metadata
license: mit
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
base_model: cmarkea/distilcamembert-base
model-index:
- name: distilcamembert-cae-no-feeling
results: []
distilcamembert-cae-no-feeling
This model is a fine-tuned version of cmarkea/distilcamembert-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7566
- Precision: 0.8160
- Recall: 0.7975
- F1: 0.7998
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 |
|---|---|---|---|---|---|---|
| 1.174 | 1.0 | 40 | 0.9568 | 0.4398 | 0.4937 | 0.3881 |
| 0.8722 | 2.0 | 80 | 0.7551 | 0.7716 | 0.7215 | 0.7120 |
| 0.5255 | 3.0 | 120 | 0.7144 | 0.7533 | 0.6962 | 0.6937 |
| 0.3102 | 4.0 | 160 | 0.6734 | 0.7711 | 0.7595 | 0.7627 |
| 0.1804 | 5.0 | 200 | 0.7566 | 0.8160 | 0.7975 | 0.7998 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2