videomae-base-finetuned-IEMOCAP_videos

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: 1.3290
• Accuracy: 0.3497

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 4070

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4698	0.1002	408	1.4911	0.2071
1.396	1.1002	816	1.3825	0.2621
1.2776	2.1002	1224	1.4461	0.2714
1.2976	3.1002	1632	1.4535	0.2632
1.34	4.1002	2040	1.4480	0.2527
1.3607	5.1002	2448	1.3579	0.3121
1.2931	6.1002	2856	1.3612	0.3159
1.3231	7.1002	3264	1.3727	0.2582
1.3491	8.1002	3672	1.4132	0.2451
1.2063	9.0978	4070	1.4588	0.2582

Framework versions

• Transformers 4.40.2
• Pytorch 2.0.1+cu117
• Datasets 2.19.1
• Tokenizers 0.19.1