-
Notifications
You must be signed in to change notification settings - Fork 0
/
config.ini
152 lines (141 loc) · 3.19 KB
/
config.ini
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
[ace2005_joint_er]
datasets = ace2005_joint_er
model_name_or_path = t5-large
num_train_epochs = 400
max_seq_length = 256
max_seq_length_eval = 512
per_device_train_batch_size = 16
per_device_eval_batch_size = 8
do_train = True
do_eval = True
do_predict = True
[conll03]
datasets = conll03
model_name_or_path = t5-large
num_train_epochs = 100
max_seq_length = 256
max_seq_length_eval = 512
per_device_train_batch_size = 16
per_device_eval_batch_size = 30
do_train = True
do_eval = True
do_predict = False
num_beams = 1
[conll04]
datasets = conll04
model_name_or_path = t5-large
num_train_epochs = 1000
max_seq_length = 256
max_seq_length_eval = 512
per_device_train_batch_size = 16
per_device_eval_batch_size = 8
do_train = True
do_eval = False
do_predict = True
[ontonotes]
datasets = ontonotes
model_name_or_path = t5-large
num_train_epochs = 300
max_seq_length = 256
max_seq_length_eval = 256
per_device_train_batch_size = 37
per_device_eval_batch_size = 16
do_train = True
do_eval = True
do_predict = False
[multi_woz]
datasets = multi_woz
model_name_or_path = t5-large
do_train = True
do_predict = True
num_train_epochs = 10
max_seq_length = 512
per_device_train_batch_size = 3
per_device_eval_batch_size = 2
overwrite_cache = True
[conll05_srl]
datasets = CoNLL2005-SRL
model_name_or_path = t5-base
num_train_epochs = 1
max_seq_length = 256
per_device_train_batch_size = 4
do_train = True
do_eval = True
do_predict = True
[conll05_srl_brown]
datasets = CoNLL2005-SRL
eval_datasets = CoNLL2005-SRL-Brown
model_name_or_path = t5-base
num_train_epochs = 1
max_seq_length = 256
per_device_train_batch_size = 4
do_train = False
do_eval = False
do_predict = True
[fewrel_meta]
datasets = FewRelFull
model_name_or_path = t5-base
num_train_epochs = 1
max_seq_length = 256
per_device_train_batch_size = 32
per_device_eval_batch_size = 8
do_train = True
do_eval = True
do_predict = False
episodes = 1
[fewrel_1shot_5way]
datasets = FewRelEpisodic
model_name_or_path = experiments/fewrel_meta-t5-large-ep1-len256-b8-train/episode1/
tokenizer_name = t5-base
num_train_epochs = 1250
max_seq_length = 256
per_device_train_batch_size = 4
do_train = True
do_eval = False
do_predict = True
episodes = 1-10
num_ways = 5
num_shots = 1
num_query = 5
[fewrel_5shot_5way]
datasets = FewRelEpisodic
model_name_or_path = experiments/fewrel_meta-t5-large-ep1-len256-b8-train/episode1/
tokenizer_name = t5-base
num_train_epochs = 250
max_seq_length = 256
per_device_train_batch_size = 4
do_train = True
do_eval = False
do_predict = True
episodes = 1-10
num_ways = 5
num_shots = 5
num_query = 5
[fewrel_1shot_10way]
datasets = FewRelEpisodic
model_name_or_path = experiments/fewrel_meta-t5-large-ep1-len256-b8-train/episode1/
tokenizer_name = t5-base
num_train_epochs = 1250
max_seq_length = 256
per_device_train_batch_size = 4
do_train = True
do_eval = False
do_predict = True
episodes = 1-10
num_ways = 10
num_shots = 1
num_query = 5
[fewrel_5shot_10way]
datasets = FewRelEpisodic
model_name_or_path = experiments/fewrel_meta-t5-large-ep1-len256-b8-train/episode1/
tokenizer_name = t5-base
num_train_epochs = 250
max_seq_length = 256
per_device_train_batch_size = 4
do_train = True
do_eval = False
do_predict = True
episodes = 1-10
num_ways = 10
num_shots = 5
num_query = 5