learning_rate: 0.0001
train_batch_size: 32
eval_batch_size: 64
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
num_epochs: 30
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Bleu Msl	Bleu Asl	Ter Msl	Ter Asl
No log	1.0	142	0.4517	4.3556	60.0932	263.6364	17.9336
No log	2.0	284	0.2911	23.2666	85.6532	55.0000	7.1819
No log	3.0	426	0.2635	7.6866	80.4693	145.5556	8.7358
0.5021	4.0	568	0.2526	36.1099	86.3805	21.6162	7.3079
0.5021	5.0	710	0.2554	29.9719	84.4134	24.9495	8.1898
0.5021	6.0	852	0.2573	73.4067	87.3570	15.9091	6.5939
0.5021	7.0	994	0.2607	75.9042	87.5621	14.0404	5.9219
0.0602	8.0	1136	0.2534	69.4208	82.2155	17.9798	7.8118
0.0602	9.0	1278	0.2540	78.9796	84.5857	13.6869	7.2659
0.0602	10.0	1420	0.2439	78.3772	85.0526	14.0909	7.6438
0.0277	11.0	1562	0.2586	79.0479	85.5579	13.5354	6.8879
0.0277	12.0	1704	0.2658	77.8620	84.0916	14.3939	9.1978
0.0277	13.0	1846	0.2593	78.6031	87.9318	12.8283	5.9219
0.0277	14.0	1988	0.2574	77.2537	86.9300	13.6364	6.6779
0.0173	15.0	2130	0.2599	77.3620	85.5624	14.6465	7.2239
0.0173	16.0	2272	0.2674	77.8456	85.5155	15.0	7.3499
0.0173	17.0	2414	0.2603	77.5280	86.8876	14.5960	7.1399
0.0104	18.0	2556	0.2718	77.0685	86.1825	15.1515	6.5939
0.0104	19.0	2698	0.2675	78.5651	86.7494	13.8384	6.5099
0.0104	20.0	2840	0.2634	79.1140	85.3241	13.5859	7.3919
0.0104	21.0	2982	0.2617	78.1030	86.0924	14.8990	7.0139
0.0076	22.0	3124	0.2625	78.8043	86.6405	13.7374	6.2999
0.0076	23.0	3266	0.2694	78.2577	86.4381	14.4444	6.2579
0.0076	24.0	3408	0.2671	78.3892	86.1058	14.1919	6.5099
0.0056	25.0	3550	0.2666	78.4721	86.1346	14.8990	6.7619
0.0056	26.0	3692	0.2670	79.2248	87.4447	13.9899	6.4259
0.0056	27.0	3834	0.2689	80.1481	87.0403	13.4343	6.4259
0.0056	28.0	3976	0.2688	80.0229	87.3402	13.5354	6.2159
0.005	29.0	4118	0.2688	79.5101	87.5201	13.5354	6.3839
0.005	30.0	4260	0.2686	79.6101	87.4768	13.4848	6.4259

vania2911/5_6kmslsamples

5_6kmslsamples