BrianLan/yolos-tiny-finetuned-600-epochs-TW-Marine-5cls-dataset
Full fine-tune vs hustvl/yolos-tiny. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: other 58/58, Embeddings 5/5, Attention 96/96, Norms 50/50
Full fine-tune vs hustvl/yolos-tiny. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: other 58/58, Embeddings 5/5, Attention 96/96, Norms 50/50