Downloads · 30 days
109
5% of all-time downloads
mbart50_ipa-to-bangla_1934_lr2
thunderboltc/mbart50_ipa-to-bangla_1934_lr2
mbart50_ipa-to-bangla_1934_lr2 is a machine learning model from thunderboltc. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
2.1K
Public
Parameters
611M
183 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors4.9 GB · 50%
At a glance
- Library
- transformers
- Model type
- mbart
- Access
- Public
- Created
- Aug 21, 2026
- Updated
- Aug 21, 2026
- SHA
- d6df8071
Base models
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. -->
mbart50_ipa-to-bangla_1934_lr2
This model is a fine-tuned version of facebook/mbart-large-50-many-to-many-mmt on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.0822
- Bleu: 23.6435
- Chrf: 40.2174
- Gen Len: 4.1587
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Chrf | Gen Len |
|---|---|---|---|---|---|---|
| 3.8182 | 1.0 | 334 | 3.1141 | 21.0224 | 45.0581 | 3.5329 |
| 2.3838 | 2.0 | 668 | 2.6427 | 16.5158 | 22.9651 | 4.1198 |
| 1.5719 | 3.0 | 1002 | 2.5787 | 17.2860 | 28.1365 | 3.9311 |
| 0.9892 | 4.0 | 1336 | 2.6323 | 13.1345 | 24.4638 | 4.1677 |
| 0.612 | 5.0 | 1670 | 2.8000 | 16.5158 | 29.2339 | 4.2006 |
| 0.4093 | 6.0 | 2004 | 2.8305 | 8.9826 | 22.7865 | 4.0659 |
| 0.278 | 7.0 | 2338 | 2.9035 | 19.3049 | 29.2339 | 4.3204 |
| 0.198 | 8.0 | 2672 | 2.9287 | 17.2860 | 29.8211 | 4.2814 |
| 0.1478 | 9.0 | 3006 | 2.9168 | 17.9652 | 33.1081 | 4.1078 |
| 0.111 | 10.0 | 3340 | 2.9489 | 19.3049 | 35.0936 | 4.1407 |
| 0.0793 | 11.0 | 3674 | 3.0087 | 24.2746 | 52.6692 | 4.1078 |
| 0.0599 | 12.0 | 4008 | 2.9698 | 22.7497 | 29.3262 | 4.3054 |
| 0.0458 | 13.0 | 4342 | 3.0226 | 23.6435 | 38.1679 | 3.9970 |
| 0.0307 | 14.0 | 4676 | 3.0436 | 24.2746 | 52.6692 | 4.1737 |
| 0.022 | 15.0 | 5010 | 3.0432 | 23.6435 | 37.3953 | 4.1257 |
| 0.0167 | 16.0 | 5344 | 3.0806 | 23.6435 | 37.3953 | 4.0180 |
| 0.0123 | 17.0 | 5678 | 3.0707 | 23.6435 | 37.3953 | 4.0120 |
| 0.0103 | 18.0 | 6012 | 3.0810 | 11.4787 | 29.7668 | 4.0599 |
| 0.0089 | 19.0 | 6346 | 3.0733 | 23.6435 | 37.3953 | 4.1677 |
| 0.0063 | 20.0 | 6680 | 3.0916 | 19.3049 | 33.1081 | 4.1617 |
| 0.005 | 21.0 | 7014 | 3.0750 | 19.3049 | 33.1081 | 4.1617 |
| 0.0039 | 22.0 | 7348 | 3.0851 | 23.6435 | 37.3953 | 4.0599 |
| 0.0033 | 23.0 | 7682 | 3.0940 | 23.6435 | 37.3953 | 4.0868 |
| 0.0028 | 24.0 | 8016 | 3.0832 | 23.6435 | 40.2174 | 4.1916 |
| 0.0025 | 25.0 | 8350 | 3.0822 | 23.6435 | 40.2174 | 4.1587 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.19.1
- Library
- transformers
- Type
- mbart
- Created
- Aug 21, 2026
- Updated
- Aug 21, 2026