OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/zle-ita/opusTCv20210807_transformer-big_2022-03-23
Benchmark: all benchmarks [download]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart] [heatmap]

ID	Language	Benchmark	Output	bleu
0	bel-ita	flores101-devtest	show	13.7
1	bel-ita	flores200-devtest	show	13.7
2	bel-ita	ntrex128	show	20.8
3	bel-ita	tatoeba-test-v2021-03-30	show	48.5
4	bel-ita	tatoeba-test-v2021-08-07	show	48.3
5	rus-ita	flores101-devtest	show	23.6
6	rus-ita	flores200-devtest	show	23.6
7	rus-ita	ntrex128	show	22
8	rus-ita	tatoeba-test-v2020-07-28	show	43.7
9	rus-ita	tatoeba-test-v2021-03-30	show	43.7
10	rus-ita	tatoeba-test-v2021-08-07	show	43.7
11	ukr-ita	flores101-devtest	show	23.5
12	ukr-ita	flores200-devtest	show	23.5
13	ukr-ita	ntrex128	show	22.2
14	ukr-ita	tatoeba-test-v2021-08-07	show	50.4
			average	30.993