metadata
library_name: transformers
license: mit
base_model: m3rg-iitd/matscibert
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: MatSciBERT_ST_DA_1000
results: []
MatSciBERT_ST_DA_1000
This model is a fine-tuned version of m3rg-iitd/matscibert on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1669
- Precision: 0.8484
- Recall: 0.8572
- F1: 0.8528
- Accuracy: 0.9724
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: 2e-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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 495 | 0.1097 | 0.8373 | 0.8310 | 0.8341 | 0.9692 |
| 0.1746 | 2.0 | 990 | 0.0968 | 0.8355 | 0.8550 | 0.8452 | 0.9720 |
| 0.0592 | 3.0 | 1485 | 0.1072 | 0.8405 | 0.8497 | 0.8451 | 0.9711 |
| 0.0316 | 4.0 | 1980 | 0.1302 | 0.8451 | 0.8468 | 0.8459 | 0.9709 |
| 0.017 | 5.0 | 2475 | 0.1426 | 0.8381 | 0.8448 | 0.8415 | 0.9702 |
| 0.0102 | 6.0 | 2970 | 0.1503 | 0.8456 | 0.8470 | 0.8463 | 0.9711 |
| 0.0058 | 7.0 | 3465 | 0.1528 | 0.8466 | 0.8509 | 0.8487 | 0.9721 |
| 0.0035 | 8.0 | 3960 | 0.1565 | 0.8459 | 0.8521 | 0.8490 | 0.9719 |
| 0.0027 | 9.0 | 4455 | 0.1592 | 0.8531 | 0.8562 | 0.8547 | 0.9728 |
| 0.0017 | 10.0 | 4950 | 0.1669 | 0.8484 | 0.8572 | 0.8528 | 0.9724 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1