OPUS-MT Dashboard

Language pair: [eng-fra] all languages
Models: [all models] [OPUS-MT] [external] [compare]
Selected: Tatoeba-MT-models/eng-fin/opusTCv20210807+nopar+ft95-sepvoc_transformer-tiny11-align_2022-01-25
Benchmark: all benchmarks [download]
Evaluation metric: bleu [spbleu][chrf][chrf++][comet]
Chart Type: [barchart] [heatmap]

orange = OPUS-MT, blue = Tatoeba-MT models, red = HPLT-MT models
green = student models, grey = external models, purple = user-contributed
render chart with [gd] [plotly]

Model Scores (selected model)

ID	Language	Benchmark	Output	bleu
0	eng-fin	flores101-devtest	show	23.7
1	eng-fin	flores200-devtest	show	23.7
2	eng-fin	newstest2015	show	23.5
3	eng-fin	newstest2016	show	24.7
4	eng-fin	newstest2017	show	27.1
5	eng-fin	newstest2018	show	17.2
6	eng-fin	newstest2019	show	23.4
7	eng-fin	newstestALL2016	show (logfile)	28.4
8	eng-fin	newstestALL2017	show (logfile)	30.8
9	eng-fin	newstestB2016	show	20.1
10	eng-fin	newstestB2017	show	22.9
11	eng-fin	ntrex128	show (logfile)	18.3
12	eng-fin	tatoeba-test-v2020-07-28	show	37.7
13	eng-fin	tatoeba-test-v2021-03-30	show	37.6
14	eng-fin	tatoeba-test-v2021-08-07	show	37.5
			average	26.440