run autoflake

.pre-commit-config.yaml

@@ -24,15 +24,15 @@ repos:
 #   - id: pyupgrade
 #     args: ['--py37-plus']
 
-## If you want to avoid flake8 errors due to unused vars or imports:
-# - repo: https://github.com/PyCQA/autoflake
-#   rev: v2.1.1
-#   hooks:
-#   - id: autoflake
-#     args: [
-#       --in-place,
-#       --remove-all-unused-imports,
-#     ]
+# If you want to avoid flake8 errors due to unused vars or imports:
+- repo: https://github.com/PyCQA/autoflake
+  rev: v2.1.1
+  hooks:
+  - id: autoflake
+    args: [
+      --in-place,
+      --remove-all-unused-imports,
+    ]
 
 # - repo: https://github.com/PyCQA/isort
 #   rev: 5.13.2

equiadapt/common/basecanonicalization.py

@@ -1,4 +1,3 @@
-from abc import ABC, abstractmethod
 from typing import Any, Dict, List, Tuple, Union
 
 import torch

equiadapt/images/canonicalization/discrete_group.py

@@ -1,5 +1,5 @@
 import math
-from typing import Any, List, Tuple, Union
+from typing import List, Tuple, Union
 
 import kornia as K
 import torch

equiadapt/images/canonicalization_networks/custom_equivariant_networks.py

@@ -1,6 +1,4 @@
-import math
 
-import kornia as K
 import torch
 import torch.nn as nn
 from .custom_group_equivariant_layers import RotationEquivariantConvLift, RotationEquivariantConv, RotoReflectionEquivariantConvLift, RotoReflectionEquivariantConv

equiadapt/nbody/canonicalization/continuous_group.py

@@ -1,12 +1,5 @@
 import torch
-import kornia as K
 from equiadapt.common.basecanonicalization import ContinuousGroupCanonicalization
-from equiadapt.common.utils import gram_schmidt
-from torch.nn.modules import Module
-from torchvision import transforms
-import math
-from torch.nn import functional as F
-from equiadapt.common.utils import gram_schmidt
 
 class ContinuousGroupNBody(ContinuousGroupCanonicalization):
     def __init__(self,

equiadapt/nbody/canonicalization_networks/euclideangraph_base_models.py

@@ -1,16 +1,11 @@
-import numpy as np
 import torch
 import torch.nn as nn
-import torch.nn.functional as F
-from torch.autograd import Variable
 import pytorch_lightning as pl
-from pytorch3d.transforms import RotateAxisAngle, Rotate, random_rotations
-import torchmetrics.functional as tmf
 import wandb
 import torch_scatter as ts
 import math
 from equiadapt.nbody.canonicalization_networks.gcl import E_GCL_vel, GCL
-from equiadapt.nbody.canonicalization_networks.vn_layers import VNLinearLeakyReLU, VNLinear, VNLeakyReLU, VNSoftplus
+from equiadapt.nbody.canonicalization_networks.vn_layers import VNLeakyReLU, VNSoftplus
 from equiadapt.nbody.canonicalization_networks.set_base_models import SequentialMultiple
 
 

equiadapt/nbody/canonicalization_networks/euclideangraph_model.py

@@ -1,14 +1,8 @@
 import torch
-import torch.nn as nn
-import torch.nn.functional as F
 import pytorch_lightning as pl
-from pytorch3d.transforms import RotateAxisAngle, Rotate, random_rotations
-import torchmetrics.functional as tmf
-import wandb
 
-from equiadapt.nbody.canonicalization_networks.vn_layers import *
 from equiadapt.nbody.canonicalization_networks.euclideangraph_base_models import EGNN_vel, GNN, VNDeepSets, BaseEuclideangraphModel, Transformer
-from canonical_network.utils import define_hyperparams, dict_to_object
+from canonical_network.utils import define_hyperparams
 
 # Input dim is 6 because location and velocity vectors are concatenated.
 NBODY_HYPERPARAMS = {

equiadapt/nbody/canonicalization_networks/gcl.py

@@ -1,7 +1,5 @@
 from torch import nn
 import torch
-import torch.nn.functional as F
-import torch_scatter as ts
 
 
 class MLP(nn.Module):

equiadapt/nbody/canonicalization_networks/image_model.py

@@ -1,7 +1,7 @@
 from torch import optim, nn
 import pytorch_lightning as pl
 import torch
-from torch.optim.lr_scheduler import OneCycleLR, MultiStepLR
+from torch.optim.lr_scheduler import MultiStepLR
 from canonical_network.models.image_networks import VanillaNetwork, EquivariantCanonizationNetwork, \
     BasicConvEncoder, Identity, PCACanonizationNetwork, RotationEquivariantConvEncoder, OptimizationCanonizationNetwork
 from canonical_network.models.resnet import resnet44

equiadapt/nbody/canonicalization_networks/image_networks.py

@@ -4,7 +4,6 @@
 import torch
 from canonical_network.models.equivariant_layers import RotationEquivariantConvLift, \
     RotoReflectionEquivariantConvLift, RotationEquivariantConv, RotoReflectionEquivariantConv
-from torchvision import transforms
 from canonical_network.models.set_base_models import SequentialMultiple
 import numpy as np
 

equiadapt/nbody/canonicalization_networks/resnet.py

@@ -27,12 +27,10 @@
 If you use this implementation in you work, please don't forget to mention the
 author, Yerlan Idelbayev.
 '''
-import torch
 import torch.nn as nn
 import torch.nn.functional as F
 import torch.nn.init as init
 
-from torch.autograd import Variable
 
 __all__ = ['ResNet', 'resnet20', 'resnet32', 'resnet44', 'resnet56', 'resnet110', 'resnet1202']
 

equiadapt/nbody/canonicalization_networks/set_base_models.py

@@ -1,4 +1,3 @@
-from sched import scheduler
 import torch
 import torch.nn as nn
 import torch.nn.functional as F

equiadapt/nbody/canonicalization_networks/set_model.py

@@ -1,13 +1,10 @@
-from collections import namedtuple
 
-from turtle import forward
 import numpy as np
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 import torch_scatter as ts
 import pytorch_lightning as pl
-import torchmetrics.functional as tmf
 from einops import rearrange
 import wandb
 

equiadapt/nbody/canonicalization_networks/vn_layers.py

@@ -1,11 +1,5 @@
-import os
-import sys
-import copy
-import math
-import numpy as np
 import torch
 import torch.nn as nn
-import torch.nn.functional as F
 
 EPS = 1e-6
 

equiadapt/pointcloud/canonicalization/continuous_group.py

@@ -2,7 +2,6 @@
 # This is meant to be a proof of concept and we are happy to receive contribution to extend this to other group actions.
 
 import torch
-import kornia as K
 from equiadapt.common.basecanonicalization import ContinuousGroupCanonicalization
 from equiadapt.common.utils import gram_schmidt
 from typing import Any, List, Tuple, Union

equiadapt/pointcloud/canonicalization_networks/equivariant_networks.py

@@ -1,6 +1,6 @@
 import torch
 import torch.nn as nn
-from equiadapt.pointcloud.canonicalization_networks.vector_neuron_layers import VNLinearLeakyReLU, VNLinear, VNMaxPool, VNBatchNorm
+from equiadapt.pointcloud.canonicalization_networks.vector_neuron_layers import VNLinearLeakyReLU, VNMaxPool, VNBatchNorm
 
 def knn(x, k):
     inner = -2 * torch.matmul(x.transpose(2, 1), x)

examples/images/classification/inference_utils.py

@@ -1,11 +1,9 @@
 import math
-from typing import Dict, Union
 
 import torch
 from omegaconf import DictConfig
 from torchvision import transforms
 
-import wandb
 
 
 def get_inference_method(canonicalizer: torch.nn.Module,

examples/images/classification/prepare/__init__.py

@@ -1,4 +0,0 @@
-from .cifar_data import CIFAR10DataModule, CIFAR100DataModule
-from .imagenet_data import ImageNetDataModule
-from .rotated_mnist_data import RotatedMNISTDataModule
-from .stl10_data import STL10DataModule

examples/images/classification/prepare/cifar_data.py

@@ -3,7 +3,7 @@
 import random
 
 import pytorch_lightning as pl
-from torch.utils.data import DataLoader, random_split
+from torch.utils.data import DataLoader
 from torchvision import transforms
 from torchvision.datasets import CIFAR10, CIFAR100
 

examples/images/classification/prepare/imagenet_data.py

@@ -1,13 +1,11 @@
 
-import os
 import random
-from typing import List
 
 import pytorch_lightning as pl
 import torch
 import torchvision
 import torchvision.transforms as transforms
-from PIL import Image, ImageOps
+from PIL import ImageOps
 from torch import nn
 
 DEFAULT_CROP_RATIO = 224/256

examples/images/classification/prepare/rotated_mnist_data.py

@@ -1,4 +1,3 @@
-import argparse
 import os
 import urllib.request as url_req
 import zipfile

examples/images/classification/prepare/stl10_data.py

@@ -3,7 +3,7 @@
 import random
 
 import pytorch_lightning as pl
-from torch.utils.data import DataLoader, random_split
+from torch.utils.data import DataLoader
 from torchvision import transforms
 from torchvision.datasets import STL10
 

examples/images/classification/train_utils.py

@@ -1,4 +1,4 @@
-from typing import Dict, Optional
+from typing import Optional
 
 import dotenv
 import pytorch_lightning as pl

examples/images/common/__init__.py

@@ -1 +0,0 @@
-from .utils import get_canonicalization_network, get_canonicalizer

examples/images/segmentation/prepare/__init__.py

@@ -1 +0,0 @@
-from .coco_data import COCODataModule

examples/images/segmentation/prepare/vision_transforms.py

@@ -2,7 +2,6 @@
 
 import torch
 from torch import Tensor, nn
-from torchvision import ops
 from torchvision.transforms import functional as F
 from torchvision.transforms import transforms as T
 

examples/images/segmentation/train_utils.py

@@ -1,4 +1,4 @@
-from typing import Dict, Optional
+from typing import Optional
 
 import dotenv
 import pytorch_lightning as pl

examples/nbody/model.py

@@ -1,11 +1,9 @@
 import torch
 import pytorch_lightning as pl
-from torch.optim.lr_scheduler import MultiStepLR
 from examples.nbody.model_utils import get_canonicalization_network, get_prediction_network, get_edges
 import torch.nn as nn
-from omegaconf import OmegaConf, DictConfig
+from omegaconf import DictConfig
 from equiadapt.nbody.canonicalization.continuous_group import ContinuousGroupNBody
-import wandb
 
 class NBodyPipeline(pl.LightningModule):
     def __init__(self, hyperparams: DictConfig):

examples/nbody/model_utils.py

@@ -1,6 +1,4 @@
 from equiadapt.nbody.canonicalization_networks.euclideangraph_base_models import EGNN_vel, VNDeepSets, GNN, Transformer
-from collections import namedtuple
-from omegaconf import OmegaConf
 import torch
 
 

examples/nbody/prepare/nbody_data.py

@@ -1,11 +1,8 @@
 import numpy as np
 import torch
-from torch.utils.data import DataLoader, Dataset
+from torch.utils.data import DataLoader
 import pytorch_lightning as pl
-import os
-import json
 import numpy as np
-from torch.utils.data import Dataset
 import pathlib
 
 SRC_PATH = pathlib.Path(__file__).parent.parent

examples/nbody/train.py

@@ -4,9 +4,8 @@
 import pytorch_lightning as pl
 import wandb
 import os
-from omegaconf import DictConfig, OmegaConf
+from omegaconf import OmegaConf
 from examples.nbody.model import NBodyPipeline
-import torch
 
 from examples.nbody.prepare.nbody_data import NBodyDataModule
 

examples/pointcloud/classification/train.py

@@ -7,7 +7,6 @@
 from omegaconf import DictConfig, OmegaConf
 import pytorch_lightning as pl
 from pytorch_lightning.loggers import WandbLogger
-from argparse import ArgumentParser
 
 from prepare import ModelNetDataModule
 from train_utils import get_model_pipeline, get_callbacks, get_checkpoint_name, get_trainer, load_envs

examples/pointcloud/common/networks.py

@@ -1,10 +1,7 @@
 from omegaconf import DictConfig
 import torch
-import numpy as np
 import torch.nn as nn
 import torch.nn.functional as F
-import pytorch_lightning as pl
-from torch.autograd import Variable
 import torch.nn.init as init
 
 def knn(x, k):

examples/pointcloud/part_segmentation/train.py

@@ -7,7 +7,6 @@
 from omegaconf import DictConfig, OmegaConf
 import pytorch_lightning as pl
 from pytorch_lightning.loggers import WandbLogger
-from argparse import ArgumentParser
 
 from prepare import ShapeNetDataModule
 from train_utils import get_model_pipeline, get_callbacks, get_checkpoint_name, get_trainer, load_envs