Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
gaoqiong
lm-evaluation-harness
Commits
207447e5
Commit
207447e5
authored
Oct 08, 2025
by
Baber
Browse files
add group
parent
36021c33
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
117 additions
and
0 deletions
+117
-0
lm_eval/tasks/mrl/mc/_generate_config.py
lm_eval/tasks/mrl/mc/_generate_config.py
+12
-0
lm_eval/tasks/mrl/mc/_global_piqa.yaml
lm_eval/tasks/mrl/mc/_global_piqa.yaml
+105
-0
No files found.
lm_eval/tasks/mrl/mc/_generate_config.py
View file @
207447e5
...
...
@@ -17,3 +17,15 @@ if __name__ == "__main__":
f
.
write
(
"include: '_template_mc'
\n
"
)
f
.
write
(
f
"task: mrl_
{
s
}
\n
"
)
f
.
write
(
f
"dataset_name:
{
s
}
\n
"
)
with
open
(
PARENT
/
"_global_piqa.yaml"
,
"w"
)
as
f
:
f
.
write
(
"group: global_piqa
\n
"
)
f
.
write
(
"task:
\n
"
)
for
s
in
subsets
:
f
.
write
(
f
" - mrl_
{
s
}
\n
"
)
f
.
write
(
"aggregate_metric_list:
\n
"
)
f
.
write
(
" - metric: acc
\n
"
)
f
.
write
(
" aggregation: mean
\n
"
)
f
.
write
(
" weight_by_size: true
\n
"
)
f
.
write
(
"metadata:
\n
"
)
f
.
write
(
" version: 1.0
\n
"
)
lm_eval/tasks/mrl/mc/_global_piqa.yaml
0 → 100644
View file @
207447e5
group
:
global_piqa
task
:
-
mrl_acq_arab
-
mrl_aeb_arab
-
mrl_afb_arab
-
mrl_als_latn
-
mrl_amh_ethi
-
mrl_apc_arab_jord
-
mrl_apc_arab_leba
-
mrl_apc_arab_pale
-
mrl_apc_arab_syri
-
mrl_arb_arab
-
mrl_arq_arab
-
mrl_ars_arab
-
mrl_ary_arab
-
mrl_arz_arab
-
mrl_azj_latn
-
mrl_bam_latn
-
mrl_bel_cyrl
-
mrl_ben_latn
-
mrl_bho_deva
-
mrl_bsk_arab
-
mrl_bul_cyrl
-
mrl_cat_latn
-
mrl_ces_latn
-
mrl_ckb_arab
-
mrl_ckm_latn
-
mrl_cmn_hans
-
mrl_cmn_hant
-
mrl_dhd_deva
-
mrl_ell_grek
-
mrl_eng_latn
-
mrl_est_latn
-
mrl_fao_latn
-
mrl_fin_latn
-
mrl_fra_latn_cana
-
mrl_fra_latn_fran
-
mrl_glg_latn
-
mrl_guj_gujr
-
mrl_hau_latn
-
mrl_haw_latn
-
mrl_heb_hebr
-
mrl_hrv_latn
-
mrl_hun_latn
-
mrl_hye_armn
-
mrl_ibo_latn
-
mrl_idu_latn
-
mrl_ind_latn
-
mrl_isl_latn
-
mrl_iso_latn
-
mrl_ita_latn
-
mrl_jav_latn
-
mrl_jpn_jpan
-
mrl_kat_geor
-
mrl_kaz_cyrl
-
mrl_kir_cyrl
-
mrl_kor_hang
-
mrl_lit_latn
-
mrl_mar_deva
-
mrl_mkd_cyrl
-
mrl_mni_beng
-
mrl_nag_latn
-
mrl_nld_latn
-
mrl_nno_latn
-
mrl_nob_latn
-
mrl_npi_deva
-
mrl_pcm_latn
-
mrl_pes_arab
-
mrl_pol_latn
-
mrl_por_latn_braz
-
mrl_por_latn_port
-
mrl_ron_latn
-
mrl_rwr_deva
-
mrl_sin_sinh
-
mrl_slk_latn
-
mrl_slk_latn_sari
-
mrl_slv_latn
-
mrl_slv_latn_cerk
-
mrl_snd_arab
-
mrl_snd_deva
-
mrl_spa_latn_peru
-
mrl_srp_cyrl
-
mrl_srp_latn
-
mrl_swe_latn
-
mrl_tam_taml
-
mrl_tgl_latn
-
mrl_tha_thai
-
mrl_tur_latn
-
mrl_uig_arab
-
mrl_ukr_cyrl
-
mrl_urd_arab
-
mrl_urd_latn
-
mrl_urh_latn
-
mrl_uzn_latn
-
mrl_vie_latn
-
mrl_yor_latn
-
mrl_yue_hant
-
mrl_zsm_latn
-
mrl_zul_latn
aggregate_metric_list
:
-
metric
:
acc
aggregation
:
mean
weight_by_size
:
true
metadata
:
version
:
1.0
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment