/
command-slurm.sh
executable file
·326 lines (275 loc) · 12.7 KB
/
command-slurm.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
#!/bin/bash
task=${task:-none}
lang=${lang:-eng}
lang2=${lang2:-eng}
lang3=${lang3:-eng}
MODEL_NAME=${MODEL_NAME:-bert}
dataset=${dataset:-wikiann}
subtask=${subtask:-nli}
prompt_lang=${prompt_lang:-zeroshot}
while [ $# -gt 0 ]; do
if [[ $1 == *"--"* ]]; then
param="${1/--/}"
declare $param="$2"
echo $1 $2 #Optional to see the parameter:value result
fi
shift
done
declare -a base_models=("bert" "xlmr")
if [[ "$MODEL_NAME" = "bert" ]]; then
base_model=${base_models[0]}
fi
if [[ "$MODEL_NAME" = "xlmr" ]]; then
base_model=${base_models[1]}
fi
if [[ "$task" = "train_udp" ]]; then
export ALL_MODELS=("UD_Armenian-ArmTDP" "UD_Norwegian-Nynorsk" "UD_Portuguese-Bosque" "UD_Italian-PoSTWITA" "UD_Old_French-SRCMF" "UD_North_Sami-Giella" "UD_Norwegian-Bokmaal" "UD_French-ParisStories" "UD_Italian-MarkIT" "UD_Chinese-GSDSimp" "UD_English-EWT" "UD_French-Rhapsodie" "UD_French-ParTUT" "UD_Classical_Chinese-Kyoto" "UD_Norwegian-NynorskLIA" "UD_Arabic-NYUAD" "UD_Portuguese-PetroGold" "UD_Italian-TWITTIRO" "UD_Turkish_German-SAGT" "UD_Maghrebi_Arabic_French-Arabizi" "UD_Portuguese-CINTIL" "UD_Ligurian-GLT" "UD_Dutch-Alpino" "UD_Western_Armenian-ArmTDP" "UD_Portuguese-GSD" "singlish" "UD_Arabic-PADT")
# export ALL_MODELS=("UD_English-EWT")
for MODEL_NAME in ${ALL_MODELS[@]}; do
efile="tr_output/${base_model}_${MODEL_NAME}_${task}.err"
ofile="tr_output/${base_model}_${MODEL_NAME}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${MODEL_NAME}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${MODEL_NAME} ${base_model}
bash install.sh --task train_udp --lang ${MODEL_NAME} --MODEL_NAME ${base_model}
# bash install.sh --task train_udp --lang UD_English-EWT --MODEL_NAME xlmr
done
fi
if [[ "$task" = "predict_udp" ]]; then
efile="tr_output/${base_model}_${task}.err"
ofile="tr_output/${base_model}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} null ${base_model}
#./command.sh --task predict_udp --MODEL_NAME bert
# bash install.sh --task ${task} --MODEL_NAME ${base_model}
fi
if [[ "$task" = "train_sdqa" || "$task" = "predict_sdqa" ]]; then
export ALL_MODELS=("all" "arabic" "bengali" "english" "finnish" "indonesian" "korean" "russian" "swahili" "telugu")
# export ALL_MODELS=("UD_English-EWT")
for MODEL_NAME in ${ALL_MODELS[@]}; do
efile="tr_output/${base_model}_${MODEL_NAME}_${task}.err"
ofile="tr_output/${base_model}_${MODEL_NAME}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${MODEL_NAME}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${MODEL_NAME} ${base_model} dev
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${MODEL_NAME} ${base_model} test
bash install.sh --task ${task} --lang ${MODEL_NAME} --MODEL_NAME ${base_model} --dataset dev
bash install.sh --task ${task} --lang ${MODEL_NAME} --MODEL_NAME ${base_model} --dataset test
done
fi
if [[ "$task" = "translate_nli" ]]; then
export ALL_TARGETS=("lmo_Latn" "ita_Latn" "fur_Latn" "scn_Latn" "srd_Latn" "vec_Latn" "azb_Arab" "azj_Latn" "tur_Latn" "kmr_Latn" "ckb_Arab" "nno_Latn" "nob_Latn" "lim_Latn" "ltz_Latn" "nld_Latn" "lvs_Latn" "ltg_Latn" "acm_Arab" "acq_Arab" "aeb_Arab" "ajp_Arab" "apc_Arab" "arb_Arab" "ars_Arab" "ary_Arab" "arz_Arab" "kab_Latn" "asm_Beng" "ben_Beng" "lij_Latn" "oci_Latn" "yue_Hant" "zho_Hans" "zho_Hant" "glg_Latn" "spa_Latn" "por_Latn" "nso_Latn" "sot_Latn")
export ALL_SRC=("ita_Latn" "azj_Latn" "ckb_Arab" "nob_Latn" "nld_Latn" "lvs_Latn" "arb_Arab" "lij_Latn" "zho_Hans" "spa_Latn" "nso_Latn" "ben_Beng")
test_hypo="tempdata/test_hypo-eng_Latn.txt"
test_premise="tempdata/test_premise-eng_Latn.txt"
# for TARGET_LANG in "${ALL_TARGETS[@]}"; do
# efile="tr_output/${TARGET_LANG}_${task}-test_hypo.err"
# ofile="tr_output/${TARGET_LANG}_${task}-test_hypo.out"
# echo ${efile}
# echo ${ofile}
# echo ${TARGET_LANG}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${TARGET_LANG} test_hypo ${test_hypo}
# efile="tr_output/${TARGET_LANG}_${task}-test_premise.err"
# ofile="tr_output/${TARGET_LANG}_${task}-test_premise.out"
# echo ${efile}
# echo ${ofile}
# echo ${TARGET_LANG}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${TARGET_LANG} test_premise ${test_premise}
# # ./install.sh --task ${task} --lang ${TARGET_LANG} --MODEL_NAME test_hypo --dataset ${test_hypo}
# done
# validation_hypo="tempdata/validation_hypo-eng_Latn.txt"
# validation_premise="tempdata/validation_premise-eng_Latn.txt"
# for TARGET_LANG in "${ALL_SRC[@]}"; do
# efile="tr_output/${TARGET_LANG}_${task}-validation_hypo.err"
# ofile="tr_output/${TARGET_LANG}_${task}-validation_hypo.out"
# echo ${efile}
# echo ${ofile}
# echo ${TARGET_LANG}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${TARGET_LANG} validation_hypo ${validation_hypo}
# done
# for TARGET_LANG in "${ALL_SRC[@]}"; do
# efile="tr_output/${TARGET_LANG}_${task}-validation_premise.err"
# ofile="tr_output/${TARGET_LANG}_${task}-validation_premise.out"
# echo ${efile}
# echo ${ofile}
# echo ${TARGET_LANG}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${TARGET_LANG} validation_premise ${validation_premise}
# done
train_hypo="tempdata/train_hypo-eng_Latn.txt"
train_premise="tempdata/train_premise-eng_Latn.txt"
for TARGET_LANG in "${ALL_SRC[@]}"; do
efile="tr_output/${TARGET_LANG}_${task}-train_premise.err"
ofile="tr_output/${TARGET_LANG}_${task}-train_premise.out"
echo ${efile}
echo ${ofile}
echo ${TARGET_LANG}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${TARGET_LANG} train_premise ${train_premise}
efile="tr_output/${TARGET_LANG}_${task}-train_hypo.err"
ofile="tr_output/${TARGET_LANG}_${task}-train_hypo.out"
echo ${efile}
echo ${ofile}
echo ${TARGET_LANG}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${TARGET_LANG} train_hypo ${train_hypo}
done
fi
if [[ "$task" = "train_ner" ]]; then
###norwegian dialects
export ALL_MODELS=("bokmaal" "nynorsk" "samnorsk")
##bokmaal:nb, ##nn:nynorsk
for MODEL_NAME in ${ALL_MODELS[@]}; do
efile="tr_output/${base_model}_${MODEL_NAME}_${task}.err"
ofile="tr_output/${base_model}_${MODEL_NAME}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${MODEL_NAME}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${MODEL_NAME} ${base_model} scripts/ner/norwegian_ner.py
done
export ALL_MODELS=("ar" "az" "ku" "tr" "hsb" "nl" "fr" "zh" "en" "mhr" "it" "de" "pa" "es" "hr" "lv" "hi" "ro" "el" "bn")
##bokmaal:nb, ##nn:nynorsk
for MODEL_NAME in ${ALL_MODELS[@]}; do
efile="tr_output/${base_model}_${MODEL_NAME}_${task}.err"
ofile="tr_output/${base_model}_${MODEL_NAME}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${MODEL_NAME}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${MODEL_NAME} ${base_model} wikiann
done
fi
if [[ "$task" = "predict_ner" ]]; then
MODEL_NAME="en"
efile="tr_output/${base_model}_${MODEL_NAME}_${task}.err"
ofile="tr_output/${base_model}_${MODEL_NAME}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${predict_ner} ${MODEL_NAME} ${base_model} wikiann
fi
if [[ "$task" = "train_did_lm" || "$task" = "predict_did_lm" ]]; then
lang="arabic"
dataset="madar"
efile="tr_output/${base_model}_${lang}_${dataset}_${task}.err"
ofile="tr_output/${base_model}_${lang}_${dataset}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${lang}
echo ${dataset}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${lang} ${base_model} ${dataset}
# ./install.sh --task train_did_lm --lang arabic --dataset madar --MODEL_NAME bert
echo
fi
# if [[ "$task" = "predict_did_lm" ]]; then
# lang="arabic"
# dataset="madar"
# efile="tr_output/${base_model}_${lang}_${dataset}_${task}.err"
# ofile="tr_output/${base_model}_${lang}_${dataset}_${task}.out"
# echo ${efile}
# echo ${ofile}
# echo ${base_model}
# echo ${lang}
# echo ${dataset}
# sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${lang} ${base_model} ${dataset}
# # ./install.sh --task train_did_lm --lang arabic --dataset madar --MODEL_NAME bert
# fi
if [[ "$task" = "train_predict_did_ml" ]]; then
lang="arabic"
dataset="madar"
efile="tr_output/${base_model}_${lang}_${dataset}_${task}.err"
ofile="tr_output/${base_model}_${lang}_${dataset}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${lang}
echo ${dataset}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${lang} ${base_model} ${dataset}
echo
fi
if [[ "$task" = "train_topic_classification_lm" || "$task" = "predict_topic_classification_lm" ]]; then
export ALL_LANGS=("eng_Latn" "ita_Latn" "azj_Latn" "ckb_Arab" "nob_Latn" "nld_Latn" "lvs_Latn" "arb_Arab" "lij_Latn" "zho_Hans" "spa_Latn" "nso_Latn")
dataset="sib"
for lang in ${ALL_LANGS[@]}; do
efile="tr_output/${base_model}_${lang}_${dataset}_${task}.err"
ofile="tr_output/${base_model}_${lang}_${dataset}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${lang}
echo ${dataset}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${lang} ${base_model} ${dataset}
# ./install.sh --task train_topic_classification_lm --lang eng_Latn --dataset sib --MODEL_NAME bert
done
fi
if [[ "$task" = "train_reading_comprehension" || "$task" = "predict_reading_comprehension" ]]; then
lang="eng_Latn"
dataset="Belebele"
efile="tr_output/${base_model}_${lang}_${dataset}_${task}.err"
ofile="tr_output/${base_model}_${lang}_${dataset}_${task}.out"
echo ${efile}
echo ${ofile}
echo ${base_model}
echo ${lang}
echo ${dataset}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${lang} ${base_model} ${dataset}
# ./install.sh --task ${task} --lang ${lang} --dataset ${dataset} --MODEL_NAME ${base_model}
# ./install.sh --task train_reading_comprehension --lang eng_Latn --dataset Belebele --MODEL_NAME bert
# echo
fi
if [[ "$task" = "llm_evaluation" ]]; then
efile="tr_output/aya-${subtask}-${prompt_lang}.err"
ofile="tr_output/aya-${subtask}-${prompt_lang}.out"
echo ${efile}
echo ${ofile}
sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${subtask} ${MODEL_NAME} ${prompt_lang}
# ./command-slurm.sh --task llm_evaluation --subtask topic --MODEL_NAME aya --prompt_lang zeroshot
# ./command-slurm.sh --task llm_evaluation --subtask topic --MODEL_NAME aya --prompt_lang in_lang
# ./install.sh --task ${task} --lang ${lang} --dataset ${dataset} --MODEL_NAME ${base_model}
# ./install.sh --task train_reading_comprehension --lang eng_Latn --dataset Belebele --MODEL_NAME bert
# echo
fi
if [[ "$task" = "llm_evaluation_gpt" ]]; then
efile="tr_output/gpt-${subtask}.err"
ofile="tr_output/gpt-${subtask}.out"
echo ${efile}
echo ${ofile}
sbatch -o ${ofile} -e ${efile} slurm/run_normal.slurm ${task} ${subtask} ${MODEL_NAME} ${prompt_lang}
# ./install.sh --task ${task} --lang ${lang} --dataset ${dataset} --MODEL_NAME ${base_model}
# ./install.sh --task train_reading_comprehension --lang eng_Latn --dataset Belebele --MODEL_NAME bert
# echo
fi
if [[ "$task" = "llm_evaluation_async_gpt4" ]]; then
efile="tr_output/gpt4-${subtask}-${prompt_lang}.err"
ofile="tr_output/gpt4-${subtask}-${prompt_lang}.out"
echo ${efile}
echo ${ofile}
sbatch -o ${ofile} -e ${efile} slurm/run_normal.slurm ${task} ${subtask} ${MODEL_NAME} ${prompt_lang}
# ./command-slurm.sh --task llm_evaluation_async_gpt4 --subtask topic --MODEL_NAME gpt4 --prompt_lang zeroshot
# ./command-slurm.sh --task llm_evaluation_async_gpt4 --subtask topic --MODEL_NAME gpt4 --prompt_lang in_language
# ./command-slurm.sh --task llm_evaluation_async_gpt4 --subtask belebele --MODEL_NAME gpt4 --prompt_lang combined
# ./command-slurm.sh --task llm_evaluation_async_gpt4 --subtask sdqa --MODEL_NAME gpt4 --prompt_lang zeroshot
# ./command-slurm.sh --task llm_evaluation_async_gpt4 --subtask sdqa --MODEL_NAME gpt4 --prompt_lang combined
# ./command-slurm.sh --task llm_evaluation_async_gpt4 --subtask sentiment --MODEL_NAME gpt4 --prompt_lang in_language
# ./install.sh --task ${task} --lang ${lang} --dataset ${dataset} --MODEL_NAME ${base_model}
# ./install.sh --task train_reading_comprehension --lang eng_Latn --dataset Belebele --MODEL_NAME bert
# echo
fi
# if [[ "$task" = "predict_topic_classification_lm" ]]; then
# lang="eng_Latn"
# dataset="sib"
# efile="tr_output/${base_model}_${lang}_${dataset}_${task}.err"
# ofile="tr_output/${base_model}_${lang}_${dataset}_${task}.out"
# echo ${efile}
# echo ${ofile}
# echo ${base_model}
# echo ${lang}
# echo ${dataset}
# # sbatch -o ${ofile} -e ${efile} slurm/run_udp.slurm ${task} ${lang} ${base_model} ${dataset}
# ./install.sh --task predict_topic_classification_lm --lang eng_Latn --dataset sib --MODEL_NAME bert
# fi