mvit_v2_s_transf_BLANK_RWF-2000_ONLY_PERSON

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

• Loss: 0.5499
• Accuracy: 0.7397
• F1: 0.7206
• Precision: 0.7909

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: 1e-05
• train_batch_size: 10
• eval_batch_size: 10
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 3040

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision
0.9144	0.0474	144	0.7011	0.5417	0.5175	0.5631
0.7346	1.0474	288	0.6937	0.575	0.5616	0.5785
0.7318	2.0474	432	0.6965	0.5417	0.5418	0.5419
0.6238	3.0474	576	0.7305	0.6083	0.5761	0.6401
0.6384	4.0474	720	0.7557	0.5417	0.5325	0.5515
0.759	5.0474	864	0.6498	0.5833	0.5824	0.5828
0.5834	6.0474	1008	0.7051	0.6333	0.6206	0.6457
0.5698	7.0474	1152	0.6900	0.6417	0.6159	0.6789
0.5315	8.0474	1296	0.6680	0.55	0.5501	0.5506
0.5826	9.0474	1440	0.6650	0.6333	0.6318	0.6334
0.579	10.0474	1584	0.7106	0.5917	0.5883	0.5992
0.6109	11.0474	1728	0.6842	0.6083	0.6022	0.6107
0.5732	12.0474	1872	0.7230	0.5333	0.5314	0.5369

Framework versions

• Transformers 4.44.2
• Pytorch 2.0.1+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1