added missing file

ava/models/blocks/blocks.py

+import torch
+from torch import nn
+from torch.autograd import Variable
+import torch.nn.functional as nnf
+
+
+class _PredictionOutput(nn.Module):
+
+    def __init__(self, input_channels, n_classes, with_softmax=False, upsample_mode='bilinear', dropout=None):
+        super().__init__()
+        self.upsample_mode = upsample_mode
+        self.with_softmax = with_softmax
+        self.n_classes = n_classes
+        self.input_channels = input_channels
+
+        self.dropout = nn.Dropout2d()
+
+        n_features = int(0.5 * input_channels + 0.5 * n_classes)
+
+        seq = [nn.Conv2d(input_channels, n_features, kernel_size=3, padding=1)]
+        if dropout is not None:
+            seq += [nn.Dropout2d(p=dropout)]
+        seq += [nn.Conv2d(n_features, n_classes, kernel_size=3, padding=1)]
+
+        self.convs = nn.Sequential(*seq)
+
+        # self.convs = nn.Sequential(
+        #     nn.Conv2d(input_channels, n_features, kernel_size=3, padding=1),
+        #     nn.Conv2d(n_features, n_classes, kernel_size=3, padding=1),
+        # )
+
+    def forward(self, x, target_size):
+        x_up = nnf.upsample(x, target_size, mode=self.upsample_mode)
+        pred = self.convs(x_up)
+
+        if self.with_softmax:
+            pred = nnf.softmax(pred, dim=1)
+
+        return pred
+
+
+class _DecoderBlock(nn.Module):
+
+    def __init__(self, input_channels, output_channels, dropout=None):
+        super().__init__()
+
+        self.input_channels = input_channels
+        self.output_channels = output_channels
+
+        self.conv = nn.Conv2d(input_channels, output_channels, kernel_size=3, padding=1)
+        self.dropout = nn.Dropout2d(p=dropout) if dropout is not None else None
+
+    def forward(self, x, skip_x):
+        x_up = nnf.upsample(x, size=(skip_x.size(2), skip_x.size(3)), mode='bilinear', align_corners=False)
+        x = torch.cat([x_up, skip_x], dim=1)
+        x = self.conv(x)
+        if self.dropout is not None:
+            x = self.dropout(x)
+
+        x = nnf.relu(x)
+        return x
\ No newline at end of file