iva_mt_wslot-m2m100_418M-en-pl-noleyzer

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: 0.3387
• Bleu: 56.3512
• Gen Len: 20.7319

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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.3611	1.0	4097	0.3088	53.0678	20.6977
0.2624	2.0	8194	0.2937	53.8146	20.5946
0.1904	3.0	12291	0.2955	53.7632	20.8701
0.1501	4.0	16388	0.2999	55.4482	20.6983
0.1079	5.0	20485	0.3074	55.5861	20.6366
0.0852	6.0	24582	0.3165	55.8774	20.6638
0.0646	7.0	28679	0.3231	55.9306	20.555
0.0464	8.0	32776	0.3294	55.9869	20.7809
0.0377	9.0	36873	0.3348	56.0674	20.7044
0.0305	10.0	40970	0.3387	56.3512	20.7319

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3