### Setup Your Own Data Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

A guide on how to set up and use your own custom climate data with the climpred library.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('custom_data')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### List Available Example Datasets

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Displays the names of all example datasets available for loading within the climpred library.

```python
load_dataset()
```

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

### Install herbie-data

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/Herbie.ipynb

Installs the herbie-data library quietly.

```python
!pip install herbie-data --quiet
```

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

### Set up Conda Environment and Install

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Commands to create a new conda environment using a provided yml file, activate it, and then perform an editable install of climpred.

```bash
conda env create -f ci/requirements/climpred-dev.yml
conda activate climpred-dev

pip install -e .
```

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

### Install Dependencies

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Installs necessary Python packages for data analysis and forecast verification using conda and pip.

```conda
#!conda install intake fsspec intake-xarray intake-thredds -c conda-forge -y
```

```conda
#!conda install eccodes cfgrib -c conda-forge -y
```

```pip
#!pip install climetlab --quiet
```

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

### Install climpred-dev Environment

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

This snippet shows how to create a conda development environment for climpred, ensuring all necessary dependencies are installed for running the examples.

```bash
conda env create -f ci/requirements/climpred-dev.yml
```

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

### Import Libraries and Load Dataset

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Imports necessary libraries such as matplotlib, xarray, and climpred, and loads an example dataset.

```python
%matplotlib inline
import matplotlib.pyplot as plt
import xarray as xr

from climpred import HindcastEnsemble
from climpred.tutorial import load_dataset
import climpred
```

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

### Install and Use pre-commit

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Instructions for installing pre-commit and its hooks into the climpred repository. pre-commit automates code beautification, maintainability checks, and syntax error detection.

```bash
pip install --user pre-commit
pre-commit install
```

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

### Install Dependencies

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Installs necessary Python packages for data handling and analysis, including intake, fsspec, intake-xarray, and climetlab.

```python
#!conda install intake fsspec intake-xarray -c conda-forge -y

```

```python
#!pip install climetlab

#!pip install climetlab_s2s_ai_challenge

```

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

### Install climpred from GitHub (clone)

Source: https://github.com/pangeo-data/climpred/blob/main/README.rst

Installs the bleeding-edge version of climpred by cloning the repository from GitHub, navigating into the directory, and installing locally.

```bash
git clone https://github.com/pangeo-data/climpred.git
cd climpred
python -m pip install . --upgrade
```

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

### Load CESM-DP-SST Dataset

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Loads the 'CESM-DP-SST' example dataset, which is a sample dataset for climate prediction analysis.

```python
initialized = climpred.tutorial.load_dataset("CESM-DP-SST")
```

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

### Install climpred from GitHub (local clone)

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/index.rst

Installs the bleeding-edge version of climpred by cloning the repository and installing locally.

```bash
git clone https://github.com/pangeo-data/climpred.git
cd climpred
pip install . --upgrade
```

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

### climpred Documentation: Setup Your Own Raw Model Output

Source: https://github.com/pangeo-data/climpred/blob/main/CHANGELOG.rst

A new demo has been added to guide users on setting up their own raw model output that is compliant with climpred. This includes instructions for using `intake-esm` with climpred for CMORized output.

```python
# Demo: Setup your own raw model output compliant to climpred
# Link: examples/misc/setup_your_own_data.html

# Demo: Using intake-esm with climpred for CMORized output
# Link: examples/misc/setup_your_own_data.html#intake-esm-for-cmorized-output
```

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

### climpred Initialization and Setup

Source: https://github.com/pangeo-data/climpred/blob/main/CHANGELOG.rst

This snippet demonstrates the basic setup for using climpred, including importing necessary libraries and configuring matplotlib for plotting. It shows how to initialize the environment for climpred operations.

```python
import climpred
from climpred import HindcastEnsemble
import matplotlib as mpl

mpl.rcdefaults()
mpl.use("Agg")
# cut border when saving (for maps)
mpl.rcParams["savefig.bbox"] = "tight"
```

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

### Install climpred from GitHub (direct)

Source: https://github.com/pangeo-data/climpred/blob/main/README.rst

Installs the bleeding-edge version of climpred directly from its GitHub repository using pip.

```bash
pip install git+https://github.com/pangeo-data/climpred.git
```

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

### DCPP Dataset Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/initialized-datasets.rst

Demonstrates how to use the DCPP dataset with climpred, referencing intake-esm for data loading and providing links to example notebooks.

```python
with intake-esm <examples/misc/setup_your_own_data.html#intake-esm-for-cmorized-output>
```

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

### Environment Setup for climpred with GPU Support

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/climpred_gpu.ipynb

Installs necessary packages for climpred, including GPU-accelerated libraries like tensorflow-gpu, pytorch-gpu, cudatoolkit, and cupy, along with the cupy-xarray integration. This setup is crucial for leveraging GPU capabilities.

```bash
#!conda create -n ML_gpu tensorflow-gpu pytorch-gpu xarray dask matplotlib nb_conda_kernels jupyterlab cudatoolkit cupy esmtools climpred -y
#!pip install git+https://github.com/jacobtomlinson/cupy-xarray.git
```

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

### NMME Dataset Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/initialized-datasets.rst

Shows how to work with the NMME dataset for seasonal predictions using climpred, with a link to seasonal examples.

```python
`seasonal SubX <examples.html#monthly-and-seasonal>`_
```

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

### Install climpred from GitHub (direct URL)

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/index.rst

Installs the bleeding-edge version of climpred directly from its GitHub repository using pip.

```bash
pip install git+https://github.com/pangeo-data/climpred.git
```

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

### Get Initialized and Observation Coordinates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Retrieves and displays the coordinates for both the initialized and observation datasets within the HindcastEnsemble. This helps verify data alignment.

```python
hindcast.get_initialized().coords
hindcast.get_observations().coords
```

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

### NWP GEFS 6h Forecasts Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Demonstrates how to work with GEFS 6-hour forecasts using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('gefs')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### GEFS Dataset Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/initialized-datasets.rst

Demonstrates the usage of the GEFS dataset for numerical weather prediction with climpred, linking to a specific example for GEFS NWP.

```python
`GEFS NWP <examples/NWP/NWP_GEFS_6h_forecasts.html>`_
```

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

### SubX Dataset Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/initialized-datasets.rst

Provides information on using the SubX dataset for subseasonal predictions with climpred, linking to relevant examples.

```python
`subseasonal SubX <examples.html#subseasonal>`_
```

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

### climpred Examples and Features

Source: https://github.com/pangeo-data/climpred/blob/main/CHANGELOG.rst

This section details various examples and new features added to the climpred library, including NWP, subseasonal forecasts, GPU acceleration, and handling biweekly aggregates.

```python
Add GEFS example <examples/NWP/NWP_GEFS_6h_forecasts.html>_ for numerical weather
prediction. (:issue:`602`, :pr:`603`) `Aaron Spring`_.
Add subseasonal daily ECMWF example <examples/subseasonal/daily-S2S-ECMWF.html>__
using `climetlab <https://github.com/ecmwf-lab/climetlab-s2s-ai-challenge>`_ to access
hindcasts from ECMWF cloud.  (:issue:`587`, :pr:`603`) `Aaron Spring`_.
Add subseasonal daily S2S example <examples/subseasonal/daily-S2S-IRIDL.html>_
accessing `S2S <http://s2sprediction.net/>`_ output on
`IRIDL <https://iridl.ldeo.columbia.edu/SOURCES/.ECMWF/.S2S/>`_ with a cookie and
working with "on-the-fly" reforecasts with ``hdate`` dimension.
(:issue:`588`, :pr:`593`) `Aaron Spring`_.
Added example climpred on GPU <examples/misc/climpred_gpu.ipynb>_. Running
:py:meth:`.PerfectModelEnsemble.verify` on GPU with `cupy-xarray
<https://github.com/xarray-contrib/cupy-xarray>`_ finishes 10x faster.
(:issue:`592`, :pr:`607`) `Aaron Spring`_.
How to work with biweekly aggregates in ``climpred``, see
`daily ECMWF example <examples/subseasonal/daily-S2S-ECMWF.html#biweekly-aggregates>`__.
(:issue:`625`, :pr:`630`) `Aaron Spring`_.
```

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

### Herbie NWP Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

An example showcasing the use of climpred with Herbie, likely for NWP data.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('herbie')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### climpred GPU Usage Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Illustrates how to utilize GPU acceleration with climpred for faster computations.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('gpu')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Perfect Model Predictability Demo

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

An example demonstrating perfect model predictability analysis with climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('perfect_model')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Install climpred with pip

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/index.rst

Installs the latest release of climpred with all optional dependencies using pip.

```bash
pip install climpred[complete]
```

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

### Daily S2S ECMWF Subseasonal Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

An example for processing daily S2S forecasts from ECMWF using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('s2s_ecmwf')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Weekly SubX Subseasonal Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Shows how to analyze weekly subseasonal forecasts from the SubX dataset with climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('subx_weekly')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### CESM-DPLE Dataset Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/initialized-datasets.rst

Highlights the use of the CESM-DPLE dataset with climpred, noting its compatibility with standard climpred examples.

```python
many standard climpred `examples <quick-start.html>`_
```

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

### Seasonal ENSO SubX Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Shows how to analyze seasonal ENSO forecasts from the SubX dataset using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('subx_seasonal')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Daily SubX Subseasonal Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Illustrates how to use climpred for daily subseasonal forecasts from the SubX dataset.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('subx')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Build Documentation Locally

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Commands to update the conda environment and build the documentation locally using make html. The built documentation will be available in the docs/build/ directory.

```bash
$ conda env update -f ci/requirements/climpred-dev.yml
$ cd docs
$ make html
```

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

### Verify Dimension Implications Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

This example demonstrates how to verify dimension implications in climate model outputs using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('verify_dim')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Fork and Clone Repository

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Steps to fork the climpred GitHub repository, clone it locally, and set up the upstream remote. It also includes creating a new branch for bug fixes or features.

```bash
git clone git@github.com:YOUR_GITHUB_USERNAME/climpred.git
cd climpred
git remote add upstream git@github.com:pangeo-data/climpred.git

# now, to fix a bug or add feature create your own branch off "main":

git checkout -b your-bugfix-feature-branch-name main
```

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

### Significance Testing Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

An example showing how to perform significance testing on climate predictions with climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('significance')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Linting Notebooks

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Loads nb_black and lab_black extensions for code linting within a Jupyter Notebook environment.

```python
# linting
%load_ext nb_black
%load_ext lab_black
```

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

### Monthly ENSO SubX Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Demonstrates climpred usage for monthly ENSO forecasts from the SubX dataset.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('subx_monthly')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Install climpred with conda

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/index.rst

Installs the latest release of climpred from the conda-forge channel.

```bash
conda install -c conda-forge climpred
```

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

### Load IPython Extensions

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Loads the nb_black and lab_black extensions for enhanced code formatting in IPython environments.

```python
# linting
%load_ext nb_black
%load_ext lab_black
```

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

### Load IPython Extensions

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Loads IPython extensions for code formatting and styling within a Jupyter Notebook environment.

```python
# linting
%load_ext nb_black
%load_ext lab_black

```

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

### S2S Dataset Examples

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/initialized-datasets.rst

Illustrates how to access and use the S2S dataset for subseasonal predictions via IRIDL and climetlab, with links to specific examples.

```python
`IRIDL <examples/subseasonal/daily-S2S-IRIDL.html>`_
```

```python
`EWC Cloud/climetlab <examples/subseasonal/daily-S2S-ECMWF.html>`_
```

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

### climpred New Example Pages

Source: https://github.com/pangeo-data/climpred/blob/main/CHANGELOG.rst

New example pages have been added to demonstrate subseasonal-to-seasonal (S2S) prediction using climpred. These examples cover calculating MJO index skill, ENSO skill as a function of lead time and initial month, and seasonal ENSO skill.

```python
# Example: Calculate the skill of the MJO index as a function of lead time
# Link: examples/subseasonal/daily-subx-example.html

# Example: Calculate ENSO skill as a function of initial month vs. lead time
# Link: examples/monseas/monthly-enso-subx-example.html
```

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

### Install climpred with pip

Source: https://github.com/pangeo-data/climpred/blob/main/README.rst

Installs the latest release of the climpred package, including optional dependencies for complete functionality, using pip.

```bash
python -m pip install climpred[complete]
```

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

### Daily S2S IRIDL Subseasonal Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Demonstrates working with daily S2S (Subseasonal to Seasonal) forecasts from IRIDL using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('s2s_iridl')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Create Biweekly Observation Aggregates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Generates biweekly aggregates for observation data using a rolling 14-day mean. The time index is adjusted to represent the starting date of each 14-day period.

```python
# 14D rolling mean
obs_biweekly = obs.rolling(time=14, center=False).mean()
obs_biweekly = obs_biweekly.isel(time=slice(13, None)).assign_coords(
    time=obs.time.isel(time=slice(None, -13))
)  # time represents first day of the biweekly aggregate
```

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

### Tropical Pacific SSTs Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

This example focuses on analyzing tropical Pacific Sea Surface Temperatures (SSTs) using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('ssts')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### Load Black Formatting Extensions

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/Herbie.ipynb

Loads nb_black and lab_black extensions for code linting and formatting in notebooks.

```python
# linting
%load_ext nb_black
%load_ext lab_black
```

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

### Diagnose Potential Predictability Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples.rst

Demonstrates how to diagnose potential predictability of climate variables using climpred.

```python
import climpred.tutorial
import xarray as xr

# Load example data
ds = climpred.tutorial.load_dataset('diagnose_predictability')

# Example analysis (replace with actual climpred usage)
print(ds)
```

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

### List Available Datasets

Source: https://github.com/pangeo-data/climpred/wiki/Sample-Data

Retrieves and displays a list of all available example datasets that can be loaded by the climpred library. This function helps users understand the data sources available for analysis.

```python
from climpred.loadutils import get_datasets
get_datasets()
>>> 'MPI-PM-DP-1D': decadal prediction ensemble area averages of SST/SSS/AMO.
>>> 'MPI-PM-DP-3D': decadal prediction ensemble lat/lon/time of SST/SSS/AMO.
>>> 'MPI-control-1D': area averages for the control run of SST/SSS.
>>> 'MPI-control-3D': lat/lon/time for the control run of SST/SSS.
>>> 'CESM-DP-SST': decadal prediction ensemble of global mean SSTs.
>>> 'CESM-DP-SSS': decadal prediction ensemble of global mean SSS.
>>> 'CESM-LE': uninitialized ensemble of global mean SSTs.
>>> 'ERSST': observations of global mean SSTs.
>>> 'FOSI-SST': reconstruction of global mean SSTs.
>>> 'FOSI-SSS': reconstruction of global mean SSS.
```

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

### Create Biweekly Forecast Aggregates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Creates biweekly aggregates for forecast data by calculating the mean over specific lead time ranges (1-14 days, 14-28 days, 28-42 days) and concatenating them. The lead dimension is then updated to represent the start of each biweekly period.

```python
# create 14D averages
forecast_w12 = forecast.sel(lead=range(1, 14)).mean(dim="lead")
forecast_w34 = forecast.sel(lead=range(14, 28)).mean(dim="lead")
forecast_w56 = forecast.sel(lead=range(28, 42)).mean(dim="lead")
forecast_biweekly = xr.concat([forecast_w12, forecast_w34, forecast_w56], dim="lead")
forecast_biweekly["lead"] = [
    1,
    14,
    28,
]  # lead represents first day of biweekly aggregate
forecast_biweekly["lead"].attrs["units"] = "days"
forecast_biweekly.coords
```

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

### Display Initialized Coordinates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Prints the coordinates of the initialized dataset. This is useful for understanding the structure and dimensions of the forecast data.

```python
initialized.coords
```

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

### Git Commit and Push

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Standard commands for committing changes locally and pushing them to the remote repository.

```shell
$ git commit -a -m "<commit message>"
$ git push -u
```

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

### Load Dataset Function

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/api/climpred.tutorial.load_dataset.rst

The `load_dataset` function from the `climpred.tutorial` module is used to load a dataset. It is part of the tutorial section, suggesting its use for examples and learning.

```python
import climpred.tutorial

dataset = climpred.tutorial.load_dataset()
```

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

### Install climpred with conda

Source: https://github.com/pangeo-data/climpred/blob/main/README.rst

Installs the latest release of the climpred package from the conda-forge channel using conda.

```bash
conda install -c conda-forge climpred
```

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

### Prepare Initial Forecast Data

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/Herbie.ipynb

Takes the first forecast from the downloaded dataset and selects the 't2m' variable, preparing it as the initial forecast data.

```python
# take the first with multiple members as forecast
init = ds[0][["t2m"]]
```

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

### Running ASV Benchmarks

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Execute performance benchmarks using the asv tool. This command runs continuous benchmarking against upstream main and the current HEAD, reporting significant changes.

```shell
$ asv continuous -f 1.1 upstream/main HEAD
```

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

### Initialize HindcastEnsemble for Smoothing

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Initializes a `HindcastEnsemble` with daily lead data, adds observations, and computes the ensemble, preparing it for subsequent smoothing operations.

```python
fct = (
    climpred.HindcastEnsemble(forecast_climetlab.drop_vars("valid_time"))  # daily lead
    .add_observations(obs)
    .compute()
)
```

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

### HindcastEnsemble Initialization and Verification

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Demonstrates initializing a HindcastEnsemble with forecast data, adding observations, computing the ensemble, and then verifying it using RMSE with specific comparison and alignment parameters. It also shows plotting the results.

```python
fct = (
    climpred.HindcastEnsemble(
        forecast_climetlab.drop_vars("valid_time").isel(lead_time=range(4))
    )
    .add_observations(obs)
    .compute()
)

metric_kwargs = dict(
    metric="rmse", comparison="e2o", alignment="same_inits", dim="init"
)

skill = fct.verify(**metric_kwargs)
skill[var].plot(col="lead", robust=True)
```

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

### Check cupy Installation

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/climpred_gpu.ipynb

Verifies the installation of the cupy library using the conda package manager. This command is useful for confirming that cupy is available in the current environment.

```bash
!conda list cupy
```

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

### Same Verifications Alignment Example

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/alignment.ipynb

An example of using the 'same_verifs' alignment strategy in climpred's verify function. This strategy ensures that the verification dates remain constant across all leads.

```python
skill = hindcast.verify(
    metric="acc", comparison="e2o", dim="init", alignment="same_verifs"
)
```

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

### Display Coordinates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Displays the coordinates of the hindcast dataset loaded using climetlab.

```python
forecast_climetlab.coords

```

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

### GPU Limitations: Bootstrap Method

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/climpred_gpu.ipynb

Demonstrates the usage of the `bootstrap` method within `PerfectModelEnsemble` on the GPU. This example is provided in the context of discussing potential GPU limitations, although the code itself shows a standard operation.

```python
pm_gpu.bootstrap(metric="pearson_r", comparison="m2m", dim=["init", "member"], iterations=10)[v]
```

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

### Import Core Libraries

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/setup_your_own_data.ipynb

Imports necessary libraries for data manipulation, plotting, and climpred functionality.

```python
%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np
import xarray as xr

import climpred
```

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

### Define Reforecast Dates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Generates a time series of dates for reforecasts, starting from January 2nd, 2020, with a weekly frequency.

```python
dates = xr.cftime_range(start="2020-01-02", freq="7D", end="2020-12-31")

```

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

### Display First Few Rows of Catalog DataFrame

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/setup_your_own_data.ipynb

Shows the beginning of the DataFrame after searching the catalog, providing a preview of the selected data.

```python
cat.df.head()
```

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

### Create and Add Observations to Hindcast Ensemble

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Creates a HindcastEnsemble object from the initialized data and adds the processed observations to it. This prepares the data for comparative analysis.

```python
hindcast = HindcastEnsemble(initialized)
hindcast = hindcast.add_observations(obs)
hindcast
```

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

### Import Libraries

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Imports core libraries for climate data analysis: climpred for prediction verification and xarray for labeled multi-dimensional array manipulation.

```python
import climpred
import xarray as xr

```

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

### Running Specific ASV Benchmark

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Execute a specific benchmark test within the ASV suite. This allows for targeted performance testing of individual components.

```shell
$ asv continuous -f 1.1 upstream/main HEAD -b benchmarks_PredictionEnsemble.GenerateHindcastEnsembleSmall.time_bootstrap
```

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

### Define Forecast Parameters

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Sets up parameters for retrieving GEFS forecasts, including initialization times, dates, ensemble members, and forecast lead times.

```python
inits_time = "0000"  # get forecasts started at 00:00
inits = ["20200829", "20200830", "20200831"]  # four initial dates
members = range(5)  # 5 members out of 20
leads = np.arange(0, 6 * 4 * 2 + 1, 6)  # 6h lead forecasts, 9 leads upto 48h
```

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

### Import Libraries

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/Herbie.ipynb

Imports necessary libraries: xarray for data manipulation, numpy for numerical operations, and climpred for forecast verification.

```python
import xarray as xr
import numpy as np

import climpred  # forecast verification
```

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

### Manual PyPI and TestPyPI Upload

Source: https://github.com/pangeo-data/climpred/blob/main/HOWTORELEASE.rst

Provides the commands to manually build the package and upload it to TestPyPI and PyPI. This is an alternative to the automated GitHub Actions workflow.

```shell
$ git clean -xfd  # remove any files not checked into git
$ python -m build  # build package
$ python -m twine upload --repository-url https://test.pypi.org/legacy dist/*  # register and push to testpypi
$ python -m twine upload dist/*  # register and push to pypi
```

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

### Generate API Documentation

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Command to generate API documentation using sphinx-autogen. This command should be run from the docs/source directory after adding new functions to api.rst.

```bash
sphinx-autogen -o api api.rst
```

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

### Prepare for Forecast Verification

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Prepares the data for forecast verification using the climpred library's `HindcastEnsemble` class. This step involves setting coordinates for lead time.

```python
# climpred.tutorial.load_dataset('s2s-ai-challenge')
# HindcastEnsemble converts `lead_time` to `lead` due to standard_name 'forecast_period'

```

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

### Adjust Lead Dimension

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-subx-example.ipynb

Adjusts the 'lead' dimension in the SubX dataset. SubX uses fractional leads (e.g., 0.5, 1.5), which are converted to integers starting from 0 to represent days since initialization.

```python
fcstds["lead"] = (fcstds["lead"] - 0.5).astype("int")
```

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

### Configure climetlab/cdsapi Logging

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Configures the logging level for climetlab or cdsapi. This example sets the logger to ERROR level, suppressing informational messages.

```python
# climetlab or cds enable logging.INFO
import logging

logger = logging.getLogger()
logger.setLevel(logging.ERROR)
```

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

### reStructuredText Quick Reference

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Links to external resources for learning reStructuredText (rst), the markup language used for climpred's documentation.

```rst
.. _rst-quickref: http://docutils.sourceforge.net/docs/user/rst/quickref.html
.. _rst-cheatsheet: https://github.com/ralsina/rst-cheatsheet/blob/master/rst-cheatsheet.rst
```

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

### Tagging a Release

Source: https://github.com/pangeo-data/climpred/blob/main/HOWTORELEASE.rst

This snippet demonstrates how to tag a new release and push it to the upstream repository. It follows semantic versioning conventions.

```shell
$ git tag -a v1.0.0 -m "Version 1.0.0"
$ git push upstream main --tags
```

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

### Get Initialized Dates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/alignment.ipynb

Prints the initialization dates from the hindcast object.

```python
print(f"initialization dates: \n{hindcast.get_initialized().init.to_index()})")
```

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

### Download and Process Forecast Data

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Iterates through specified initializations, leads, and members to download GEFS forecast data using intake-xarray and cfgrib, applying filters for GRIB file parsing. Includes error handling for failed downloads.

```python
for init in inits:
    for lead in leads:
        for member in members:
            try:
                url = f"https://www.ncei.noaa.gov/thredds/fileServer/model-gefs-003/202008/{init}/gens-a_3_{init}_{inits_time}_{str(lead).zfill(3)}_{str(member).zfill(2)}.grb2"
                # print(f'download init = {init}, lead = {lead}, member = {member}')
                intake_xarray.NetCDFSource(
                    f"simplecache::{url}",
                    xarray_kwargs=dict(
                        engine="cfgrib",
                        backend_kwargs=dict(
                            filter_by_keys={
                                "typeOfLevel": "heightAboveGround",
                                "cfVarName": "t2m",
                            }
                        ),
                    ),
                ).to_dask()
            except Exception as e:
                print("failed", type(e).__name__, e)
```

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

### Changelog Entry Format

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Format for creating new entries in the CHANGELOG.rst file, detailing changes, pull request numbers, and author names.

```rst
<description> (:pr:`#<pull request number>`) `<author's names>`_
```

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

### Open GRIB Files with cfgrib

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Demonstrates how to open a GRIB file using intake-xarray and cfgrib, specifying backend arguments to filter by level type and variable name. This is necessary to correctly parse GRIB data.

```python
# how to open grib files: https://github.com/ecmwf/cfgrib/issues/170
intake_xarray.NetCDFSource(
    "simplecache::https://www.ncei.noaa.gov/thredds/fileServer/model-gefs-003/202008/20200831/gens-a_3_20200831_1800_000_20.grb2",
    xarray_kwargs=dict(
        engine="cfgrib",
        backend_kwargs=dict(
            filter_by_keys={"typeOfLevel": "heightAboveGround", "shortName": "2t"}
        ),
    ),
).to_dask().coords
```

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

### Code Coverage and Reporting

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Measure code coverage during testing and report the results. This helps ensure that new contributions are adequately tested.

```shell
coverage run --source climpred -m py.test
coverage report
coveralls
```

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

### Create HindcastEnsemble and Add Observations

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-subx-example.ipynb

Initializes a climpred HindcastEnsemble object with the forecast data and adds the observed data to it.

```python
hindcast = HindcastEnsemble(fcstds).add_observations(obsds)
hindcast
```

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

### Install climpred Dependencies

Source: https://github.com/pangeo-data/climpred/blob/main/requirements_upstream.txt

This snippet lists the primary Python dependencies for the climpred project, specifying their installation source via git repositories. These packages cover data handling, numerical computation, and scientific visualization.

```python
bias_correction @ git+https://github.com/pankajkarman/bias_correction
bottleneck @ git+https://github.com/pydata/bottleneck
cftime @ git+https://github.com/Unidata/cftime
dask @ git+https://github.com/dask/dask
nc-time-axis @ git+https://github.com/SciTools/nc-time-axis
xarray @ git+https://github.com/pydata/xarray
xclim @ git+https://github.com/Ouranosinc/xclim
xrft @ git+https://github.com/xgcm/xrft
xskillscore @ git+https://github.com/xarray-contrib/xskillscore
```

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

### Load and Prepare Dataset for GPU

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/climpred_gpu.ipynb

Loads the same sample datasets as the CPU example but converts them to use CuPy arrays via `.as_cupy()` for GPU acceleration. This prepares the data for `PerfectModelEnsemble` operations on the GPU.

```python
import cupy_xarray

v = "tos"
ds3d = load_dataset("MPI-PM-DP-3D")[v].as_cupy()
ds3d.lead.attrs["unit"] = "years"
control3d = load_dataset("MPI-control-3D")[v].as_cupy()

pm_gpu = PerfectModelEnsemble(ds3d)
pm_gpu = pm_gpu.add_control(control3d)
```

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

### Download Observations with climetlab

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Downloads observational data from the Copernicus Climate Data Store (CDS) using the climetlab library. It specifies the dataset, product type, time, grid, parameter, and date range for the download. The data is then converted to an xarray DataArray and squeezed.

```python
obs = (
    climetlab.load_source(
        "cds",
        "reanalysis-era5-single-levels",
        product_type="reanalysis",
        time=["00:00", "06:00", "12:00", "18:00"],
        grid=[1.0, 1.0],
        param="2t",
        date=[
            "2020-08-29",
            "2020-08-30",
            "2020-08-31",
            "2020-09-01",
            "2020-09-02",
            "2020-09-03",
        ],
    )
    .to_xarray()
    .squeeze(drop=True)
).drop_vars("valid_time")
```

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

### climpred Documentation Additions

Source: https://github.com/pangeo-data/climpred/blob/main/CHANGELOG.rst

Describes documentation additions for climpred, including an examples notebook for temporal and spatial smoothing, documentation for computing metrics over specified dimensions, an organized API documentation structure, a page detailing HindcastEnsemble and PerfectModelEnsemble objects, and new pages for publications and helpful links.

```python
- Add examples notebook for
  `temporal and spatial smoothing <examples/smoothing.html>`_. (:pr:`244`)
  `Aaron Spring`_
- Add documentation for computing a metric over a
  `specified dimension <comparisons.html#compute-over-dimension>`_.
  (:pr:`244`) `Aaron Spring`_
- Update `API <api.html>`_ to be more organized with individual function/class pages.
  (:pr:`243`) `Riley X. Brady`_.
- Add `page <prediction-ensemble-object.html>`_ describing the
  :py:class:`.HindcastEnsemble` and
  :py:class:`.PerfectModelEnsemble` objects more clearly.
  (:pr:`243`) `Riley X. Brady`_
- Add page for `publications <publications.html>`_ and
  `helpful links <helpful-links.html>`_. (:pr:`270`) `Riley X. Brady`_.
```

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

### Initialize Lead Attribute Units

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/quick-start.ipynb

Sets the units for the 'lead' attribute in the initialized dataset to 'years'. This is a common step in preparing forecast data for analysis.

```python
initialized["lead"].attrs["units"] = "years"
```

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

### Running Tests with Pytest

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/contributing.rst

Execute all tests in the climpred package using pytest. This includes running doctests on modules, excluding specific test directories.

```shell
pytest climpred
pytest --doctest-modules climpred --ignore climpred/tests
```

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

### Combine Forecast and Observation Biweekly Aggregates

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

Initializes a `HindcastEnsemble` with the pre-aggregated biweekly forecast data and adds the biweekly observation data.

```python
fct_biweekly = climpred.HindcastEnsemble(forecast_biweekly).add_observations(
    obs_biweekly
)
```

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

### Load Hindcast Data with Preprocessing

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/setup_your_own_data.ipynb

Loads hindcast data using `load_hindcast`, specifying initializations, members, and a preprocessing function to handle data format.

```python
# lead_offset because yearmean output
%time ds = load_hindcast(inits=range(1961, 1965), members=range(1, 3), preprocess=preprocess_1var, get_path=get_path)
```

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

### Import Libraries and Load Data

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/smoothing.ipynb

Imports necessary libraries (climpred, matplotlib, xarray) and loads sample sea surface temperature data for PerfectModelEnsemble and HindcastEnsemble examples.

```python
%matplotlib inline
from climpred import PerfectModelEnsemble, HindcastEnsemble
from climpred.tutorial import load_dataset
import matplotlib.pylab as plt
import xarray as xr
```

```python
# Sea surface temperature
v = "tos"
initialized3d = load_dataset("MPI-PM-DP-3D")[v]
control3d = load_dataset("MPI-control-3D")[v]
```

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

### Initialize HindcastEnsemble and Add Observations

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/Herbie.ipynb

Initializes a climpred.HindcastEnsemble object with the prepared initial forecast data and adds the prepared observation data to it. The data is expanded to include 'time' and 'step' dimensions.

```python
hindcast = climpred.HindcastEnsemble(
    init.expand_dims(["time", "step"])
).add_observations(obs)

hindcast
```

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

### Load Dataset from NetCDF

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Loads the previously saved GEFS forecast data from a NetCDF file into an xarray Dataset, making it ready for analysis and skill calculation.

```python
init = xr.open_dataset("tmp_GEFS_a.nc")
```

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

### Initialize fsspec Caching

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/NWP/NWP_GEFS_6h_forecasts.ipynb

Configures fsspec to cache downloaded files locally, improving performance for repeated access. It specifies the cache directory and ensures original filenames are preserved.

```python
import intake
import fsspec  # caching downloads

# specify caching location, where to store files to with their original names
fsspec.config.conf["simplecache"] = {
    "cache_storage": "../my_caching_folder",
    "same_names": True,
}
import intake_xarray  # access files hosted via THREDDS
import climetlab  # download ERA5

import xarray as xr
import numpy as np

import climpred  # forecast verification
```

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

### Import Preprocessing Utilities

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/misc/setup_your_own_data.ipynb

Imports specific preprocessing functions from climpred for loading hindcast data and setting integer time axes.

```python
from climpred.preprocessing.shared import load_hindcast, set_integer_time_axis
from climpred.preprocessing.mpi import get_path
```

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

### Verify and Plot Climate Forecasts

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/examples/subseasonal/daily-S2S-ECMWF.ipynb

This snippet demonstrates how to verify climate model forecasts against observations using the 'rps' metric and 'm2o' comparison. It also shows how to select specific lead times and plot the results, with different category edges for observations and forecasts.

```python
fct.smooth(dict(lead=14), how="mean").verify(
    metric="rps",
    comparison="m2o",
    alignment="same_inits",
    dim=["member", "init"],
    category_edges=(obs_edges, model_edges),
).sel(lead=["1.0-14.0", "14.0-27.0", "28.0-41.0"]).t2m.plot(col="lead", robust=True)
```

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

### Load climpred Tutorial Dataset and Set Up Plotting

Source: https://github.com/pangeo-data/climpred/blob/main/docs/source/alignment.ipynb

Loads a sample dataset using climpred's tutorial functions and configures matplotlib for plotting. It also imports necessary libraries for data analysis and handling warnings.

```python
from climpred import HindcastEnsemble
from climpred.tutorial import load_dataset
from esmtools.stats import rm_trend

import matplotlib.pyplot as plt

plt.style.use("fivethirtyeight")
%matplotlib inline

import numpy as np

import warnings
```