Update encoder

Author: qubvel

segmentation_models_pytorch/encoders/efficientnet.py

@@ -31,8 +31,6 @@
 
 
 class EfficientNetEncoder(EfficientNet, EncoderMixin):
-    _is_torch_scriptable = False
-
     def __init__(
         self,
         stage_idxs: List[int],
@@ -41,7 +39,7 @@ def __init__(
         depth: int = 5,
         output_stride: int = 32,
     ):
-        if depth > 5 or depth < 1:
+        if depth > 5 or depth < 2:
             raise ValueError(
                 f"{self.__class__.__name__} depth should be in range [1, 5], got {depth}"
             )
@@ -50,11 +48,13 @@ def __init__(
         super().__init__(blocks_args, global_params)
 
         self._stage_idxs = stage_idxs
+        self._out_indexes = [x - 1 for x in stage_idxs]
         self._depth = depth
         self._in_channels = 3
         self._out_channels = out_channels
         self._output_stride = output_stride
 
+        self._drop_connect_rate = self._global_params.drop_connect_rate
         del self._fc
 
     def get_stages(self) -> Dict[int, Sequence[torch.nn.Module]]:
@@ -63,17 +63,6 @@ def get_stages(self) -> Dict[int, Sequence[torch.nn.Module]]:
             32: [self._blocks[self._stage_idxs[2] :]],
         }
 
-    def apply_blocks(
-        self, x: torch.Tensor, start_idx: int, end_idx: int
-    ) -> torch.Tensor:
-        drop_connect_rate = self._global_params.drop_connect_rate
-
-        for block_number in range(start_idx, end_idx):
-            drop_connect_prob = drop_connect_rate * block_number / len(self._blocks)
-            x = self._blocks[block_number](x, drop_connect_prob)
-
-        return x
-
     def forward(self, x: torch.Tensor) -> List[torch.Tensor]:
         features = [x]
 
@@ -83,21 +72,19 @@ def forward(self, x: torch.Tensor) -> List[torch.Tensor]:
             x = self._swish(x)
             features.append(x)
 
-        if self._depth >= 2:
-            x = self.apply_blocks(x, 0, self._stage_idxs[0])
-            features.append(x)
+        depth = 1
+        for i, block in enumerate(self._blocks):
+            drop_connect_prob = self._drop_connect_rate * i / len(self._blocks)
+            x = block(x, drop_connect_prob)
 
-        if self._depth >= 3:
-            x = self.apply_blocks(x, self._stage_idxs[0], self._stage_idxs[1])
-            features.append(x)
+            if i in self._out_indexes:
+                features.append(x)
+                depth += 1
 
-        if self._depth >= 4:
-            x = self.apply_blocks(x, self._stage_idxs[1], self._stage_idxs[2])
-            features.append(x)
+            if not torch.jit.is_scripting() and depth > self._depth:
+                break
 
-        if self._depth >= 5:
-            x = self.apply_blocks(x, self._stage_idxs[2], len(self._blocks))
-            features.append(x)
+        features = features[: self._depth + 1]
 
         return features