videomae-timesformer-surf-analytics

This model is a fine-tuned version of facebook/timesformer-base-finetuned-k400 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6192
• Accuracy: 0.8382

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 925

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4491	0.2011	186	0.6939	0.7386
0.5627	1.2011	372	0.6806	0.7759
0.5189	2.2011	558	0.6510	0.8174
0.2503	3.2011	744	0.6732	0.8174
0.0159	4.1957	925	0.6192	0.8382

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.0+cpu
• Datasets 2.19.2
• Tokenizers 0.19.1