videomae-base-finetuned-FT_2PT_3PT

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8099
• Accuracy: 0.7290

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1320

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9402	0.0508	67	0.9318	0.5484
0.7357	1.0508	134	0.8826	0.5871
0.9663	2.0508	201	0.8852	0.5484
0.6539	3.0508	268	0.7543	0.6645
0.5231	4.0508	335	0.7400	0.6323
0.5684	5.0508	402	1.2109	0.6
0.5058	6.0508	469	1.3525	0.5935
0.3816	7.0508	536	0.7959	0.6710
0.5652	8.0508	603	0.8605	0.6581
0.6047	9.0508	670	0.9432	0.6194
0.4412	10.0508	737	1.0180	0.5742
0.3621	11.0508	804	1.3228	0.6065
0.521	12.0508	871	1.0286	0.6258
0.3356	13.0508	938	1.0859	0.6774
0.3303	14.0508	1005	0.9708	0.6710
0.3837	15.0508	1072	0.9323	0.6903
0.4748	16.0508	1139	0.8099	0.7290
0.3888	17.0508	1206	0.9717	0.7161
0.248	18.0508	1273	0.9719	0.6968
0.4058	19.0356	1320	0.9779	0.6968

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.1+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1