-
Notifications
You must be signed in to change notification settings - Fork 19
/
run_k_fold.sh
174 lines (164 loc) · 4.38 KB
/
run_k_fold.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
#!/bin/bash
# usual
python roberta_k_fold.py \
--pre_train_path roberta_large \
--model_name BertBase \
--output_dir usual_k_fold_model/roberta_large \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_large.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path roberta_base \
--model_name BertBase \
--output_dir usual_k_fold_model/roberta_base \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_base.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path roberta_wwm_ext_large \
--model_name BertBase \
--output_dir usual_k_fold_model/roberta_wwm_ext_large \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_wwm_ext_large.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path roberta_large \
--model_name BertLSTMAttention \
--output_dir usual_k_fold_model/roberta_large_lstm_attention \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_large_lstm_attention.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path roberta_large \
--model_name BertAttention \
--output_dir usual_k_fold_model/roberta_large_attention \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_large_attention.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path uer_large \
--model_name BertUER \
--output_dir usual_k_fold_model/uer_large \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir uer_large.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path virus_k_fold_model/roberta_wwm_ext \
--model_name BertTransferLearning \
--output_dir usual_k_fold_model/roberta_wwm_ext_transfer_learning1 \
--data_dir k_fold/usual \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_wwm_ext_transfer_learning1.log \
--k_fold 5 \
--do_train
# virus
python roberta_k_fold.py \
--pre_train_path roberta_large \
--model_name BertBase \
--output_dir virus_k_fold_model/roberta_large \
--data_dir k_fold/virus \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_large.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path usual_k_fold_model/roberta_wwm_ext \
--model_name BertTransferLearning \
--output_dir virus_k_fold_model/roberta_wwm_ext_transfer_learning \
--data_dir k_fold/virus \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_wwm_ext_transfer_learning.log \
--k_fold 5 \
--do_train
python roberta_k_fold.py \
--pre_train_path usual_k_fold_model/roberta_wwm_ext \
--model_name BertTransferLearning \
--output_dir virus_k_fold_model/roberta_wwm_ext_transfer_learning_pseudo \
--data_dir k_fold/virus \
--max_seq_length 140 \
--train_batch_size 8 \
--gradient_accumulation_steps 8 \
--num_train_epochs 20 \
--learning_rate 1e-5 \
--warmup_rate 0.3 \
--weight_decay 0.02 \
--attack fgm \
--log_dir roberta_wwm_ext_transfer_learning_pseudo.log \
--k_fold 5 \
--use_pseudo_data pseudo \
--do_train