crcrpar

from __future__ import print_function
import collections
import os

import numpy as np

import chainer
import chainer.functions as F
import chainer.links as L

crcrpar

"""Modified VGG16 to compute perceptual loss.

This class is mostly copied from pytorch/examples.
See, fast_neural_style in https://github.com/pytorch/examples.
"""
import torch
from torchvision import models


class VGG_OUTPUT(object):

crcrpar

import torch
import torch.nn as nn
from torchvision.models import vgg19


"""
sequential (
    # block_1
  (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (1): ReLU (inplace)

crcrpar

from __future__ import print_function
import collections
import os

import numpy
try:
    from PIL import Image
    available = True
except ImportError as e:
    available = False

crcrpar

import numpy

import chainer
from chainer import configuration
from chainer import cuda
from chainer import functions
from chainer import initializers
from chainer import link
from chainer.utils import argument
from chainer import variable

crcrpar

from __future__ import print_function
import argparse
from datetime import datetime as dt

import numpy as np

import onnx_caffe2.backend

import torch
import torch.nn as nn

crcrpar

FROM caffe2ai/caffe2:c2v0.8.1.cpu.min.ubuntu16.04
LABEL maintainer="aaronmarkham@fb.com"

# caffe2 install with cpu support

RUN apt-get update && apt-get install -y --no-install-recommends \
    libgflags-dev \
    libgtest-dev \
    libiomp-dev \
    libleveldb-dev \

crcrpar

FROM nvidia/cuda:9.0-cudnn7-devel-ubuntu16.04

RUN apt-get update && apt-get install -y --no-install-recommends \
         build-essential \
         cmake \
         git \
         curl \
         vim \
         ca-certificates \
         libjpeg-dev \

crcrpar

import chainer
from chainer.backends import cuda
import chainer.link_hooks
from chainer.link_hooks import _ForwardPreprocessCallbackArgs
from chainer import variable


def _get_axis(ndim, axis):
    axes = [axis]
    for i in range(ndim):

crcrpar

import torch.nn as nn
from torch.nn.utils import spectral_norm


if __name__ == '__main__':
    layer = nn.Conv2d(3, 15, 3, 1, 1)
    layer = spectral_norm(layer)

    torch.save(layer.state_dict(), 'tmp.pth')