Attention-based_Human_Activity_Recognition_with_3-axis_Accelerometer_Data_Conversion/train_rp.py at main · mnseong/Attention-based_Human_Activity_Recognition_with_3-axis_Accelerometer_Data_Conversion · GitHub

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
import torch
import numpy as np
import torch.nn as nn
from tqdm import tqdm
from torch.utils.data import DataLoader
from sklearn.metrics import f1_score
import pandas as pd
from dataset import RPDataset
from sklearn.preprocessing import LabelEncoder
import random
import albumentations as A
from sklearn.model_selection import train_test_split
from model import EfficientNet, ViT, ResNet, AttResNet, VGG
import warnings
import os
warnings.filterwarnings('ignore')

device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')

CFG = {
    'IMG_SIZE':224,
    'EPOCHS':30,
    'LEARNING_RATE':3e-4,
    'BATCH_SIZE':32,
    'SEED':42
}

def seed_everything(seed):
    random.seed(seed)
    os.environ['PYTHONHASHSEED'] = str(seed)
    np.random.seed(seed)
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.backends.cudnn.deterministic = True
    torch.backends.cudnn.benchmark = True

seed_everything(CFG['SEED'])
def train_func(model, optimizer, scheduler, device):
    model.to(device)
    #     criterion = FocalLoss().to(device)
    criterion = nn.CrossEntropyLoss().to(device)

    best_val_score = 0
    best_model = None

    for epoch in range(1, CFG['EPOCHS'] + 1):
        model.train()
        train_loss = []

        val_loss = []
        preds, trues = [], []

        for i, data in enumerate(tqdm(RP_train_loader)):
            images, labels = data

            images = images.to(device)
            labels = labels.to(device)

            optimizer.zero_grad()

            output = model(images)
            loss = criterion(output, labels)

            loss.backward()
            optimizer.step()

            train_loss.append(loss.item())

        model.eval()

        with torch.no_grad():
            for i, data in enumerate(tqdm(RP_val_loader)):
                images, labels = data

                images = images.to(device)
                labels = labels.to(device)

                logit = model(images)
                loss = criterion(logit, labels)

                val_loss.append(loss.item())

                preds += logit.argmax(1).detach().cpu().numpy().tolist()
                trues += labels.detach().cpu().numpy().tolist()

            _val_loss = np.mean(val_loss)

        _val_score = f1_score(trues, preds, average='micro')

        _train_loss = np.mean(train_loss)
        print(
            f'Epoch [{epoch}], Train Loss : [{_train_loss:.5f}] Val Loss : [{_val_loss:.5f}] Val F1 : [{_val_score:.5f}]')

        if scheduler is not None:
            scheduler.step(_val_score)

        if best_val_score < _val_score:
            best_val_score = _val_score
            torch.save(model, 'save_model/0429_eff_224_mfcc.pth')


train_df = pd.read_csv('data/mfcc_train_data.csv', index_col = 0)

RP_tfms = A.Compose([
    A.Resize(width=CFG['IMG_SIZE'], height=CFG['IMG_SIZE']),
    A.Normalize()
], p=1)

le = LabelEncoder()
le = le.fit(train_df['action'])
train_df['action'] = le.transform(train_df['action'])

train, val, _, _ = train_test_split(train_df, train_df['action'], test_size=0.1, random_state=CFG['SEED'], stratify=train_df['action'])
train['img_path'] = train['img_path'].apply(lambda x : x.replace('./ETRI_data_RP_png', '../ETRIdata'))
val['img_path'] = val['img_path'].apply(lambda x : x.replace('./ETRI_data_RP_png', '../ETRIdata'))


RP_train_dataset = RPDataset(df=train, rp_path_list=train['img_path'].values, label_list=train['action'].values, tfms=RP_tfms)
RP_train_loader = DataLoader(RP_train_dataset, batch_size = CFG['BATCH_SIZE'], shuffle=True, num_workers=0)

RP_val_dataset = RPDataset(df=val, rp_path_list=val['img_path'].values, label_list=val['action'].values, tfms=RP_tfms)
RP_val_loader = DataLoader(RP_val_dataset, batch_size=CFG['BATCH_SIZE'], shuffle=False, num_workers=0)

_model = EfficientNet(4)
_model = _model.to(device)
_model.eval()
_optimizer = torch.optim.Adam(params=_model.parameters(), lr=CFG["LEARNING_RATE"])
_scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(_optimizer, mode='max', factor=0.5, patience=2, threshold_mode='abs', min_lr=1e-8, verbose=True)

train_func(_model, _optimizer, _scheduler, device)