-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathinference_cnn_classifier.py
157 lines (130 loc) · 6.16 KB
/
inference_cnn_classifier.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
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
import torch
import torch.nn as nn
from torch import optim
import numpy as np
import pandas as pd
import pickle
from sklearn.model_selection import KFold
from src.model_cnn import *
from src.data_multi_input import *
import argparse
import datetime
from os import listdir
from os.path import isdir, join, basename, dirname
import pdb
parser = argparse.ArgumentParser(description='Parser for inference.')
parser.add_argument('-k', '--k_folds', type=int, default=5,
help='number of folds for cross validation (default: 5)')
parser.add_argument('-b', '--batch_size', type=int, default=4,
help='size of batch (default: 20)')
# parser.add_argument('--embedding_dim', type=int, default=24,
# help='embedding dimension (default: 24)')
parser.add_argument('--species', type=bool, default=True,
help='add species feature (default: True)')
parser.add_argument('--hidden_dim', type=int, default=128,
help='hidden dimension (default: 64)')
parser.add_argument('--data_path', default='./data/210729_drop.csv',
help='path for train dataframe (default: ./data/210729_drop.csv')
parser.add_argument('--key', default='F primer',
help='sequence type for prediction (default: F primer)')
parser.add_argument('--model_path',
help='path for trained encoder (default: latest model)')
parser.add_argument('--word_dict', default='./data/word_dict.pkl',
help='path for word dict (default: ./data/word_dict.pkl)')
parser.add_argument('--kernel_size', type=int, default=3,
help='conv kernel size (default: 3)')
# parser.add_argument('--debug', type=bool, default=False,
# help='debug mode (default: False)')
parser.add_argument('--max_len', type=int, default=40,
help='max sequence length (default: 40)')
parser.add_argument('--target_name', type=str, default='ct',
help='target name to train model (default: ct)')
args = parser.parse_args()
if __name__ == '__main__':
gpu = 0
device = torch.device(f"cuda:{gpu}" if torch.cuda.is_available() else "cpu")
k_folds = args.k_folds
batch_size = args.batch_size # same as len(testloader) ?
# emb_dim = args.embedding_dim
hidn_dim = args.hidden_dim
key = args.key
spec = args.species
model_root_path = './model/cnn'
result_name = f'./result/result.csv'
# debug_name = '_debug' if args.debug else ''
# result_name = f'./result/result{debug_name}.csv'
# Load recent model
load_model_path = args.model_path
if load_model_path is None:
model_dirs = [d for d in listdir(model_root_path) if isdir(join(model_root_path, d))]
model_dirs.sort()
cur_date = model_dirs[-1]
load_model_path = f'{model_root_path}/{cur_date}/'
# result_name = f'./result/result_{cur_date}.csv'
result_name = join(load_model_path, f'result_{cur_date}.csv')
else:
# cur_date = basename(dirname(load_model_path))
cur_date = load_model_path.split('/')[-1]
# result_name = f'./result/result_{cur_date}.csv'
result_name = join(load_model_path, f'result_{cur_date}.csv')
print(f'Load cnn model at {load_model_path}')
torch.manual_seed(42)
dataset_df = pd.read_csv(args.data_path)
# dataset_test = dataset_df.to_numpy()
word2index_dict = {'A': 0, 'T': 1, 'G': 2, 'C': 3}
# word2index_dict = {'A': 0, 'T': 1, 'G': 2, 'C': 3, 'R':4, 'Y':5, 'M':6, 'K':7}
# with open(args.word_dict,'rb') as f:
# word2index_dict = pickle.load(f)
vocab_size = len(word2index_dict)
dataset = Dataset_FRP(dataset_df, key, target_name=args.target_name)
dataset.set_max_seq_len(args.max_len)
testloader = get_loader_CNN_infer(dataset, batch_size, key, word2index_dict, is_test=True)
result = []
for fold in list(range(k_folds)):
model = MultiInputCNN(args.max_len, vocab_size, hidn_dim, device, args.kernel_size).to(device)
model.eval()
path_cnn = join(load_model_path, f'model_fold{fold}.pth')
fold_result = np.array([])
checkpoint_cnn = torch.load(path_cnn)
model.load_state_dict(checkpoint_cnn)
activation = nn.Sigmoid()
for i, data in enumerate(testloader, 0):
inputs, species = data
if spec == True:
outputs = model(inputs, species)
else:
outputs = model(inputs)
outputs = activation(outputs)
output_cpu = outputs.detach().cpu().numpy()
# classifier
output_cpu = np.where(output_cpu > 0.5, 1, 0)
# fold_result += list(output_cpu)
fold_result = np.concatenate((fold_result, output_cpu), axis=0) if fold_result.size else output_cpu
result.append(fold_result)
if len(result) == 0:
print(f'No model exists')
else:
# print()
# print(np.asarray(result).shape)
# print(np.squeeze(result))
# best results from fold (int value is needed)
# mean = np.mean(result, axis=0)
med = np.median(result, axis=0).astype(np.int32)
print(np.squeeze(med))
dataset_df[args.target_name + '_pred'] = med
# print(result_name)
dataset_df.to_csv(result_name, index=0)
len_data = len(dataset_df.label)
num_normal = sum(dataset_df[args.target_name + '_pred'] == 0)
num_correct = sum(dataset_df.label == dataset_df[args.target_name + '_pred'])
print(f'# of 0 prediction (normal ct) : {num_normal}\n' + \
f'# of correct prediction : {num_correct}\n# of incorrect prediction : {len_data - num_correct}\n' + \
f'Accuarcy : {round(num_correct/len_data, 2)}')
# Save no nan data
no_nan_df = dataset_df[dataset_df[args.target_name + '_pred'] == 0]
# pdb.set_trace()
no_nan_df['ct'] = np.where(no_nan_df['ct'].isnull(), 40, no_nan_df['ct'])
# pdb.set_trace()
tmp_str = args.data_path.replace('.csv', '')
no_nan_df.to_csv(f'{tmp_str}_no_nan.csv', index=0)
print('Inference Completed')