Merge pull request #793 from henrykironde/PR777-new

Update docs to Panda's pydata_sphinx_theme

Author: henrykironde

.github/workflows/Conda-app.yml

@@ -52,8 +52,8 @@ jobs:
       - name: Check style
         run: yapf -d --recursive deepforest/ --style=.style.yapf
 
-      - name: Check notebook style 
-        run: nbqa yapf --in-place docs/examples/*.ipynb --style=.style.yapf
+      - name: Check notebook style
+        run: nbqa yapf --in-place docs/user_guide/examples/*.ipynb --style=.style.yapf
 
       - name: Check notebook build
         run: pytest --nbmake docs/**/*_test.ipynb

.gitignore

@@ -19,4 +19,5 @@ public/
 tests/__pycache__
 tests/data/*
 .vscode/
-
+*ipynb_checkpoints/
+docs/user_guide/deepforestr.md

AUTHORS.rst

@@ -21,13 +21,5 @@ Primary Authors
 Other Contributors
 ==================
 
-The incomplete list of individuals below have provided patches or otherwise
-contributed to the project.
 Check GitHub for recent `contributors <https://github.com/weecology/DeepForest/graphs/contributors>`_.
-We would like to thank everyone who has contributed to the project in any way.
-
-* `Henry Senyondo <https://github.com/henrykironde>`_
-
-* `Dingyi Fang <https://github.com/dingyif>`_
-
-* `Shawn Taylor <https://github.com/sdtaylor>`_
+We would like to thank everyone who has contributed to the project in any way.

HISTORY.rst

@@ -1,54 +1,68 @@
-=====================
-DeepForest Change Log
-=====================
+====================
+DeepForest Changelog
+====================
 
-**1.3.5**
+Version 1.3.5 (Date: Mar 12, 2024)
+----------------------------------
 
-Add Support for multiple annotation types (point, box, polygon).
-Create a utilities.download_ArcGIS_REST function to download tiles from ArcGIS REST API. For example to download NAIP imagery.
+- **New Feature:** Added support for multiple annotation types including point, box, and polygon.
+- **New Utility:** Created ``utilities.download_ArcGIS_REST`` function to download tiles from the ArcGIS REST API (e.g., NAIP imagery).
 
-**1.3.3**
+Version 1.3.3 (Date: Mar 12, 2024)
+----------------------------------
 
-* Allow for annotations_file to be none in split_raster, for use in data preprocessing.
+- **Enhancement:** ``split_raster`` now allows ``annotations_file`` to be ``None``, enabling flexibility during data preprocessing.
 
-**1.3.0**
+Version 1.3.0 (Date: Dec 3, 2023)
+----------------------------------
 
-* Removed IoU_Callback to better align with pytorch-lightning API, see https://github.com/Lightning-AI/pytorch-lightning/issues/19101
-* Refactored evaluate code to use pytorch-lightning evaluation loop directly to calculate results frame during training.
-* Refactored image_callbacks. Now uses predictions directly, you do not need to specify the root dir or csv file, since it assumed to be evaluation file.
+- **Deprecation:** Removed ``IoU_Callback`` for better alignment with the PyTorch Lightning API (see `issue <https://github.com/Lightning-AI/pytorch-lightning/issues/19101>`_).
+- **Refactor:** Evaluation code now leverages the PyTorch Lightning evaluation loop for result calculation during training.
+- **Refactor:** Simplified ``image_callbacks`` by using predictions directly. No need to specify the root directory or CSV file, as the evaluation file is assumed.
 
-**1.1.3**
+Version 1.1.3 (Date: Nov 9, 2021)
+----------------------------------
 
-Add box coordinates to evaluate results frame.
+- **Enhancement:** Added box coordinates to the evaluation results frame for better result tracking.
 
-**1.1.2**
+Version 1.1.2 (Date: Sep 30, 2021)
+----------------------------------
 
-If there was more than one class, the precision in the class_recall.csv was computed incorrectly.
+- **Bug Fix:** Fixed incorrect precision calculation in ``class_recall.csv`` when multiple classes were present.
 
-**1.1.1**
+Version 1.1.1 (Date: Sep 14, 2021)
+----------------------------------
 
-Update to project_boxes to include an output for predict_tile and predict_image, the function was split into two. annotations_to_shapefile reverses shapefile_to_annotations. Thanks to @sdtaylor for this contribution.
+- **Update:** ``project_boxes`` now includes output options for both ``predict_tile`` and ``predict_image``.
+- **New Feature:** Introduced ``annotations_to_shapefile``, which reverses ``shapefile_to_annotations`` functionality.
+  Thanks to @sdtaylor for this contribution.
 
-**1.1.0**
+Version 1.1.0 (Date: Aug 5, 2021)
+----------------------------------
 
-1.
-Empty frames are now allowed by passing annotations with 0's for all coords. A single row for each blank image.
+1. **Enhancement:** Empty frames are now allowed by passing annotations with 0's for all coordinates. Example format:
+   ::
 
-image_path, 0,0,0,0, "Tree"
+     image_path, 0, 0, 0, 0, "Tree"
 
-2.
-A check_release function was implemented to reduce github rate limit issues. on use_release(), the local model will be used if check_release = False
+2. **New Feature:** Introduced ``check_release`` to reduce GitHub rate limit issues. When using ``use_release()``, the local model will be used if ``check_release = False``.
 
-**1.0.9**
+Version 1.0.9 (Date: Jul 14, 2021)
+----------------------------------
 
-Minor update to improve windows users default dtype, thanks to @ElliotSalisbury
+- **Enhancement:** Improved default dtype for Windows users, thanks to @ElliotSalisbury for the contribution.
 
-**1.0.0**
+Version 1.0.0 (Date: Jun 7, 2021)
+----------------------------------
 
-Major update to replace tensorflow backend with pytorch. 
+- **Major Update:** Transitioned from TensorFlow to a PyTorch backend, enhancing performance and flexibility.
 
-**0.1.30**
-Bug fixes to allow learning rate monitoring and decay on val_classification_loss
+Version 0.1.34 (Date: )
+-----------------------
 
-**0.1.34**
-Profiled the dataset and evaluation code for performance. Evaluation should be much faster now.
+- **Optimization:** Profiled dataset and evaluation code, significantly improving evaluation performance.
+
+Version 0.1.30 (Date: )
+-----------------------
+
+- **Bug Fix:** Resolved issues to allow learning rate monitoring and decay based on ``val_classification_loss``.

deepforest/model.py

@@ -246,7 +246,7 @@ def configure_optimizers(self):
                                                                min_lr=0,
                                                                eps=1e-08)
 
-        #Monitor rate is val data is used
+        # Monitor rate is val data is used
         return {'optimizer': optimizer, 'lr_scheduler': scheduler, "monitor": 'val_loss'}
 
     def dataset_confusion(self, loader):

dev_requirements.txt

@@ -34,6 +34,7 @@ slidingwindow
 sphinx
 sphinx_markdown_tables
 sphinx_rtd_theme
+sphinx-design
 supervision
 tensorboard
 torch