print("=================FLAGS==================")
for k, v in args.__dict__.items():
print('{}: {}'.format(k, v))
print("========================================")
# seed
args.cuda = torch.cuda.is_available()
torch.manual_seed(args.seed)
if args.cuda:
torch.cuda.manual_seed(args.seed)
# data loader and model
train_loader, test_loader = dataset.get(batch_size=args.batch_size,
data_root=args.data_root,
num_workers=1)
model = model.svhn(n_channel=args.channel)
model = torch.nn.DataParallel(model, device_ids=range(args.ngpu))
if args.cuda:
model.cuda()
# optimizer
optimizer = optim.Adam(model.parameters(), lr=args.lr, weight_decay=args.wd)
decreasing_lr = list(map(int, args.decreasing_lr.split(',')))
print('decreasing_lr: ' + str(decreasing_lr))
best_acc, old_file = 0, None
t_begin = time.time()
try:
for epoch in range(args.epochs):
model.train()
if epoch in decreasing_lr:
optimizer.param_groups[0]['lr'] *= 0.1
print("========================================")
# seed
args.cuda = torch.cuda.is_available()
torch.manual_seed(args.seed)
if args.cuda:
torch.cuda.manual_seed(args.seed)
# data loader and model
train_loader, test_loader = dataset_digits.get(batch_size=args.batch_size,
csv_path=args.csv_path,
data_root=args.data_root,
num_workers=0)
model = model.svhn(n_channel=args.channel,
pretrained=args.use_pretrained,
local_model=args.local_model)
model = torch.nn.DataParallel(model, device_ids=range(args.ngpu))
if args.cuda:
model.cuda()
# optimizer
optimizer = optim.Adam(model.parameters(), lr=args.lr, weight_decay=args.wd)
decreasing_lr = list(map(int, args.decreasing_lr.split(',')))
print('decreasing_lr: ' + str(decreasing_lr))
best_acc, old_file = 0, None
best_loss = 50000
t_begin = time.time()
try:
for epoch in range(args.epochs):
from torch.utils.data import DataLoader
from torch.utils.data import Dataset
import torch.nn.functional as F
from scipy import ndimage
from PIL import Image
import matplotlib.pyplot as plt
from sklearn.metrics import average_precision_score, classification_report
from PIL import Image
use_cuda = torch.cuda.is_available()
model_svhn = model.svhn(
32,
pretrained="Local",
local_model=
"C:\\Users\\fcalcagno\\Documents\\pytorch-playground_local\\svhn\\log\\latest.pth"
)
class MyDataset(Dataset):
def __init__(self, transform=None, target_transform=None):
self.data = [(cv2.imread(file), file) for file in glob.glob(
"C:\\Users\\fcalcagno\\Documents\\pytorch-playground_local\\svhn\\testingimages\\test\\*.png"
)]
self.transform = transform
def __getitem__(self, index):
img1 = self.data[index][0]
img1 = cv2.resize(img1, (32, 32))