args.gpu,
rnn=rnn,
pre_computed_patterns=None)
if args.gpu:
print("Cuda!")
model.to_cuda(model)
state_dict = torch.load(args.input_model)
else:
state_dict = torch.load(args.input_model,
map_location=lambda storage, loc: storage)
# Loading model
model.load_state_dict(state_dict)
interpret_documents(model, args.batch_size, dev_data, dev_text, args.ofile,
args.max_doc_len)
return 0
if __name__ == '__main__':
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
parents=[soft_pattern_arg_parser(),
general_arg_parser()])
parser.add_argument("--ofile", help="Output file", required=True)
sys.exit(main(parser.parse_args()))
else:
rnn = None
model = SoftPatternClassifier(pattern_specs, mlp_hidden_dim, num_mlp_layers, num_classes, embeddings, vocab,
semiring, args.bias_scale_param, args.gpu, rnn=rnn, pre_computed_patterns=None,
no_sl=args.no_sl, shared_sl=args.shared_sl, no_eps=args.no_eps,
eps_scale=args.eps_scale, self_loop_scale=args.self_loop_scale)
if args.gpu:
state_dict = torch.load(args.input_model)
else:
state_dict = torch.load(args.input_model, map_location=lambda storage, loc: storage)
model.load_state_dict(state_dict)
if args.gpu:
model.to_cuda(model)
visualize_patterns(model, dev_data, dev_text, args.k_best, args.max_doc_len, num_padding_tokens)
return 0
if __name__ == '__main__':
parser = argparse.ArgumentParser(description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
parents=[soft_pattern_arg_parser(), general_arg_parser()])
parser.add_argument("-k", "--k_best", help="Number of nearest neighbor phrases", type=int, default=5)
sys.exit(main(parser.parse_args()))
if model_save_dir is not None:
if not os.path.exists(model_save_dir):
os.makedirs(model_save_dir)
print("Training with", model_file_prefix)
train(train_data,
dev_data,
model,
num_classes,
model_save_dir,
args.num_iterations,
model_file_prefix,
args.learning_rate,
args.batch_size,
args.scheduler,
gpu=args.gpu,
clip=args.clip,
debug=args.debug,
dropout=args.dropout,
word_dropout=args.word_dropout,
patience=args.patience)
if __name__ == '__main__':
parser = \
argparse.ArgumentParser(description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
parents=[training_arg_parser(), general_arg_parser()])
main(parser.parse_args())
if model_save_dir is not None:
if not os.path.exists(model_save_dir):
os.makedirs(model_save_dir)
print("Training with", model_file_prefix)
train(train_data,
dev_data,
model,
num_classes,
model_save_dir,
args.num_iterations,
model_file_prefix,
args.learning_rate,
args.batch_size,
args.scheduler,
gpu=args.gpu,
clip=args.clip,
debug=args.debug,
dropout=dropout,
word_dropout=args.word_dropout,
patience=args.patience)
if __name__ == '__main__':
parser = \
argparse.ArgumentParser(description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
parents=[lstm_arg_parser(), mlp_arg_parser(), training_arg_parser(), general_arg_parser()])
main(parser.parse_args())
"--num_cnn_layers",
help="Number of MLP layers",
type=int,
default=2)
p.add_argument("-z",
"--window_size",
help="Size of window of CNN",
type=int,
default=3)
p.add_argument("-o",
"--pooling",
help="Type of pooling to use [max, sum, avg]",
type=str,
default="max")
return p
def pooling_cnn_arg_parser():
p = ArgumentParser(add_help=False,
parents=[cnn_arg_parser(),
mlp_arg_parser()])
return p
if __name__ == '__main__':
parser = \
ArgumentParser(description=__doc__,
formatter_class=ArgumentDefaultsHelpFormatter,
parents=[pooling_cnn_arg_parser(), training_arg_parser(), general_arg_parser()])
main(parser.parse_args())