feat: expand api documentation and document legacy

docs/chapters/plantseg_legacy/installation.md

@@ -0,0 +1,107 @@
+# Installation
+
+## Prerequisites for Conda package
+
+* Linux, Windows, macOS (not all features are available on macOS)
+* (Optional) Nvidia GPU with official Nvidia drivers installed for GPU acceleration
+
+## Install Mamba
+
+The easiest way to install PlantSeg is by using the [conda (Anaconda)](https://www.anaconda.com/) or
+[mamba (Miniforge)](https://mamba.readthedocs.io/en/latest/index.html) package manager. We recommend using `mamba` because it is faster and usually more consistent than `conda`.
+
+=== "Linux"
+
+    To download Miniforge open a terminal and type:
+
+    ```bash
+    curl -L -O "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-$(uname)-$(uname -m).sh"
+    bash Miniforge3-$(uname)-$(uname -m).sh
+    ```
+
+    Then install by typing:
+
+    ```bash
+    bash Miniforge3-$(uname)-$(uname -m).sh
+    ```
+
+    and follow the installation instructions.
+    Please refer to the [Miniforge repo](https://github.com/conda-forge/miniforge) for more information, troubleshooting and uninstallation instructions.
+    The miniforge installation file `Miniforge3-*.sh` can be deleted now.
+
+=== "Windows/macOS"
+    The first step required to use the pipeline is installing mamba. The installation can be done by downloading the installer from the [Miniforge repo](https://github.com/conda-forge/miniforge). There you can find the download links for the latest version of Miniforge, troubleshooting and uninstallation instructions.
+
+## Install PlantSeg using Mamba
+
+PlantSeg can be installed directly by executing in the terminal (or PowerShell on Windows). For `conda` users, the command is identical, just replace `mamba` with `conda`.
+
+=== "Linux"
+
+    * NVIDIA GPU version, CUDA=12.x
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c nvidia -c conda-forge pytorch pytorch-cuda=12.1 plant-seg=1.8.1 --no-channel-priority
+        ```
+
+    * NVIDIA GPU version, CUDA=11.x
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c nvidia -c conda-forge pytorch pytorch-cuda=11.8 plant-seg=1.8.1 --no-channel-priority
+        ```
+
+    * CPU version
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c nvidia -c conda-forge pytorch cpuonly plant-seg=1.8.1 --no-channel-priority
+        ```
+
+=== "Windows"
+
+    * NVIDIA GPU version, CUDA=12.x
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c nvidia -c conda-forge pytorch pytorch-cuda=12.1 nifty=1.2.1=*_4 plant-seg=1.8.1 --no-channel-priority
+        ```
+
+    * NVIDIA GPU version, CUDA=11.x
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c nvidia -c conda-forge pytorch pytorch-cuda=11.8 nifty=1.2.1=*_4 plant-seg=1.8.1 --no-channel-priority
+        ```
+
+    * CPU version
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c nvidia -c conda-forge pytorch cpuonly nifty=1.2.1=*_4 plant-seg=1.8.1 --no-channel-priority
+        ```
+
+=== "macOS"
+
+    * Apple silicon version
+
+        ```bash
+        mamba create -n plant-seg -c pytorch -c conda-forge python=3.11 pytorch::pytorch plant-seg=1.8.1 --no-channel-priority
+        ```
+
+If you used older versions of PlantSeg, please delete the old config files in `~/.plantseg_models/configs/` after installing new PlantSeg.
+
+The above command will create new conda environment `plant-seg` together with all required dependencies.
+
+Please refer to the [PyTorch](https://pytorch.org/get-started/locally/) website for more information on the available versions of PyTorch and the required CUDA version. The GPU version of Pytorch will also work on CPU only machines but has a much larger installation on disk.
+
+## Optional dependencies
+
+Certain compressed TIFF files (e.g., Zlib, ZSTD, LZMA formats) require additional codecs to be processed correctly by PlantSeg. To handle such files, install the `imagecodecs` package:
+
+```bash
+conda activate plant-seg
+pip install imagecodecs
+```
+
+If you plan to use SimpleITK-based watershed segmentation, you will need to install `SimpleITK` as an additional dependency:
+
+```bash
+conda activate plant-seg
+pip install SimpleITK
+```

docs/chapters/python_api/tasks/dataprocessing_tasks.md

@@ -0,0 +1,41 @@
+# Import and export tasks
+
+## Gaussian smoothing task
+
+::: plantseg.tasks.dataprocessing_tasks.gaussian_smoothing_task
+
+## Image cropping task
+
+::: plantseg.tasks.dataprocessing_tasks.image_cropping_task
+
+
+## Image rescale to shape task
+
+::: plantseg.tasks.dataprocessing_tasks.image_rescale_to_shape_task
+
+
+## Image rescale to voxel size task
+
+::: plantseg.tasks.dataprocessing_tasks.image_rescale_to_voxel_size_task
+
+## Set image voxel size task
+
+::: plantseg.tasks.dataprocessing_tasks.set_voxel_size_task
+
+## Label Postprocessing
+
+## Remove false positives task
+
+::: plantseg.tasks.dataprocessing_tasks.remove_false_positives_by_foreground_probability_task
+
+## Fix Over/Under segmentation task
+
+::: plantseg.tasks.dataprocessing_tasks.fix_over_under_segmentation_from_nuclei_task
+
+## Set biggest object as background task
+
+::: plantseg.tasks.dataprocessing_tasks.set_biggest_instance_to_zero_task
+
+## Relabel task
+
+::: plantseg.tasks.dataprocessing_tasks.relabel_segmentation_task

docs/chapters/python_api/tasks/io_tasks.md

@@ -0,0 +1,9 @@
+# Import and export tasks
+
+## Import task
+
+::: plantseg.tasks.io_tasks.import_image_task
+
+## Export task
+
+::: plantseg.tasks.io_tasks.export_image_task

docs/chapters/python_api/tasks/prediction_tasks.md

@@ -0,0 +1,5 @@
+# Neural network prediction tasks
+
+## UNet prediction task
+
+::: plantseg.tasks.prediction_tasks.unet_prediction_task

docs/chapters/python_api/tasks/segmentation_tasks.md

@@ -0,0 +1,13 @@
+# Segmentation tasks
+
+## Distance transform watershed task
+
+::: plantseg.tasks.segmentation_tasks.dt_watershed_task
+
+## Cluster segmentation task
+
+::: plantseg.tasks.segmentation_tasks.clustering_segmentation_task
+
+## Lifted Multicut task
+
+::: plantseg.tasks.segmentation_tasks.lmc_segmentation_task

plantseg/tasks/io_tasks.py

@@ -30,18 +30,8 @@ def import_image_task(
         input_path (Path): path to the image file
         semantic_type (str): semantic type of the image (raw, segmentation, prediction)
         stack_layout (str): stack layout of the image (3D, 2D, 2D_time)
-        image_name (str | Noinput_path = inputs[input_schema["name"]]
-    input_paths = parse_import_image_task(input_path)
-    list_inputs.extend(input_paths)ne): name of the image (if None, the file name will be used)
-        key (str | None):"export_directory": RunTimeInput(
-            allowed_types=['str'],
-            description="Output directory path where the image will be saved",
-            headless_default=None,
-            user_input_required=True,
-        ),
-        "name_pattern": RunTimeInput(
-            allowed_types=['str'], description="Output file name", headless_default=None, user_input_required=False
-        ), key for the image (used only for h5 and zarr formats)
+        image_name (str): name of the image, if None the name will be the same as the file name
+        key (str | None): key for the image (used only for h5 and zarr formats)
         m_slicing (str | None): m_slicing of the image (None, time, z, y, x)
     """