utils.py

# Copyright (c) 2015-present, Facebook, Inc.
# All rights reserved.
"""
Misc functions, including distributed helpers.
Mostly copy-paste from torchvision references.
# Modified by Chaodong Xiao.
"""
import io
import os
import time
from collections import defaultdict, deque
import datetime

import torch
import torch.distributed as dist
from typing import Any, NewType
from torch.autograd import Function


class SmoothedValue(object):
    """Track a series of values and provide access to smoothed values over a
    window or the global series average.
    """

    def __init__(self, window_size=20, fmt=None):
        if fmt is None:
            fmt = "{median:.4f} ({global_avg:.4f})"
        self.deque = deque(maxlen=window_size)
        self.total = 0.0
        self.count = 0
        self.fmt = fmt

    def update(self, value, n=1):
        self.deque.append(value)
        self.count += n
        self.total += value * n

    def synchronize_between_processes(self):
        """
        Warning: does not synchronize the deque!
        """
        if not is_dist_avail_and_initialized():
            return
        t = torch.tensor([self.count, self.total], dtype=torch.float64, device='cuda')
        dist.barrier()
        dist.all_reduce(t)
        t = t.tolist()
        self.count = int(t[0])
        self.total = t[1]

    @property
    def median(self):
        d = torch.tensor(list(self.deque))
        return d.median().item()

    @property
    def avg(self):
        d = torch.tensor(list(self.deque), dtype=torch.float32)
        return d.mean().item()

    @property
    def global_avg(self):
        return self.total / self.count

    @property
    def max(self):
        return max(self.deque)

    @property
    def value(self):
        return self.deque[-1]

    def __str__(self):
        return self.fmt.format(
            median=self.median,
            avg=self.avg,
            global_avg=self.global_avg,
            max=self.max,
            value=self.value)


class MetricLogger(object):
    def __init__(self, delimiter="\t"):
        self.meters = defaultdict(SmoothedValue)
        self.delimiter = delimiter

    def update(self, **kwargs):
        for k, v in kwargs.items():
            if isinstance(v, torch.Tensor):
                v = v.item()
            assert isinstance(v, (float, int))
            self.meters[k].update(v)

    def __getattr__(self, attr):
        if attr in self.meters:
            return self.meters[attr]
        if attr in self.__dict__:
            return self.__dict__[attr]
        raise AttributeError("'{}' object has no attribute '{}'".format(
            type(self).__name__, attr))

    def __str__(self):
        loss_str = []
        for name, meter in self.meters.items():
            loss_str.append(
                "{}: {}".format(name, str(meter))
            )
        return self.delimiter.join(loss_str)

    def synchronize_between_processes(self):
        for meter in self.meters.values():
            meter.synchronize_between_processes()

    def add_meter(self, name, meter):
        self.meters[name] = meter

    def log_every(self, iterable, print_freq, header=None):
        i = 0
        if not header:
            header = ''
        start_time = time.time()
        end = time.time()
        iter_time = SmoothedValue(fmt='{avg:.4f}')
        data_time = SmoothedValue(fmt='{avg:.4f}')
        space_fmt = ':' + str(len(str(len(iterable)))) + 'd'
        log_msg = [
            header,
            '[{0' + space_fmt + '}/{1}]',
            'eta: {eta}',
            '{meters}',
            'time: {time}',
            'data: {data}'
        ]
        if torch.cuda.is_available():
            log_msg.append('max mem: {memory:.0f}')
        log_msg = self.delimiter.join(log_msg)
        MB = 1024.0 * 1024.0
        for obj in iterable:
            data_time.update(time.time() - end)
            yield obj
            iter_time.update(time.time() - end)
            if i % print_freq == 0 or i == len(iterable) - 1:
                eta_seconds = iter_time.global_avg * (len(iterable) - i)
                eta_string = str(datetime.timedelta(seconds=int(eta_seconds)))
                if torch.cuda.is_available():
                    print(log_msg.format(
                        i, len(iterable), eta=eta_string,
                        meters=str(self),
                        time=str(iter_time), data=str(data_time),
                        memory=torch.cuda.max_memory_allocated() / MB))
                else:
                    print(log_msg.format(
                        i, len(iterable), eta=eta_string,
                        meters=str(self),
                        time=str(iter_time), data=str(data_time)))
            i += 1
            end = time.time()
        total_time = time.time() - start_time
        total_time_str = str(datetime.timedelta(seconds=int(total_time)))
        print('{} Total time: {} ({:.4f} s / it)'.format(
            header, total_time_str, total_time / len(iterable)))


def _load_checkpoint_for_ema(model_ema, checkpoint):
    """
    Workaround for ModelEma._load_checkpoint to accept an already-loaded object
    """
    mem_file = io.BytesIO()
    torch.save({'state_dict_ema':checkpoint}, mem_file)
    mem_file.seek(0)
    model_ema._load_checkpoint(mem_file)


def setup_for_distributed(is_master):
    """
    This function disables printing when not in master process
    """
    import builtins as __builtin__
    builtin_print = __builtin__.print

    def print(*args, **kwargs):
        force = kwargs.pop('force', False)
        if is_master or force:
            builtin_print(*args, **kwargs)

    __builtin__.print = print


def is_dist_avail_and_initialized():
    if not dist.is_available():
        return False
    if not dist.is_initialized():
        return False
    return True


def get_world_size():
    if not is_dist_avail_and_initialized():
        return 1
    return dist.get_world_size()


def get_rank():
    if not is_dist_avail_and_initialized():
        return 0
    return dist.get_rank()


def is_main_process():
    return get_rank() == 0


def save_on_master(*args, **kwargs):
    if is_main_process():
        torch.save(*args, **kwargs)


def init_distributed_mode(args):
    if 'RANK' in os.environ and 'WORLD_SIZE' in os.environ:
        args.rank = int(os.environ["RANK"])
        args.world_size = int(os.environ['WORLD_SIZE'])
        args.gpu = int(os.environ['LOCAL_RANK'])
    elif 'SLURM_PROCID' in os.environ:
        args.rank = int(os.environ['SLURM_PROCID'])
        args.gpu = args.rank % torch.cuda.device_count()
    else:
        print('Not using distributed mode')
        args.distributed = False
        return

    args.distributed = True

    torch.cuda.set_device(args.gpu)
    args.dist_backend = 'nccl'
    print('| distributed init (rank {}): {}'.format(
        args.rank, args.dist_url), flush=True)
    torch.distributed.init_process_group(backend=args.dist_backend, init_method=args.dist_url,
                                         world_size=args.world_size, rank=args.rank)
    torch.distributed.barrier()
    setup_for_distributed(args.rank == 0)




""" Quantization """

BinaryTensor = NewType('BinaryTensor', torch.Tensor)  # A type where each element is in {-1, 1}

def binary_sign(x: torch.Tensor) -> BinaryTensor:
    """Return -1 if x < 0, 1 if x >= 0."""
    return x.sign() + (x == 0).type(torch.float)  


class STESign(Function):
    """
    Binarize tensor using sign function.
    Straight-Through Estimator (STE) is used to approximate the gradient of sign function.
    """

    @staticmethod
    def forward(ctx: Any, x: torch.Tensor) -> BinaryTensor: 
        """
        Return a Sign tensor.

        Args:
            ctx: context
            x: input tensor

        Returns:
            Sign(x) = (x>=0) - (x<0)
            Output type is float tensor where each element is either -1 or 1.
        """
        ctx.save_for_backward(x)
        sign_x = binary_sign(x)
        return sign_x

    @staticmethod
    def backward(ctx: Any, grad_output: torch.Tensor) -> torch.Tensor:
        """
        Compute gradient using STE.

        Args:
            ctx: context
            grad_output: gradient w.r.t. output of Sign

        Returns:
            Gradient w.r.t. input of the Sign function
        """
        x, = ctx.saved_tensors
        grad_input = grad_output.clone()
        grad_input[x.gt(1)] = 0
        grad_input[x.lt(-1)] = 0
        return grad_input

binarize = STESign.apply    


class SymQuantizer(Function):
    """
    uniform quantization
    """
    @staticmethod
    def forward(ctx, input, clip_val, num_bits, layerwise=False):
        """
        :param ctx:
        :param input: tensor to be quantized
        :param clip_val: clip val
        :param num_bits: number of bits
        :return: quantized tensor
        """
        ctx.save_for_backward(input, clip_val)
        
        if layerwise:
            max_input = torch.max(torch.abs(input)).expand_as(input)
        else:
            assert input.ndimension() == 4
            max_input = (
                    torch.max(torch.abs(input), dim=-2, keepdim=True)[0]
                    .expand_as(input)
                    .detach()
                )

        s = (2 ** (num_bits - 1) - 1) / (max_input + 1e-6)

        output = torch.round(input * s).div(s + 1e-6)

        return output
    
    @staticmethod
    def backward(ctx, grad_output):
        """
        :param ctx: saved non-clipped full-precision tensor and clip_val
        :param grad_output: gradient ert the quantized tensor
        :return: estimated gradient wrt the full-precision tensor
        """
        input, clip_val = ctx.saved_tensors 
        grad_input = grad_output.clone()
        grad_input[input.ge(clip_val[1])] = 0
        grad_input[input.le(clip_val[0])] = 0
        return grad_input, None, None, None

symquantize = SymQuantizer.apply