Merge branch 'master' into ohinds-warmstart

.pre-commit-config.yaml

@@ -12,7 +12,7 @@ repos:
     -   id: end-of-file-fixer
     -   id: trailing-whitespace
 -   repo: https://github.com/psf/black
-    rev: 23.7.0
+    rev: 23.9.1
     hooks:
     -   id: black
 -   repo: https://github.com/PyCQA/flake8

README.md

@@ -66,7 +66,7 @@ the Apache 2.0 license. It was started under the support of NIH R01 EB020470.
 
 ### Augmentation methods
 #### Spatial Transforms
-[Center crop](), [Spacial Constant Padding](), [Random Crop](), [Resize](), [Random flip (left and right)]()
+[Center crop](), [Spatial Constant Padding](), [Random Crop](), [Resize](), [Random flip (left and right)]()
 
 #### Intensity Transforms
 [Add gaussian noise](), [Min-Max intensity scaling](), [Costom intensity scaling](), [Intensity masking](), [Contrast adjustment]()

nobrainer/bayesian_utils.py

@@ -64,7 +64,7 @@ def default_loc_scale_fn(
           safe to use when doing asynchronous distributed training. The default
           (`None`) is to use the `tf.get_variable` default.
         weightnorm: An optional (boolean) to activate weightnorm for the mean
-        kernal.
+        kernel.
 
     Returns
     ----------

nobrainer/models/bayesian_vnet.py

@@ -28,9 +28,9 @@ def down_stage(
     kld: a func to compute KL Divergence loss, default is set None.
     KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use.
@@ -83,11 +83,11 @@ def up_stage(
     kld: a func to compute KL Divergence loss, default is set None.
         KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
     filters: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
@@ -153,17 +153,17 @@ def end_stage(
     kld: a func to compute KL Divergence loss, default is set None.
         KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
     n_classes: int, for binary class use the value 1.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
 
     Result
     ----------
-    prediction probablities
+    prediction probabilities
     """
     conv = tfp.layers.Convolution3DFlipout(
         n_classes,
@@ -216,12 +216,12 @@ def bayesian_vnet(
         a value of 1.
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size(int): size of the kernal of conv layers
+    kernal_size(int): size of the kernel of conv layers
     activation(str): all tf.keras.activations are allowed
     kld: a func to compute KL Divergence loss, default is set None.
         KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
     See Bayesian Utils for more options for kld, prior_fn and kernal_posterior_fn
     activation: str or optimizer object, the non-linearity to use. All

nobrainer/models/bayesian_vnet_semi.py

@@ -24,7 +24,7 @@ def down_stage(inputs, filters, kernel_size=3, activation="relu", padding="SAME"
     inputs: tf.layer for encoding stage.
     filters: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
@@ -61,11 +61,11 @@ def up_stage(
     kld: a func to compute KL Divergence loss, default is set None.
         KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
     filters: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
@@ -131,17 +131,17 @@ def end_stage(
     kld: a func to compute KL Divergence loss, default is set None.
         KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
     n_classes: int, for binary class use the value 1.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
 
     Result
     ----------
-    prediction probablities.
+    prediction probabilities.
     """
     conv = tfp.layers.Convolution3DFlipout(
         n_classes,
@@ -195,12 +195,12 @@ def bayesian_vnet_semi(
         a value of 1.
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size(int): size of the kernal of conv layers
+    kernal_size(int): size of the kernel of conv layers
     activation(str): all tf.keras.activations are allowed
     kld: a func to compute KL Divergence loss, default is set None.
         KLD can be set as (lambda q, p, ignore: kl_lib.kl_divergence(q, p))
     prior_fn: a func to initialize priors distributions
-    kernel_posterior_fn:a func to initlaize kernal posteriors
+    kernel_posterior_fn:a func to initlaize kernel posteriors
         (loc, scale and weightnorms)
     See Bayesian Utils for more options for kld, prior_fn and kernal_posterior_fn
     activation: str or optimizer object, the non-linearity to use. All

nobrainer/models/highresnet.py

@@ -16,7 +16,7 @@ def highresnet(
     https://arxiv.org/abs/1707.01992
     Args:
     n_classes(int): number of classes
-    input_shape(tuple):four ints representating the shape of 3D input
+    input_shape(tuple):four ints representing the shape of 3D input
     activation(str): all tf.keras.activations are allowed
     dropout_rate(int): [0,1].
     """

nobrainer/models/unet.py

@@ -19,7 +19,7 @@ def unet(
     https://arxiv.org/abs/1606.06650
     Args:
     n_classes(int): number of classes
-    input_shape(tuple):four ints representating the shape of 3D input
+    input_shape(tuple):four ints representing the shape of 3D input
     activation(str): all tf.keras.activations are allowed
     batch_size(int): batch size.
     """

nobrainer/models/vnet.py

@@ -23,7 +23,7 @@ def down_stage(inputs, filters, kernel_size=3, activation="relu", padding="SAME"
     inputs: tf.layer for encoding stage.
     filters: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
@@ -48,7 +48,7 @@ def up_stage(inputs, skip, filters, kernel_size=3, activation="relu", padding="S
     inputs: tf.layer for encoding stage.
     filters: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
@@ -80,14 +80,14 @@ def end_stage(inputs, n_classes=1, kernel_size=3, activation="relu", padding="SA
     ----------
     inputs: tf.model layer.
     n_classes: int, for binary class use the value 1.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use
 
     Result
     ----------
-    prediction probablities
+    prediction probabilities
     """
     conv = Conv3D(
         filters=n_classes,
@@ -123,7 +123,7 @@ def vnet(
         a value of 1.
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
     activation: str or optimizer object, the non-linearity to use. All
         tf.activations are allowed to use

nobrainer/models/vox2vox.py

@@ -26,11 +26,11 @@ def vox_gan(
         a value of 1.
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    g_kernal_size: int, size of the kernal for generator. Default kernal size
+    g_kernal_size: int, size of the kernel for generator. Default kernel size
         is set to be 4.
     g_filters: int, number of filters for generator. default is set 64.
     g_norm: str, to set batch or instance norm.
-    d_kernal_size: int, size of the kernal for discriminator. Default kernal size
+    d_kernal_size: int, size of the kernel for discriminator. Default kernel size
         is set to be 4.
     d_filters: int, number of filters for discriminator. default is set 64.
     d_norm: str, to set batch or instance norm.
@@ -65,7 +65,7 @@ def Vox_generator(n_classes, input_shape, n_filters=64, kernel_size=4, norm="bat
         a value of 1.
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 4.
     n_filters: int, number of filters. default is set 64.
     norm: str, to set batch or instance norm.
@@ -195,7 +195,7 @@ def Vox_discriminator(input_shape, n_filters=64, kernel_size=4, norm="batch"):
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
     n_filters: int, number of filters. default is set 64.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 4.
     norm: str, to set batch or instance norm.
 
@@ -275,7 +275,7 @@ def Vox_ensembler(n_classes, input_shape, kernel_size=3, **kwargs):
         a value of 1.
     input_shape: list or tuple of four ints, the shape of the input data. Omit
         the batch dimension, and include the number of channels.
-    kernal_size: int, size of the kernal of conv layers. Default kernal size
+    kernal_size: int, size of the kernel of conv layers. Default kernel size
         is set to be 3.
 
     Returns

pyproject.toml

@@ -12,3 +12,6 @@ force_sort_within_sections = true
 reverse_relative = true
 sort_relative_in_force_sorted_sections = true
 known_first_party = ["nobrainer"]
+
+[tool.codespell]
+ignore-words-list = "nd"