marian-finetuned-smash-up-en-to-fr

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-fr on an unknown dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 32
eval_batch_size: 64
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Bleu
No log	1.0	32	0.5544	0.003	67.7307
No log	2.0	64	0.4248	0.003	72.2999
No log	3.0	96	0.3800	0.003	74.1053
No log	4.0	128	0.3590	0.003	74.9361
No log	5.0	160	0.3445	0.003	75.5260
No log	6.0	192	0.3370	0.003	75.1898
No log	7.0	224	0.3332	0.003	75.6648
No log	8.0	256	0.3301	0.003	75.7235
No log	9.0	288	0.3289	0.003	75.9833
No log	10.0	320	0.3278	0.003	75.8408

Loading…