jatinmehra/Accident-Detection-using-Dashcam
Full fine-tune vs MCG-NJU/videomae-large-finetuned-kinetics. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Norms 98/98, Embeddings 2/2, Attention 168/168, other 96/96
Full fine-tune vs MCG-NJU/videomae-large-finetuned-kinetics. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Norms 98/98, Embeddings 2/2, Attention 168/168, other 96/96