metadata

license: apache-2.0
base_model: facebook/convnextv2-base-22k-224
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: convnextv2-base-22k-224-finetuned-tekno24
    results: []

convnextv2-base-22k-224-finetuned-tekno24

This model is a fine-tuned version of facebook/convnextv2-base-22k-224 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.9468
Accuracy: 0.5932
F1: 0.5674
Precision: 0.5709
Recall: 0.5932

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
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 12
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.2755	0.9963	68	1.2008	0.4904	0.3910	0.4577	0.4904
1.1711	1.9927	136	1.0651	0.5354	0.4766	0.4865	0.5354
1.1599	2.9890	204	1.0533	0.5418	0.5077	0.5275	0.5418
1.1595	4.0	273	1.0423	0.5455	0.5414	0.5691	0.5455
1.096	4.9963	341	1.0160	0.5611	0.5463	0.5419	0.5611
1.0592	5.9927	409	0.9847	0.5767	0.5415	0.5485	0.5767
1.0706	6.9890	477	0.9868	0.5886	0.5836	0.5862	0.5886
1.0404	8.0	546	0.9758	0.5868	0.5737	0.5695	0.5868
1.0059	8.9963	614	0.9468	0.5932	0.5674	0.5709	0.5932
0.965	9.9927	682	0.9565	0.5932	0.5804	0.5858	0.5932
0.9362	10.9890	750	0.9466	0.5886	0.5778	0.5768	0.5886
0.9334	11.9560	816	0.9442	0.5859	0.5700	0.5692	0.5859

Framework versions

Transformers 4.42.4
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.19.1