OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/sla-por/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	bel-por	flores101-devtest	show	17.2
1	bel-por	flores200-devtest	show	17.2
2	bul-por	flores101-devtest	show	33.3
3	bul-por	flores200-devtest	show	33.3
4	ces-por	flores101-devtest	show	33
5	ces-por	flores200-devtest	show	33
6	hrv-por	flores101-devtest	show	31.6
7	hrv-por	flores200-devtest	show	31.6
8	mkd-por	flores101-devtest	show	35
9	mkd-por	flores200-devtest	show	35
10	pol-por	flores101-devtest	show	26.3
11	pol-por	flores200-devtest	show	26.3
12	pol-por	tatoeba-test-v2020-07-28	show	43.3
13	pol-por	tatoeba-test-v2021-03-30	show	43.3
14	pol-por	tatoeba-test-v2021-08-07	show	43.4
15	rus-por	flores101-devtest	show	30.8
16	rus-por	flores200-devtest	show	30.8
17	rus-por	tatoeba-test-v2021-08-07	show	42.8
18	slv-por	flores101-devtest	show	30.7
19	slv-por	flores200-devtest	show	30.7
20	srp_Cyrl-por	flores101-devtest	show	36.2
21	srp_Cyrl-por	flores200-devtest	show	36.2
22	ukr-por	flores101-devtest	show	31.8
23	ukr-por	flores200-devtest	show	31.8
24	ukr-por	tatoeba-test-v2021-08-07	show	44.1
			average	33.148