combined_marianmt_sanlish_to_bangla_1234

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-mul 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 with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 25
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Bleu	Chrf	Meteor	Bertscore F1
3.4867	1.0	246	2.4953	0.3013	11.3201	0.0297	0.725
2.4133	2.0	492	2.1473	0.6426	13.563	0.0506	0.749
2.0753	3.0	738	1.9103	1.921	19.4409	0.1044	0.777
1.7616	4.0	984	1.7607	3.6539	23.9	0.142	0.7939
1.5058	5.0	1230	1.6992	3.4247	27.293	0.1882	0.8078
1.3095	6.0	1476	1.6391	4.6868	29.0131	0.204	0.815
1.1616	7.0	1722	1.5981	6.5137	31.0419	0.2256	0.821
1.0289	8.0	1968	1.5805	7.1741	33.3236	0.2333	0.8257
0.9136	9.0	2214	1.5644	10.2041	34.4361	0.2499	0.8311
0.8246	10.0	2460	1.5826	9.4203	34.8978	0.2498	0.8303
0.7421	11.0	2706	1.5625	8.6133	35.331	0.2567	0.8326
0.6666	12.0	2952	1.5989	7.9675	35.9283	0.2655	0.8345
0.605	13.0	3198	1.5774	8.5414	36.5257	0.2707	0.8344
0.5581	14.0	3444	1.5885	9.8217	37.8949	0.2779	0.8389
0.5159	15.0	3690	1.6101	10.3227	37.5904	0.2828	0.8384
0.4817	16.0	3936	1.6000	11.1169	38.2473	0.2918	0.8404
0.4378	17.0	4182	1.6189	10.0945	38.8729	0.2903	0.8441
0.4151	18.0	4428	1.6312	9.92	38.5636	0.2941	0.8411
0.3888	19.0	4674	1.6346	10.4983	38.6156	0.2991	0.8426
0.3645	20.0	4920	1.6442	10.5683	38.9307	0.3012	0.8443
0.3447	21.0	5166	1.6457	10.4767	38.8105	0.3043	0.8436
0.3277	22.0	5412	1.6523	11.544	39.756	0.3081	0.8453
0.3227	23.0	5658	1.6586	11.0236	38.6142	0.2986	0.8427
0.3095	24.0	5904	1.6593	11.0408	38.7821	0.3002	0.8426
0.3072	25.0	6150	1.6589	10.9934	38.8597	0.3027	0.8431