VideoMAE-v2 (Base) — HMDB51 (51 classes) Fine-tuned Checkpoint

Contents

• best_videomaev2_base_51cls.pt — raw checkpoint (torch.load)
• checkpoint_meta.json — metadata (best epoch, best val accuracy, etc.)
• training_results.csv — epoch-level train/val accuracy
• label2id.json / id2label.json — label maps (if available)

Training results

• Best validation accuracy: 0.8626 (epoch 10)

Download & load

Notes

• This repo stores a raw PyTorch checkpoint (not necessarily a save_pretrained() Transformers folder).
• Use the included metadata + label maps to reconstruct the training/inference code you used in Kaggle.