videomae-large_14class_UCFCrime

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

• eval_loss: 2.1336
• eval_confusion_matrix: {'confusion_matrix': array([[ 5, 6, 0, 11, 5, 1, 17, 1, 0, 0, 0, 6, 0, 0], [ 25, 76, 0, 9, 53, 0, 36, 1, 24, 1, 0, 13, 5, 0], [ 0, 0, 31, 0, 2, 0, 0, 0, 0, 0, 0, 0, 9, 7], [ 11, 21, 0, 52, 3, 0, 40, 2, 1, 1, 0, 0, 19, 1], [ 21, 9, 0, 1, 113, 0, 21, 2, 0, 30, 0, 1, 12, 43], [ 1, 2, 13, 0, 0, 39, 1, 0, 0, 1, 0, 0, 12, 1], [ 3, 15, 1, 23, 8, 0, 58, 0, 0, 6, 0, 23, 1, 4], [ 0, 0, 0, 0, 0, 0, 0, 430, 0, 0, 0, 0, 0, 0], [ 3, 11, 1, 2, 1, 2, 3, 0, 25, 3, 0, 4, 7, 0], [ 0, 19, 0, 9, 17, 0, 18, 57, 0, 22, 0, 56, 0, 0], [ 0, 11, 0, 6, 9, 0, 3, 3, 2, 0, 5, 3, 24, 0], [ 0, 5, 0, 0, 0, 0, 5, 11, 0, 9, 0, 40, 0, 0], [ 0, 40, 2, 19, 9, 0, 9, 44, 3, 0, 1, 30, 120, 11], [ 0, 4, 1, 12, 2, 1, 0, 0, 0, 7, 0, 11, 6, 68]])}
• eval_runtime: 888.8746
• eval_samples_per_second: 2.459
• eval_steps_per_second: 1.23
• step: 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: 2
• eval_batch_size: 2
• 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: 5560

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2