videomae-base-finetuned-yt_short_classification-2

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: 0.4733
• Accuracy: 0.7818
• 0 Precision: 0.7333
• 0 Recall: 0.8515
• 0 F1-score: 0.7880
• 0 Support: 6322.0
• 1 Precision: 0.8419
• 1 Recall: 0.7186
• 1 F1-score: 0.7753
• 1 Support: 6957.0
• Accuracy F1-score: 0.7818
• Macro avg Precision: 0.7876
• Macro avg Recall: 0.7850
• Macro avg F1-score: 0.7817
• Macro avg Support: 13279.0
• Weighted avg Precision: 0.7902
• Weighted avg Recall: 0.7818
• Weighted avg F1-score: 0.7814
• Weighted avg Support: 13279.0

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: 16
• eval_batch_size: 16
• 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: 4120

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	0 Precision	0 Recall	0 F1-score	0 Support	1 Precision	1 Recall	1 F1-score	1 Support	Accuracy F1-score	Macro avg Precision	Macro avg Recall	Macro avg F1-score	Macro avg Support	Weighted avg Precision	Weighted avg Recall	Weighted avg F1-score	Weighted avg Support
0.6122	0.1002	413	0.7143	0.6551	0.5925	0.8828	0.7091	6322.0	0.8080	0.4482	0.5766	6957.0	0.6551	0.7002	0.6655	0.6428	13279.0	0.7054	0.6551	0.6396	13279.0
0.6904	1.1002	826	0.5800	0.6909	0.8170	0.4519	0.5819	6322.0	0.6458	0.9080	0.7548	6957.0	0.6909	0.7314	0.6800	0.6683	13279.0	0.7273	0.6909	0.6725	13279.0
0.5489	2.1002	1239	0.5122	0.7555	0.7450	0.7395	0.7422	6322.0	0.7648	0.7700	0.7674	6957.0	0.7555	0.7549	0.7547	0.7548	13279.0	0.7554	0.7555	0.7554	13279.0
0.4979	3.1002	1652	0.5434	0.7443	0.6785	0.8798	0.7662	6322.0	0.8505	0.6212	0.7180	6957.0	0.7443	0.7645	0.7505	0.7421	13279.0	0.7686	0.7443	0.7409	13279.0
0.5141	4.1002	2065	0.4793	0.7723	0.7482	0.7866	0.7669	6322.0	0.7966	0.7594	0.7775	6957.0	0.7723	0.7724	0.7730	0.7722	13279.0	0.7735	0.7723	0.7725	13279.0
0.4472	5.1002	2478	0.4673	0.7798	0.7398	0.8290	0.7819	6322.0	0.8255	0.7351	0.7777	6957.0	0.7798	0.7827	0.7820	0.7798	13279.0	0.7847	0.7798	0.7797	13279.0
0.4108	6.1002	2891	0.4491	0.7952	0.7715	0.8096	0.7901	6322.0	0.8188	0.7821	0.8000	6957.0	0.7952	0.7951	0.7958	0.7950	13279.0	0.7963	0.7952	0.7953	13279.0
0.3756	7.1002	3304	0.4955	0.7773	0.7472	0.8045	0.7748	6322.0	0.8090	0.7526	0.7798	6957.0	0.7773	0.7781	0.7786	0.7773	13279.0	0.7796	0.7773	0.7774	13279.0
0.3147	8.1002	3717	0.5889	0.7318	0.6523	0.9348	0.7684	6322.0	0.9023	0.5472	0.6813	6957.0	0.7318	0.7773	0.7410	0.7249	13279.0	0.7833	0.7318	0.7228	13279.0
0.4019	9.0978	4120	0.4733	0.7818	0.7333	0.8515	0.7880	6322.0	0.8419	0.7186	0.7753	6957.0	0.7818	0.7876	0.7850	0.7817	13279.0	0.7902	0.7818	0.7814	13279.0

Framework versions

• Transformers 4.46.3
• Pytorch 2.0.0+cu117
• Datasets 3.1.0
• Tokenizers 0.20.3