videomae-base-finetuned-cricket_shot_detection_12_latest

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: 1.4676
• Accuracy: 0.6316
• F1: 0.6366
• Recall: 0.6316
• Precision: 0.7675

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-06
• 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: 576

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
1.7629	0.1267	73	1.8203	0.2105	0.1815	0.2105	0.1974
1.5269	1.1267	146	1.7099	0.2632	0.2731	0.2632	0.5474
1.2084	2.1267	219	1.5873	0.5789	0.5797	0.5789	0.7570
1.1034	3.1267	292	1.5001	0.5789	0.5797	0.5789	0.7570
0.9862	4.1267	365	1.4676	0.6316	0.6366	0.6316	0.7675
0.891	5.1267	438	1.4256	0.6316	0.6366	0.6316	0.7675
0.6645	6.1267	511	1.4055	0.6316	0.6476	0.6316	0.8184
0.7161	7.1128	576	1.4041	0.6316	0.6366	0.6316	0.7675

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.19.1