-
Notifications
You must be signed in to change notification settings - Fork 0
/
tests.py
97 lines (81 loc) · 4.36 KB
/
tests.py
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
export CUDA_VISIBLE_DEVICES=0,2,3
nohup python run_modeling.py --batch_size 40 \
--data "data/COVID-QA_cleaned_final.json" \
--model_name "navteca/roberta-base-squad2" \
--dte_lookup_table_fp "NN-DTE-to-navteca-roberta-base-squad2.pkl" \
--lr 1e-5 \
--n_epochs 1 \
--max_len 384 \
--n_stride 196 \
--warmup_proportion 0.1 \
--use_kge T \
--fancy_concat T \
--n_neg_records 4 \
--gpus 3 6 \
--seed 16 \
--port 42013 > result_biobert_kge.out &
nohup python run_modeling.py --batch_size 40 \
--data "data/COVID-QA_cleaned_final.json" \
--model_name "ktrapeznikov/scibert_scivocab_uncased_squad_v2" \
--dte_lookup_table_fp "Mikolov++_to_ktrapeznikov_scibert_scivocab_uncased_squad_v2.pkl" \
--lr 2e-5 \
--n_epochs 1 \
--max_len 384 \
--n_stride 196 \
--warmup_proportion 0.1 \
--use_kge T \
--random_kge T\
--fancy_concat T \
--n_neg_records 4 \
--gpus 3 4 \
--seed 16 \
--port 42020 > result_scibert_kge.out &
nohup python run_modeling.py --batch_size 40 \
--data "data/COVID-QA_cleaned_final.json" \
--model_name "navteca/roberta-base-squad2" \
--dte_lookup_table_fp "NN-DTE-to-navteca-roberta-base-squad2.pkl" \
--lr 2e-5 \
--n_epochs 1 \
--max_len 384 \
--n_stride 196 \
--warmup_proportion 0.1 \
--use_kge T \
--random_kge T \
--fancy_concat T \
--n_neg_records 4 \
--gpus 0 2 3 \
--seed 16 \
--port 42068 > result_roberta.out &
python run_modeling.py --batch_size 40 \
--model_name "ktrapeznikov/biobert_v1.1_pubmed_squad_v2" \
--dte_lookup_table_fp "NN-DTE-to-phiyodr-bert-base-finetuned-squad2.pkl" \
--lr 2e-5 \
--n_epochs 2 \
--max_len 384 \
--n_stride 196 \
--warmup_proportion 0.1 \
--n_neg_records 4 \
--gpus 1 2 \
--seed 16 \
--port 42090
export CUDA_VISIBLE_DEVICES=3,6
nohup python run_modeling.py --batch_size 40 \
--model_name "ktrapeznikov/scibert_scivocab_uncased_squad_v2" \
--dte_lookup_table_fp "NN-DTE-to-ktrapeznikov-scibert_scivocab_uncased_squad_v2.pkl" \
--lr 2e-5 \
--n_epochs 1 \
--max_len 384 \
--n_stride 196 \
--warmup_proportion 0.1 \
--n_neg_records 4 \
--concat_kge True \
--use_kge True \
--use_dict False \
--gpus 3 6 \
--seed 16 \
--port 42056 > results_all.out &
python pykg2vec_tune.py -mn TransE -ds UMLS_KG_MT -dsp /home/Train_KGE/UMLS_KG_MT-original -device cuda
python pykg2vec_train.py -mn TransE -ds UMLS_KG_MT -dsp /home/Train_KGE/UMLS_KG_MT-original -device cuda
python pykg2vec_train.py -mn TransE -ds UMLS_KG_MT -dsp /home/Train_KGE/UMLS_KG_MT-original -device cuda -b 2769 -l 10 -k 170 -l1 True -lr 0.030143630391557222 -mg 0.2616645450619097 -opt sgd
python pykg2vec_train.py -mn DistMult -ds UMLS_KG_MT -dsp /home/Train_KGE/UMLS_KG_MT -device cuda -b 3037 -l 10 -k 135 -lr 0.024743441143928905 -lmda 1.384683308307553e-05 -opt rms
'batch_size': 451, 'epochs': 10, 'hidden_size': 79, 'l1_flag': True, 'learning_rate': 0.0023120564129837724, 'margin': 0.23949684100171093, 'optimizer': 'rms'