---
library_name: peft
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-en-mr
tags:
- generated_from_trainer
model-index:
- name: english-marathi-colloquial-translator-updated
  results: []
---

<!-- 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. -->

# english-marathi-colloquial-translator-updated

This model is a fine-tuned version of [Helsinki-NLP/opus-mt-en-mr](https://huggingface.co./Helsinki-NLP/opus-mt-en-mr) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6189

## 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: 0.0003
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step  | Validation Loss |
|:-------------:|:------:|:-----:|:---------------:|
| 0.5693        | 0.08   | 500   | 0.6740          |
| 0.8813        | 0.16   | 1000  | 0.6623          |
| 0.4563        | 0.24   | 1500  | 0.6546          |
| 0.306         | 0.32   | 2000  | 0.6500          |
| 0.4356        | 0.4    | 2500  | 0.6460          |
| 0.9248        | 0.48   | 3000  | 0.6441          |
| 0.5172        | 0.56   | 3500  | 0.6410          |
| 0.6833        | 0.64   | 4000  | 0.6391          |
| 0.6665        | 0.72   | 4500  | 0.6365          |
| 0.6056        | 0.8    | 5000  | 0.6360          |
| 0.7969        | 0.88   | 5500  | 0.6338          |
| 1.0205        | 0.96   | 6000  | 0.6325          |
| 0.7191        | 1.04   | 6500  | 0.6312          |
| 0.7727        | 1.12   | 7000  | 0.6300          |
| 0.7991        | 1.2    | 7500  | 0.6287          |
| 0.4923        | 1.28   | 8000  | 0.6286          |
| 0.4553        | 1.3600 | 8500  | 0.6270          |
| 0.9829        | 1.44   | 9000  | 0.6265          |
| 0.8621        | 1.52   | 9500  | 0.6257          |
| 0.6201        | 1.6    | 10000 | 0.6242          |
| 0.9138        | 1.6800 | 10500 | 0.6234          |
| 0.5501        | 1.76   | 11000 | 0.6233          |
| 0.7267        | 1.8400 | 11500 | 0.6232          |
| 0.7323        | 1.92   | 12000 | 0.6223          |
| 1.0083        | 2.0    | 12500 | 0.6222          |
| 0.3809        | 2.08   | 13000 | 0.6220          |
| 0.9979        | 2.16   | 13500 | 0.6214          |
| 0.5853        | 2.24   | 14000 | 0.6211          |
| 0.6319        | 2.32   | 14500 | 0.6207          |
| 0.7518        | 2.4    | 15000 | 0.6206          |
| 0.6834        | 2.48   | 15500 | 0.6198          |
| 0.246         | 2.56   | 16000 | 0.6198          |
| 0.5533        | 2.64   | 16500 | 0.6194          |
| 0.5569        | 2.7200 | 17000 | 0.6193          |
| 0.5367        | 2.8    | 17500 | 0.6189          |
| 0.5825        | 2.88   | 18000 | 0.6189          |
| 0.8336        | 2.96   | 18500 | 0.6189          |


### Framework versions

- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.3.1
- Tokenizers 0.21.0