-
Notifications
You must be signed in to change notification settings - Fork 0
/
pe_run_InCaseLaw.py
34 lines (29 loc) · 997 Bytes
/
pe_run_InCaseLaw.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
import pe_app
ENCODER_ID = 'law-ai/InCaseLawBERT' # id from HuggingFace
MODEL_REFERENCE = 'InCaseLaw'
MAX_SEQUENCE_LENGTH = 512
EMBEDDING_DIM = 768
LEARNING_RATE = 1e-5
BATCH_SIZE = 16
DROPOUT_RATE = 0.2
DATASET = '7_roles' # '7_roles' or '4_roles'
N_EPOCHS = 4
COMBINATION = 'C' # S for sum, C for concatenation
train_params = {}
train_params['max_seq_len'] = MAX_SEQUENCE_LENGTH
train_params['learning_rate'] = LEARNING_RATE
train_params['n_epochs'] = N_EPOCHS
train_params['batch_size'] = BATCH_SIZE
train_params['encoder_id'] = ENCODER_ID
train_params['dataset'] = DATASET
train_params['model_reference'] = MODEL_REFERENCE
train_params['dropout_rate'] = DROPOUT_RATE
train_params['embedding_dim'] = EMBEDDING_DIM
train_params['combination'] = COMBINATION
train_params['weight_decay'] = 1e-3
train_params['eps'] = 1e-8
#train_params['n_documents'] = 1
train_params['use_dev_set'] = False
train_params['n_iterations'] = 5
train_params['use_mock'] = False
pe_app.evaluate_BERT(train_params)