Action_model

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the action_class dataset. It achieves the following results on the evaluation set:

• Loss: 0.6087
• Accuracy: 0.8436

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2783	0.3731	100	1.2065	0.7153
0.9907	0.7463	200	0.8331	0.8102
0.8428	1.1194	300	0.7278	0.8260
0.7442	1.4925	400	0.6576	0.8172
0.6749	1.8657	500	0.6087	0.8436

Framework versions

• Transformers 4.41.1
• Pytorch 2.1.2
• Datasets 2.19.1
• Tokenizers 0.19.1