From c6d44acf0155ae92cfc912253c86a2e81de9d49c Mon Sep 17 00:00:00 2001
From: "Daniel.Frisinghelli" <daniel.frisinghelli@eurac.edu>
Date: Mon, 31 Aug 2020 10:59:07 +0200
Subject: [PATCH] Added cross-references for documentation.
---
pysegcnn/core/graphics.py | 86 ++++++------
pysegcnn/core/logging.py | 4 +-
pysegcnn/core/predict.py | 57 ++++----
pysegcnn/core/trainer.py | 142 +++++++++++++-------
pysegcnn/core/transforms.py | 214 ++++++++++++++++++-----------
pysegcnn/core/utils.py | 222 +++++++++++++++----------------
pysegcnn/main/eval.py | 14 +-
pysegcnn/main/train.py | 19 ++-
pysegcnn/preprocessing/sparcs.py | 2 +
9 files changed, 432 insertions(+), 328 deletions(-)
diff --git a/pysegcnn/core/graphics.py b/pysegcnn/core/graphics.py
index a4f693b..7c9755e 100644
--- a/pysegcnn/core/graphics.py
+++ b/pysegcnn/core/graphics.py
@@ -37,15 +37,15 @@ def contrast_stretching(image, alpha=5):
Parameters
----------
- image : `numpy.ndarray`
- the input image.
+ image : :py:class:`numpy.ndarray`
+ The input image.
alpha : `int`, optional
- The level of the percentiles. The default is 5.
+ The level of the percentiles. The default is `5`.
Returns
-------
- norm : `numpy.ndarray`
- the stretched image.
+ norm : :py:class:`numpy.ndarray`
+ The contrast-stretched image.
"""
# compute upper and lower percentiles defining the range of the stretch
@@ -68,14 +68,14 @@ def running_mean(x, w):
Parameters
----------
- x : array_like
+ x : `array_like`
The sequence to compute a running mean on.
w : `int`
The window length of the running mean.
Returns
-------
- rm : `numpy.ndarray`
+ rm : :py:class:`numpy.ndarray`
The running mean of the sequence ``x``.
"""
@@ -90,8 +90,8 @@ def plot_sample(x, use_bands, labels, y=None, y_pred=None, figsize=(10, 10),
Parameters
----------
- x : `numpy.ndarray` or `torch.Tensor`, (b, h, w)
- Array containing the raw data of the tile, shape=(bands, height, width)
+ x : :py:class:`numpy.ndarray` or :py:class:`torch.Tensor`, (b, h, w)
+ Array containing the data of the tile, shape=(bands, height, width).
use_bands : `list` of `str`
List describing the order of the bands in ``x``.
labels : `dict` [`int`, `dict`]
@@ -101,31 +101,31 @@ def plot_sample(x, use_bands, labels, y=None, y_pred=None, figsize=(10, 10),
A named color (`str`).
``'label'``
The name of the class label (`str`).
- y : `numpy.ndarray` or `torch.Tensor` or `None`, optional
+ y : :py:class:`numpy.ndarray` or :py:class:`torch.Tensor`, optional
Array containing the ground truth of tile ``x``, shape=(height, width).
- The default is None.
- y_pred : `numpy.ndarray` or `torch.Tensor` or `None`, optional
+ The default is `None`, i.e. the ground truth is not plotted.
+ y_pred : :py:class:`numpy.ndarray` or :py:class:`torch.Tensor`, optional
Array containing the prediction for tile ``x``, shape=(height, width).
- The default is None.
+ The default is `None`, i.e. the prediction is not plotted.
figsize : `tuple`, optional
- The figure size in centimeters. The default is (10, 10).
+ The figure size in centimeters. The default is `(10, 10)`.
bands : `list` [`str`], optional
- The bands to build the FCC. The default is ['nir', 'red', 'green'].
+ The bands to build the FCC. The default is `['nir', 'red', 'green']`.
state : `str` or `None`, optional
Filename to save the plot to. ``state`` should be an existing model
- state file ending with '.pt'. The default is None, i.e. plot is not
- saved to disk.
- outpath : `str` or `pathlib.Path`, optional
- Output path. The default is 'pysegcnn/main/_samples'.
+ state file ending with `'.pt'`. The default is `None`, i.e. the plot is
+ not saved to disk.
+ outpath : `str` or :py:class:`pathlib.Path`, optional
+ Output path. The default is `'pysegcnn/main/_samples'`.
alpha : `int`, optional
The level of the percentiles to increase constrast in the FCC.
- The default is 0, i.e. no stretching.
+ The default is `0`, i.e. no stretching is applied.
Returns
-------
- fig : `matplotlib.figure.Figure`
+ fig : :py:class:`matplotlib.figure.Figure`
The figure handle.
- ax : `numpy.ndarray` [`matplotlib.axes._subplots.AxesSubplot`]
+ ax : :py:class:`numpy.ndarray`
An array of the axes handles.
"""
@@ -199,7 +199,7 @@ def plot_confusion_matrix(cm, labels, normalize=True,
Parameters
----------
- cm : `numpy.ndarray`
+ cm : :py:class:`numpy.ndarray`
The confusion matrix.
labels : `dict` [`int`, `dict`]
The label dictionary. The keys are the values of the class labels
@@ -209,23 +209,23 @@ def plot_confusion_matrix(cm, labels, normalize=True,
``'label'``
The name of the class label (`str`).
normalize : `bool`, optional
- Whether to normalize the confusion matrix. The default is True.
+ Whether to normalize the confusion matrix. The default is `True`.
figsize : `tuple`, optional
- The figure size in centimeters. The default is (10, 10).
+ The figure size in centimeters. The default is `(10, 10)`.
cmap : `str`, optional
- A colormap in `matplotlib.pyplot.colormaps()`. The default is 'Blues'.
- state_file : `str` or `None` or `pathlib.Path`, optional
+ A matplotlib colormap. The default is `'Blues'`.
+ state_file : `str` or `None` or :py:class:`pathlib.Path`, optional
Filename to save the plot to. ``state`` should be an existing model
- state file ending with '.pt'. The default is None, i.e. plot is not
- saved to disk.
- outpath : `str` or `pathlib.Path`, optional
- Output path. The default is 'pysegcnn/main/_graphics/'.
+ state file ending with `'.pt'`. The default is `None`, i.e. the plot is
+ not saved to disk.
+ outpath : `str` or :py:class:`pathlib.Path`, optional
+ Output path. The default is `'pysegcnn/main/_graphics/'`.
Returns
-------
- fig : `matplotlib.figure.Figure`
+ fig : :py:class:`matplotlib.figure.Figure`
The figure handle.
- ax : `matplotlib.axes._subplots.AxesSubplot`
+ ax : :py:class:`matplotlib.axes._subplots.AxesSubplot`
The axes handle.
"""
@@ -302,23 +302,23 @@ def plot_loss(state_file, figsize=(10, 10), step=5,
Parameters
----------
- state_file : `str` or `pathlib.Path`
+ state_file : `str` or :py:class:`pathlib.Path`
The model state file. Model state files are stored in
- pysegcnn/main/_models.
+ `pysegcnn/main/_models`.
figsize : `tuple`, optional
- The figure size in centimeters. The default is (10, 10).
+ The figure size in centimeters. The default is `(10, 10)`.
step : `int`, optional
- The step of epochs for the x-axis labels. The default is 5, i.e. label
- each fifth epoch.
+ The step to label epochs on the x-axis labels. The default is `5`, i.e.
+ label each fifth epoch.
colors : `list` [`str`], optional
- A list of four named colors supported by `matplotlib`.
- The default is ['lightgreen', 'green', 'skyblue', 'steelblue'].
- outpath : `str` or `pathlib.Path`, optional
- Output path. The default is 'pysegcnn/main/_graphics/'.
+ A list of four named colors supported by matplotlib.
+ The default is `['lightgreen', 'green', 'skyblue', 'steelblue']`.
+ outpath : `str` or :py:class:`pathlib.Path`, optional
+ Output path. The default is `'pysegcnn/main/_graphics/'`.
Returns
-------
- fig : `matplotlib.figure.Figure`
+ fig : :py:class:`matplotlib.figure.Figure`
The figure handle.
"""
diff --git a/pysegcnn/core/logging.py b/pysegcnn/core/logging.py
index 916a4f0..c7c6464 100644
--- a/pysegcnn/core/logging.py
+++ b/pysegcnn/core/logging.py
@@ -20,7 +20,7 @@ import pathlib
# the logging configuration dictionary
def log_conf(logfile):
- """Set basic logging configuration passed to `logging.config.dictConfig`.
+ """Set basic logging configuration.
See the logging `docs`_ for a detailed description of the configuration
dictionary.
@@ -30,7 +30,7 @@ def log_conf(logfile):
Parameters
----------
- logfile : `str` or `pathlib.Path`
+ logfile : `str` or :py:class:`pathlib.Path`
The file to save the logs to.
Returns
diff --git a/pysegcnn/core/predict.py b/pysegcnn/core/predict.py
index ff7455e..2489aeb 100644
--- a/pysegcnn/core/predict.py
+++ b/pysegcnn/core/predict.py
@@ -34,14 +34,13 @@ LOGGER = logging.getLogger(__name__)
def _get_scene_tiles(ds, scene_id):
- """Return the tiles of the scene with id = ``scene_id``.
+ """Return the tiles of the scene with id ``scene_id``.
Parameters
----------
- ds : `pysegcnn.core.split.RandomSubset` or
- `pysegcnn.core.split.SceneSubset`
- An instance of `~pysegcnn.core.split.RandomSubset` or
- `~pysegcnn.core.split.SceneSubset`.
+ ds : :py:class:`pysegcnn.core.split.CustomSubset`
+ A instance of a subclass of
+ :py:class:`pysegcnn.core.split.CustomSubset`.
scene_id : `str`
A valid scene identifier.
@@ -77,27 +76,27 @@ def predict_samples(ds, model, cm=False, plot=False, **kwargs):
Parameters
----------
- ds : `pysegcnn.core.split.RandomSubset` or
- `pysegcnn.core.split.SceneSubset`
- An instance of `~pysegcnn.core.split.RandomSubset` or
- `~pysegcnn.core.split.SceneSubset`.
- model : `pysegcnn.core.models.Network`
- An instance of `~pysegcnn.core.models.Network`.
+ ds : :py:class:`pysegcnn.core.split.RandomSubset` or
+ :py:class:`pysegcnn.core.split.SceneSubset`
+ An instance of :py:class:`pysegcnn.core.split.RandomSubset` or
+ :py:class:`pysegcnn.core.split.SceneSubset`.
+ model : :py:class:`pysegcnn.core.models.Network`
+ An instance of :py:class:`pysegcnn.core.models.Network`.
cm : `bool`, optional
- Whether to compute the confusion matrix. The default is False.
+ Whether to compute the confusion matrix. The default is `False`.
plot : `bool`, optional
Whether to plot a false color composite, ground truth and model
- prediction for each sample. The default is False.
+ prediction for each sample. The default is `False`.
**kwargs
Additional keyword arguments passed to
- `pysegcnn.core.graphics.plot_sample`.
+ :py:func:`pysegcnn.core.graphics.plot_sample`.
Raises
------
TypeError
Raised if ``ds`` is not an instance of
- `~pysegcnn.core.split.RandomSubset` or
- `~pysegcnn.core.split.SceneSubset`.
+ :py:class:`pysegcnn.core.split.RandomSubset` or
+ :py:class:`pysegcnn.core.split.SceneSubset`.
Returns
-------
@@ -109,9 +108,10 @@ def predict_samples(ds, model, cm=False, plot=False, **kwargs):
The ground truth
``'prediction'``
Model prediction
- conf_mat : `numpy.ndarray`
+ conf_mat : :py:class:`numpy.ndarray`
The confusion matrix. Note that the confusion matrix ``conf_mat`` is
- only computed if ``cm`` = True.
+ only computed if ``cm=True``.
+
"""
# check whether the dataset is a valid subset, i.e.
# an instance of pysegcnn.core.split.SceneSubset or
@@ -183,26 +183,26 @@ def predict_scenes(ds, model, scene_id=None, cm=False, plot=False, **kwargs):
Parameters
----------
- ds : `pysegcnn.core.split.SceneSubset`
- An instance of `~pysegcnn.core.split.SceneSubset`.
- model : `pysegcnn.core.models.Network`
- An instance of `~pysegcnn.core.models.Network`.
+ ds : :py:class:`pysegcnn.core.split.SceneSubset`
+ An instance of :py:class:`pysegcnn.core.split.SceneSubset`.
+ model : :py:class:`pysegcnn.core.models.Network`
+ An instance of :py:class:`pysegcnn.core.models.Network`.
scene_id : `str` or `None`
A valid scene identifier.
cm : `bool`, optional
- Whether to compute the confusion matrix. The default is False.
+ Whether to compute the confusion matrix. The default is `False`.
plot : `bool`, optional
Whether to plot a false color composite, ground truth and model
- prediction for each scene. The default is False.
+ prediction for each scene. The default is `False`.
**kwargs
Additional keyword arguments passed to
- `pysegcnn.core.graphics.plot_sample`.
+ :py:func:`pysegcnn.core.graphics.plot_sample`.
Raises
------
TypeError
Raised if ``ds`` is not an instance of
- `~pysegcnn.core.split.SceneSubset`.
+ :py:class:`pysegcnn.core.split.SceneSubset`.
Returns
-------
@@ -214,9 +214,10 @@ def predict_scenes(ds, model, scene_id=None, cm=False, plot=False, **kwargs):
The ground truth
``'prediction'``
Model prediction
- conf_mat : `numpy.ndarray`
+ conf_mat : :py:class:`numpy.ndarray`
The confusion matrix. Note that the confusion matrix ``conf_mat`` is
- only computed if ``cm`` = True.
+ only computed if ``cm=True``.
+
"""
# check whether the dataset is a valid subset, i.e. an instance of
# pysegcnn.core.split.SceneSubset
diff --git a/pysegcnn/core/trainer.py b/pysegcnn/core/trainer.py
index 24f56fe..1057831 100644
--- a/pysegcnn/core/trainer.py
+++ b/pysegcnn/core/trainer.py
@@ -102,17 +102,17 @@ class DatasetConfig(BaseConfig):
seed : `int`
The random seed. Used to split the dataset into training,
validation and test set. Useful for reproducibility.
- sort : `bool`, optional
+ sort : `bool`
Whether to chronologically sort the samples. Useful for time series
data. The default is `False`.
- transforms : `list`, optional
+ transforms : `list`
List of :py:class:`pysegcnn.core.transforms.Augment` instances.
Each item in ``transforms`` generates a distinct transformed
version of the dataset. The total dataset is composed of the
original untransformed dataset together with each transformed
version of it. If ``transforms=[]``, only the original dataset is
used. The default is `[]`.
- pad : `bool`, optional
+ pad : `bool`
Whether to center pad the input image. Set ``pad=True``, if the
images are not evenly divisible by the ``tile_size``. The image
data is padded with a constant padding value of zero. For each
@@ -206,13 +206,13 @@ class SplitConfig(BaseConfig):
tvratio : `float`
The ratio of training data to validation data, e.g. ``tvratio=0.8``
means 80% training, 20% validation.
- date : `str`, optional
+ date : `str`
A date. Used if ``split_mode='date'``. The default is `yyyymmdd`.
- dateformat : `str`, optional
+ dateformat : `str`
The format of ``date``. ``dateformat`` is used by
:py:func:`datetime.datetime.strptime' to parse ``date`` to a
:py:class:`datetime.datetime` object. The default is `'%Y%m%d'`.
- drop : `float`, optional
+ drop : `float`
Whether to drop samples (during training only) with a fraction of
pixels equal to the constant padding value >= ``drop``. ``drop=0``
means, do not drop any samples. The default is `0`.
@@ -403,49 +403,49 @@ class ModelConfig(BaseConfig):
The name of the optimizer to update the model weights.
loss_name : `str`
The name of the loss function measuring the model error.
- skip_connection : `bool`, optional
+ skip_connection : `bool`
Whether to apply skip connections. The default is `True`.
- kwargs: `dict`, optional
+ kwargs: `dict`
The configuration for each convolution in the model. The default is
`{'kernel_size': 3, 'stride': 1, 'dilation': 1}`.
- batch_size : `int`, optional
+ batch_size : `int`
The model batch size. Determines the number of samples to process
before updating the model weights. The default is `64`.
- checkpoint : `bool`, optional
+ checkpoint : `bool`
Whether to resume training from an existing model checkpoint. The
default is `False`.
- transfer : `bool`, optional
+ transfer : `bool`
Whether to use a model for transfer learning on a new dataset. If True,
the model architecture of ``pretrained_model`` is adjusted to a new
dataset. The default is `False`.
- pretrained_model : `str`, optional
+ pretrained_model : `str`
The name of the pretrained model to use for transfer learning.
The default is `''`.
- lr : `float`, optional
+ lr : `float`
The learning rate used by the gradient descent algorithm.
The default is `0.001`.
- early_stop : `bool`, optional
+ early_stop : `bool`
Whether to apply `Early Stopping`_. The default is `False`.
- mode : `str`, optional
- The mode of the early stopping. Depends on the metric measuring
+ mode : `str`
+ The early stopping mode. Depends on the metric measuring
performance. When using model loss as metric, use ``mode='min'``,
however, when using accuracy as metric, use ``mode='max'``. For now,
only ``mode='max'`` is supported. Only used if ``early_stop=True``.
The default is `'max'`.
- delta : `float`, optional
+ delta : `float`
Minimum change in early stopping metric to be considered as an
improvement. Only used if ``early_stop=True``. The default is `0`.
- patience : `int`, optional
+ patience : `int`
The number of epochs to wait for an improvement in the early stopping
metric. If the model does not improve over more than ``patience``
epochs, quit training. Only used if ``early_stop=True``. The default is
`10`.
- epochs : `int`, optional
+ epochs : `int`
The maximum number of epochs to train. The default is `50`.
- nthreads : `int`, optional
+ nthreads : `int`
The number of cpu threads to use during training. The default is
:py:func:`torch.get_num_threads()`.
- save : `bool`, optional
+ save : `bool`
Whether to save the model state to disk. Model states are saved in
pysegcnn/main/_models. The default is `True`.
model_class : :py:class:`pysegcnn.core.models.Network`
@@ -853,31 +853,41 @@ class EvalConfig(BaseConfig):
test : `bool` or `None`
Whether to evaluate the model on the training(``test=None``), the
validation (``test=False``) or the test set (``test=True``).
- predict_scene : `bool`, optional
+ predict_scene : `bool`
The model prediction order. If False, the samples (tiles) of a dataset
are predicted in any order and the scenes are not reconstructed.
If True, the samples (tiles) are ordered according to the scene they
belong to and a model prediction for each entire reconstructed scene is
returned. The default is `False`.
- plot_samples : `bool`, optional
+ plot_samples : `bool`
Whether to save a plot of false color composite, ground truth and model
prediction for each sample (tile). Only used if ``predict_scene=False``
. The default is `False`.
- plot_scenes : `bool`, optional
+ plot_scenes : `bool`
Whether to save a plot of false color composite, ground truth and model
prediction for each entire scene. Only used if ``predict_scene=True``.
The default is `False`.
- plot_bands : `list` [`str`], optional
+ plot_bands : `list` [`str`]
The bands to build the false color composite. The default is
`['nir', 'red', 'green']`.
- cm : `bool`, optional
+ cm : `bool`
Whether to compute and plot the confusion matrix. The default is `True`
.
- figsize : `tuple`, optional
+ figsize : `tuple`
The figure size in centimeters. The default is `(10, 10)`.
- alpha : `int`, optional
+ alpha : `int`
The level of the percentiles for contrast stretching of the false color
compsite. The default is `0`, i.e. no stretching.
+ base_path : :py:class:`pathlib.Path`
+ Root path to store model output.
+ sample_path : :py:class:`pathlib.Path`
+ Path to store plots of model predictions for single samples.
+ scenes_path : :py:class:`pathlib.Path`
+ Path to store plots of model predictions for entire scenes.
+ perfmc_path : :py:class:`pathlib.Path`
+ Path to store plots of model performance, e.g. confusion matrix.
+ models_path : :py:class:`pathlib.Path`
+ Path to search for model state files, i.e. pretrained models.
"""
@@ -920,10 +930,6 @@ class EvalConfig(BaseConfig):
self.models_path = self.base_path.joinpath('_models')
self.state_file = self.models_path.joinpath(self.state_file)
- # write initialization string to log file
- LogConfig.init_log('{}: ' + 'Evaluating model: {}.'.format(
- self.state_file.name))
-
@dataclasses.dataclass
class LogConfig(BaseConfig):
@@ -931,11 +937,14 @@ class LogConfig(BaseConfig):
Generate the model log file.
- Parameters
+ Attributes
----------
state_file : :py:class:`pathlib.Path`
Path to a model state file.
-
+ log_path : :py:class:`pathlib.Path`
+ Path to store model logs.
+ log_file : :py:class:`pathlib.Path`
+ Path to the log file of the model ``state_file``.
"""
state_file: pathlib.Path
@@ -991,7 +1000,7 @@ class NetworkTrainer(BaseConfig):
:py:class:`pysegcnn.core.models.Network` on a dataset of type
:py:class:`pysegcnn.core.dataset.ImageDataset`.
- Parameters
+ Attributes
----------
model : :py:class:`pysegcnn.core.models.Network`
The model to train. An instance of
@@ -1003,35 +1012,38 @@ class NetworkTrainer(BaseConfig):
The loss function to compute the model error. An instance of
:py:class:`torch.nn.Module`.
train_dl : :py:class:`torch.utils.data.DataLoader`
- The training :py:class:`torch.utils.data.DataLoader` instance.
+ The training :py:class:`torch.utils.data.DataLoader` instance build
+ from an instance of :py:class:`pysegcnn.core.split.CustomSubset`.
valid_dl : :py:class:`torch.utils.data.DataLoader`
- The validation :py:class:`torch.utils.data.DataLoader` instance.
+ The validation :py:class:`torch.utils.data.DataLoader` instance build
+ from an instance of :py:class:`pysegcnn.core.split.CustomSubset`.
test_dl : :py:class:`torch.utils.data.DataLoader`
- The test :py:class:`torch.utils.data.DataLoader` instance.
+ The test :py:class:`torch.utils.data.DataLoader` instance build from an
+ instance of :py:class:`pysegcnn.core.split.CustomSubset`.
state_file : :py:class:`pathlib.Path`
Path to save the model state.
- epochs : `int`, optional
+ epochs : `int`
The maximum number of epochs to train. The default is `1`.
- nthreads : `int`, optional
+ nthreads : `int`
The number of cpu threads to use during training. The default is
:py:func:`torch.get_num_threads()`.
- early_stop : `bool`, optional
+ early_stop : `bool`
Whether to apply `Early Stopping`_. The default is `False`.
- mode : `str`, optional
- The mode of the early stopping. Depends on the metric measuring
+ mode : `str`
+ The early stopping mode. Depends on the metric measuring
performance. When using model loss as metric, use ``mode='min'``,
however, when using accuracy as metric, use ``mode='max'``. For now,
only ``mode='max'`` is supported. Only used if ``early_stop=True``.
The default is `'max'`.
- delta : `float`, optional
+ delta : `float`
Minimum change in early stopping metric to be considered as an
improvement. Only used if ``early_stop=True``. The default is `0`.
- patience : `int`, optional
+ patience : `int`
The number of epochs to wait for an improvement in the early stopping
metric. If the model does not improve over more than ``patience``
epochs, quit training. Only used if ``early_stop=True``. The default is
`10`.
- checkpoint_state : `dict` [`str`, :py:class:`numpy.ndarray`], optional
+ checkpoint_state : `dict` [`str`, :py:class:`numpy.ndarray`]
A model checkpoint for ``model``. If specified, ``checkpoint_state``
should be a dictionary with keys:
``'ta'``
@@ -1043,9 +1055,25 @@ class NetworkTrainer(BaseConfig):
``'vl'``
The loss on the validation set (:py:class:`numpy.ndarray`).
The default is `{}`.
- save : `bool`, optional
+ save : `bool`
Whether to save the model state to ``state_file``. The default is
`True`.
+ device : `str`
+ The device to train the model on, i.e. `cpu` or `cuda`.
+ max_accuracy : `float`
+ Maximum accuracy on the validation dataset.
+ es : `None` or :py:class:`pysegcnn.core.trainer.EarlyStopping`
+ The early stopping instance if ``early_stop=True``, else `None`.
+ training_state : `dict` [`str`, `numpy.ndarray`]
+ The training state dictionary with keys:
+ ``'ta'``
+ The accuracy on the training set (:py:class:`numpy.ndarray`).
+ ``'tl'``
+ The loss on the training set (:py:class:`numpy.ndarray`).
+ ``'va'``
+ The accuracy on the validation set (:py:class:`numpy.ndarray`).
+ ``'vl'``
+ The loss on the validation set (:py:class:`numpy.ndarray`).
.. _Early Stopping:
https://en.wikipedia.org/wiki/Early_stopping
@@ -1363,6 +1391,24 @@ class EarlyStopping(object):
.. _Early Stopping:
https://en.wikipedia.org/wiki/Early_stopping
+ Attributes
+ ----------
+ mode : `str`
+ The early stopping mode.
+ best : `float`
+ Best metric score.
+ min_delta : `float`
+ Minimum change in early stopping metric to be considered as an
+ improvement.
+ patience : `int`
+ The number of epochs to wait for an improvement.
+ is_better : `function`
+ Function indicating whether the metric improved.
+ early_stop : `bool`
+ Whether the early stopping criterion is met.
+ counter : `int`
+ The counter advancing each time a metric does not improve.
+
"""
def __init__(self, mode='max', best=0, min_delta=0, patience=10):
@@ -1371,7 +1417,7 @@ class EarlyStopping(object):
Parameters
----------
mode : `str`, optional
- The mode of the early stopping. Depends on the metric measuring
+ The early stopping mode. Depends on the metric measuring
performance. When using model loss as metric, use ``mode='min'``,
however, when using accuracy as metric, use ``mode='max'``. For
now, only ``mode='max'`` is supported. Only used if
diff --git a/pysegcnn/core/transforms.py b/pysegcnn/core/transforms.py
index 0045c5a..4d3c27d 100644
--- a/pysegcnn/core/transforms.py
+++ b/pysegcnn/core/transforms.py
@@ -34,17 +34,14 @@ class Transform(object):
Parameters
----------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The image to transform.
Raises
------
NotImplementedError
- Raised if `~pysegcnn.core.transforms.Transform` is not inherited.
-
- Returns
- -------
- None.
+ Raised if :py:class:`pysegcnn.core.transforms.Transform` is not
+ inherited.
"""
raise NotImplementedError
@@ -54,6 +51,12 @@ class VariantTransform(Transform):
"""Base class for a spatially variant transformation.
Transformation on the ground truth required.
+
+ Attributes
+ ----------
+ invariant : `bool`
+ Whether the transformation is spatially invariant.
+
"""
def __init__(self):
@@ -66,6 +69,12 @@ class InvariantTransform(Transform):
"""Base class for a spatially invariant transformation.
Transformation on the ground truth not required.
+
+ Attributes
+ ----------
+ invariant : `bool`
+ Whether the transformation is spatially invariant.
+
"""
def __init__(self):
@@ -77,18 +86,24 @@ class InvariantTransform(Transform):
class FlipLr(VariantTransform):
"""Flip an image horizontally.
- Parameters
+ Attributes
----------
- p : `float`, optional
- The probability to apply the transformation. The default is 0.5.
-
- Returns
- -------
- None.
+ p : `float`
+ The probability to apply the transformation.
+ applied : `bool`
+ Whether the transformation was applied.
"""
def __init__(self, p=0.5):
+ """Initialize.
+
+ Parameters
+ ----------
+ p : `float`, optional
+ The probability to apply the transformation. The default is `0.5`.
+
+ """
super().__init__()
# the probability to apply the transformation
self.p = p
@@ -98,12 +113,12 @@ class FlipLr(VariantTransform):
Parameters
----------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The image to transform.
Returns
-------
- transform : `numpy.ndarray`
+ transform : :py:class:`numpy.ndarray`
The transformed image.
"""
@@ -131,18 +146,24 @@ class FlipLr(VariantTransform):
class FlipUd(VariantTransform):
"""Flip an image vertically.
- Parameters
+ Attributes
----------
- p : `float`, optional
- The probability to apply the transformation. The default is 0.5.
-
- Returns
- -------
- None.
+ p : `float`
+ The probability to apply the transformation.
+ applied : `bool`
+ Whether the transformation was applied.
"""
def __init__(self, p=0.5):
+ """Initialize.
+
+ Parameters
+ ----------
+ p : `float`, optional
+ The probability to apply the transformation. The default is `0.5`.
+
+ """
super().__init__()
# the probability to apply the transformation
self.p = p
@@ -152,12 +173,12 @@ class FlipUd(VariantTransform):
Parameters
----------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The image to transform.
Returns
-------
- transform : `numpy.ndarray`
+ transform : :py:class:`numpy.ndarray`
The transformed image.
"""
@@ -183,26 +204,35 @@ class FlipUd(VariantTransform):
class Rotate(VariantTransform):
- """Rotate an image by ``angle``.
+ """Rotate an image in the spatial plane.
- The image is rotated in the spatial plane.
+ .. important::
- If the input array has more then two dimensions, the spatial dimensions are
- assumed to be the last two dimensions of the array.
+ If the input array has more then two dimensions, the spatial dimensions
+ are assumed to be the last two dimensions of the array.
- Parameters
+ Attributes
----------
+ p : `float`
+ The probability to apply the transformation.
+ applied : `bool`
+ Whether the transformation was applied.
angle : `float`
The rotation angle in degrees.
- p : `float`, optional
- The probability to apply the transformation. The default is 0.5.
- Returns
- -------
- None.
"""
def __init__(self, angle, p=0.5):
+ """Initialize.
+
+ Parameters
+ ----------
+ angle : `float`
+ The rotation angle in degrees.
+ p : `float`, optional
+ The probability to apply the transformation. The default is `0.5`.
+
+ """
super().__init__()
# the rotation angle
@@ -216,13 +246,14 @@ class Rotate(VariantTransform):
Parameters
----------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The image to transform.
Returns
-------
- transform : `numpy.ndarray`
+ transform : :py:class:`numpy.ndarray`
The transformed image.
+
"""
if np.random.random(1) < self.p:
@@ -260,35 +291,27 @@ class Rotate(VariantTransform):
class Noise(InvariantTransform):
"""Add gaussian noise to an image.
- Valid modes are:
+ Supported modes are:
- 'add': image = image + noise
- 'speckle' : image = image + image * noise
+ - 'add' : ``image = image + noise``
+ - 'speckle' : ``image = image + image * noise``
- Parameters
+ Attributes
----------
+ modes : `list` [`str`]
+ The supported modes.
mode : `str`
The mode to add the noise.
- mean : `float`, optional
- The mean of the gaussian distribution from which the noise is sampled.
- The default is 0.
- var : `float`, optional
- The variance of the gaussian distribution from which the noise is
- sampled. The default is 0.05.
- p : `float`, optional
- The probability to apply the transformation. The default is 0.5.
- exclude : `list` [`float`] or `list` [`int`], optional
- Values for which the noise is not added. Useful for pixels resulting
- from image padding. The default is [].
-
- Raises
- ------
- ValueError
- Raised if ``mode`` is not supported.
-
- Returns
- -------
- None.
+ mean : `float`
+ The mean of the gaussian distribution.
+ var : `float`
+ The variance of the gaussian distribution.
+ p : `float`
+ The probability to apply the transformation.
+ applied : `bool`
+ Whether the transformation was applied.
+ exclude : `list` [`float`]
+ Values for which the noise is not added.
"""
@@ -296,6 +319,30 @@ class Noise(InvariantTransform):
modes = ['add', 'speckle']
def __init__(self, mode, mean=0, var=0.05, p=0.5, exclude=[]):
+ """Initialize.
+
+ Parameters
+ ----------
+ mode : `str`
+ The mode to add the noise.
+ mean : `float`, optional
+ The mean of the gaussian distribution from which the noise is
+ sampled. The default is `0`.
+ var : `float`, optional
+ The variance of the gaussian distribution from which the noise is
+ sampled. The default is `0.05`.
+ p : `float`, optional
+ The probability to apply the transformation. The default is `0.5`.
+ exclude : `list` [`float`] or `list` [`int`], optional
+ Values for which the noise is not added. Useful for pixels
+ resulting from image padding. The default is `[]`.
+
+ Raises
+ ------
+ ValueError
+ Raised if ``mode`` is not supported.
+
+ """
super().__init__()
# check which kind of noise to apply
@@ -320,13 +367,13 @@ class Noise(InvariantTransform):
Parameters
----------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The image to transform.
Returns
-------
- transform : `numpy.ndarray`
- The transformed image
+ transform : :py:class:`numpy.ndarray`
+ The transformed image.
"""
if np.random.random(1) < self.p:
@@ -372,19 +419,24 @@ class Augment(object):
Container class applying each transformation in ``transforms`` in order.
- Parameters
+ Attributes
----------
transforms : `list` or `tuple`
- A sequence of instances of `pysegcnn.core.transforms.VariantTransform`
- or `pysegcnn.core.transforms.InvariantTransform`.
-
- Returns
- -------
- None.
+ The transformations to apply.
"""
def __init__(self, transforms):
+ """Initialize.
+
+ Parameters
+ ----------
+ transforms : `list` or `tuple`
+ A list of instances of
+ :py:class:`pysegcnn.core.transforms.VariantTransform`
+ or :py:class:`pysegcnn.core.transforms.InvariantTransform`.
+
+ """
assert isinstance(transforms, (list, tuple))
self.transforms = transforms
@@ -396,16 +448,16 @@ class Augment(object):
Parameters
----------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The input image.
- gt : `numpy.ndarray`
+ gt : :py:class:`numpy.ndarray`
The corresponding ground truth of ``image``.
Returns
-------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The transformed image.
- gt : `numpy.ndarray`
+ gt : :py:class:`numpy.ndarray`
The transformed ground truth.
"""
@@ -432,17 +484,17 @@ class Augment(object):
return image, gt
def __repr__(self):
- """Representation of `~pysegcnn.core.transforms.Augment`.
+ """Representation.
Returns
-------
- repr : `str`
+ fs : `str`
Representation string.
"""
- fstring = self.__class__.__name__ + '('
+ fs = self.__class__.__name__ + '('
for t in self.transforms:
- fstring += '\n'
- fstring += ' {0}'.format(t)
- fstring += '\n)'
- return fstring
+ fs += '\n'
+ fs += ' {0}'.format(t)
+ fs += '\n)'
+ return fs
diff --git a/pysegcnn/core/utils.py b/pysegcnn/core/utils.py
index 84d2b3c..8cd73e8 100644
--- a/pysegcnn/core/utils.py
+++ b/pysegcnn/core/utils.py
@@ -38,76 +38,70 @@ SUFFIXES = ['toa_ref', 'toa_rad', 'toa_brt']
def img2np(path, tile_size=None, tile=None, pad=False, cval=0):
- """Read an image to a `numpy.ndarray`.
+ r"""Read an image to a :py:class:`numpy.ndarray`.
If ``tile_size`` is not `None`, the input image is divided into square
- tiles of size (``tile_size``, ``tile_size``). If the image is not evenly
- divisible and ``pad`` = False, a `ValueError` is raised. However, if
- ``pad`` = True, center padding with constant value ``cval`` is applied.
+ tiles of size ``(tile_size, tile_size)``. If the image is not evenly
+ divisible and ``pad=False``, a ``ValueError`` is raised. However, if
+ ``pad=True``, center padding with constant value ``cval`` is applied.
The tiling works as follows:
- (Padded) Input image:
-
- ------------------------------------------------
- | | | | |
- | tile_00 | tile_01 | ... | tile_0n |
- | | | | |
- |----------------------------------------------|
- | | | | |
- | tile_10 | tile_11 | ... | tile_1n |
- | | | | |
- |----------------------------------------------|
- | | | | |
- | ... | ... | ... | ... |
- | | | | |
- |----------------------------------------------|
- | | | | |
- | tile_m0 | tile_m1 | ... | tile_mn |
- | | | | |
- ------------------------------------------------
-
- where m = n. Each tile has its id, which starts at 0 in the topleft corner
- of the input image, i.e. tile_00 has id=0, and increases along the width
- axis, i.e. tile_0n has id=n, tile_10 has id=n+1, ..., tile_mn has
- id=(m * n) - 1.
-
- If ``tile`` is an integer, only the tile with id = ``tile`` is returned.
+ +-----------+-----------+-----------+-----------+
+ | | | | |
+ | tile_00 | tile_01 | ... | tile_0n |
+ | | | | |
+ +-----------+-----------+-----------+-----------+
+ | | | | |
+ | tile_10 | tile_11 | ... | tile_1n |
+ | | | | |
+ +-----------+-----------+-----------+-----------+
+ | | | | |
+ | ... | ... | ... | ... |
+ | | | | |
+ +-----------+-----------+-----------+-----------+
+ | | | | |
+ | tile_m0 | tile_m1 | ... | tile_mn |
+ | | | | |
+ +-----------+-----------+-----------+-----------+
+
+ where :math:`m = n`. Each tile has its id, which starts at `0` in the
+ topleft corner of the input image, i.e. `tile_00` has :math:`id=0`, and
+ increases along the width axis, i.e. `tile_0n` has :math:`id=n`, `tile_10`
+ has :math:`id=n+1`, ..., `tile_mn` has :math:`id=(m \\cdot n) - 1`.
+
+ If ``tile`` is an integer, only the tile with ``id=tile`` is returned.
Parameters
----------
- path : `str` or `None` or `numpy.ndarray`
+ path : `str` or `None` or :py:class:`numpy.ndarray`
The image to read.
tile_size : `None` or `int`, optional
- The size of a tile. The default is None.
+ The size of a tile. The default is `None`.
tile : `int`, optional
- The tile id. The default is None.
+ The tile id. The default is `None`.
pad : `bool`, optional
- Whether to center pad the input image. The default is False.
+ Whether to center pad the input image. The default is `False`.
cval : `float`, optional
- The constant padding value. The default is 0.
+ The constant padding value. The default is `0`.
Raises
------
FileNotFoundError
Raised if ``path`` is a path that does not exist.
TypeError
- Raised if ``path`` is not `str` or `None` or `numpy.ndarray`.
+ Raised if ``path`` is not `str` or `None` or :py:class:`numpy.ndarray`.
Returns
-------
- image : `numpy.ndarray`
+ image : :py:class:`numpy.ndarray`
The image array. The output shape is:
+ - `(tiles, bands, tile_size, tile_size)` if ``tile_size`` is not
+ `None`. If the image does only have one band,
+ `(tiles, tile_size, tile_size)`
- if ``tile_size`` is not `None`:
- shape=(tiles, bands, tile_size, tile_size)
- if the image does only have one band:
- shape=(tiles, tile_size, tile_size)
-
- else:
- shape=(bands, height, width)
- if the image does only have one band:
- shape=(height, width)
+ - `(bands, height, width)` if ``tile_size=None``. If the image does
+ only have one band, `(height, width)`.
"""
# check the type of path
@@ -252,12 +246,12 @@ def is_divisible(img_size, tile_size, pad=False):
tile_size : `int`
The size of the tile.
pad : `bool`, optional
- Whether to center pad the input image. The default is False.
+ Whether to center pad the input image. The default is `False`.
Raises
------
ValueError
- Raised if the image is not evenly divisible and ``pad`` = False.
+ Raised if the image is not evenly divisible and ``pad=False``.
Returns
-------
@@ -334,7 +328,7 @@ def check_tile_extend(img_size, topleft, tile_size):
-------
nrows : `int`
Number of rows of the tile within the image.
- ncols : TYPE
+ ncols : `int`
Number of columns of the tile within the image.
"""
@@ -382,7 +376,7 @@ def tile_topleft_corner(img_size, tile_size):
-------
indices : `dict`
The keys of ``indices`` are the tile ids (`int`) and the values are the
- topleft corners (`tuple` = (y, x)) of the tiles.
+ topleft corners (y, x) of the tiles.
"""
# check if the image is divisible into square tiles of size
@@ -411,14 +405,14 @@ def reconstruct_scene(tiles):
Parameters
----------
- tiles : `torch.Tensor` or `numpy.ndarray`
- The tiled image, shape=(tiles, bands, tile_size, tile_size) or
- shape=(tiles, tile_size, tile_size).
+ tiles : :py:class:`torch.Tensor` or :py:class:`numpy.ndarray`
+ The tiled image, shape: `(tiles, bands, tile_size, tile_size)` or
+ `(tiles, tile_size, tile_size)`.
Returns
-------
- image : `numpy.ndarray`
- The reconstructed image.
+ image : :py:class:`numpy.ndarray`
+ The reconstructed image, shape: `(bands, height, width)`.
"""
# convert to numpy array
@@ -453,9 +447,9 @@ def accuracy_function(outputs, labels):
Parameters
----------
- outputs : `torch.Tensor` or array_like
+ outputs : :py:class:`torch.Tensor` or `array_like`
The model prediction.
- labels : `torch.Tensor` or array_like
+ labels : :py:class:`torch.Tensor` or `array_like`
The ground truth.
Returns
@@ -666,13 +660,13 @@ def doy2date(year, doy):
Parameters
----------
year : `int`
- The year
+ The year.
doy : `int`
- The day of the year
+ The day of the year.
Returns
-------
- date : `datetime.datetime`
+ date : :py:class:`datetime.datetime`
The converted date.
"""
# convert year/day of year to a datetime object
@@ -689,8 +683,8 @@ def item_in_enum(name, enum):
----------
name : `str`
Name of the item.
- enum : `enum.Enum`
- An instance of `enum.Enum`.
+ enum : :py:class:`enum.Enum`
+ An instance of :py:class:`enum.Enum`.
Raises
------
@@ -718,25 +712,25 @@ def destack_tiff(image, outpath=None, overwrite=False, remove=False,
"""Destack a TIFF with more than one band into a TIFF file for each band.
Each band in ``image`` is saved to ``outpath`` as distinct TIFF file.
- The default filenames are: "filename(``image``) + _B(i).tif", where i is
- the respective number of each band in ``image``.
+ The default filenames are: `"filename(``image``) + _B(i).tif"`, where `i`
+ is the respective number of each band in ``image``.
Parameters
----------
- image : `str` or `pathlib.Path`
+ image : `str` or :py:class:`pathlib.Path`
The TIFF to destack.
outpath : `str`, optional
- Path to save the output TIFF files. The default is None. If None,
+ Path to save the output TIFF files. The default is `None`. If `None`,
``outpath`` is the path to ``image``.
remove : `bool`, optional
Whether to remove ``image`` from disk after destacking. The default is
- False.
+ `False`.
overwrite : `bool`, optional
- Whether to overwrite existing TIFF files.
+ Whether to overwrite existing TIFF files. The default is `False`.
suffix : `str`, optional
String to append to the filename of ``image``. If specified, the TIFF
- filenames for each band in ``image`` are, "filename(``image``) +
- + _B(i)_ + ``suffix``.tif". The default is ''.
+ filenames for each band in ``image`` are, `"filename(``image``) +
+ + _B(i)_ + ``suffix``.tif"`. The default is `''`.
Raises
------
@@ -841,39 +835,33 @@ def standard_eo_structure(source_path, target_path, overwrite=False,
The directory tree in ``source_path`` is modified to the following
structure in ``target_path``:
- target_path/
- scene_id_1/
- files matching scene_id_1
- scene_id_2/
- files matching scene_id_2
- .
- .
- .
- scene_id_n/
- files matching scene_id_n
+ - target_path/
+ - scene_id_1/
+ - files matching scene_id_1
+ - scene_id_2/
+ - files matching scene_id_2
+ - ...
+ - scene_id_n/
+ - files matching scene_id_n
Parameters
----------
- source_path : `str` or `pathlib.Path`
+ source_path : `str` or :py:class:`pathlib.Path`
Path to the remote sensing dataset.
- target_path : `str` or `pathlib.Path`
+ target_path : `str` or :py:class:`pathlib.Path`
Path to save the restructured dataset.
overwrite : `bool`, optional
Whether to overwrite existing files in ``target_path``.
- The default is True.
+ The default is `False`.
move : `bool`, optional
Whether to move the files from ``source_path`` to ``target_path``. If
- True, files in ``source_path`` are moved to ``target_path``, if False,
- files in ``source_path`` are copied to ``target_path``. The default is
- False.
+ `True`, files in ``source_path`` are moved to ``target_path``, if
+ `False`, files in ``source_path`` are copied to ``target_path``. The
+ default is `False`.
parser : `function`, optional
The scene identifier parsing function. Depends on the sensor of the
- dataset. See e.g., `pysegcnn.core.utils.parse_landsat_scene`.
-
- Returns
- -------
- None.
+ dataset. See e.g., :py:func:`pysegcnn.core.utils.parse_landsat_scene`.
"""
# create a directory for each scene
@@ -931,17 +919,18 @@ def extract_archive(inpath, outpath, overwrite=False):
Parameters
----------
- inpath : `str` or `pathlib.Path`
+ inpath : `str` or :py:class:`pathlib.Path`
Path to an archive.
- outpath : `str` or `pathlib.Path`
+ outpath : `str` or :py:class:`pathlib.Path`
Path to save extracted files.
overwrite : `bool`, optional
- Whether to overwrite existing extracted files.
+ Whether to overwrite existing extracted files. The default is `False`.
Returns
-------
- subdir : str
- path to the extracted files
+ target : :py:class:`pathlib.Path`
+ Path to the extracted files.
+
"""
inpath = pathlib.Path(inpath)
@@ -984,7 +973,7 @@ def read_landsat_metadata(file):
Parameters
----------
- file : `str` or `pathlib.Path`
+ file : `str` or :py:class:`pathlib.Path`
Path to a Landsat *_MTL.txt file.
Raises
@@ -997,6 +986,7 @@ def read_landsat_metadata(file):
metadata : `dict`
The metadata text file as dictionary, where each line is a (key, value)
pair.
+
"""
file = pathlib.Path(file)
# check if the metadata file exists
@@ -1032,7 +1022,8 @@ def get_radiometric_constants(metadata):
Parameters
----------
metadata : `dict`
- The dictionary returned by ``read_landsat_metadata``.
+ The dictionary returned by
+ :py:func:`pysegcnn.core.utils.read_landsat_metadata`.
Returns
-------
@@ -1040,6 +1031,7 @@ def get_radiometric_constants(metadata):
Radiometric rescaling factors of the OLI sensor.
tir : `dict`
Thermal conversion constants of the TIRS sensor.
+
"""
# regular expression patterns matching the radiometric rescaling factors
oli_pattern = re.compile('(RADIANCE|REFLECTANCE)_(MULT|ADD)_BAND_\\d{1,2}')
@@ -1064,41 +1056,41 @@ def landsat_radiometric_calibration(scene, outpath=None, exclude=[],
Convert the Landsat OLI bands to top of atmosphere radiance or reflectance
and the TIRS bands to top of atmosphere brightness temperature.
- Conversion is performed following the `equations`_ provided by the USGS.
+ .. important::
+
+ Conversion is performed following the `equations`_ provided by the
+ USGS.
The filename of each band is extended by one of the following suffixes,
depending on the type of the radiometric calibration:
- 'toa_ref': top of atmosphere reflectance
- 'toa_rad': top of atmopshere radiance
- 'toa_brt': top of atmosphere brightness temperature
+ - `'toa_ref'`: top of atmosphere reflectance
+ - `'toa_rad'`: top of atmopshere radiance
+ - `'toa_brt'`: top of atmosphere brightness temperature
Parameters
----------
- scene : `str` or `pathlib.Path`
+ scene : `str` or :py:class:`pathlib.Path`
Path to a Landsat scene in digital number format.
- outpath : `str` or `pathlib.Path`, optional
- Path to save the calibrated images. The default is None, which means
- saving to ``scene``.
+ outpath : `str` or :py:class:`pathlib.Path`, optional
+ Path to save the calibrated images. The default is `None`, which means
+ saving in the same directory ``scene``.
exclude : `list` [`str`], optional
- Bands to exclude from the radiometric calibration. The default is [].
+ Bands to exclude from the radiometric calibration. The default is `[]`.
radiance : `bool`, optional
- Whether to calculate top of atmosphere radiance. The default is False,
- which means calculating top of atmopshere reflectance.
+ Whether to calculate top of atmosphere radiance. The default is `False`
+ , which means calculating top of atmopshere reflectance.
overwrite : `bool`, optional
- Whether to overwrite the calibrated images. The default is False.
+ Whether to overwrite the calibrated images. The default is `False`.
remove_raw : `bool`, optional
- Whether to remove the raw digitial number images. The default is True.
+ Whether to remove the raw digitial number images. The default is `True`
+ .
Raises
------
FileNotFoundError
Raised if ``scene`` does not exist.
- Returns
- -------
- None.
-
.. _equations:
https://www.usgs.gov/land-resources/nli/landsat/using-usgs-landsat-level-1-data-product
diff --git a/pysegcnn/main/eval.py b/pysegcnn/main/eval.py
index 9197a88..4dc8849 100644
--- a/pysegcnn/main/eval.py
+++ b/pysegcnn/main/eval.py
@@ -2,10 +2,15 @@
Steps to run a model evaluation:
- (1) Configure the dictionary 'eval_config' in pysegcnn/main/config.py
- (2) Save pysegcnn/main/config.py
- (3) In a terminal, navigate to the repository's root directory
- (4) run "python pysegcnn/main/eval.py"
+ 1. Configure the dictionary ``eval_config`` in
+ :py:mod:`pysegcnn.main.config.py`
+ 2. Save :py:mod:`pysegcnn.main.config.py`
+ 3. In a terminal, navigate to the repository's root directory
+ 4. Run
+
+ .. code-block:: bash
+
+ python pysegcnn/main/eval.py
License
@@ -42,6 +47,7 @@ if __name__ == '__main__':
# initialize logging
log = LogConfig(ec.state_file)
dictConfig(log_conf(log.log_file))
+ log.init_log('{}: ' + 'Evaluating model: {}.'.format(ec.state_file.name))
# load the model state
model, _, model_state = Network.load(ec.state_file)
diff --git a/pysegcnn/main/train.py b/pysegcnn/main/train.py
index 6d75060..bdc23b5 100644
--- a/pysegcnn/main/train.py
+++ b/pysegcnn/main/train.py
@@ -2,13 +2,18 @@
Steps to launch a model run:
- (1) Configure the model run in pysegcnn/main/config.py
- (i) configure the dataset : dictionary 'dataset_config'
- (j) configure the dataset split: dictionary 'split_config'
- (k) configure the model : dictionary 'model_config'
- (2) Save pysegcnn/main/config.py
- (3) In a terminal, navigate to the repository's root directory
- (4) run "python pysegcnn/main/train.py"
+ 1. Configure the model run in :py:mod:`pysegcnn.main.config.py`
+ - configure the dataset: ``dataset_config``
+ - configure the split : ``split_config``
+ - configure the model : ``model_config``
+ 2. Save :py:mod:`pysegcnn.main.config.py`
+ 3. In a terminal, navigate to the repository's root directory
+ 4. Run
+
+ .. code-block:: bash
+
+ python pysegcnn/main/train.py
+
License
-------
diff --git a/pysegcnn/preprocessing/sparcs.py b/pysegcnn/preprocessing/sparcs.py
index 0f12f8a..e298e28 100644
--- a/pysegcnn/preprocessing/sparcs.py
+++ b/pysegcnn/preprocessing/sparcs.py
@@ -2,6 +2,8 @@
After downloading the Sparcs dataset, from the repository's root directory run:
+.. code-block:: bash
+
python pysegcnn/preprocessing/sparcs.py
The above command will print the usage of the script.
--
GitLab