, country, s
     submission_data = {
         "id": submission_id,
         "timestamp": timestamp,
-        "image_filename": image_path,
+        "image_filename": os.path.basename(image_path) if image_path else None,
         "cultural_relevance": text_answer,
         "continent": multiple_choice,
         "city": city,
@@ -48,9 +73,37 @@ def process_submission(input_img, text_answer, multiple_choice, city, country, s
     with open(json_path, "w") as f:
         json.dump(submission_data, f, indent=2)
     
+    # Upload to Hugging Face Dataset if token is available
+    if HF_TOKEN and DATASET_CREATED:
+        try:
+            api = HfApi()
+            # Upload the JSON data
+            api.upload_file(
+                path_or_fileobj=json_path,
+                path_in_repo=f"submissions/{json_filename}",
+                repo_id=DATASET_NAME,
+                repo_type="dataset",
+                token=HF_TOKEN
+            )
+            # Upload the image if it exists
+            if image_path and os.path.exists(image_path):
+                api.upload_file(
+                    path_or_fileobj=image_path,
+                    path_in_repo=f"images/{os.path.basename(image_path)}",
+                    repo_id=DATASET_NAME,
+                    repo_type="dataset",
+                    token=HF_TOKEN
+                )
+            print(f"Submission {submission_id} uploaded to Hugging Face Dataset")
+        except Exception as e:
+            print(f"Error uploading to dataset: {e}")
+    
     # Return values to display in the interface
     return input_img, f"Your text response: {text_answer}", f"Your selected option: {multiple_choice}", f"Location: {city}, {country}", f"SE Asia relevance: {se_asia_relevance}", f"Cultural knowledge source: {culture_knowledge}", f"Native caption: {native_caption}", f"English caption: {english_caption}"
 
+# Initialize the dataset
+setup_hf_dataset()
+
 gradio_app = gr.Interface(
     process_submission,
     inputs=[