OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/cat+oci+spa-eng/opusTCv20210807+bt_transformer-big_2022-03-13
Benchmark: all benchmarks [download]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart] [heatmap]

ID	Language	Benchmark	Output	bleu
0	cat-eng	flores101-devtest	show	45.4
1	cat-eng	flores200-devtest	show	45.4
2	cat-eng	ntrex128	show (logfile)	38.3
3	cat-eng	tatoeba-test-v2020-07-28	show	57.3
4	cat-eng	tatoeba-test-v2021-03-30	show	57.3
5	cat-eng	tatoeba-test-v2021-08-07	show	57.3
6	oci-eng	flores101-devtest	show	37.5
7	oci-eng	flores200-devtest	show	37.5
8	oci-eng	tatoeba-test-v2020-07-28	show	19.3
9	oci-eng	tatoeba-test-v2021-03-30	show	19.3
10	oci-eng	tatoeba-test-v2021-08-07	show	19.3
11	spa-eng	flores101-devtest	show	29.9
12	spa-eng	flores200-devtest	show	29.9
13	spa-eng	newssyscomb2009	show	30.8
14	spa-eng	newstest2008	show (logfile)	27.9
15	spa-eng	newstest2009	show	30.2
16	spa-eng	newstest2010	show	36.8
17	spa-eng	newstest2011	show	34.7
18	spa-eng	newstest2012	show	38.6
19	spa-eng	newstest2013	show	35.3
20	spa-eng	ntrex128	show (logfile)	40.4
21	spa-eng	tatoeba-test-v2020-07-28	show	61.4
22	spa-eng	tatoeba-test-v2021-03-30	show	61.4
23	spa-eng	tatoeba-test-v2021-08-07	show	62.3
24	spa-eng	tico19-test	show	51.8
			average	40.212