### Install Miniforge on Windows

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniforge for Windows (x86_64) using PowerShell. It downloads the installer, executes it silently, and then cleans up the installer file.

```powershell
Invoke-WebRequest -Uri "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Windows-x86_64.exe" -OutFile "$env:UserProfile/Downloads/Miniforge3-Windows-x86_64.exe"; Start-Process -FilePath "$env:UserProfile/Downloads/Miniforge3-Windows-x86_64.exe" -ArgumentList "/InstallationType=JustMe /RegisterPython=1 /S" -Wait; Remove-Item -Path "$env:UserProfile/Downloads/Miniforge3-Windows-x86_64.exe"
```

--------------------------------

### Pip Installation from Repository (Specific Branch)

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Installs a specific branch (e.g., 'develop') of the sleap package from its GitHub repository using Pip.

```shell
pip install git+https://github.com/murthylab/sleap.git@develop
```

--------------------------------

### Install Miniforge on Linux

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniforge for Linux (x86_64) using bash. It downloads the installer script, makes it executable, runs it, removes the script, and initializes conda for the current shell.

```bash
curl -fsSL --compressed https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-x86_64.sh -o "~/Downloads/Miniforge3-Linux-x86_64.sh" && chmod +x "~/Downloads/Miniforge3-Linux-x86_64.sh" && "~/Downloads/Miniforge3-Linux-x86_64.sh" -b -p ~/miniforge3 && rm "~/Downloads/Miniforge3-Linux-x86_64.sh" && ~/miniforge3/bin/conda init "$(basename "${SHELL}")" && source "$HOME/.$(basename "${SHELL}")rc"
```

--------------------------------

### Install SLEAP

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Training_and_inference_on_an_example_dataset.ipynb

Installs the SLEAP library and its dependencies using pip. It first uninstalls potentially conflicting OpenCV versions and then installs SLEAP with the recommended PyPI package.

```shell
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"
```

--------------------------------

### Install Miniconda on Linux

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniconda for Linux (x86_64) using bash. It creates a directory, downloads the installer script, executes it, removes the script, and initializes conda for the current shell.

```bash
mkdir -p ~/miniconda3 && wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh && bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3 && rm ~/miniconda3/miniconda.sh && ~/miniconda3/bin/conda init "$(basename "${SHELL}")" && source "$HOME/.$(basename "${SHELL}")rc"
```

--------------------------------

### Clone SLEAP Repository

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Clones the SLEAP source code from GitHub and navigates into the cloned directory. This is the first step for installing SLEAP from source.

```bash
git clone https://github.com/talmolab/sleap && cd sleap
```

--------------------------------

### Development Installation from Repository

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Installs sleap from its GitHub repository for development purposes. This involves cloning the repo, creating a dedicated Conda environment ('sleap_dev'), and installing the package in editable mode.

```shell
git clone https://github.com/murthylab/sleap.git
cd sleap
conda env create -n sleap_dev -f environment.yml
conda activate sleap_dev
pip install -r dev_requirements.txt
pip install --no-deps -e .
```

--------------------------------

### Pip Installation from Repository (Windows Dependencies)

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Installs sleap from the repository, specifically addressing Windows users who need to install 'tensorflow-gpu' and 'Shapely' via Conda first. It then installs sleap using Pip.

```shell
conda env create -n sleap_env python=3.6 tensorflow-gpu=2 Shapely
conda activate sleap_env
pip install git+https://github.com/murthylab/sleap.git
```

--------------------------------

### Pip Installation from Test PyPi

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Installs the sleap package from the test PyPI repository. This method requires creating a Conda environment first and then using Pip to install from the specified index URL.

```shell
conda env create -n sleap_env python=3.6
conda activate sleap_env
pip install --extra-index-url https://test.pypi.org/simple/ sleap
```

--------------------------------

### Launch SLEAP GUI

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Launches the SLEAP graphical labeling interface. This command is used to verify that the GUI component of SLEAP is installed and functioning correctly.

```bash
sleap-label
```

--------------------------------

### Install Miniforge on Mac (Intel)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniforge for macOS on Intel (x86_64) using bash. It downloads the installer script, makes it executable, runs it, removes the script, and initializes conda for the current shell.

```bash
curl -fsSL --compressed https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-x86_64.sh -o "~/Downloads/Miniforge3-MacOSX-x86_64.sh" && chmod +x "~/Downloads/Miniforge3-MacOSX-x86_64.sh" && "~/Downloads/Miniforge3-MacOSX-x86_64.sh" -b -p ~/miniforge3 && rm "~/Downloads/Miniforge3-MacOSX-x86_64.sh" && ~/miniforge3/bin/conda init "$(basename "${SHELL}")" && source "$HOME/.$(basename "${SHELL}")rc"
```

--------------------------------

### List Installed Conda Packages

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Activates the SLEAP conda environment and lists all installed packages with their versions. This is crucial for reporting installation problems, as it provides essential dependency information.

```bash
conda list
```

--------------------------------

### Download Sample Training Data

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Training_and_inference_on_an_example_dataset.ipynb

Downloads a sample dataset for demonstration purposes. It installs the 'tree' utility, downloads a zip file containing the dataset, extracts it, and then displays the directory structure.

```shell
!apt-get install tree
!wget -O dataset.zip https://github.com/talmolab/sleap-datasets/releases/download/dm-courtship-v1/drosophila-melanogaster-courtship.zip
!mkdir dataset
!unzip dataset.zip -d dataset
!rm dataset.zip
!tree dataset
```

--------------------------------

### Install Miniconda on Windows

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniconda for Windows (x86_64) using PowerShell. It downloads the latest installer, executes it silently, and then removes the downloaded installer file.

```powershell
curl https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe -o miniconda.exe; Start-Process -FilePath ".\miniconda.exe" -ArgumentList "/S" -Wait; del miniconda.exe
```

--------------------------------

### Install Miniconda on Mac (Intel)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniconda for macOS on Intel (x86_64) using bash. It downloads the installer script, makes it executable, runs it, removes the script, and initializes conda for the current shell.

```bash
curl -fsSL --compressed https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-x86_64.sh -o "~/Downloads/Miniconda3-latest-MacOSX-x86_64.sh" && chmod +x "~/Downloads/Miniconda3-latest-MacOSX-x86_64.sh" && "~/Downloads/Miniconda3-latest-MacOSX-x86_64.sh" -b -u -p ~/miniconda3 && rm "~/Downloads/Miniconda3-latest-MacOSX-x86_64.sh" && ~/miniconda3/bin/conda init "$(basename "${SHELL}")" && source "$HOME/.$(basename "${SHELL}")rc"
```

--------------------------------

### Install SLEAP via Pip

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs SLEAP version 1.4.1 from PyPI using pip. This command should be run after activating the appropriate conda environment.

```bash
pip install sleap[pypi]==1.4.1
```

--------------------------------

### Install Miniforge on Mac (Apple Silicon)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniforge for macOS on Apple Silicon (arm64) using bash. It downloads the installer script, makes it executable, runs it, removes the script, and initializes conda for the current shell.

```bash
curl -fsSL --compressed https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-arm64.sh -o "~/Downloads/Miniforge3-MacOSX-arm64.sh" && chmod +x "~/Downloads/Miniforge3-MacOSX-arm64.sh" && "~/Downloads/Miniforge3-MacOSX-arm64.sh" -b -p ~/miniforge3 && rm "~/Downloads/Miniforge3-MacOSX-arm64.sh" && ~/miniforge3/bin/conda init "$(basename "${SHELL}")" && source "$HOME/.$(basename "${SHELL}")rc"
```

--------------------------------

### Import and Verify SLEAP Installation

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_resumable_training.ipynb

Imports the SLEAP library and calls functions to display version information and system summary. This step confirms the installation was successful and provides diagnostic details about the system's configuration.

```Python
import sleap
sleap.versions()
sleap.system_summary()

```

--------------------------------

### Install SLEAP from Source (Mac OS)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Creates a conda environment named 'sleap' from the 'environment_mac.yml' file for Mac OS installations. This is part of the 'conda from source' installation method.

```bash
conda env create -f environment_mac.yml -n sleap
```

--------------------------------

### Install Miniconda on Mac (Apple Silicon)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs Miniconda for macOS on Apple Silicon (arm64) using bash. It downloads the installer script, makes it executable, runs it, removes the script, and initializes conda for the current shell.

```bash
curl -fsSL --compressed https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh -o "~/Downloads/Miniconda3-latest-MacOSX-arm64.sh" && chmod +x "~/Downloads/Miniconda3-latest-MacOSX-arm64.sh" && "~/Downloads/Miniconda3-latest-MacOSX-arm64.sh" -b -u -p ~/miniconda3 && rm "~/Downloads/Miniconda3-latest-MacOSX-arm64.sh" && ~/miniconda3/bin/conda init "$(basename "${SHELL}")" && source "$HOME/.$(basename "${SHELL}")rc"
```

--------------------------------

### Install Conda Package

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Creates a Conda environment named 'sleap_env' with the sleap package installed. This environment can be used to run the 'sleap-label' application.

```shell
conda create -n sleap_env -c defaults -c sleap sleap python=3.6 -y
```

--------------------------------

### List Conda Environments

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Lists all conda environments currently installed on the system. This is useful for verifying environment names before activation or troubleshooting.

```bash
conda env list
```

--------------------------------

### Download Example Data

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Post_inference_tracking.ipynb

This snippet downloads an example `.mp4` video file and a `.slp` file containing pre-computed predictions using `wget`. These files are used as input for the tracking demonstration.

```bash
!wget -q -O fly_clip.mp4 https://github.com/talmolab/sleap-tutorial-uo/blob/main/data/fly_clip.mp4?raw=true
!wget -q -O predictions.slp https://github.com/talmolab/sleap-tutorial-uo/blob/main/data/predictions.slp?raw=true
```

--------------------------------

### Install SLEAP from Source (Windows/Linux CPU)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Creates a conda environment named 'sleap' from the 'environment_no_cuda.yml' file, suitable for CPU-only execution. This is part of the 'conda from source' installation method.

```bash
conda env create -f environment_no_cuda.yml -n sleap
```

--------------------------------

### Check SLEAP and System Versions

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Post_inference_tracking.ipynb

This snippet imports the SLEAP library and calls `sleap.versions()` and `sleap.system_summary()` to display information about the installed SLEAP version and the system environment. This is useful for debugging and verifying the setup.

```python
import sleap
sleap.versions()
sleap.system_summary()
```

--------------------------------

### Verify SLEAP Installation and Versions

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Checks if SLEAP is installed correctly in non-interactive environments by importing the library and calling the versions() function. It displays SLEAP version along with its key dependencies like TensorFlow, Numpy, and Python.

```python
python -c "import sleap; sleap.versions()"
```

--------------------------------

### Install SLEAP and Download Data

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Model_evaluation.ipynb

Installs the SLEAP library and necessary dependencies, then downloads and unzips sample data for model evaluation.

```shell
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"
!apt -qq install tree
!wget -q https://storage.googleapis.com/sleap-data/reference/flies13/td_fast.210505_012601.centered_instance.n%3D1800.zip
!unzip -qq -o -d "td_fast.210505_012601.centered_instance.n=1800" "td_fast.210505_012601.centered_instance.n=1800.zip"
```

--------------------------------

### Continuous Integration and Coverage Reporting

Source: https://github.com/talmolab/sleap/wiki/Developer-Guide

Details the CI setup using AppVeyor and coverage reporting via Coveralls with coverage.py integration. This ensures code quality and test coverage are maintained.

```yaml
# Example AppVeyor configuration snippet (conceptual)
# build: ...
# test:
#   - commands:
#       - coverage run -m pytest
#       - coveralls
```

```python
# Example coveralls-python configuration (conceptual)
# .coveralls.yml
repo_token: YOUR_REPO_TOKEN
```

--------------------------------

### Run SLEAP Diagnostics

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Executes the SLEAP diagnostic tool to identify potential system issues or configuration problems. This command helps gather information about your environment that can be useful for troubleshooting.

```bash
sleap-diagnostic
```

--------------------------------

### Re-install Development Package

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Re-runs the editable installation command for the sleap package. This is necessary if the directory path changes or submodules are added/modified.

```shell
pip install --no-deps -e .
```

--------------------------------

### Start Training

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_resumable_training.ipynb

Initiates the training process using the configured `Trainer`. This command starts the model training, allowing it to improve over time based on the provided data and configuration.

```Python
trainer.train()

```

--------------------------------

### Create Conda Environment for Pip Install (CPU)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Creates a new conda environment named 'sleap' with Python 3.7 and pip, suitable for installing SLEAP via pip on CPU-only systems.

```bash
conda create --name sleap pip python=3.7.12
```

--------------------------------

### Install SLEAP with Conda

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Commands to create a new conda environment named 'sleap' with SLEAP version 1.4.1. These commands include necessary dependencies from specified channels and are tailored for different operating systems.

```Windows and Linux
conda create -y -n sleap -c conda-forge -c nvidia -c sleap/label/dev -c sleap -c anaconda sleap=1.4.1
```

```Mac OS
conda create -y -n sleap -c conda-forge -c anaconda -c sleap sleap=1.4.1
```

--------------------------------

### SLEAP Training Script Example

Source: https://github.com/talmolab/sleap/blob/develop/docs/guides/custom-training.md

An example of a shell script command that can be used to initiate SLEAP training with custom profiles and dataset packages.

```bash
#!/bin/bash

# Path to the custom training profile
CUSTOM_PROFILE="path/to/custom/profile.json"

# Path to the SLEAP dataset package
DATASET_PACKAGE="path/to/dataset.pkg.slp"

# Execute the sleap-train command
sleap-train "${CUSTOM_PROFILE}" "${DATASET_PACKAGE}"
```

--------------------------------

### Install SLEAP from Source (Windows/Linux NVIDIA GPU)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Creates a conda environment named 'sleap' from the 'environment.yml' file, which includes dependencies for NVIDIA GPU support. This is part of the 'conda from source' installation method.

```bash
conda env create -f environment.yml -n sleap
```

--------------------------------

### Configure Conda Solver to libmamba

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Steps to update conda, install the libmamba solver, and set it as the default. This is recommended for older Anaconda installations to leverage faster dependency resolution.

```bash
conda update -n base conda
conda install -n base conda-libmamba-solver
conda config --set solver libmamba
```

--------------------------------

### Install SLEAP and Dependencies

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Post_inference_tracking.ipynb

This snippet installs the SLEAP library and its dependencies using pip. It also includes commands to uninstall potentially conflicting OpenCV packages. This is a prerequisite for running the tracking examples.

```bash
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"
```

--------------------------------

### Install SLEAP

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Training_and_inference_using_Google_Drive.ipynb

Installs the SLEAP library and its dependencies using pip. It first uninstalls potentially conflicting OpenCV packages to ensure a clean installation.

```python
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"
```

--------------------------------

### Install SLEAP and Dependencies

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_resumable_training.ipynb

Installs the SLEAP library and its dependencies using pip. It also uninstalls conflicting OpenCV packages to ensure a clean installation, which is crucial for setting up the environment correctly.

```Python
# This should take care of all the dependencies on colab:
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"


# But to do it locally, we'd recommend the conda package (available on Windows + Linux):
# conda create -n sleap -c sleap -c conda-forge -c nvidia sleap

```

--------------------------------

### Install SLEAP and Dependencies

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Data_structures.ipynb

Installs SLEAP and necessary dependencies using pip. Includes commands to uninstall conflicting packages and install a specific version of SLEAP with PyPI support.

```python
# This should take care of all the dependencies on colab:
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"
```

--------------------------------

### Install SLEAP via Conda (Windows/Linux GPU)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs SLEAP version 1.4.1 with GPU support using conda. This method requires an NVIDIA GPU and updated drivers. It creates a new conda environment named 'sleap'.

```bash
conda create -y -n sleap -c conda-forge -c nvidia -c sleap/label/dev -c sleap -c anaconda sleap=1.4.1
```

--------------------------------

### List Conda Environments

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

A utility command to display all available conda environments on the system. This helps users identify which environments exist before performing operations like removal or activation.

```bash
conda env list
```

--------------------------------

### SLEAP Training Profile Examples

Source: https://github.com/talmolab/sleap/blob/develop/docs/guides/remote.md

Examples of built-in SLEAP training profile filenames that can be specified when running training. These files define training parameters like learning rate and augmentation methods.

```text
baseline_medium_rf.bottomup.json
```

--------------------------------

### Install SLEAP and Dependencies

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_realtime_inference.ipynb

Installs the SLEAP library and its dependencies using pip. It also includes commands to uninstall conflicting OpenCV packages to ensure proper installation.

```python
# This should take care of all the dependencies on colab:
!pip uninstall -qqq -y opencv-python opencv-contrib-python
!pip install -qqq "sleap[pypi]>=1.3.3"


# But to do it locally, we'd recommend the conda package (available on Windows + Linux):
# conda create -n sleap -c sleap -c conda-forge -c nvidia sleap
```

--------------------------------

### Install Linux xcb Plugin Dependency

Source: https://github.com/talmolab/sleap/wiki/Problems---workarounds

Install the 'libx11-xcb1' package on Ubuntu-based Linux systems using apt-get. This library is often required by Qt's xcb platform plugin when it's not bundled with PySide2.

```Shell
sudo apt-get install libx11-xcb1
```

--------------------------------

### Pre-commit Hook Management

Source: https://github.com/talmolab/sleap/wiki/Developer-Guide

Utilizes pre-commit to automate code quality checks before commits. This includes running formatters like 'black' and linters. A guide for specific integration with black is available.

```bash
pre-commit install
pre-commit run --all-files
```

--------------------------------

### Configure LD_LIBRARY_PATH for SLEAP on Linux

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

This set of commands configures the LD_LIBRARY_PATH environment variable for SLEAP on Linux systems to ensure CUDA libraries are found. It involves creating activation and deactivation scripts within the conda environment for persistent setup.

```bash
mkdir -p $CONDA_PREFIX/etc/conda/activate.d
echo '#!/bin/sh' >> $CONDA_PREFIX/etc/conda/activate.d/sleap_activate.sh
echo 'export SLEAP_OLD_LD_LIBRARY_PATH=$LD_LIBRARY_PATH' >> $CONDA_PREFIX/etc/conda/activate.d/sleap_activate.sh
echo 'export LD_LIBRARY_PATH=$CONDA_PREFIX/lib:$LD_LIBRARY_PATH' >> $CONDA_PREFIX/etc/conda/activate.d/sleap_activate.sh
source $CONDA_PREFIX/etc/conda/activate.d/sleap_activate.sh
```

```bash
mkdir -p $CONDA_PREFIX/etc/conda/deactivate.d
echo '#!/bin/sh' >> $CONDA_PREFIX/etc/conda/deactivate.d/sleap_deactivate.sh
echo 'export LD_LIBRARY_PATH=$SLEAP_OLD_LD_LIBRARY_PATH' >> $CONDA_PREFIX/etc/conda/deactivate.d/sleap_deactivate.sh
```

--------------------------------

### Build Sleap Documentation Locally

Source: https://github.com/talmolab/sleap/wiki/Updating-website

This snippet contains the shell commands required to build the Sleap project's documentation locally. It includes navigating to the correct directory and initiating the Sphinx build process.

```shell
cd sleap/docs
make html
```

--------------------------------

### Launch SLEAP GUI (Command Line)

Source: https://github.com/talmolab/sleap/blob/develop/docs/tutorials/new-project.md

Launches the SLEAP graphical user interface from the command line. This command should be run after activating the SLEAP environment. The GUI allows users to manage projects, load videos, and label data.

```shell
sleap-label
```

--------------------------------

### Initialize SLEAP Trainer

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_resumable_training.ipynb

Creates a SLEAP `Trainer` instance from the specified configuration object. The `Trainer` handles the backend mechanics for setting up and managing the training process.

```Python
trainer = sleap.nn.training.Trainer.from_config(cfg)

```

--------------------------------

### Activate Conda Environment

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Activates the 'sleap' conda environment. This isolates the SLEAP installation and its dependencies from other projects or the base system environment.

```bash
conda activate sleap
```

--------------------------------

### Download Training Data

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_resumable_training.ipynb

Downloads a sample training dataset package (`.pkg.slp`) using `curl` and lists the files in the current directory. This prepares the necessary data for the training process.

```Python
# !curl -L --output labels.pkg.slp https://www.dropbox.com/s/b990gxjt3d3j3jh/210205.sleap_wt_gold.13pt.pkg.slp?dl=1
!curl -L --output labels.pkg.slp https://storage.googleapis.com/sleap-data/datasets/wt_gold.13pt/tracking_split2/train.pkg.slp
!ls -lah

```

--------------------------------

### Install SLEAP via Conda (Mac OS CPU)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Installs SLEAP version 1.4.1 for Mac OS, which operates in CPU mode. This command creates a new conda environment named 'sleap' without GPU-specific dependencies.

```bash
conda create -y -n sleap -c conda-forge -c anaconda -c sleap sleap=1.4.1
```

--------------------------------

### Activate SLEAP Environment

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Activates the SLEAP conda environment. This command is used after installing SLEAP with conda to ensure the correct environment is active for using the software.

```bash
conda activate sleap
```

--------------------------------

### Uninstall SLEAP Environment

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Command to remove an existing SLEAP conda environment. It's recommended to install SLEAP in a fresh environment when updating to avoid dependency conflicts.

```bash
conda env remove -n sleap
```

--------------------------------

### Create Conda Environment for Pip Install (NVIDIA GPU)

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Creates a new conda environment named 'sleap' with Python 3.7, pip, CUDA Toolkit v11.3, and cuDNN v8.2, specifically for installing SLEAP via pip with GPU support.

```bash
conda create --name sleap pip python=3.7.12 cudatoolkit=11.3 cudnn=8.2 -c conda-forge -c nvidia
```

--------------------------------

### View Inference Predictions Directory

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Training_and_inference_on_an_example_dataset.ipynb

Shows the directory structure of the sample dataset, including the location where the inference predictions file (`.predictions.slp`) is saved.

```shell
!tree dataset/drosophila-melanogaster-courtship
```

--------------------------------

### Downgrade TensorFlow for CUDA Compatibility

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Provides a command to install a specific older version of TensorFlow (2.6.3) which is known to be compatible with CUDA Toolkit v11.3 and cuDNN v8.2. This addresses potential detection issues with newer TensorFlow versions.

```bash
pip install tensorflow==2.6.3
```

--------------------------------

### Update Conda Package

Source: https://github.com/talmolab/sleap/wiki/Installation-methods

Updates the sleap package to the latest version within an existing Conda environment.

```shell
conda update -c sleap sleap
```

--------------------------------

### Train Centered-Instance Model

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Training_and_inference_on_an_example_dataset.ipynb

Trains a centered-instance model for pose estimation using the SLEAP command-line interface. It uses a different training profile, the same dataset labels, a distinct run name, and the video path.

```shell
!sleap-train baseline_medium_rf.topdown.json "dataset/drosophila-melanogaster-courtship/courtship_labels.slp" --run_name "courtship.topdown_confmaps" --video-paths "dataset/drosophila-melanogaster-courtship/20190128_113421.mp4"
```

--------------------------------

### Import SLEAP and System Information

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Interactive_and_realtime_inference.ipynb

Imports the SLEAP library, disables GPU memory preallocation for better resource management, and prints system and SLEAP version information.

```python
import sleap
sleap.disable_preallocation()  # This initializes the GPU and prevents TensorFlow from filling the entire GPU memory
sleap.versions()
sleap.system_summary()
```

--------------------------------

### Check SLEAP GPU Detection

Source: https://github.com/talmolab/sleap/blob/develop/docs/installation.md

Verifies if SLEAP can detect available GPUs on the system using a Python command. This is a primary check to confirm GPU enablement within the SLEAP environment.

```bash
python -c "import sleap; sleap.system_summary()"
```

--------------------------------

### Download Trained Models

Source: https://github.com/talmolab/sleap/blob/develop/docs/notebooks/Training_and_inference_on_an_example_dataset.ipynb

Zips the 'models/' directory containing the trained models and then uses Google Colab's file download utility to allow the user to download the archive.

```python
# Zip up the models directory
!zip -r trained_models.zip models/

# Download.
from google.colab import files
files.download("/content/trained_models.zip")
```