Merge pull request #104 from CosmoStat/develop

Minor Patch Release 2.0.1
CosmoStat · Dec 15, 2023 · 87e0c8e · 87e0c8e
2 parents 146725f + 0cffc54
commit 87e0c8e
Show file tree

Hide file tree

Showing 13 changed files with 64 additions and 38 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -7,7 +7,7 @@ on:
   pull_request:
     branches:
       - main
-
+      - develop
 
 jobs:
   test-full:

diff --git a/docs/source/conf.py b/docs/source/conf.py
@@ -13,7 +13,7 @@
 project = "wf-psf"
 copyright = "2023, CosmoStat"
 author = "CosmoStat"
-release = "2.0.0"
+release = "2.0.1"
 
 # -- General configuration ---------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration

diff --git a/docs/source/configuration.md b/docs/source/configuration.md
@@ -30,7 +30,7 @@ Next, we shall describe each configuration file.
 (data_config)=
 ## Data Configuration
 
-The file [data_config.yaml](https://github.com/CosmoStat/wf-psf/blob/dummy_main/config/data_config.yaml) stores the metadata for generating training and test data sets or retrieving existing ones.  A set of training and test data is provided in the `data/coherent_euclid_dataset` directory.  `Wavediff` will automatically retrieve the dataset within its directory tree.  In the field `data_dir`, the user should specify the sub-path to the data directory as in the example below.  The actual name of the dataset file is provided as an entry to the field `file`. However, new training and test data sets can be produced with the parameters in the file, which *should be* provided to the `simPSF` code although not at present (implementation upgrade pending).  
+The file [data_config.yaml](https://github.com/CosmoStat/wf-psf/blob/main/config/data_config.yaml) stores the metadata for generating training and test data sets or retrieving existing ones.  A set of training and test data is provided in the `data/coherent_euclid_dataset` directory.  `Wavediff` will automatically retrieve the dataset within its directory tree.  In the field `data_dir`, the user should specify the sub-path to the data directory as in the example below.  The actual name of the dataset file is provided as an entry to the field `file`. However, new training and test data sets can be produced with the parameters in the file, which *should be* provided to the `simPSF` code although not at present (implementation upgrade pending).  
 
 ```
 # Training and test data sets for training and/or metrics evaluation
@@ -54,7 +54,7 @@ data:
 (training_config)=
 ## Training Configuration
 
-The file [training_config.yaml](https://github.com/CosmoStat/wf-psf/blob/dummy_main/config/training_config.yaml) is used to configure the settings for the training pipeline task. The first line contains the parent key `training`. All of the following child keys are treated as values of the `training` key. Above each child key a description is provided. Below is an abridged example of this:
+The file [training_config.yaml](https://github.com/CosmoStat/wf-psf/blob/main/config/training_config.yaml) is used to configure the settings for the training pipeline task. The first line contains the parent key `training`. All of the following child keys are treated as values of the `training` key. Above each child key a description is provided. Below is an abridged example of this:
 
 ```
 training:
@@ -90,7 +90,7 @@ Training hyperparameters, defined by the parent key: `training_hparams`, include
 (metrics_config)=
 ## Metrics Configuration
 
-The [metrics_config.yaml](https://github.com/CosmoStat/wf-psf/blob/dummy_main/config/metrics_config.yaml) file stores the configuration parameters for the WaveDiff pipeline to perform computations of the four metrics listed in the table on a trained PSF model, as applied in {cite:t}`Liaudat:23`.
+The [metrics_config.yaml](https://github.com/CosmoStat/wf-psf/blob/main/config/metrics_config.yaml) file stores the configuration parameters for the WaveDiff pipeline to perform computations of the four metrics listed in the table on a trained PSF model, as applied in {cite:t}`Liaudat:23`.
 
 | Metric type | Description |
 | --- | ----------- |
@@ -213,7 +213,7 @@ The `metrics` package is run using [TensorFlow](https://www.tensorflow.org) to r
 (plotting_config)=
 ## Plot Configuration
 
-The [plotting_config.yaml](https://github.com/CosmoStat/wf-psf/blob/dummy_main/config/plotting_config.yaml) file stores the configuration parameters for the WaveDiff pipeline to generate plots for the metrics listed in the {ref}`metrics settings table <metrics_settings>` for each data set.
+The [plotting_config.yaml](https://github.com/CosmoStat/wf-psf/blob/main/config/plotting_config.yaml) file stores the configuration parameters for the WaveDiff pipeline to generate plots for the metrics listed in the {ref}`metrics settings table <metrics_settings>` for each data set.
 
 An example of the contents of the `plotting_config.yaml` file is shown below.
 
@@ -283,7 +283,7 @@ plotting_params:
 (master_config_file)=
 ## Master Configuration
 
-The `configs.yaml` file is the master configuration file that is used to define all of the pipeline tasks to be submitted and executed by `WaveDiff` during runtime. In this file, the user lists the processing tasks (one or more) to be performed by setting the values of the associated configuration variables `{pipeline_task}_conf` and the name of the configuration file `{pipeline_task}_config.yaml`.  See an example below to configure `WaveDiff` to launch a sequence of runs to train models 1...n with their respective configurations given in the files `training_config_{id}.yaml`.
+The [configs.yaml](https://github.com/CosmoStat/wf-psf/blob/main/config/configs.yaml) file is the master configuration file that is used to define all of the pipeline tasks to be submitted and executed by `WaveDiff` during runtime. In this file, the user lists the processing tasks (one or more) to be performed by setting the values of the associated configuration variables `{pipeline_task}_conf` and the name of the configuration file `{pipeline_task}_config.yaml`.  See an example below to configure `WaveDiff` to launch a sequence of runs to train models 1...n with their respective configurations given in the files `training_config_{id}.yaml`.
 
 ```
 ---

diff --git a/docs/source/dependencies.md b/docs/source/dependencies.md
@@ -8,6 +8,7 @@ Third-party software packages required by WaveDiff are installed automatically (
 |--------------|---------------------------------------------------|
 | [numpy](https://numpy.org/)                  | {cite:t}`harris:20` | 
 | [scipy](https://scipy.org)                   | {cite:t}`SciPy-NMeth:20` |
+| [keras](https://keras.io)                    | {cite:t}`chollet:2015keras`|
 | [tensorflow](https://www.tensorflow.org)     | {cite:t}`tensorflow:15` |
 | [tensorflow-addons](https://www.tensorflow.org/addons) |{cite:t}`tensorflow:15` |
 | [tensorflow-estimator](https://www.tensorflow.org/api_docs/python/tf/estimator) |{cite:t}`tensorflow:15` |
@@ -17,4 +18,5 @@ Third-party software packages required by WaveDiff are installed automatically (
 | [galsim](http://galsim-developers.github.io/GalSim/_build/html/index.html#) |    {cite:t}`rowe:15`        |
 | [astropy](https://www.astropy.org) |   {cite:t}`astropy:13,astropy:18`, <br>{cite:t}`astropy:22`       |
 | [matplotlib](https://matplotlib.org) |   {cite:t}`Hunter:07`    |
+| [pandas](https://pandas.pydata.org)  | {cite:t}`mckinney:2010pandas`   |
 | [seaborn](https://seaborn.pydata.org)   |    {cite:t}`Waskom:21`        |
diff --git a/docs/source/installation.md b/docs/source/installation.md
@@ -5,6 +5,8 @@ WaveDiff is a software written in Python and uses the [TensorFlow](https://www.t
 
 Note: You may want to set up a dedicated Python environment for running WaveDiff using e.g. [Conda](https://docs.conda.io/en/latest/).  You can use the minimal installer [Miniconda](https://docs.conda.io/projects/miniconda/en/latest/) to set up the environment in which we run the command below to install the subset of packages needed for running WaveDiff.
 
+In the [WaveDiff repository](https://github.com/CosmoStat/wf-psf.git), we provide in the file `environment.yml` the specific environment used in the testing of `WaveDiff`. We recommend users to work with this environment for the current release. 
+
 ## Installation Steps
 
 Clone the repository:

diff --git a/docs/source/refs.bib b/docs/source/refs.bib
@@ -126,6 +126,14 @@ @article{Liaudat:23
 journal = {Inverse Problems},
 }
 
+@online{chollet:2015keras, 
+  title={Keras}, 
+  author={Chollet, Francois and others}, 
+  year={2015}, 
+  publisher={GitHub}, 
+  url={https://github.com/fchollet/keras}, 
+}
+
 @misc{clark:15, 
   title={Pillow (PIL Fork) Documentation}, 
   author={Clark, Alex}, 
@@ -196,6 +204,16 @@ @ARTICLE{SciPy-NMeth:20
   doi     = {10.1038/s41592-019-0686-2},
 }
 
+@inproceedings{mckinney:2010pandas, 
+  title={Data structures for statistical computing in python}, 
+  author={McKinney, Wes and others}, 
+  booktitle={Proceedings of the 9th Python in Science Conference}, 
+  volume={445}, 
+  pages={51--56}, 
+  year={2010}, 
+  organization={Austin, TX} 
+}
+
 @article{opencv_library:08, 
     author = {Bradski, G.}, 
     citeulike-article-id = {2236121}, 

diff --git a/docs/source/wf_psf.info.rst b/docs/source/wf_psf.info.rst
diff --git a/docs/source/wf_psf.plotting.plot_optimisation_metrics.rst b/docs/source/wf_psf.plotting.plot_optimisation_metrics.rst
diff --git a/docs/source/wf_psf.plotting.rst b/docs/source/wf_psf.plotting.rst
@@ -12,5 +12,4 @@ Submodules
 .. toctree::
    :maxdepth: 4
 
-   wf_psf.plotting.plot_optimisation_metrics
    wf_psf.plotting.plots_interface
diff --git a/docs/source/wf_psf.rst b/docs/source/wf_psf.rst
@@ -25,6 +25,5 @@ Submodules
 
 .. toctree::
    :maxdepth: 4
-
-   wf_psf.info
+
    wf_psf.run
diff --git a/environment.yml b/environment.yml
@@ -0,0 +1,19 @@
+name: wavediff-env
+channels:
+  - conda-forge
+dependencies:
+  - keras=2.9.0=pyhd8ed1ab_0
+  - scipy=1.11.4=py310h2b794db_0
+  - tensorflow=2.9.1=cpu_py310h6ecea76_0
+  - pip=23.3.1
+  - pip:
+      - astropy>=6.0.0
+      - galsim>=2.5.1
+      - matplotlib>=3.8.2
+      - opencv-python>=4.8.1.78
+      - pandas>=2.1.4
+      - pillow>=10.1.0
+      - seaborn>=0.13.0
+      - tensorflow-addons>=0.23.0
+      - tensorflow-estimator>=2.15.0
+      - zernike>=0.0.32
diff --git a/pyproject.toml b/pyproject.toml
@@ -11,21 +11,22 @@ maintainers = [
 
 description = 'A software framework to perform Differentiable wavefront-based PSF modelling.'
 dependencies = [
-    "numpy>=1.19.2",
-    "scipy>=1.11.2",
-    "tensorflow>=2.9.1",
-    "tensorflow-addons>=0.12.1",
-    "tensorflow-estimator>=2.9.0",
-    "zernike==0.0.31",
-    "opencv-python>=4.5.1.48",
-    "pillow>=9.5.0",
-    "galsim>=2.4.11",
-    "astropy>=5.3.3",
-    "matplotlib>=3.3.2",
-    "seaborn>=0.12.2",
+    "numpy",
+    "scipy",
+    "keras==2.9.0",
+    "tensorflow==2.9.1",
+    "tensorflow-addons>=0.23.0",
+    "tensorflow-estimator",
+    "zernike",
+    "opencv-python",
+    "pillow",
+    "galsim",
+    "astropy",
+    "matplotlib",
+    "seaborn",
 ]
 
-version = "2.0.0"
+version = "2.0.1"
 
 [project.optional-dependencies]
 docs = [

diff --git a/src/wf_psf/training/train.py b/src/wf_psf/training/train.py
@@ -362,7 +362,7 @@ def train(
             learning_rate_non_param=training_handler.learning_rate_non_params[
                 current_cycle - 1
             ],
-            n_epochs_param=training_handler.n_epochs_non_params[current_cycle - 1],
+            n_epochs_param=training_handler.n_epochs_params[current_cycle - 1],
             n_epochs_non_param=training_handler.n_epochs_non_params[current_cycle - 1],
             param_optim=param_optim,
             non_param_optim=non_param_optim,