Merge pull request #76 from v0lta/format-everything

Apply formatting everywhere

Author: v0lta

docs/conf.py

@@ -12,14 +12,15 @@
 #
 import os
 import sys
-sys.path.insert(0, os.path.abspath('../src'))
+
+sys.path.insert(0, os.path.abspath("../src"))
 
 
 # -- Project information -----------------------------------------------------
 
-project = 'PyTorch-Wavelet-Toolbox'
-copyright = '2022, Moritz Wolter'
-author = 'Moritz Wolter'
+project = "PyTorch-Wavelet-Toolbox"
+copyright = "2022, Moritz Wolter"
+author = "Moritz Wolter"
 
 
 # -- General configuration ---------------------------------------------------
@@ -28,27 +29,27 @@
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
 extensions = [
-    'sphinx.ext.napoleon',
-    'sphinx.ext.autodoc',
-    'sphinx.ext.intersphinx',
-    'sphinx.ext.todo',
-    'sphinx.ext.coverage',
-    'sphinx.ext.viewcode',
+    "sphinx.ext.napoleon",
+    "sphinx.ext.autodoc",
+    "sphinx.ext.intersphinx",
+    "sphinx.ext.todo",
+    "sphinx.ext.coverage",
+    "sphinx.ext.viewcode",
 ]
 
 napoleon_google_docstring = True
 # napoleon_use_admonition_for_examples = True
 
 # Add any paths that contain templates here, relative to this directory.
-templates_path = ['_templates']
+templates_path = ["_templates"]
 
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
 # This pattern also affects html_static_path and html_extra_path.
-exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
+exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
 
 # document __init__ in the docpages
-autoclass_content = 'both'
+autoclass_content = "both"
 
 
 # -- Options for HTML output -------------------------------------------------
@@ -57,26 +58,26 @@
 # a list of builtin themes.
 #
 
-html_theme = 'alabaster'
+html_theme = "alabaster"
 html_sidebars = {
-    '**': [
-        'about.html',
-        'navigation.html',
-        'searchbox.html',
+    "**": [
+        "about.html",
+        "navigation.html",
+        "searchbox.html",
     ]
 }
 
 html_theme_options = {
-    'github_user': 'v0lta',
-    'github_repo': 'PyTorch-Wavelet-Toolbox',
-    'github_banner': 'false',
-    'show_related': 'true',
-    'page_width': 'auto',
-    'sidebar_width': '250px'
+    "github_user": "v0lta",
+    "github_repo": "PyTorch-Wavelet-Toolbox",
+    "github_banner": "false",
+    "show_related": "true",
+    "page_width": "auto",
+    "sidebar_width": "250px",
 }
 
 
-html_favicon = 'favicon/favicon.ico'
+html_favicon = "favicon/favicon.ico"
 
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,

docs/favicon/plot_shannon.py

@@ -1,19 +1,16 @@
-import numpy as np
 import matplotlib.pyplot as plt
+import numpy as np
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     bandwidth = 1
     center = 0
     grid_values = np.linspace(-6, 6, 500)
     shannon = (
         np.sqrt(bandwidth)
-        * (
-            np.sin(np.pi * bandwidth * grid_values)
-            / (np.pi * bandwidth * grid_values)
-        )
+        * (np.sin(np.pi * bandwidth * grid_values) / (np.pi * bandwidth * grid_values))
         * np.exp(1j * 2 * np.pi * center * grid_values)
     )
     plt.plot(shannon, linewidth=20.0)
     plt.axis("off")
     plt.savefig("shannon.png")
-    plt.show()
+    plt.show()

examples/continuous_signal_analysis/cwt_chirp_analysis.py

@@ -1,8 +1,9 @@
-import torch
-import numpy as np
-import ptwt
 import matplotlib.pyplot as plt
+import numpy as np
 import scipy.signal as signal
+import torch
+
+import ptwt
 
 if __name__ == "__main__":
     t = np.linspace(-2, 2, 800, endpoint=False)

examples/deepfake_analysis/packet_plot.py

@@ -1,14 +1,14 @@
 import os
 from itertools import product
-from tqdm import tqdm
 
-from PIL import Image
+import matplotlib.pyplot as plt
 import numpy as np
-import torch
-import ptwt
 import pywt
+import torch
+from PIL import Image
+from tqdm import tqdm
 
-import matplotlib.pyplot as plt
+import ptwt
 
 
 def get_freq_order(level: int):
@@ -78,13 +78,13 @@ def generate_frequency_packet_image(packet_array: np.ndarray, degree: int):
 
 
 def load_image(path_to_file: str) -> torch.Tensor:
-    
     image = Image.open(path_to_file)
     tensor = torch.from_numpy(np.nan_to_num(np.array(image), posinf=255, neginf=0))
     return tensor
 
+
 def process_images(tensor: torch.Tensor, paths: list) -> torch.Tensor:
-    tensor = torch.mean(tensor/255., -1)
+    tensor = torch.mean(tensor / 255.0, -1)
     packets = ptwt.WaveletPacket2D(tensor, pywt.Wavelet("Haar"))
 
     packet_list = []
@@ -103,10 +103,10 @@ def load_images(path: str) -> list:
             path = os.path.join(root, name)
             packets = load_image(path)
             image_list.append(packets)
-    return image_list 
+    return image_list
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     frequency_path, natural_path = get_freq_order(level=3)
     print("Loading ffhq images:")
     ffhq_images = load_images("./ffhq_style_gan/source_data/A_ffhq")
@@ -120,7 +120,6 @@ def load_images(path: str) -> list:
     del ffhq_images
     del ffhq_packets
 
-
     print("Loading style-gan images")
     gan_images = load_images("./ffhq_style_gan/source_data/B_stylegan")
     print("processing style-gan")
@@ -136,7 +135,7 @@ def load_images(path: str) -> list:
     plot_ffhq = generate_frequency_packet_image(mean_packets_ffhq, 3)
     plot_gan = generate_frequency_packet_image(mean_packets_gan, 3)
 
-    fig = plt.figure(figsize=(9,3))
+    fig = plt.figure(figsize=(9, 3))
     fig.add_subplot(1, 2, 1)
     plt.imshow(plot_ffhq, vmax=1.5, vmin=-7)
     plt.title("real")
@@ -151,9 +150,9 @@ def load_images(path: str) -> list:
     plt.colorbar()
     plt.show()
 
-    plt.plot(torch.mean(mean_packets_ffhq, (1, 2)).flatten().numpy(), label='real')
-    plt.plot(torch.mean(mean_packets_gan, (1, 2)).flatten().numpy(), label='fake')
-    plt.xlabel('mean packets')
-    plt.ylabel('magnitude')
+    plt.plot(torch.mean(mean_packets_ffhq, (1, 2)).flatten().numpy(), label="real")
+    plt.plot(torch.mean(mean_packets_gan, (1, 2)).flatten().numpy(), label="fake")
+    plt.xlabel("mean packets")
+    plt.ylabel("magnitude")
     plt.legend()
     plt.show()

examples/network_compression/mnist_compression.py

@@ -3,17 +3,19 @@
 # based on https://github.com/pytorch/examples/blob/master/mnist/main.py
 
 import argparse
+import collections
+
+import matplotlib.pyplot as plt
 import numpy as np
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 import torch.optim as optim
-import collections
-import matplotlib.pyplot as plt
-from torchvision import datasets, transforms
 from torch.optim.lr_scheduler import StepLR
-from wavelet_linear import WaveletLayer
 from torch.utils.tensorboard.writer import SummaryWriter
+from torchvision import datasets, transforms
+from wavelet_linear import WaveletLayer
+
 from ptwt.wavelets_learnable import ProductFilter
 
 

examples/network_compression/wavelet_linear.py

@@ -1,9 +1,10 @@
 # Originally created by moritz ([email protected])
 # at https://github.com/v0lta/Wavelet-network-compression/blob/master/wavelet_learning/wavelet_linear.py
-import torch
 import numpy as np
-from torch.nn.parameter import Parameter
 import pywt
+import torch
+from torch.nn.parameter import Parameter
+
 from ptwt.conv_transform import wavedec, waverec
 
 

examples/speed_tests/timeitconv_1d.py

@@ -1,11 +1,13 @@
-import pywt
-import ptwt
-import torch
-import numpy as np
 import time
 from typing import NamedTuple
 
 import matplotlib.pyplot as plt
+import numpy as np
+import pywt
+import torch
+
+import ptwt
+
 
 class WaveletTuple(NamedTuple):
     """Replaces namedtuple("Wavelet", ("dec_lo", "dec_hi", "rec_lo", "rec_hi"))."""
@@ -24,15 +26,15 @@ def _set_up_wavelet_tuple(wavelet, dtype):
         torch.tensor(wavelet.rec_hi).type(dtype),
     )
 
+
 def _jit_wavedec_fun(data, wavelet):
     return ptwt.wavedec(data, wavelet, "periodic", level=10)
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     length = 1e6
     repetitions = 100
 
-
     pywt_time_cpu = []
     ptwt_time_cpu = []
     ptwt_time_gpu = []
@@ -56,10 +58,10 @@ def _jit_wavedec_fun(data, wavelet):
 
     wavelet = _set_up_wavelet_tuple(pywt.Wavelet("db5"), torch.float32)
     jit_wavedec = torch.jit.trace(
-            _jit_wavedec_fun,
-            (data, wavelet),
-            strict=False,
-        )
+        _jit_wavedec_fun,
+        (data, wavelet),
+        strict=False,
+    )
 
     for _ in range(repetitions):
         data = np.random.randn(32, int(length)).astype(np.float32)
@@ -69,7 +71,6 @@ def _jit_wavedec_fun(data, wavelet):
         end = time.perf_counter()
         ptwt_time_cpu_jit.append(end - start)
 
-
     for _ in range(repetitions):
         data = np.random.randn(32, int(length)).astype(np.float32)
         data = torch.from_numpy(data).cuda()
@@ -82,10 +83,10 @@ def _jit_wavedec_fun(data, wavelet):
 
     wavelet = _set_up_wavelet_tuple(pywt.Wavelet("db5"), torch.float32)
     jit_wavedec = torch.jit.trace(
-            _jit_wavedec_fun,
-            (data.cuda(), wavelet),
-            strict=False,
-        )
+        _jit_wavedec_fun,
+        (data.cuda(), wavelet),
+        strict=False,
+    )
 
     for _ in range(repetitions):
         data = np.random.randn(32, int(length)).astype(np.float32)
@@ -95,14 +96,24 @@ def _jit_wavedec_fun(data, wavelet):
         res = jit_wavedec(data, wavelet)
         torch.cuda.synchronize()
         end = time.perf_counter()
-        ptwt_time_gpu_jit.append(end-start)
+        ptwt_time_gpu_jit.append(end - start)
 
     print("1d fwt results")
-    print(f"1d-pywt-cpu    :{np.mean(pywt_time_cpu):5.5f} +- {np.std(pywt_time_cpu):5.5f}")
-    print(f"1d-ptwt-cpu    :{np.mean(ptwt_time_cpu):5.5f} +- {np.std(ptwt_time_cpu):5.5f}")    
-    print(f"1d-ptwt-cpu-jit:{np.mean(ptwt_time_cpu_jit):5.5f} +- {np.std(ptwt_time_cpu_jit):5.5f}")
-    print(f"1d-ptwt-gpu    :{np.mean(ptwt_time_gpu):5.5f} +- {np.std(ptwt_time_gpu):5.5f}")
-    print(f"1d-ptwt-gpu-jit:{np.mean(ptwt_time_gpu_jit):5.5f} +- {np.std(ptwt_time_gpu_jit):5.5f}")
+    print(
+        f"1d-pywt-cpu    :{np.mean(pywt_time_cpu):5.5f} +- {np.std(pywt_time_cpu):5.5f}"
+    )
+    print(
+        f"1d-ptwt-cpu    :{np.mean(ptwt_time_cpu):5.5f} +- {np.std(ptwt_time_cpu):5.5f}"
+    )
+    print(
+        f"1d-ptwt-cpu-jit:{np.mean(ptwt_time_cpu_jit):5.5f} +- {np.std(ptwt_time_cpu_jit):5.5f}"
+    )
+    print(
+        f"1d-ptwt-gpu    :{np.mean(ptwt_time_gpu):5.5f} +- {np.std(ptwt_time_gpu):5.5f}"
+    )
+    print(
+        f"1d-ptwt-gpu-jit:{np.mean(ptwt_time_gpu_jit):5.5f} +- {np.std(ptwt_time_gpu_jit):5.5f}"
+    )
     # plt.semilogy(pywt_time_cpu, label='pywt-cpu')
     # plt.semilogy(ptwt_time_cpu, label='ptwt-cpu')
     # plt.semilogy(ptwt_time_cpu_jit, label='ptwt-cpu-jit')
@@ -112,12 +123,24 @@ def _jit_wavedec_fun(data, wavelet):
     # plt.xlabel('repetition')
     # plt.ylabel('runtime [s]')
     # plt.show()
-    time_stack = np.stack([pywt_time_cpu, ptwt_time_cpu, ptwt_time_cpu_jit, ptwt_time_gpu, ptwt_time_gpu_jit], -1)
+    time_stack = np.stack(
+        [
+            pywt_time_cpu,
+            ptwt_time_cpu,
+            ptwt_time_cpu_jit,
+            ptwt_time_gpu,
+            ptwt_time_gpu_jit,
+        ],
+        -1,
+    )
     plt.boxplot(time_stack)
-    plt.yscale('log')
-    plt.xticks([1,2,3,4,5], ["pywt-cpu", "ptwt-cpu", "ptwt-cpu-jit", "ptwt-gpu", "ptwt-gpu-jit"])
+    plt.yscale("log")
+    plt.xticks(
+        [1, 2, 3, 4, 5],
+        ["pywt-cpu", "ptwt-cpu", "ptwt-cpu-jit", "ptwt-gpu", "ptwt-gpu-jit"],
+    )
     plt.xticks(rotation=20)
-    plt.ylabel('runtime [s]')
-    plt.title('DWT-1D')
-    plt.savefig('./figs/timeitconv1d.png')
+    plt.ylabel("runtime [s]")
+    plt.title("DWT-1D")
+    plt.savefig("./figs/timeitconv1d.png")
     # plt.show()