unet3d.py

#!/usr/bin/env python

"""

Purpose : 

"""

import torch
import torch.nn as nn
import torch.utils.data

__author__ = "Kartik Prabhu, Mahantesh Pattadkal, and Soumick Chatterjee"
__copyright__ = "Copyright 2020, Faculty of Computer Science, Otto von Guericke University Magdeburg, Germany"
__credits__ = ["Kartik Prabhu", "Mahantesh Pattadkal", "Soumick Chatterjee"]
__license__ = "GPL"
__version__ = "1.0.0"
__maintainer__ = "Soumick Chatterjee"
__email__ = "[email protected]"
__status__ = "Production"


class ConvBlock(nn.Module):
    """
    Convolution Block
    """

    def __init__(self, in_channels, out_channels, k_size=3, stride=1, padding=1, bias=True, dropout_rate=None):
        super(ConvBlock, self).__init__()
        if bool(dropout_rate):
            self.conv = nn.Sequential(
                nn.Conv3d(in_channels=in_channels, out_channels=out_channels, kernel_size=k_size,
                        stride=stride, padding=padding, bias=bias),
                nn.BatchNorm3d(num_features=out_channels),
                nn.LeakyReLU(inplace=True),
                nn.Dropout3d(p=dropout_rate), #This changes the order in the sequential model
                nn.Conv3d(in_channels=out_channels, out_channels=out_channels, kernel_size=k_size,
                        stride=stride, padding=padding, bias=bias),
                nn.BatchNorm3d(num_features=out_channels),
                nn.LeakyReLU(inplace=True)
            )
        else:
            self.conv = nn.Sequential(
                nn.Conv3d(in_channels=in_channels, out_channels=out_channels, kernel_size=k_size,
                        stride=stride, padding=padding, bias=bias),
                nn.BatchNorm3d(num_features=out_channels),
                nn.LeakyReLU(inplace=True),
                nn.Conv3d(in_channels=out_channels, out_channels=out_channels, kernel_size=k_size,
                        stride=stride, padding=padding, bias=bias),
                nn.BatchNorm3d(num_features=out_channels),
                nn.LeakyReLU(inplace=True)
            )

    def forward(self, x):
        x = self.conv(x)
        return x


class UpConv(nn.Module):
    """
    Up Convolution Block
    """

    # def __init__(self, in_ch, out_ch):
    def __init__(self, in_channels, out_channels, k_size=3, stride=1, padding=1, bias=True):
        super(UpConv, self).__init__()
        self.up = nn.Sequential(
            nn.Upsample(scale_factor=2),
            nn.Conv3d(in_channels=in_channels, out_channels=out_channels, kernel_size=k_size,
                      stride=stride, padding=padding, bias=bias),
            nn.BatchNorm3d(num_features=out_channels),
            nn.LeakyReLU(inplace=True))

    def forward(self, x):
        x = self.up(x)
        return x


class UNet(nn.Module):
    """
    UNet - Basic Implementation
    Input _ [batch * channel(# of channels of each image) * depth(# of frames) * height * width].
    Paper : https://arxiv.org/abs/1505.04597
    """

    def __init__(self, in_ch=1, out_ch=1, init_features=64, dropout_rate=None):
        super(UNet, self).__init__()

        n1 = init_features  
        filters = [n1, n1 * 2, n1 * 4, n1 * 8, n1 * 16]  # 64,128,256,512,1024

        self.Maxpool1 = nn.MaxPool3d(kernel_size=2, stride=2)
        self.Maxpool2 = nn.MaxPool3d(kernel_size=2, stride=2)
        self.Maxpool3 = nn.MaxPool3d(kernel_size=2, stride=2)
        self.Maxpool4 = nn.MaxPool3d(kernel_size=2, stride=2)

        self.Conv1 = ConvBlock(in_ch, filters[0], dropout_rate=dropout_rate)
        self.Conv2 = ConvBlock(filters[0], filters[1], dropout_rate=dropout_rate)
        self.Conv3 = ConvBlock(filters[1], filters[2], dropout_rate=dropout_rate)
        self.Conv4 = ConvBlock(filters[2], filters[3], dropout_rate=dropout_rate)
        self.Conv5 = ConvBlock(filters[3], filters[4], dropout_rate=dropout_rate)

        self.Up5 = UpConv(filters[4], filters[3])
        self.Up_conv5 = ConvBlock(filters[4], filters[3], dropout_rate=dropout_rate)

        self.Up4 = UpConv(filters[3], filters[2])
        self.Up_conv4 = ConvBlock(filters[3], filters[2], dropout_rate=dropout_rate)

        self.Up3 = UpConv(filters[2], filters[1])
        self.Up_conv3 = ConvBlock(filters[2], filters[1], dropout_rate=dropout_rate)

        self.Up2 = UpConv(filters[1], filters[0])
        self.Up_conv2 = ConvBlock(filters[1], filters[0], dropout_rate=dropout_rate)

        self.Conv = nn.Conv3d(filters[0], out_ch, kernel_size=1, stride=1, padding=0)

    # self.active = torch.nn.Sigmoid()

    def forward(self, x):
        # print("unet")
        # print(x.shape)
        # print(padded.shape)

        e1 = self.Conv1(x)
        # print("conv1:")
        # print(e1.shape)

        e2 = self.Maxpool1(e1)
        e2 = self.Conv2(e2)
        # print("conv2:")
        # print(e2.shape)

        e3 = self.Maxpool2(e2)
        e3 = self.Conv3(e3)
        # print("conv3:")
        # print(e3.shape)

        e4 = self.Maxpool3(e3)
        e4 = self.Conv4(e4)
        # print("conv4:")
        # print(e4.shape)

        e5 = self.Maxpool4(e4)
        e5 = self.Conv5(e5)
        # print("conv5:")
        # print(e5.shape)

        d5 = self.Up5(e5)
        # print("d5:")
        # print(d5.shape)
        # print("e4:")
        # print(e4.shape)
        d5 = torch.cat((e4, d5), dim=1)
        d5 = self.Up_conv5(d5)
        # print("upconv5:")
        # print(d5.size)

        d4 = self.Up4(d5)
        # print("d4:")
        # print(d4.shape)
        d4 = torch.cat((e3, d4), dim=1)
        d4 = self.Up_conv4(d4)
        # print("upconv4:")
        # print(d4.shape)
        d3 = self.Up3(d4)
        d3 = torch.cat((e2, d3), dim=1)
        d3 = self.Up_conv3(d3)
        # print("upconv3:")
        # print(d3.shape)
        d2 = self.Up2(d3)
        d2 = torch.cat((e1, d2), dim=1)
        d2 = self.Up_conv2(d2)
        # print("upconv2:")
        # print(d2.shape)
        out = self.Conv(d2)
        # print("out:")
        # print(out.shape)
        # d1 = self.active(out)

        return [out]


class UNetDeepSup(nn.Module):
    """
    UNet - Basic Implementation
    Input _ [batch * channel(# of channels of each image) * depth(# of frames) * height * width].
    Paper : https://arxiv.org/abs/1505.04597
    """

    def __init__(self, in_ch=1, out_ch=1, init_features=64, dropout_rate=None):
        super(UNetDeepSup, self).__init__()

        n1 = init_features
        filters = [n1, n1 * 2, n1 * 4, n1 * 8, n1 * 16]  # 64,128,256,512,1024

        self.Maxpool1 = nn.MaxPool3d(kernel_size=2, stride=2)
        self.Maxpool2 = nn.MaxPool3d(kernel_size=2, stride=2)
        self.Maxpool3 = nn.MaxPool3d(kernel_size=2, stride=2)
        self.Maxpool4 = nn.MaxPool3d(kernel_size=2, stride=2)

        self.Conv1 = ConvBlock(in_ch, filters[0], dropout_rate=dropout_rate)
        self.Conv2 = ConvBlock(filters[0], filters[1], dropout_rate=dropout_rate)
        self.Conv3 = ConvBlock(filters[1], filters[2], dropout_rate=dropout_rate)
        self.Conv4 = ConvBlock(filters[2], filters[3], dropout_rate=dropout_rate)
        self.Conv5 = ConvBlock(filters[3], filters[4], dropout_rate=dropout_rate)

        # 1x1x1 Convolution for Deep Supervision
        self.Conv_d3 = ConvBlock(filters[1], 1, dropout_rate=None)
        self.Conv_d4 = ConvBlock(filters[2], 1, dropout_rate=None)

        self.Up5 = UpConv(filters[4], filters[3])
        self.Up_conv5 = ConvBlock(filters[4], filters[3], dropout_rate=dropout_rate)

        self.Up4 = UpConv(filters[3], filters[2])
        self.Up_conv4 = ConvBlock(filters[3], filters[2], dropout_rate=dropout_rate)

        self.Up3 = UpConv(filters[2], filters[1])
        self.Up_conv3 = ConvBlock(filters[2], filters[1], dropout_rate=dropout_rate)

        self.Up2 = UpConv(filters[1], filters[0])
        self.Up_conv2 = ConvBlock(filters[1], filters[0], dropout_rate=dropout_rate)

        self.Conv = nn.Conv3d(filters[0], out_ch, kernel_size=1, stride=1, padding=0)

        for submodule in self.modules():
            submodule.register_forward_hook(self.nan_hook)

    # self.active = torch.nn.Sigmoid()

    def nan_hook(self, module, inp, output):
        for i, out in enumerate(output):
            nan_mask = torch.isnan(out)
            if nan_mask.any():
                print("In", self.__class__.__name__)
                torch.save(inp, '/nfs1/sutrave/outputs/nan_values_input/inp_2_Nov.pt')
                raise RuntimeError(" classname " + self.__class__.__name__ + "i " + str(
                    i) + f" module: {module} classname {self.__class__.__name__} Found NAN in output {i} at indices: ",
                                   nan_mask.nonzero(), "where:", out[nan_mask.nonzero()[:, 0].unique(sorted=True)])

    def forward(self, x):
        # print("unet")
        # print(x.shape)
        # print(padded.shape)

        e1 = self.Conv1(x)
        # print("conv1:")
        # print(e1.shape)

        e2 = self.Maxpool1(e1)
        e2 = self.Conv2(e2)
        # print("conv2:")
        # print(e2.shape)

        e3 = self.Maxpool2(e2)
        e3 = self.Conv3(e3)
        # print("conv3:")
        # print(e3.shape)

        e4 = self.Maxpool3(e3)
        e4 = self.Conv4(e4)
        # print("conv4:")
        # print(e4.shape)

        e5 = self.Maxpool4(e4)
        e5 = self.Conv5(e5)
        # print("conv5:")
        # print(e5.shape)

        d5 = self.Up5(e5)
        # print("d5:")
        # print(d5.shape)
        # print("e4:")
        # print(e4.shape)
        d5 = torch.cat((e4, d5), dim=1)
        d5 = self.Up_conv5(d5)
        # print("upconv5:")
        # print(d5.size)

        d4 = self.Up4(d5)
        # print("d4:")
        # print(d4.shape)
        d4 = torch.cat((e3, d4), dim=1)
        d4 = self.Up_conv4(d4)
        d4_out = self.Conv_d4(d4)

        # print("upconv4:")
        # print(d4.shape)
        d3 = self.Up3(d4)
        d3 = torch.cat((e2, d3), dim=1)
        d3 = self.Up_conv3(d3)
        d3_out = self.Conv_d3(d3)

        # print("upconv3:")
        # print(d3.shape)
        d2 = self.Up2(d3)
        d2 = torch.cat((e1, d2), dim=1)
        d2 = self.Up_conv2(d2)
        # print("upconv2:")
        # print(d2.shape)
        out = self.Conv(d2)
        # print("out:")
        # print(out.shape)
        # d1 = self.active(out)

        return [out, d3_out, d4_out]