Edit model card

videomae-base-finetuned-good-gestureUnitV11

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:

  • Accuracy: 0.9310
  • Loss: 0.3025
  • Accuracy Gunit: 0.8571
  • Accuracy Nothing: 1.0

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: 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: 80

Training results

Training Loss Epoch Step Accuracy Validation Loss Accuracy Gunit Accuracy Nothing
0.6685 0.2125 17 0.6071 0.6297 1.0 0.2903
0.5348 1.2125 34 0.8214 0.4292 0.6 1.0
0.3395 2.2125 51 0.7857 0.4852 0.88 0.7097
0.3607 3.2125 68 0.8214 0.4507 0.6 1.0
0.2436 4.15 80 0.7857 0.4433 0.76 0.8065

Framework versions

  • Transformers 4.41.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
1
Safetensors
Model size
86.2M params
Tensor type
F32
·
This model can be loaded on Inference API (serverless).

Finetuned from