# BSD 3-Clause License
#
# Copyright (c) 2017 xxxx
# All rights reserved.
# Copyright 2021 Huawei Technologies Co., Ltd
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright notice, this
#   list of conditions and the following disclaimer.
#
# * Redistributions in binary form must reproduce the above copyright notice,
#   this list of conditions and the following disclaimer in the documentation
#   and/or other materials provided with the distribution.
#
# * Neither the name of the copyright holder nor the names of its
#   contributors may be used to endorse or promote products derived from
#   this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
# ============================================================================

import argparse
import builtins
import os
import random
import shutil
import time
import warnings

import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
import torch.distributed as dist
import torch.optim
import torch.multiprocessing as mp
import torch.utils.data
import torch.utils.data.distributed
import torchvision.transforms as transforms
import torchvision.datasets as datasets
import torchvision.models as models

import apex
from apex import amp
  
model_names = sorted(name for name in models.__dict__
    if name.islower() and not name.startswith("__")
    and callable(models.__dict__[name]))

parser = argparse.ArgumentParser(description='PyTorch ImageNet Training')
parser.add_argument('data', metavar='DIR',
                    help='path to dataset')
parser.add_argument('-a', '--arch', metavar='ARCH', default='resnet50',
                    choices=model_names,
                    help='model architecture: ' +
                        ' | '.join(model_names) +
                        ' (default: resnet50)')
parser.add_argument('-j', '--workers', default=32, type=int, metavar='N',
                    help='number of data loading workers (default: 32)')
parser.add_argument('--epochs', default=100, type=int, metavar='N',
                    help='number of total epochs to run')
parser.add_argument('--start-epoch', default=0, type=int, metavar='N',
                    help='manual epoch number (useful on restarts)')
parser.add_argument('-b', '--batch-size', default=256, type=int,
                    metavar='N',
                    help='mini-batch size (default: 256), this is the total '
                         'batch size of all GPUs on the current node when '
                         'using Data Parallel or Distributed Data Parallel')
parser.add_argument('--lr', '--learning-rate', default=30., type=float,
                    metavar='LR', help='initial learning rate', dest='lr')
parser.add_argument('--schedule', default=[60, 80], nargs='*', type=int,
                    help='learning rate schedule (when to drop lr by a ratio)')
parser.add_argument('--momentum', default=0.9, type=float, metavar='M',
                    help='momentum')
parser.add_argument('--wd', '--weight-decay', default=0., type=float,
                    metavar='W', help='weight decay (default: 0.)',
                    dest='weight_decay')
parser.add_argument('-p', '--print-freq', default=10, type=int,
                    metavar='N', help='print frequency (default: 10)')
parser.add_argument('--resume', default='', type=str, metavar='PATH',
                    help='path to latest checkpoint (default: none)')
parser.add_argument('-e', '--evaluate', dest='evaluate', action='store_true',
                    help='evaluate model on validation set')
parser.add_argument('--world-size', default=-1, type=int,
                    help='number of nodes for distributed training')
parser.add_argument('--rank', default=-1, type=int,
                    help='node rank for distributed training')
parser.add_argument('--dist-url', default='tcp://224.66.41.62:23456', type=str,
                    help='url used to set up distributed training')
parser.add_argument('--dist-backend', default='nccl', type=str,
                    help='distributed backend')
parser.add_argument('--seed', default=None, type=int,
                    help='seed for initializing training. ')
parser.add_argument('--gpu', default=None, type=int,
                    help='GPU or NPU id to use.')
parser.add_argument('--multiprocessing-distributed', action='store_true',
                    help='Use multi-processing distributed training to launch '
                         'N processes per node, which has N GPUs. This is the '
                         'fastest way to use PyTorch for either single node or '
                         'multi node data parallel training')

parser.add_argument('--pretrained', default='', type=str,
                    help='path to moco pretrained checkpoint')

best_acc1 = 0

# amp
parser.add_argument("--amp", action="store_true",
                    help="Use amp to train the model")
parser.add_argument("--loss_scale", default=32.0,
                    type=float, help="amp setting: loss scale, default 32.0")
parser.add_argument("--opt_level", default="O1", type=str,
                    help="amp setting: opt level, default O1")

# device
parser.add_argument("--device", default="npu", type=str,
                    help="device: npu or gpu or cpu")
parser.add_argument('--nDevices', default=8, type=int,
                    help='number of GPUs or NPUs')

def main():
    args = parser.parse_args()

    if args.seed is not None:
        random.seed(args.seed)
        torch.manual_seed(args.seed)
        cudnn.deterministic = True
        warnings.warn('You have chosen to seed training. '
                      'This will turn on the CUDNN deterministic setting, '
                      'which can slow down your training considerably! '
                      'You may see unexpected behavior when restarting '
                      'from checkpoints.')

    if args.gpu is not None:
        warnings.warn('You have chosen a specific GPU or NPU. This will completely '
                      'disable data parallelism.')

    if args.dist_url == "env://" and args.world_size == -1:
        args.world_size = int(os.environ["WORLD_SIZE"])

    args.distributed = args.world_size > 1 or args.multiprocessing_distributed
    print(args)

    ngpus_per_node = args.nDevices
    if args.multiprocessing_distributed:
        # Since we have ngpus_per_node processes per node, the total world_size
        # needs to be adjusted accordingly
        args.world_size = ngpus_per_node * args.world_size
        # Use torch.multiprocessing.spawn to launch distributed processes: the
        # main_worker process function
        mp.spawn(main_worker, nprocs=ngpus_per_node, args=(ngpus_per_node, args))
    else:
        # Simply call main_worker function
        main_worker(args.gpu, ngpus_per_node, args)


def main_worker(gpu, ngpus_per_node, args):
    global best_acc1
    args.gpu = gpu

    # suppress printing if not master
    if args.multiprocessing_distributed and args.gpu != 0:
        def print_pass(*args):
            pass
        builtins.print = print_pass

    if args.gpu is not None:
        print("Use {}: {} for training".format(args.device, args.gpu))

    if args.distributed:
        if args.dist_url == "env://" and args.rank == -1:
            args.rank = int(os.environ["RANK"])
        if args.multiprocessing_distributed:
            # For multiprocessing distributed training, rank needs to be the
            # global rank among all the processes
            args.rank = args.rank * ngpus_per_node + gpu
        dist.init_process_group(backend=args.dist_backend, init_method=args.dist_url,
                                world_size=args.world_size, rank=args.rank)
    # create model
    print("=> creating model '{}'".format(args.arch))
    model = models.__dict__[args.arch]()

    # freeze all layers but the last fc
    for name, param in model.named_parameters():
        if name not in ['fc.weight', 'fc.bias']:
            param.requires_grad = False
    # init the fc layer
    model.fc.weight.data.normal_(mean=0.0, std=0.01)
    model.fc.bias.data.zero_()

    # load from pre-trained, before DistributedDataParallel constructor
    if args.pretrained:
        if os.path.isfile(args.pretrained):
            print("=> loading checkpoint '{}'".format(args.pretrained))
            checkpoint = torch.load(args.pretrained, map_location="cpu")

            # rename moco pre-trained keys
            state_dict = checkpoint['state_dict']
            for k in list(state_dict.keys()):
                # retain only encoder_q up to before the embedding layer
                if k.startswith('module.encoder_q') and not k.startswith('module.encoder_q.fc'):
                    # remove prefix
                    state_dict[k[len("module.encoder_q."):]] = state_dict[k]
                # delete renamed or unused k
                del state_dict[k]

            args.start_epoch = 0
            msg = model.load_state_dict(state_dict, strict=False)
            assert set(msg.missing_keys) == {"fc.weight", "fc.bias"}

            print("=> loaded pre-trained model '{}'".format(args.pretrained))
        else:
            print("=> no checkpoint found at '{}'".format(args.pretrained))

    loc = "cpu"
    device = torch.device(loc)
    if args.device == "gpu":
        loc = "cuda:{}".format(args.gpu)
        device = torch.device(loc)
        torch.cuda.set_device(device)
    elif args.device == "npu":
        loc = "npu:{}".format(args.gpu)
        device = torch.device(loc)
        torch.npu.set_device(device)

    if args.distributed:
        # For multiprocessing distributed, DistributedDataParallel constructor
        # should always set the single device scope, otherwise,
        # DistributedDataParallel will use all available devices.
        if args.gpu is not None:
            model.to(device)
            # When using a single GPU per process and per
            # DistributedDataParallel, we need to divide the batch size
            # ourselves based on the total number of GPUs we have
            args.batch_size = int(args.batch_size / ngpus_per_node)
            args.workers = int((args.workers + ngpus_per_node - 1) / ngpus_per_node)
        else:
            model.to(device)
            # DistributedDataParallel will divide and allocate batch_size to all
            # available GPUs if device_ids are not set
    elif args.gpu is not None:
        model.to(device)
        # comment out the following line for debugging
        # raise NotImplementedError("Only DistributedDataParallel is supported.")
    else:
        # # DataParallel will divide and allocate batch_size to all available GPUs
        # if args.arch.startswith('alexnet') or args.arch.startswith('vgg'):
        #     model.features = torch.nn.DataParallel(model.features)
        #     model.cuda()
        # else:
        #     model = torch.nn.DataParallel(model).cuda()

        # AllGather implementation (batch shuffle, queue update, etc.) in
        # this code only supports DistributedDataParallel.
        raise NotImplementedError("Only DistributedDataParallel is supported.")
    print(model)

    # define loss function (criterion) and optimizer
    criterion = nn.CrossEntropyLoss().to(device)

    if args.amp:
        if args.device == 'npu':
            print(torch.npu.synchronize(), "amp.register_half_function, bmm and einsum")
        amp.register_half_function(torch, 'bmm')
        amp.register_half_function(torch, 'einsum')

    # optimize only the linear classifier
    parameters = list(filter(lambda p: p.requires_grad, model.parameters()))
    assert len(parameters) == 2  # fc.weight, fc.bias
    if args.device == "npu":
        optimizer = apex.optimizers.NpuFusedSGD(parameters, args.lr,
                                    momentum=args.momentum,
                                    weight_decay=args.weight_decay)
    else:
        optimizer = torch.optim.SGD(parameters, args.lr,
                                    momentum=args.momentum,
                                    weight_decay=args.weight_decay)

    if args.amp:
        if args.device == 'npu':
            model, optimizer = amp.initialize(model, optimizer, opt_level=args.opt_level, loss_scale=args.loss_scale, combine_grad=True)
        else:
            model, optimizer = amp.initialize(model, optimizer, opt_level=args.opt_level, loss_scale=args.loss_scale)
    
    if args.distributed:
        if args.gpu is not None:
            model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[device])
        else:
            model = torch.nn.parallel.DistributedDataParallel(model)

    # optionally resume from a checkpoint
    if args.resume:
        if os.path.isfile(args.resume):
            print("=> loading checkpoint '{}'".format(args.resume))
            if args.gpu is None:
                checkpoint = torch.load(args.resume)
            else:
                # Map model to be loaded to specified single gpu.
                checkpoint = torch.load(args.resume, map_location=loc)
            args.start_epoch = checkpoint['epoch']
            best_acc1 = checkpoint['best_acc1']
            if args.gpu is not None:
                # best_acc1 may be from a checkpoint from a different GPU
                best_acc1 = best_acc1.to(args.gpu)
            model.load_state_dict(checkpoint['state_dict'])
            optimizer.load_state_dict(checkpoint['optimizer'])
            print("=> loaded checkpoint '{}' (epoch {})"
                  .format(args.resume, checkpoint['epoch']))
        else:
            print("=> no checkpoint found at '{}'".format(args.resume))

    cudnn.benchmark = True

    # Data loading code
    traindir = os.path.join(args.data, 'train')
    valdir = os.path.join(args.data, 'val')
    normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                     std=[0.229, 0.224, 0.225])

    train_dataset = datasets.ImageFolder(
        traindir,
        transforms.Compose([
            transforms.RandomResizedCrop(224),
            transforms.RandomHorizontalFlip(),
            transforms.ToTensor(),
            normalize,
        ]))

    if args.distributed:
        train_sampler = torch.utils.data.distributed.DistributedSampler(train_dataset)
    else:
        train_sampler = None

    train_loader = torch.utils.data.DataLoader(
        train_dataset, batch_size=args.batch_size, shuffle=(train_sampler is None),
        num_workers=args.workers, pin_memory=True, sampler=train_sampler)

    val_loader = torch.utils.data.DataLoader(
        datasets.ImageFolder(valdir, transforms.Compose([
            transforms.Resize(256),
            transforms.CenterCrop(224),
            transforms.ToTensor(),
            normalize,
        ])),
        batch_size=args.batch_size, shuffle=False,
        num_workers=args.workers, pin_memory=True)

    if args.evaluate:
        validate(val_loader, model, criterion, args, device)
        return

    for epoch in range(args.start_epoch, args.epochs):
        if args.distributed:
            train_sampler.set_epoch(epoch)
        adjust_learning_rate(optimizer, epoch, args)

        # train for one epoch
        train(train_loader, model, criterion, optimizer, epoch, args, device)

        # evaluate on validation set
        acc1 = validate(val_loader, model, criterion, args, device)

        # remember best acc@1 and save checkpoint
        is_best = acc1 > best_acc1
        best_acc1 = max(acc1, best_acc1)

        if not args.multiprocessing_distributed or (args.multiprocessing_distributed
                and args.rank % ngpus_per_node == 0):
            save_checkpoint({
                'epoch': epoch + 1,
                'arch': args.arch,
                'state_dict': model.state_dict(),
                'best_acc1': best_acc1,
                'optimizer' : optimizer.state_dict(),
            }, is_best)
            if epoch == args.start_epoch:
                sanity_check(model.state_dict(), args.pretrained)


def train(train_loader, model, criterion, optimizer, epoch, args, device):
    batch_time = AverageMeter('Time', ':6.3f')
    data_time = AverageMeter('Data', ':6.3f')
    losses = AverageMeter('Loss', ':.4e')
    top1 = AverageMeter('Acc@1', ':6.2f')
    top5 = AverageMeter('Acc@5', ':6.2f')
    FPS = AverageMeter('FPS', ':6.2f')
    progress = ProgressMeter(
        len(train_loader),
        [batch_time, data_time, losses, top1, top5, FPS],
        prefix="Epoch: [{}]".format(epoch))

    """
    Switch to eval mode:
    Under the protocol of linear classification on frozen features/models,
    it is not legitimate to change any part of the pre-trained model.
    BatchNorm in train mode may revise running mean/std (even if it receives
    no gradient), which are part of the model parameters too.
    """
    model.eval()

    if args.device == 'npu':
        torch.npu.synchronize()
    elif args.device == 'cuda':
        torch.cuda.synchronize()
    end = time.time()
    prof_flag = True
    for i, (images, target) in enumerate(train_loader):
        # measure data loading time
        if args.device == 'npu':
            torch.npu.synchronize()
        elif args.device == 'cuda':
            torch.cuda.synchronize()
        data_time.update(time.time() - end)

        if args.gpu is not None:
            images = images.to(device, non_blocking=True)
        target = target.to(device, non_blocking=True)

        if not args.distributed:
            if i == 5 and args.device == "npu":
                with torch.autograd.profiler.profile(use_npu=True) as prof:
                    # compute output
                    output = model(images)
                    loss = criterion(output, target)

                    # acc1/acc5 are (K+1)-way contrast classifier accuracy
                    # measure accuracy and record loss
                    acc1, acc5 = accuracy(output, target, topk=(1, 5))
                    losses.update(loss.item(), images.size(0))
                    top1.update(acc1[0], images.size(0))
                    top5.update(acc5[0], images.size(0))

                    # compute gradient and do SGD step
                    optimizer.zero_grad()
                    if args.amp:
                        with amp.scale_loss(loss, optimizer) as scaled_loss:
                            scaled_loss.backward()
                    else:
                        loss.backward()
                    optimizer.step()
                prof_file = open("npu_output_lincls.txt", 'w')
                print(prof.key_averages().table(sort_by="self_cpu_time_total"), file=prof_file)
                prof.export_chrome_trace("npu_output_lincls.prof")
                prof_flag = False
            elif i == 5 and args.device == "gpu" and prof_flag:
                with torch.autograd.profiler.profile(use_cuda=True) as prof:
                    # compute output
                    output = model(images)
                    loss = criterion(output, target)

                    # acc1/acc5 are (K+1)-way contrast classifier accuracy
                    # measure accuracy and record loss
                    acc1, acc5 = accuracy(output, target, topk=(1, 5))
                    losses.update(loss.item(), images.size(0))
                    top1.update(acc1[0], images.size(0))
                    top5.update(acc5[0], images.size(0))

                    # compute gradient and do SGD step
                    optimizer.zero_grad()
                    if args.amp:
                        with amp.scale_loss(loss, optimizer) as scaled_loss:
                            scaled_loss.backward()
                    else:
                        loss.backward()
                    optimizer.step()
                prof_file = open("cuda_output_lincls.txt", 'w')
                print(prof.key_averages().table(sort_by="self_cpu_time_total"), file=prof_file)
                prof.export_chrome_trace("cuda_output_lincls.prof")
                prof_flag = False
        else:
            # compute output
            output = model(images)
            loss = criterion(output, target)

            # acc1/acc5 are (K+1)-way contrast classifier accuracy
            # measure accuracy and record loss
            acc1, acc5 = accuracy(output, target, topk=(1, 5))
            losses.update(loss.item(), images.size(0))
            top1.update(acc1[0], images.size(0))
            top5.update(acc5[0], images.size(0))

            # compute gradient and do SGD step
            optimizer.zero_grad()
            if args.amp:
                with amp.scale_loss(loss, optimizer) as scaled_loss:
                    scaled_loss.backward()
            else:
                loss.backward()
            optimizer.step()

        # measure elapsed time
        if args.device == 'npu':
            torch.npu.synchronize()
        elif args.device == 'cuda':
            torch.cuda.synchronize()            
        batch_time.update(time.time() - end)
        if args.device == 'npu':
            torch.npu.synchronize()
        elif args.device == 'cuda':
            torch.cuda.synchronize()           
        end = time.time()
        FPS.update(args.batch_size * args.nDevices / batch_time.avg)

        if i % args.print_freq == 0:
            progress.display(i)


def validate(val_loader, model, criterion, args, device):
    batch_time = AverageMeter('Time', ':6.3f')
    losses = AverageMeter('Loss', ':.4e')
    top1 = AverageMeter('Acc@1', ':6.2f')
    top5 = AverageMeter('Acc@5', ':6.2f')
    progress = ProgressMeter(
        len(val_loader),
        [batch_time, losses, top1, top5],
        prefix='Test: ')

    # switch to evaluate mode
    model.eval()

    with torch.no_grad():
        end = time.time()
        for i, (images, target) in enumerate(val_loader):
            if args.gpu is not None:
                images = images.to(device, non_blocking=True)
            target = target.to(device, non_blocking=True)

            # compute output
            output = model(images)
            loss = criterion(output, target)

            # measure accuracy and record loss
            acc1, acc5 = accuracy(output, target, topk=(1, 5))
            losses.update(loss.item(), images.size(0))
            top1.update(acc1[0], images.size(0))
            top5.update(acc5[0], images.size(0))

            # measure elapsed time
            batch_time.update(time.time() - end)
            end = time.time()

            if i % args.print_freq == 0:
                progress.display(i)

        # TODO: this should also be done with the ProgressMeter
        print(' * Acc@1 {top1.avg:.3f} Acc@5 {top5.avg:.3f}'
              .format(top1=top1, top5=top5))

    return top1.avg


def save_checkpoint(state, is_best, filename='model_lincls_latest.pth.tar'):
    torch.save(state, filename)
    if is_best:
        shutil.copyfile(filename, 'model_lincls_best.pth.tar')


def sanity_check(state_dict, pretrained_weights):
    """
    Linear classifier should not change any weights other than the linear layer.
    This sanity check asserts nothing wrong happens (e.g., BN stats updated).
    """
    print("=> loading '{}' for sanity check".format(pretrained_weights))
    checkpoint = torch.load(pretrained_weights, map_location="cpu")
    state_dict_pre = checkpoint['state_dict']

    for k in list(state_dict.keys()):
        # only ignore fc layer
        if 'fc.weight' in k or 'fc.bias' in k:
            continue

        # name in pretrained model
        k_pre = 'module.encoder_q.' + k[len('module.'):] \
            if k.startswith('module.') else 'module.encoder_q.' + k

        assert ((state_dict[k].cpu() == state_dict_pre[k_pre]).all()), \
            '{} is changed in linear classifier training.'.format(k)

    print("=> sanity check passed.")


class AverageMeter(object):
    """Computes and stores the average and current value"""
    def __init__(self, name, fmt=':f'):
        self.name = name
        self.fmt = fmt
        self.reset()

    def reset(self):
        self.val = 0
        self.avg = 0
        self.sum = 0
        self.count = 0

    def update(self, val, n=1):
        self.val = val
        self.sum += val * n
        self.count += n
        self.avg = self.sum / self.count

    def __str__(self):
        fmtstr = '{name} {val' + self.fmt + '} ({avg' + self.fmt + '})'
        return fmtstr.format(**self.__dict__)


class ProgressMeter(object):
    def __init__(self, num_batches, meters, prefix=""):
        self.batch_fmtstr = self._get_batch_fmtstr(num_batches)
        self.meters = meters
        self.prefix = prefix

    def display(self, batch):
        entries = [self.prefix + self.batch_fmtstr.format(batch)]
        entries += [str(meter) for meter in self.meters]
        print('\t'.join(entries))

    def _get_batch_fmtstr(self, num_batches):
        num_digits = len(str(num_batches // 1))
        fmt = '{:' + str(num_digits) + 'd}'
        return '[' + fmt + '/' + fmt.format(num_batches) + ']'


def adjust_learning_rate(optimizer, epoch, args):
    """Decay the learning rate based on schedule"""
    lr = args.lr
    for milestone in args.schedule:
        lr *= 0.1 if epoch >= milestone else 1.
    for param_group in optimizer.param_groups:
        param_group['lr'] = lr


def accuracy(output, target, topk=(1,)):
    """Computes the accuracy over the k top predictions for the specified values of k"""
    with torch.no_grad():
        maxk = max(topk)
        batch_size = target.size(0)

        _, pred = output.topk(maxk, 1, True, True)
        pred = pred.t()
        correct = pred.eq(target.view(1, -1).expand_as(pred))

        res = []
        for k in topk:
            correct_k = correct[:k].view(-1).float().sum(0, keepdim=True)
            res.append(correct_k.mul_(100.0 / batch_size))
        return res


if __name__ == '__main__':
    main()