videomae-finetuned_39

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

• Loss: 4.2762
• Accuracy: 0.3457

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.2
• training_steps: 54160

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5558	0.05	2709	4.2762	0.3457
0.5573	1.05	5418	5.4086	0.3340
0.5828	2.05	8127	5.5041	0.3205
0.5675	3.05	10836	6.0256	0.2988
0.5578	4.05	13545	6.6278	0.3198
0.3451	5.05	16254	6.5753	0.3027
0.4555	6.05	18963	6.7402	0.3112
0.5271	7.05	21672	7.1148	0.3037
0.3353	8.05	24381	6.9036	0.3254
0.3414	9.05	27090	6.6694	0.3281
0.3493	10.05	29799	6.8241	0.3345
0.2228	11.05	32508	7.3151	0.3240
0.0894	12.05	35217	7.5526	0.3269
0.2138	13.05	37926	7.0634	0.3276
0.295	14.05	40635	6.9772	0.3323
0.1493	15.05	43344	7.3242	0.3340
0.1968	16.05	46053	7.3585	0.3359
0.151	17.05	48762	7.4257	0.3413
0.0955	18.05	51471	7.5148	0.3386
0.0552	19.05	54160	7.4669	0.3401

Framework versions

• Transformers 4.38.1
• Pytorch 2.0.1+cu118
• Datasets 2.18.0
• Tokenizers 0.15.2