iva_mt_wslot-m2m100_418M-en-ja

This model is a fine-tuned version of facebook/m2m100_418M on the iva_mt_wslot dataset. It achieves the following results on the evaluation set:

• Loss: 0.0153
• Bleu: 66.503
• Gen Len: 20.9519

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: 7
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.0185	1.0	2017	0.0164	63.4304	20.6499
0.0134	2.0	4034	0.0150	64.827	20.666
0.0104	3.0	6051	0.0146	64.465	21.2155
0.0079	4.0	8068	0.0148	64.8578	20.7915
0.0062	5.0	10085	0.0149	65.9149	21.0718
0.005	6.0	12102	0.0151	66.2905	20.8766
0.004	7.0	14119	0.0153	66.503	20.9519

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3

Citation

If you use this model, please cite the following:

@article{Sowanski2023SlotLI,
  title={Slot Lost in Translation? Not Anymore: A Machine Translation Model for Virtual Assistants with Type-Independent Slot Transfer},
  author={Marcin Sowanski and Artur Janicki},
  journal={2023 30th International Conference on Systems, Signals and Image Processing (IWSSIP)},
  year={2023},
  pages={1-5}
}