metadata
base_model: mHossain/bengali_pos_v1_200000
tags:
- generated_from_trainer
datasets:
- pos_tag_100k
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bengali_pos_v1_300000
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: pos_tag_100k
type: pos_tag_100k
config: conll2003
split: validation
args: conll2003
metrics:
- name: Precision
type: precision
value: 0.7824197234815842
- name: Recall
type: recall
value: 0.7854805341357909
- name: F1
type: f1
value: 0.783947141194579
- name: Accuracy
type: accuracy
value: 0.839060352367862
bengali_pos_v1_300000
This model is a fine-tuned version of mHossain/bengali_pos_v1_200000 on the pos_tag_100k dataset. It achieves the following results on the evaluation set:
- Loss: 0.6558
- Precision: 0.7824
- Recall: 0.7855
- F1: 0.7839
- Accuracy: 0.8391
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.3708 | 1.0 | 22500 | 0.6313 | 0.7688 | 0.7733 | 0.7711 | 0.8304 |
0.3099 | 2.0 | 45000 | 0.6491 | 0.7770 | 0.7789 | 0.7780 | 0.8353 |
0.3127 | 3.0 | 67500 | 0.6558 | 0.7824 | 0.7855 | 0.7839 | 0.8391 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0