metadata

license: apache-2.0
base_model: ai-forever/ruBert-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - recall
  - precision
  - f1
model-index:
  - name: training_results
    results: []

training_results

This model is a fine-tuned version of ai-forever/ruBert-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.1162
Accuracy: 0.7193
Recall: 0.7342
Precision: 0.7262
F1: 0.7271

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
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1
No log	1.0	200	0.7908	0.7251	0.6546	0.6651	0.6511
No log	2.0	400	0.9332	0.6988	0.6728	0.7010	0.6670
0.6701	3.0	600	1.0779	0.7427	0.7348	0.7912	0.7473
0.6701	4.0	800	1.2092	0.7427	0.6989	0.7237	0.7037
0.1446	5.0	1000	1.5001	0.7281	0.7049	0.7784	0.7231
0.1446	6.0	1200	1.5468	0.7368	0.7078	0.7808	0.7256
0.1446	7.0	1400	1.7923	0.7193	0.7284	0.7360	0.7230
0.0354	8.0	1600	1.7583	0.7339	0.7357	0.7462	0.7357
0.0354	9.0	1800	1.7942	0.7485	0.7203	0.7576	0.7308
0.0259	10.0	2000	1.9056	0.7310	0.7059	0.7337	0.7143
0.0259	11.0	2200	2.0351	0.7018	0.7047	0.6977	0.6952
0.0259	12.0	2400	1.6337	0.7602	0.7246	0.7861	0.7338
0.0262	13.0	2600	1.9012	0.7251	0.7056	0.7611	0.7148
0.0262	14.0	2800	2.0006	0.7339	0.7022	0.7625	0.7166
0.0302	15.0	3000	2.2857	0.6842	0.7072	0.6874	0.6766
0.0302	16.0	3200	2.0855	0.7310	0.7168	0.7557	0.7232
0.0302	17.0	3400	2.1281	0.7398	0.6868	0.7629	0.7132
0.0321	18.0	3600	2.1162	0.7193	0.7342	0.7262	0.7271

Framework versions

Transformers 4.34.0
Pytorch 2.1.0+cu121
Datasets 2.14.5
Tokenizers 0.14.1