marianmt_sanlish_to_english_1934_lr2

This model is a fine-tuned version of Helsinki-NLP/opus-mt-mul-en on the None 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: 8
eval_batch_size: 8
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: 25
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Bleu	Chrf
3.9700	1.0	195	3.1817	22.9575	38.3609
2.6273	2.0	390	2.9092	33.5211	32.4808
2.0781	3.0	585	2.8153	15.6197	37.7766
1.6715	4.0	780	2.7877	21.3644	25.5606
1.3545	5.0	975	2.7951	21.8342	37.1458
1.0990	6.0	1170	2.8105	21.8342	37.1458
0.8755	7.0	1365	2.8481	21.8342	37.1458
0.7095	8.0	1560	2.9156	21.8342	37.1458
0.5776	9.0	1755	2.9351	27.7762	43.8892
0.4579	10.0	1950	2.9752	21.8342	37.1458
0.3740	11.0	2145	3.0392	27.7762	43.8892
0.3013	12.0	2340	3.0604	27.7762	43.8892
0.2581	13.0	2535	3.0998	27.7762	43.8892
0.2146	14.0	2730	3.1433	27.7762	43.8892
0.1860	15.0	2925	3.1458	27.7762	43.8892
0.1628	16.0	3120	3.1808	27.7762	43.8892
0.1363	17.0	3315	3.1835	27.7762	43.8892
0.1258	18.0	3510	3.1991	27.7762	43.8892
0.1131	19.0	3705	3.2020	27.7762	43.8892
0.1022	20.0	3900	3.2272	27.7762	43.8892
0.0904	21.0	4095	3.2479	27.7762	43.8892
0.0859	22.0	4290	3.2544	27.7762	43.8892
0.0784	23.0	4485	3.2655	27.7762	43.8892
0.0732	24.0	4680	3.2671	27.7762	43.8892
0.0732	25.0	4875	3.2683	27.7762	43.8892