sergeipetrov's picture
update model card README.md
4c48944
metadata
license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: convnextv2-base-22k-384-finetuned
    results: []

convnextv2-base-22k-384-finetuned

This model is a fine-tuned version of facebook/convnextv2-base-22k-384 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3257
  • Accuracy: 0.9611
  • F1: 0.9510
  • Precision: 0.9714
  • Recall: 0.9315

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: 0.00015
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3.0

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
No log 1.0 2 0.5986 0.8167 0.8092 0.7 0.9589
No log 2.0 4 0.3945 0.9611 0.9510 0.9714 0.9315
No log 3.0 6 0.3257 0.9611 0.9510 0.9714 0.9315

Framework versions

  • Transformers 4.28.1
  • Pytorch 2.0.0
  • Datasets 2.12.0
  • Tokenizers 0.13.3