You can deploy this model and start earning money today!

JuliusFx/
dyu-fr-helsinki_v2_xh

Name: dyu-fr-helsinki_v2_xh
Availability: InStock

safetensors

marian

generated_from_trainer

Helsinki-NLP/opus-mt-xh-fr

finetune

apache-2.0

Views

No views yet

dyu-fr-helsinki_v2_xh

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

Loss: 5.7627
Bleu score: 0.1038

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: 20
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu score
4.7699	1.0	596	3.9281	0.0
4.1092	2.0	1192	3.8960	0.0
3.8056	3.0	1788	3.8476	0.0
3.5267	4.0	2384	3.8732	0.0
3.3131	5.0	2980	3.8689	0.0
2.8882	6.0	3576	3.8581	0.0693
2.7001	7.0	4172	3.9001	0.0541
2.5432	8.0	4768	3.8264	0.0569
2.4091	9.0	5364	3.8740	0.0794
2.2451	10.0	5960	3.9506	0.0617
1.9826	11.0	6556	3.9691	0.0718
1.8533	12.0	7152	3.9986	0.0668
1.748	13.0	7748	4.0533	0.0554
1.6282	14.0	8344	4.0454	0.0482
1.5518	15.0	8940	4.1565	0.0605
1.3517	16.0	9536	4.1506	0.0706
1.2488	17.0	10132	4.1920	0.0547
1.1755	18.0	10728	4.2185	0.0537
1.1012	19.0	11324	4.3805	0.0500
1.0373	20.0	11920	4.3064	0.0514
0.9021	21.0	12516	4.3818	0.0503
0.8232	22.0	13112	4.4830	0.0552
0.7745	23.0	13708	4.4410	0.0977
0.7202	24.0	14304	4.4905	0.0541
0.6774	25.0	14900	4.5591	0.0529
0.6265	26.0	15496	4.5700	0.0541
0.534	27.0	16092	4.5780	0.0590
0.4996	28.0	16688	4.6759	0.0664
0.4647	29.0	17284	4.7150	0.0594
0.4344	30.0	17880	4.7752	0.0590
0.4068	31.0	18476	4.8266	0.0634
0.3456	32.0	19072	4.8717	0.0498
0.32	33.0	19668	4.8170	0.0545
0.3023	34.0	20264	4.8654	0.0858
0.2817	35.0	20860	4.8642	0.0613
0.2608	36.0	21456	4.9754	0.0889
0.2293	37.0	22052	4.9754	0.0537
0.2091	38.0	22648	5.0749	0.0665
0.2011	39.0	23244	5.0712	0.1087
0.1877	40.0	23840	5.0571	0.0625
0.1768	41.0	24436	5.1075	0.0958
0.1576	42.0	25032	5.1775	0.1019
0.1456	43.0	25628	5.1977	0.0510
0.1365	44.0	26224	5.2568	0.0940
0.1318	45.0	26820	5.2515	0.0567
0.1238	46.0	27416	5.2735	0.0923
0.1102	47.0	28012	5.3140	0.0638
0.1044	48.0	28608	5.3494	0.0874
0.097	49.0	29204	5.4621	0.0578
0.0943	50.0	29800	5.4158	0.0906
0.0916	51.0	30396	5.4666	0.0428
0.0863	52.0	30992	5.4664	0.0547
0.0775	53.0	31588	5.4710	0.0547
0.0756	54.0	32184	5.5460	0.0474
0.0716	55.0	32780	5.4745	0.0615
0.0701	56.0	33376	5.5189	0.0765
0.0675	57.0	33972	5.4933	0.0507
0.0605	58.0	34568	5.5229	0.0843
0.0587	59.0	35164	5.5023	0.0906
0.0562	60.0	35760	5.6262	0.0923
0.0544	61.0	36356	5.5985	0.0940
0.0533	62.0	36952	5.6278	0.0923
0.0496	63.0	37548	5.5495	0.0859
0.0469	64.0	38144	5.5974	0.0815
0.0468	65.0	38740	5.6458	0.0977
0.0442	66.0	39336	5.6121	0.1060
0.0434	67.0	39932	5.7019	0.1060
0.0411	68.0	40528	5.7208	0.0977
0.0394	69.0	41124	5.6680	0.0956
0.0396	70.0	41720	5.7240	0.0905
0.0376	71.0	42316	5.7103	0.0813
0.0358	72.0	42912	5.6782	0.1060
0.0336	73.0	43508	5.7993	0.1056
0.033	74.0	44104	5.6958	0.1014
0.0327	75.0	44700	5.6557	0.0874
0.0324	76.0	45296	5.6959	0.0829
0.0307	77.0	45892	5.7028	0.0978
0.0314	78.0	46488	5.7380	0.0996
0.0289	79.0	47084	5.7078	0.0996
0.0293	80.0	47680	5.7234	0.0996
0.0281	81.0	48276	5.6458	0.0831
0.0273	82.0	48872	5.7549	0.0845
0.0262	83.0	49468	5.7560	0.0996
0.0254	84.0	50064	5.7187	0.0996
0.0254	85.0	50660	5.7721	0.0996
0.025	86.0	51256	5.7724	0.0996
0.0245	87.0	51852	5.7187	0.0996
0.0245	88.0	52448	5.7559	0.0996
0.0245	89.0	53044	5.7402	0.1038
0.0231	90.0	53640	5.7659	0.0996
0.0222	91.0	54236	5.7475	0.1038
0.0227	92.0	54832	5.7590	0.1038
0.0224	93.0	55428	5.7617	0.1038
0.0218	94.0	56024	5.7454	0.1038
0.021	95.0	56620	5.7353	0.1038
0.0211	96.0	57216	5.7410	0.1038
0.0207	97.0	57812	5.7465	0.1038
0.0212	98.0	58408	5.7473	0.1038
0.0208	99.0	59004	5.7605	0.1038
0.0208	100.0	59600	5.7627	0.1038

Framework versions

Transformers 4.44.0
Pytorch 2.1.0+cu118
Datasets 2.20.0
Tokenizers 0.19.1

JuliusFx/dyu-fr-helsinki_v2_xh

dyu-fr-helsinki_v2_xh

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

JuliusFx/
dyu-fr-helsinki_v2_xh