metadata
license: mit
base_model: ai-forever/ruElectra-medium
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/ruElectra-medium on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.6537
- Accuracy: 0.6901
- Recall: 0.6451
- Precision: 0.6599
- F1: 0.6390
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 | 100 | 1.3590 | 0.5643 | 0.3617 | 0.3821 | 0.3270 |
| No log | 2.0 | 200 | 0.9903 | 0.6637 | 0.5263 | 0.5238 | 0.5058 |
| No log | 3.0 | 300 | 0.9370 | 0.6842 | 0.5254 | 0.5367 | 0.5185 |
| No log | 4.0 | 400 | 0.9366 | 0.7047 | 0.5982 | 0.5655 | 0.5675 |
| 0.9611 | 5.0 | 500 | 1.0894 | 0.6901 | 0.5707 | 0.5656 | 0.5529 |
| 0.9611 | 6.0 | 600 | 1.1565 | 0.7018 | 0.5834 | 0.5569 | 0.5649 |
| 0.9611 | 7.0 | 700 | 1.1471 | 0.7076 | 0.5887 | 0.5565 | 0.5687 |
| 0.9611 | 8.0 | 800 | 1.2477 | 0.7281 | 0.6326 | 0.7122 | 0.6341 |
| 0.9611 | 9.0 | 900 | 1.3606 | 0.7310 | 0.6556 | 0.7163 | 0.6484 |
| 0.1529 | 10.0 | 1000 | 1.7044 | 0.6725 | 0.6059 | 0.6230 | 0.5964 |
| 0.1529 | 11.0 | 1100 | 1.5851 | 0.7193 | 0.6600 | 0.6571 | 0.6548 |
| 0.1529 | 12.0 | 1200 | 1.7624 | 0.6959 | 0.6463 | 0.6714 | 0.6457 |
| 0.1529 | 13.0 | 1300 | 1.9156 | 0.6988 | 0.6312 | 0.6636 | 0.6360 |
| 0.1529 | 14.0 | 1400 | 1.8304 | 0.7251 | 0.6525 | 0.6899 | 0.6586 |
| 0.0417 | 15.0 | 1500 | 1.9549 | 0.7164 | 0.6442 | 0.6758 | 0.6485 |
| 0.0417 | 16.0 | 1600 | 1.9306 | 0.7398 | 0.6569 | 0.7047 | 0.6639 |
| 0.0417 | 17.0 | 1700 | 2.1130 | 0.6959 | 0.6591 | 0.6904 | 0.6556 |
| 0.0417 | 18.0 | 1800 | 1.9658 | 0.7368 | 0.6312 | 0.7479 | 0.6545 |
| 0.0417 | 19.0 | 1900 | 2.0108 | 0.7281 | 0.6497 | 0.7180 | 0.6605 |
| 0.0149 | 20.0 | 2000 | 2.0183 | 0.7368 | 0.6757 | 0.7038 | 0.6832 |
| 0.0149 | 21.0 | 2100 | 2.1543 | 0.7222 | 0.7085 | 0.6745 | 0.6824 |
| 0.0149 | 22.0 | 2200 | 1.9347 | 0.7485 | 0.6518 | 0.7867 | 0.6722 |
| 0.0149 | 23.0 | 2300 | 1.8752 | 0.7690 | 0.6852 | 0.7686 | 0.7024 |
| 0.0149 | 24.0 | 2400 | 2.0048 | 0.7544 | 0.6834 | 0.7379 | 0.6966 |
| 0.0111 | 25.0 | 2500 | 2.0534 | 0.7515 | 0.6635 | 0.7640 | 0.6841 |
| 0.0111 | 26.0 | 2600 | 2.0457 | 0.7368 | 0.6503 | 0.6918 | 0.6586 |
| 0.0111 | 27.0 | 2700 | 2.1561 | 0.7368 | 0.6657 | 0.6990 | 0.6678 |
| 0.0111 | 28.0 | 2800 | 2.1431 | 0.7398 | 0.6590 | 0.6734 | 0.6604 |
| 0.0111 | 29.0 | 2900 | 2.3783 | 0.7135 | 0.6544 | 0.6643 | 0.6509 |
| 0.0103 | 30.0 | 3000 | 2.3847 | 0.7251 | 0.6368 | 0.7351 | 0.6597 |
| 0.0103 | 31.0 | 3100 | 2.2030 | 0.7427 | 0.7017 | 0.7082 | 0.7023 |
| 0.0103 | 32.0 | 3200 | 2.4123 | 0.7368 | 0.6679 | 0.6974 | 0.6697 |
| 0.0103 | 33.0 | 3300 | 2.2644 | 0.7398 | 0.6760 | 0.7428 | 0.6902 |
| 0.0103 | 34.0 | 3400 | 2.3744 | 0.7339 | 0.6847 | 0.7080 | 0.6800 |
| 0.0135 | 35.0 | 3500 | 2.1573 | 0.7485 | 0.6933 | 0.6932 | 0.6867 |
| 0.0135 | 36.0 | 3600 | 2.1728 | 0.7515 | 0.6649 | 0.7606 | 0.6802 |
| 0.0135 | 37.0 | 3700 | 2.0993 | 0.7719 | 0.6859 | 0.7705 | 0.6972 |
| 0.0135 | 38.0 | 3800 | 2.6537 | 0.6901 | 0.6451 | 0.6599 | 0.6390 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.14.1