Upload kaggle_auto.ipynb with huggingface_hub

kaggle_auto.ipynb

@@ -0,0 +1,81 @@
+# 🎵 Spotify Genre Classifier - Automatic Training
+# 
+# Kaggle Notebook: Just run this single cell!
+# 
+# Setup:
+# 1. Add secrets in Kaggle: HF_TOKEN and HF_USERNAME
+# 2. Run this cell
+# 3. Wait for training to complete (~15-20 min on GPU)
+# 4. Model will be saved and optionally pushed to HF Hub
+
+print("=" * 60)
+print("  Spotify Genre Classifier - Automatic Training")
+print("=" * 60)
+
+# Step 1: Get secrets
+from kaggle_secrets import UserSecretsClient
+import os
+
+user_secrets = UserSecretsClient()
+os.environ['HF_TOKEN'] = user_secrets.get_secret("HF_TOKEN")
+os.environ['HF_USERNAME'] = user_secrets.get_secret("HF_USERNAME")
+
+print(f"\n✓ Logged in as: {os.environ['HF_USERNAME']}")
+
+# Step 2: Install dependencies
+print("\n📦 Installing dependencies...")
+!pip install -q transformers datasets accelerate evaluate scikit-learn python-dotenv tqdm
+
+# Step 3: Clone repo
+print("\n📥 Cloning training repo...")
+!git clone https://huggingface.co/maxxcarl/spotify-training
+%cd spotify-training
+
+# Step 4: Check GPU
+print("\n🔍 Checking GPU...")
+import torch
+if torch.cuda.is_available():
+    print(f"✓ GPU: {torch.cuda.get_device_name(0)}")
+else:
+    print("⚠ No GPU - using CPU")
+
+# Step 5: Run training
+print("\n" + "=" * 60)
+print("  Starting Training")
+print("=" * 60)
+
+!python src/training_pipeline.py gpt2_spotify
+
+# Step 6: Test model
+print("\n" + "=" * 60)
+print("  Testing Model")
+print("=" * 60)
+
+!python test_model.py outputs/final_model
+
+# Step 7: Push to Hub (optional)
+print("\n" + "=" * 60)
+print("  Push to Hugging Face Hub?")
+print("=" * 60)
+
+from huggingface_hub import login
+
+hf_token = os.environ['HF_TOKEN']
+username = os.environ['HF_USERNAME']
+
+login(token=hf_token)
+
+repo_name = "spotify-genre-classifier"
+print(f"\nPushing to: {username}/{repo_name}")
+
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+model = AutoModelForSequenceClassification.from_pretrained("./outputs/final_model")
+tokenizer = AutoTokenizer.from_pretrained("./outputs/final_model")
+
+model.push_to_hub(f"{username}/{repo_name}")
+tokenizer.push_to_hub(f"{username}/{repo_name}")
+
+print(f"\n✅ Complete!")
+print(f"📊 Model: https://huggingface.co/{username}/{repo_name}")
+print(f"🌐 Space: https://huggingface.co/spaces/{username}/pool")