You can deploy this model and start earning money today!

cshim-cmu/
k30

Name: k30
Availability: InStock

pytorch

marian

generated_from_trainer

apache-2.0

Views

No views yet

k30

This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-fi on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5489
Bleu: 0.6841
Chrf: 25.4223
Gen Len: 41.0744

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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Chrf	Gen Len
1.2907	0.09	1000	1.0516	0.9283	11.7645	14.1811
1.0684	0.18	2000	0.9490	0.8798	11.533	16.8561
0.9676	0.27	3000	0.8863	0.5757	11.1388	29.1087
0.9079	0.36	4000	0.8401	0.2088	11.1723	37.4085
0.8646	0.45	5000	0.8086	0.1558	11.8155	44.8551
0.8163	0.54	6000	0.7860	0.2941	12.7228	50.8431
0.793	0.63	7000	0.7685	0.3	13.5452	52.5644
0.7595	0.72	8000	0.7545	0.3107	14.482	50.3109
0.7473	0.81	9000	0.7427	0.2853	14.8197	52.3994
0.7258	0.9	10000	0.7332	0.2517	15.1279	52.5553
0.7071	0.99	11000	0.7249	0.1917	15.2298	54.4427
0.6906	1.08	12000	0.7171	0.1168	15.2917	57.1861
0.688	1.18	13000	0.7103	0.1183	15.2671	58.4869
0.6755	1.27	14000	0.7041	0.1194	15.45	58.0372
0.66	1.36	15000	0.6985	0.1229	15.5324	58.0352
0.6527	1.45	16000	0.6924	0.2327	15.9639	59.498
0.6467	1.54	17000	0.6876	0.2417	16.0874	57.839
0.6345	1.63	18000	0.6827	0.2501	16.4667	56.3249
0.6316	1.72	19000	0.6786	0.2272	16.5233	60.7022
0.6185	1.81	20000	0.6744	0.2431	16.6877	57.5654
0.6192	1.9	21000	0.6703	0.1254	16.6404	59.3702
0.6003	1.99	22000	0.6669	0.1637	17.2389	54.5292
0.6071	2.08	23000	0.6627	0.2801	17.5863	54.8873
0.5918	2.17	24000	0.6590	0.2808	17.7327	54.333
0.5867	2.26	25000	0.6556	0.2605	18.1009	53.9819
0.5847	2.35	26000	0.6524	0.4114	18.2943	52.2455
0.5791	2.44	27000	0.6494	0.297	18.4928	51.7656
0.5683	2.53	28000	0.6464	0.2975	18.636	51.1861
0.5713	2.62	29000	0.6440	0.3344	18.7266	53.4849
0.5627	2.71	30000	0.6411	0.3163	18.932	50.2525
0.5649	2.8	31000	0.6385	0.3404	19.3834	48.6157
0.5582	2.89	32000	0.6353	0.3887	19.5944	49.6982
0.5511	2.98	33000	0.6328	0.4004	19.8008	48.1046
0.5531	3.07	34000	0.6304	0.3838	19.8527	49.5915
0.5435	3.16	35000	0.6278	0.3752	20.0612	49.7807
0.5336	3.25	36000	0.6257	0.4452	20.1601	47.2455
0.5289	3.34	37000	0.6233	0.4305	20.2813	47.9266
0.5245	3.43	38000	0.6210	0.3889	20.6298	47.9125
0.5271	3.53	39000	0.6194	0.5911	20.7396	46.666
0.5246	3.62	40000	0.6170	0.4446	20.9203	48.7384
0.5229	3.71	41000	0.6147	0.4973	21.0565	47.8793
0.5187	3.8	42000	0.6127	0.5665	21.0825	48.4859
0.515	3.89	43000	0.6107	0.5109	21.4095	46.7012
0.5143	3.98	44000	0.6092	0.4426	21.1644	47.9547
0.5157	4.07	45000	0.6073	0.5276	21.3193	49.4567
0.5079	4.16	46000	0.6056	0.3953	21.2488	49.1358
0.5081	4.25	47000	0.6034	0.4722	21.7613	46.837
0.5005	4.34	48000	0.6017	0.5514	21.6184	47.9849
0.4979	4.43	49000	0.6004	0.5843	21.891	47.2746
0.4955	4.52	50000	0.5984	0.5818	21.7768	46.834
0.4968	4.61	51000	0.5967	0.6015	22.0726	46.5523
0.4876	4.7	52000	0.5953	0.5473	22.0663	45.1227
0.484	4.79	53000	0.5936	0.6827	22.2377	45.1298
0.482	4.88	54000	0.5920	0.6917	22.1828	44.9396
0.4833	4.97	55000	0.5904	0.5064	22.0199	47.0845
0.4872	5.06	56000	0.5888	0.5249	22.1373	47.2274
0.4785	5.15	57000	0.5875	0.6014	22.2523	46.1147
0.4729	5.24	58000	0.5858	0.5947	22.5401	46.4256
0.4786	5.33	59000	0.5844	0.639	22.803	45.2153
0.4812	5.42	60000	0.5830	0.6011	22.9732	45.0272
0.4715	5.51	61000	0.5824	0.5704	23.1058	44.4044
0.4706	5.6	62000	0.5806	0.5077	22.9898	46.5865
0.4682	5.69	63000	0.5797	0.6452	23.1412	43.8823
0.4622	5.79	64000	0.5783	0.6197	22.9263	46.6911
0.4561	5.88	65000	0.5771	0.6415	23.2077	45.336
0.4559	5.97	66000	0.5758	0.6878	23.1664	44.7787
0.4673	6.06	67000	0.5743	0.6022	23.2732	44.2384
0.4609	6.15	68000	0.5735	0.5477	23.043	45.5262
0.4557	6.24	69000	0.5723	0.6173	23.4316	42.4044
0.4594	6.33	70000	0.5713	0.5941	23.6243	43.7817
0.4585	6.42	71000	0.5702	0.5709	23.6328	41.8521
0.4515	6.51	72000	0.5687	0.5895	23.8766	42.3169
0.4436	6.6	73000	0.5680	0.6076	23.8065	42.9829
0.439	6.69	74000	0.5671	0.5831	23.9339	42.7555
0.4413	6.78	75000	0.5659	0.6361	24.1313	42.0795
0.4331	6.87	76000	0.5649	0.6271	24.227	41.8722
0.4453	6.96	77000	0.5639	0.6496	24.1779	41.2978
0.4441	7.05	78000	0.5632	0.6544	24.3191	41.0634
0.4347	7.14	79000	0.5624	0.6712	24.3874	41.6207
0.4382	7.23	80000	0.5615	0.6385	24.3813	42.1569
0.4375	7.32	81000	0.5607	0.5959	24.505	42.8592
0.4328	7.41	82000	0.5596	0.6244	24.7194	42.5392
0.4322	7.5	83000	0.5588	0.6218	24.6816	40.3078
0.4327	7.59	84000	0.5580	0.6393	24.4116	41.8873
0.4346	7.68	85000	0.5572	0.652	24.8885	40.501
0.4324	7.77	86000	0.5565	0.6477	25.0305	40.6388
0.4233	7.86	87000	0.5558	0.6859	24.787	41.2012
0.419	7.95	88000	0.5552	0.6624	25.0463	41.5262
0.4203	8.04	89000	0.5545	0.7753	25.1134	41.6026
0.4226	8.14	90000	0.5537	0.6754	25.1785	40.6167
0.4236	8.23	91000	0.5528	0.7879	25.4603	41.1408
0.4171	8.32	92000	0.5526	0.6759	25.2511	40.7525
0.42	8.41	93000	0.5519	0.68	25.4983	39.8994
0.4197	8.5	94000	0.5509	0.7101	25.3426	40.831
0.4176	8.59	95000	0.5505	0.6733	25.2469	41.5634
0.4146	8.68	96000	0.5498	0.6816	25.4491	41.1258
0.4169	8.77	97000	0.5493	0.6513	25.2509	41.1982
0.4136	8.86	98000	0.5489	0.6841	25.4223	41.0744

Framework versions

Transformers 4.28.1
Pytorch 2.0.0+cu117
Datasets 2.11.0
Tokenizers 0.13.3

cshim-cmu/k30

k30

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

cshim-cmu/
k30