metadata
license: mit
base_model: beomi/KcELECTRA-base-v2022
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: KcELECTRA-base-v2022-KEmoFact-0925
results: []
KcELECTRA-base-v2022-KEmoFact-0925
This model is a fine-tuned version of beomi/KcELECTRA-base-v2022 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0671
- Precision: 0.2272
- Recall: 0.2813
- F1: 0.2514
- Accuracy: 0.7244
- Jaccard Scores: 0.7016
- Cls Accuracy: 0.6433
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | Jaccard Scores | Cls Accuracy |
|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 414 | 1.0860 | 0.1223 | 0.1235 | 0.1229 | 0.7113 | 0.5185 | 0.4574 |
| 1.405 | 2.0 | 828 | 0.9584 | 0.1670 | 0.2076 | 0.1851 | 0.7275 | 0.6476 | 0.6001 |
| 0.8557 | 3.0 | 1242 | 0.9666 | 0.1887 | 0.2312 | 0.2078 | 0.7286 | 0.6656 | 0.6134 |
| 0.6123 | 4.0 | 1656 | 1.0142 | 0.2122 | 0.2538 | 0.2311 | 0.7321 | 0.6794 | 0.6346 |
| 0.4582 | 5.0 | 2070 | 1.0391 | 0.2069 | 0.2549 | 0.2284 | 0.7260 | 0.6787 | 0.6358 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3