metadata

library_name: transformers
license: apache-2.0
base_model: c14kevincardenas/beit-large-patch16-384-limb-person-crop
tags:
  - image-sequence-classification
  - vision
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: finetuned-beit-limb-seq-t4-2heads-2layers-1e-4lr
    results: []

finetuned-beit-limb-seq-t4-2heads-2layers-1e-4lr

This model is a fine-tuned version of c14kevincardenas/beit-large-patch16-384-limb-person-crop on the c14kevincardenas/beta_caller_284_person_crop_seq_withlimb_4 dataset. It achieves the following results on the evaluation set:

Loss: 0.3905
Accuracy: 0.9198

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: 0.0001
train_batch_size: 32
eval_batch_size: 32
seed: 2014
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 250
num_epochs: 20.0
mixed_precision_training: Native AMP
label_smoothing_factor: 0.05

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.728	1.0	148	0.4692	0.8934
0.4398	2.0	296	0.4154	0.9174
0.4817	3.0	444	0.4183	0.9114
0.4336	4.0	592	0.4365	0.9006
0.4266	5.0	740	0.4001	0.9186
0.4375	6.0	888	0.4102	0.9090
0.3893	7.0	1036	0.4297	0.9078
0.4142	8.0	1184	0.3987	0.9222
0.4409	9.0	1332	0.3905	0.9198
0.3839	10.0	1480	0.3984	0.9126
0.3798	11.0	1628	0.4075	0.9138
0.4059	12.0	1776	0.3997	0.9114
0.3785	13.0	1924	0.4169	0.9114
0.3823	14.0	2072	0.4268	0.9102
0.3601	15.0	2220	0.4115	0.9150
0.3725	16.0	2368	0.3957	0.9246
0.3321	17.0	2516	0.4084	0.9138
0.3718	18.0	2664	0.4134	0.9102
0.3605	19.0	2812	0.4041	0.9150
0.3167	20.0	2960	0.4074	0.9186

Framework versions

Transformers 4.45.2
Pytorch 2.5.0+cu124
Datasets 3.0.1
Tokenizers 0.20.1