Upload kaggle_single_cell.py with huggingface_hub

kaggle_single_cell.py

@@ -0,0 +1,56 @@
+# Kaggle Auto-Training Notebook
+# Copy this entire file into a single Kaggle notebook cell
+
+print("=" * 60)
+print("  Spotify Genre Classifier - Automatic Training")
+print("=" * 60)
+
+# 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']}")
+
+# Install
+print("\n📦 Installing...")
+!pip install -q transformers datasets accelerate evaluate scikit-learn python-dotenv tqdm
+
+# Clone
+print("\n📥 Cloning...")
+!git clone https://huggingface.co/maxxcarl/spotify-training
+%cd spotify-training
+
+# Check GPU
+print("\n🔍 GPU:")
+import torch
+if torch.cuda.is_available():
+    print(f"✓ {torch.cuda.get_device_name(0)}")
+else:
+    print("⚠ CPU only")
+
+# Train
+print("\n🚀 Training...")
+!python src/training_pipeline.py gpt2_spotify
+
+# Test
+print("\n📈 Testing...")
+!python test_model.py outputs/final_model
+
+# Push to Hub
+print("\n💾 Pushing to Hub...")
+from huggingface_hub import login
+login(token=os.environ['HF_TOKEN'])
+
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+model = AutoModelForSequenceClassification.from_pretrained("./outputs/final_model")
+tokenizer = AutoTokenizer.from_pretrained("./outputs/final_model")
+
+username = os.environ['HF_USERNAME']
+model.push_to_hub(f"{username}/spotify-genre-classifier")
+tokenizer.push_to_hub(f"{username}/spotify-genre-classifier")
+
+print(f"\n✅ Done! https://huggingface.co/{username}/spotify-genre-classifier")