refactor: ➖ Remove funlib dependency.

setup.cfg

@@ -2,9 +2,9 @@
 name = LeibNetz
 version = 0.2.0
 author = Jeff Rhoades, Larissa Heinrich
-author_email = [email protected], [email protected]
+author_email = [email protected]
 url = https://github.com/janelia-cellmap/LeibNetz
-description = A lightweight and modular library for rapidly developing and constructing PyTorch models deep learning.
+description = A lightweight and modular library for rapidly developing and constructing PyTorch models for deep learning.
 long_description = file: README.md
 long_description_content_type = text/markdown
 keywords = image-segmentation, convolutional-neural-networks, deep-learning, pytorch

src/leibnetz/leibnet.py

@@ -1,11 +1,12 @@
-from typing import Iterable, Optional, Sequence, Tuple, Union
+from typing import Iterable, Sequence, Tuple
 import networkx as nx
 from torch import device
 import torch
 from torch.nn import Module
 import numpy as np
 from leibnetz.nodes import Node
-from funlib.learn.torch.models.conv4d import Conv4d
+
+# from funlib.learn.torch.models.conv4d import Conv4d
 
 # from model_opt.apis import optimize
 
@@ -43,29 +44,26 @@ def __init__(
             self.apply(
                 lambda m: (
                     torch.nn.init.kaiming_normal_(m.weight, mode="fan_out")
-                    if isinstance(m, torch.nn.Conv2d)
-                    or isinstance(m, torch.nn.Conv3d)
-                    or isinstance(m, Conv4d)
+                    if isinstance(m, torch.nn.Conv2d) or isinstance(m, torch.nn.Conv3d)
+                    # or isinstance(m, Conv4d)
                     else None
                 )
             )
         elif initialization == "xavier":
             self.apply(
                 lambda m: (
                     torch.nn.init.xavier_normal_(m.weight)
-                    if isinstance(m, torch.nn.Conv2d)
-                    or isinstance(m, torch.nn.Conv3d)
-                    or isinstance(m, Conv4d)
+                    if isinstance(m, torch.nn.Conv2d) or isinstance(m, torch.nn.Conv3d)
+                    # or isinstance(m, Conv4d)
                     else None
                 )
             )
         elif initialization == "orthogonal":
             self.apply(
                 lambda m: (
                     torch.nn.init.orthogonal_(m.weight)
-                    if isinstance(m, torch.nn.Conv2d)
-                    or isinstance(m, torch.nn.Conv3d)
-                    or isinstance(m, Conv4d)
+                    if isinstance(m, torch.nn.Conv2d) or isinstance(m, torch.nn.Conv3d)
+                    # or isinstance(m, Conv4d)
                     else None
                 )
             )

src/leibnetz/nodes/node_ops.py

@@ -1,6 +1,7 @@
 from torch import nn
 import numpy as np
-from funlib.learn.torch.models.conv4d import Conv4d
+
+# from funlib.learn.torch.models.conv4d import Conv4d
 
 
 class ConvPass(nn.Module):
@@ -95,10 +96,13 @@ def __init__(
             self.dims = len(kernel_size)
 
             try:
-                conv = {2: nn.Conv2d, 3: nn.Conv3d, 4: Conv4d}[self.dims]
+                # TODO: Implement Conv4d
+                # conv = {2: nn.Conv2d, 3: nn.Conv3d, 4: Conv4d}[self.dims]
+                conv = {2: nn.Conv2d, 3: nn.Conv3d}[self.dims]
             except KeyError:
                 raise ValueError(
-                    f"Only 2D, 3D and 4D convolutions are supported, not {self.dims}D"
+                    # f"Only 2D, 3D and 4D convolutions are supported, not {self.dims}D"
+                    f"Only 2D and 3D convolutions are supported, not {self.dims}D"
                 )
 
             layers.append(

src/leibnetz/nodes/resample_ops.py

@@ -1,7 +1,6 @@
-import math
 from torch import nn
-import torch
-from funlib.learn.torch.models.conv4d import Conv4d
+
+# from funlib.learn.torch.models.conv4d import Conv4d
 
 from logging import getLogger
 
@@ -52,7 +51,15 @@ def __init__(
 
         self.dims = len(kernel_sizes)
 
-        conv = {2: nn.Conv2d, 3: nn.Conv3d, 4: Conv4d}[self.dims]
+        try:
+            # TODO: Implement Conv4d
+            # conv = {2: nn.Conv2d, 3: nn.Conv3d, 4: Conv4d}[self.dims]
+            conv = {2: nn.Conv2d, 3: nn.Conv3d}[self.dims]
+        except KeyError:
+            raise ValueError(
+                # f"Only 2D, 3D and 4D convolutions are supported, not {self.dims}D"
+                f"Only 2D and 3D convolutions are supported, not {self.dims}D"
+            )
 
         try:
             layers.append(