'''Train CIFAR10 with PyTorch.'''
import torch
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
import torch.backends.cudnn as cudnn

import torchvision
import torchvision.transforms as transforms

import os
import argparse

from models.mobilenet import MobileNet1
from utils import progress_bar
from replace import replace_sq
from collections import OrderedDict
# from lsq_int import Input_Quantizer
from lsq_sq import Input_Quantizer
from replace_int import replace_int


parser = argparse.ArgumentParser(description='PyTorch CIFAR10 Training')
parser.add_argument('--lr', default=0.1, type=float, help='learning rate')
parser.add_argument('--resume', '-r', default=None, type=str,
                    help='resume from checkpoint')
parser.add_argument('--dir', default='default', type=str,
                    help='save dir name')
parser.add_argument('--test', default=False, action='store_true',
                    help='test version or not')
parser.add_argument('--qat', default=False, action='store_true',
                    help='qat version or not')
args = parser.parse_args()


# Training
def train(epoch):
    print('\nEpoch: %d' % epoch)
    net.train()
    train_loss = 0
    correct = 0
    total = 0
    for batch_idx, (inputs, targets) in enumerate(trainloader):
        inputs, targets = inputs.to(device), targets.to(device)
        optimizer.zero_grad()
        outputs = net(inputs)
        loss = criterion(outputs, targets)
        loss.backward()
        optimizer.step()

        train_loss += loss.item()
        _, predicted = outputs.max(1)
        total += targets.size(0)
        correct += predicted.eq(targets).sum().item()
        progress_bar(batch_idx, len(trainloader), 'Loss: %.3f | Acc: %.3f%% (%d/%d)'
                     % (train_loss/(batch_idx+1), 100.*correct/total, correct, total))


def test(epoch):
    global best_acc
    dir_name = args.dir
    net.eval()
    test_loss = 0
    correct = 0
    total = 0
    with torch.no_grad():
        for batch_idx, (inputs, targets) in enumerate(testloader):

            inputs, targets = inputs.to(device), targets.to(device)
            outputs = net(inputs)
            loss = criterion(outputs, targets)

            test_loss += loss.item()
            _, predicted = outputs.max(1)
            total += targets.size(0)
            correct += predicted.eq(targets).sum().item()
            progress_bar(batch_idx, len(testloader), 'Loss: %.3f | Acc: %.3f%% (%d/%d)'
                         % (test_loss/(batch_idx+1), 100.*correct/total, correct, total))

    # Save checkpoint.
    acc = 100.*correct/total
    if acc > best_acc:
        print('Saving..')
        state = {
            'net': net.state_dict(),
            'acc': acc,
            'epoch': epoch,
        }
        if not os.path.isdir(dir_name):
            os.mkdir(dir_name)
        torch.save(state, f'./{dir_name}/ckpt.pth')
        best_acc = acc
    print('*** best Test Accuracy: ', best_acc)

if __name__ == '__main__':
    device = 'cuda' if torch.cuda.is_available() else 'cpu'
    best_acc = 0  # best test accuracy
    start_epoch = 0  # start from epoch 0 or last checkpoint epoch

    # Data
    print('==> Preparing data..')
    transform_train = transforms.Compose([
        transforms.RandomCrop(32, padding=4),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)),
    ])

    transform_test = transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)),
    ])

    trainset = torchvision.datasets.CIFAR10(
        root='./data', train=True, download=True, transform=transform_train)
    trainloader = torch.utils.data.DataLoader(
        trainset, batch_size=256, shuffle=True, num_workers=4)

    testset = torchvision.datasets.CIFAR10(
        root='./data', train=False, download=True, transform=transform_test)
    testloader = torch.utils.data.DataLoader(
        testset, batch_size=100, shuffle=False, num_workers=4)

    classes = ('plane', 'car', 'bird', 'cat', 'deer',
            'dog', 'frog', 'horse', 'ship', 'truck')

    # Model
    print('==> Building model..')
    net = MobileNet1(3, 10)
    net = net.to(device)

    if args.qat:
        net = replace_sq(model=net)
        net = nn.Sequential(Input_Quantizer(abit=8, dequantize=True),
                            net)

    if args.resume:
        # Load checkpoint.
        print('==> Resuming from checkpoint..')
        assert os.path.isfile(args.resume), 'Error: no checkpoint directory found!'
        checkpoint = torch.load(args.resume)
        new_state_dict = OrderedDict()
        for k, v in checkpoint['net'].items():
            k = k.replace("module.", "")
            new_state_dict[k] = v
        net.load_state_dict(new_state_dict)
        best_acc = 0.0
        start_epoch = 0

    print(net)
    # replace_int(net)

    if device == 'cuda':
        net = torch.nn.DataParallel(net)
        cudnn.benchmark = True
        net.cuda()

    criterion = nn.CrossEntropyLoss()
    optimizer = optim.SGD(net.parameters(), lr=args.lr,
                        momentum=0.9, weight_decay=5e-4)
    scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=200)


    for epoch in range(start_epoch, start_epoch+200):
        if args.test:
            test(epoch)
            break
        else:
            train(epoch)
            test(epoch)
            scheduler.step()