def main():
device = torch.device('cpu')
model = EfficientNet(1.0, 1.0)
weights = torch.load('efficientnet_torch.pth',
map_location=torch.device('cpu'))
model.load_state_dict(weights)
model.eval()
config = CONFIG()
optimizer = optim.SGD(model.parameters(),
lr=LEARNING_RATE,
momentum=config.momentum,
weight_decay=config.weight_decay)
frame_rate_calc = 1
freq = getTickFrequency()
image_path = []
writing = 0
phoneWithHand = 0
others = 0
sleep = 0
image_path.append("./others/")
image_path.append("./phoneWithHand/")
image_path.append("./sleep/")
image_path.append("./writing/")
for j in range(0, 4):
sumsum = 0
others = 0
phoneWithHand = 0
sleep = 0
writing = 0
for i in range(0, 100):
image = Image.open(image_path[j] + str(i) + '.jpg')
trans = torchvision.transforms.ToTensor()
image = trans(image)
data = image.unsqueeze(0)
t1 = getTickCount()
scores = model(data)
t2 = getTickCount()
time1 = (t2 - t1) / freq
frame_rate_calc = 1 / time1
sumsum += frame_rate_calc
pred = scores.data.max(1)[1]
sumsum /= 100
print('others: ', end='')
print(others)
print('phoneWithHand: ', end='')
print(phoneWithHand)
print('writing: ', end='')
print(writing)
print('sleep: ', end='')
print(sleep)
print('Average FPS: ', end='')
print(sumsum)
pred = scores.data.max(1)[1]
test_correct += pred.eq(target.data).cpu().sum()
print("Predicted {} out of {} correctly".format(test_correct,
total_examples))
return 100.0 * test_correct / (float(total_examples))
if __name__ == '__main__':
torch.cuda.device(0)
model = EfficientNet(1.0, 1.0)
config = CONFIG()
model = model.cuda()
avg_loss = list()
best_accuracy = 0.0
optimizer = optim.SGD(model.parameters(),
lr=LEARNING_RATE,
momentum=config.momentum,
weight_decay=config.weight_decay)
train_acc, val_acc = list(), list()
for i in range(1, EPOCHS + 1):
train_acc.append(train(i))
val_acc.append(val())
save_model(model, i)
print(train_acc)
print(val_acc)