README.md

metadata

license: mit
base_model: facebook/m2m100_418M
tags:
  - generated_from_trainer
model-index:
  - name: output
    results: []

output

This model is a fine-tuned version of facebook/m2m100_418M on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1802

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
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
3.7737	0.11	100	0.4133
0.3772	0.23	200	0.2586
0.2884	0.34	300	0.2401
0.2732	0.46	400	0.2320
0.285	0.57	500	0.2212
0.2375	0.69	600	0.2143
0.232	0.8	700	0.2113
0.2468	0.91	800	0.2028
0.2222	1.03	900	0.1996
0.1715	1.14	1000	0.1971
0.1442	1.26	1100	0.1949
0.1726	1.37	1200	0.1936
0.1359	1.49	1300	0.1932
0.1613	1.6	1400	0.1912
0.1599	1.71	1500	0.1886
0.1431	1.83	1600	0.1856
0.1607	1.94	1700	0.1829
0.1269	2.06	1800	0.1867
0.0838	2.17	1900	0.1837
0.0982	2.29	2000	0.1850
0.0936	2.4	2100	0.1831
0.1009	2.51	2200	0.1832
0.0889	2.63	2300	0.1824
0.0956	2.74	2400	0.1813
0.0967	2.86	2500	0.1807
0.1075	2.97	2600	0.1802

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2