from __future__ import print_function

import argparse

import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.autograd import Variable
from torchvision import datasets, transforms, models

import models_densenet
import sys
#from pytorch.models import *

# Training settings
parser = argparse.ArgumentParser(description='PyTorch MNIST Example')
parser.add_argument('--batch-size', type=int, default=64, metavar='N',
                    help='input batch size for training (default: 64)')
parser.add_argument('--test-batch-size', type=int, default=1000, metavar='N',
                    help='input batch size for testing (default: 1000)')
parser.add_argument('--epochs', type=int, default=10, metavar='N',
                    help='number of epochs to train (default: 10)')
parser.add_argument('--lr', type=float, default=0.01, metavar='LR',
                    help='learning rate (default: 0.01)')
parser.add_argument('--momentum', type=float, default=0.5, metavar='M',
                    help='SGD momentum (default: 0.5)')
parser.add_argument('--no-cuda', action='store_true', default=False,
                    help='enables CUDA training')
parser.add_argument('--seed', type=int, default=1, metavar='S',
                    help='random seed (default: 1)')
parser.add_argument('--log-interval', type=int, default=100, metavar='N',
                    help='how many batches to wait before logging training status')
parser.add_argument('--dropout', type=bool, default=False,required=False)
parser.add_argument('--data_root', type=str, default='/data/pytorch/', help='path to data')
args = parser.parse_args()
args.cuda = not args.no_cuda and torch.cuda.is_available()

torch.manual_seed(args.seed)
if args.cuda:
    torch.cuda.manual_seed(args.seed)


kwargs = {'num_workers': 1, 'pin_memory':True} if args.cuda else {}

transform = 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)),
])

trn_data = datasets.CIFAR100(args.data_root+'/cifar100/', train=True, download=True, transform=transform)
tst_data = datasets.CIFAR100(args.data_root+'/cifar100/', train=False, transform=transform)

#print(vars(trn_data))

train_loader = torch.utils.data.DataLoader(trn_data, batch_size=args.batch_size, shuffle=True, **kwargs)
test_loader = torch.utils.data.DataLoader(tst_data, batch_size=args.batch_size, shuffle=True, **kwargs)

def adjust_learning_rate(optimizer, epoch):
    """Sets the learning rate to the initial LR decayed by 10 every 30 epochs"""
    lr = args.lr * (0.1 ** (epoch // 50))
    for param_group in optimizer.param_groups:
        param_group['lr'] = lr

#model = Cifar10Quick(dropout=args.dropout)
model = models.densenet121(num_classes=100)
if args.cuda:
    model.cuda()

optimizer = optim.Adam(model.parameters(), lr=args.lr)
A = None
A_test = None

def train(epoch):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        if args.cuda:
            data, target = data.float().cuda(), target.cuda()
        data, target = Variable(data), Variable(target)
        output, feats, t1, t2 = model(data)
        ce_loss = F.nll_loss(feats, target)
        total_loss = ce_loss
        optimizer.zero_grad()
        total_loss.backward()
        optimizer.step()
        if batch_idx % args.log_interval == 0:
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}, SimplexLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                100. * batch_idx / len(train_loader), ce_loss.data[0], simpl_loss.data[0]))
def test(epoch):
    model.eval()
    test_loss = 0
    correct = 0
    for data, target in test_loader:
        if args.cuda:
            data, target = data.float().cuda(), target.cuda()
        data, target = Variable(data, volatile=True), Variable(target.long()).long()
        output, feats, t1, t2 = model(data)
        test_loss += F.nll_loss(feats, target).data[0]
        pred = feats.data.max(1)[1] # get the index of the max log-probability
        correct += pred.eq(target.data).cpu().sum()

    test_loss = test_loss
    test_loss /= len(test_loader) # loss function already averages over batch size
    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))

for epoch in range(1, args.epochs + 1):
    train(epoch)
    adjust_learning_rate(optimizer, epoch)
    test(epoch)