Downloads · 30 days
20
22% of all-time downloads
bengali_pos_v1
mHossain/bengali_pos_v1
bengali_pos_v1 is a token classification model from mHossain. Use it when you need labels on individual words, such as names. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
92
Public
Parameters
110M
1.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors440 MB · 100%
At a glance
- Task
- Token Classification
- Library
- transformers
- Model type
- electra
- Access
- Public
- Created
- Dec 30, 2023
- Updated
- Dec 31, 2023
- SHA
- 2e4b87a3
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
bengali_pos_v1
This model is a fine-tuned version of csebuetnlp/banglabert on the pos_tag_100k dataset. It achieves the following results on the evaluation set:
- Loss: 0.6322
- Precision: 0.7541
- Recall: 0.7567
- F1: 0.7554
- Accuracy: 0.8182
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 |
|---|---|---|---|---|---|---|---|
| 0.7003 | 1.0 | 20000 | 0.6762 | 0.7281 | 0.7339 | 0.7310 | 0.8000 |
| 0.6097 | 2.0 | 40000 | 0.6277 | 0.7481 | 0.7481 | 0.7481 | 0.8135 |
| 0.5062 | 3.0 | 60000 | 0.6322 | 0.7541 | 0.7567 | 0.7554 | 0.8182 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
- Task
- Token Classification
- Library
- transformers
- Type
- electra
- Created
- Dec 30, 2023
- Updated
- Dec 31, 2023