videomae-diving48-multilabel-finetuned

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

• Loss: 1.3969
• F1 Macro: 0.2636
• Precision Macro: 0.2180
• Recall Macro: 0.4145
• Exact Match Ratio: 0.0003
• Hamming Accuracy: 0.7557

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: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch 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: 25544
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro	Precision Macro	Recall Macro	Exact Match Ratio	Hamming Accuracy
1.3045	0.1250	3194	1.4774	0.1609	0.1518	0.2915	0.0	0.7280
0.8881	1.1250	6388	1.4002	0.2035	0.1952	0.3239	0.0	0.7537
1.0683	2.1250	9582	1.4017	0.2014	0.1999	0.3470	0.0	0.7403
0.7672	3.1250	12776	1.4316	0.2280	0.1893	0.3459	0.0022	0.7566
0.8529	4.1250	15970	1.4307	0.2333	0.2011	0.3484	0.0003	0.7650
1.0022	5.1250	19164	1.4566	0.2367	0.2095	0.3404	0.0005	0.7734
1.3422	6.1250	22358	1.4297	0.2602	0.2127	0.3922	0.0005	0.7634
1.0641	7.1247	25544	1.3969	0.2636	0.2180	0.4145	0.0003	0.7557

Framework versions

• Transformers 4.51.3
• Pytorch 2.1.0+cu118
• Datasets 3.6.0
• Tokenizers 0.21.1