metadata

license: cc-by-nc-sa-4.0
base_model: InstaDeepAI/nucleotide-transformer-v2-250m-multi-species
tags:
  - generated_from_trainer
metrics:
  - f1
  - matthews_correlation
  - accuracy
model-index:
  - name: gut_1024b-finetuned-lora-v2-250m-multi-species
    results: []

gut_1024b-finetuned-lora-v2-250m-multi-species

This model is a fine-tuned version of InstaDeepAI/nucleotide-transformer-v2-250m-multi-species on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.4815
F1: 0.8414
Matthews Correlation: 0.5610
Accuracy: 0.7880
F1 Score: 0.8414

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: 0.0005
train_batch_size: 8
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Matthews Correlation	Accuracy	F1 Score
0.682	0.02	100	0.5545	0.8132	0.4597	0.7369	0.8132
0.6379	0.04	200	0.6119	0.7498	0.4244	0.7154	0.7498
0.5973	0.05	300	0.5226	0.8221	0.5154	0.7707	0.8221
0.5451	0.07	400	0.5159	0.8244	0.5010	0.7521	0.8244
0.5538	0.09	500	0.5538	0.8102	0.5043	0.7648	0.8102
0.549	0.11	600	0.5220	0.8258	0.5188	0.7715	0.8258
0.4887	0.12	700	0.4940	0.8330	0.5317	0.7728	0.8330
0.4893	0.14	800	0.4951	0.8352	0.5519	0.7872	0.8352
0.4794	0.16	900	0.5008	0.8443	0.5687	0.7893	0.8443
0.5437	0.18	1000	0.4815	0.8414	0.5610	0.7880	0.8414

Framework versions

Transformers 4.38.1
Pytorch 2.1.0+cu121
Datasets 2.17.1
Tokenizers 0.15.2