OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/sla-fin/opusTCv20210807_transformer-big_2022-09-15
Benchmark: all benchmarks [download]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart] [heatmap]

ID	Language	Benchmark	Output	bleu
0	bul-fin	flores101-devtest	show	18.7
1	bul-fin	flores200-devtest	show	18.7
2	bul-fin	ntrex128	show	14.4
3	ces-fin	flores101-devtest	show	19.2
4	ces-fin	flores200-devtest	show	19.2
5	ces-fin	ntrex128	show	15.3
6	hrv-fin	flores101-devtest	show	18.5
7	hrv-fin	flores200-devtest	show	18.5
8	hrv-fin	ntrex128	show	16
9	pol-fin	flores101-devtest	show	16.1
10	pol-fin	flores200-devtest	show	16.1
11	pol-fin	ntrex128	show	14.1
12	pol-fin	tatoeba-test-v2021-03-30	show	43.6
13	pol-fin	tatoeba-test-v2021-08-07	show	43.7
14	rus-fin	flores101-devtest	show	17.1
15	rus-fin	flores200-devtest	show	17.1
16	rus-fin	ntrex128	show	12.9
17	rus-fin	tatoeba-test-v2021-08-07	show	43.8
18	slv-fin	flores101-devtest	show	18.3
19	slv-fin	flores200-devtest	show	18.3
20	slv-fin	ntrex128	show	15
21	srp_Cyrl-fin	flores101-devtest	show	18.5
22	srp_Cyrl-fin	flores200-devtest	show	18.5
23	srp_Cyrl-fin	ntrex128	show	11.2
24	srp_Latn-fin	ntrex128	show	14.2
25	ukr-fin	flores101-devtest	show	18.2
26	ukr-fin	flores200-devtest	show	18.2
27	ukr-fin	ntrex128	show	13.3
			average	19.525