From ec78da33819d61c983328e6941fe89d6c63fe174 Mon Sep 17 00:00:00 2001 From: Taylor Mandelbaum Date: Mon, 26 Jan 2026 20:16:20 -0500 Subject: [PATCH 1/6] Update installation instructions in README.md Added installation instructions for 'uv' and extra options for 'extremeweatherbench'. --- README.md | 27 +++++++++++++++++++++------ 1 file changed, 21 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index d90aff16..215f731f 100644 --- a/README.md +++ b/README.md @@ -39,19 +39,34 @@ We welcome your involvement! The success of a benchmark suite rests on communit # Installing EWB -Currently, the easiest way to install EWB is using the ```pip``` command: +Currently, the easiest way to install EWB is using the ```pip``` command or ```uv```: ```shell -$ pip install git+https://github.com/brightbandtech/ExtremeWeatherBench.git +$ pip install extremeweatherbench + +# Add to an existing uv virtual environment +$ uv add extremeweatherbench +``` + +If you'd like to install the most recent updates to EWB: + +```shell +$ pip install git+https://github.com/brightbandtech/ExtremeWeatherBench.git ``` -It is highly recommend to use [uv](https://docs.astral.sh/uv/) if possible: +For extra installation options: ```shell -$ git clone https://github.com/brightbandtech/ExtremeWeatherBench.git -$ cd ExtremeWeatherBench -$ uv sync +# For running the data_prep modules: +$ pip install "extremeweatherbench[data-prep]" + +# To include dask: +$ pip install "extremeweatherbench[multiprocessing]" + +# For both: +$ pip install "extremeweatherbench[complete]" ``` + # How to Run EWB Running EWB on sample data (included) is straightforward. From 3f34298c40c7f8cb5aa5aba762b47518679ca32f Mon Sep 17 00:00:00 2001 From: Taylor Mandelbaum Date: Tue, 27 Jan 2026 15:35:01 -0500 Subject: [PATCH 2/6] Update README to remove pre-release notices Removed pre-release comment --- README.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/README.md b/README.md index 215f731f..e933ceba 100644 --- a/README.md +++ b/README.md @@ -2,10 +2,6 @@ [![Documentation Status](https://readthedocs.org/projects/extremeweatherbench/badge/?version=latest)](https://extremeweatherbench.readthedocs.io/en/latest/?badge=latest) -**EWB is currently in limited pre-release. Bugs are likely to occur for now.** - -**v1.0 to be published alongside EWB preprint.** - [Read our blog post here](https://www.brightband.com/blog/extreme-weather-bench) As AI weather models are growing in popularity, we need a standardized set of community driven tests that evaluate the models across a wide variety of high-impact hazards. Extreme Weather Bench (EWB) builds on the successful work of WeatherBench and introduces a set of high-impact weather events, spanning across multiple spatial and temporal scales and different parts of the weather spectrum. We provide data to use for testing, standard metrics for evaluation by forecasters worldwide for each of the phenomena, as well as impact-based metrics. EWB is a community system and will be adding additional phenomena, test cases and metrics in collaboration with the worldwide weather and forecast verification community. From c861a9bb05d81753a0f2815dde7f6677de7c63dc Mon Sep 17 00:00:00 2001 From: Taylor Mandelbaum Date: Tue, 27 Jan 2026 15:35:55 -0500 Subject: [PATCH 3/6] simplify installation instructions --- README.md | 6 ------ 1 file changed, 6 deletions(-) diff --git a/README.md b/README.md index e933ceba..4929b891 100644 --- a/README.md +++ b/README.md @@ -55,12 +55,6 @@ For extra installation options: ```shell # For running the data_prep modules: $ pip install "extremeweatherbench[data-prep]" - -# To include dask: -$ pip install "extremeweatherbench[multiprocessing]" - -# For both: -$ pip install "extremeweatherbench[complete]" ``` # How to Run EWB From e34a79d5b6b568371190d24461e86552d5f32f07 Mon Sep 17 00:00:00 2001 From: Taylor Mandelbaum Date: Tue, 27 Jan 2026 15:41:22 -0500 Subject: [PATCH 4/6] formatting and update references in README.md --- README.md | 45 ++++++++++++++++++++++++--------------------- 1 file changed, 24 insertions(+), 21 deletions(-) diff --git a/README.md b/README.md index 4929b891..e2cec9c6 100644 --- a/README.md +++ b/README.md @@ -7,7 +7,7 @@ As AI weather models are growing in popularity, we need a standardized set of community driven tests that evaluate the models across a wide variety of high-impact hazards. Extreme Weather Bench (EWB) builds on the successful work of WeatherBench and introduces a set of high-impact weather events, spanning across multiple spatial and temporal scales and different parts of the weather spectrum. We provide data to use for testing, standard metrics for evaluation by forecasters worldwide for each of the phenomena, as well as impact-based metrics. EWB is a community system and will be adding additional phenomena, test cases and metrics in collaboration with the worldwide weather and forecast verification community. # Events -EWB has cases broken down by multiple event types within `src/extremeweatherbench/data/events.yaml` between 2020 and 2024. EWB case studies are documented [here](docs/events/AllCaseStudies.md). +EWB has cases broken down by multiple event types within `src/extremeweatherbench/data/events.yaml` between 2020 and 2024. EWB case studies are documented [here](docs/events/AllCaseStudies.md). ## Available: @@ -22,7 +22,8 @@ EWB has cases broken down by multiple event types within `src/extremeweatherbenc # EWB paper and talks -* AMS 2025 talk: https://ams.confex.com/ams/105ANNUAL/meetingapp.cgi/Paper/451220 +* AMS 2025 talk: [Amy](https://ams.confex.com/ams/105ANNUAL/meetingapp.cgi/Paper/451220) +* AMS 2026 talks: [Amy](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477140), [Taylor](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477141) * EWB paper is in preparation and will be submitted in late 2025 # How do I suggest new data, metrics, or otherwise get involved? @@ -35,12 +36,12 @@ We welcome your involvement! The success of a benchmark suite rests on communit # Installing EWB -Currently, the easiest way to install EWB is using the ```pip``` command or ```uv```: +Currently, the easiest way to install EWB is using ```pip``` or ```uv```: ```shell $ pip install extremeweatherbench -# Add to an existing uv virtual environment +# Or, add to an existing uv virtual environment $ uv add extremeweatherbench ``` @@ -53,58 +54,60 @@ $ pip install git+https://github.com/brightbandtech/ExtremeWeatherBench.git For extra installation options: ```shell -# For running the data_prep modules: +# For running the data prep modules: $ pip install "extremeweatherbench[data-prep]" +$ uv add "extremeweatherbench[data-prep]" ``` # How to Run EWB Running EWB on sample data (included) is straightforward. -## Using command line initialization: - -```shell -$ ewb --default -``` -**Note**: this will run every event type, case, target source, and metric for the individual event type as they become available for GFS initialized FourCastNetv2. It is expected a full evaluation will take some time, even on a large VM. ## Using Jupyter Notebook or a Script: ```python -from extremeweatherbench import cases, inputs, metrics, evaluate, utils +import extremeweatherbench as ewb # Load in a forecast; here, we load in GFS initialized FCNv2 from the CIRA MLWP archive with a default variable built-in for convenience -fcnv2_heatwave_forecast = defaults.cira_fcnv2_heatwave_forecast +fcnv2_heatwave_forecast = ewb.defaults.cira_fcnv2_heatwave_forecast # Load in ERA5 with another default convenience variable -era5_heatwave_target = defaults.era5_heatwave_target +era5_heatwave_target = ewb.defaults.era5_heatwave_target # EvaluationObjects are used to evaluate a single forecast source against a single target source with a defined event type. Event types are declared with each case. One or more metrics can be evaluated with each EvaluationObject. heatwave_evaluation_list = [ - inputs.EvaluationObject( + ewb.inputs.EvaluationObject( event_type="heat_wave", metric_list=[ - metrics.MaximumMeanAbsoluteError(), - metrics.RootMeanSquaredError(), - metrics.MaximumLowestMeanAbsoluteError(), + ewb.metrics.MaximumMeanAbsoluteError(), + ewb.metrics.RootMeanSquaredError(), + ewb.metrics.MaximumLowestMeanAbsoluteError(), ], target=era5_heatwave_target, forecast=fcnv2_heatwave_forecast, ), ] # Load in the EWB default list of event cases -case_metadata = cases.load_ewb_events_yaml_into_case_list() +case_metadata = ewb.cases.load_ewb_events_yaml_into_case_list() # Create the evaluation class, with cases and evaluation objects declared -ewb_instance = evaluate.ExtremeWeatherBench( +ewb_instance = ewb.evaluation( case_metadata=case_metadata, evaluation_objects=heatwave_evaluation_list, ) # Execute a parallel run and return the evaluation results as a pandas DataFrame -heatwave_outputs = ewb_instance.run( +heatwave_outputs = ewb_instance.run_evaluation( parallel_config={'n_jobs':16} # Uses 16 jobs with the loky backend as default ) # Save the results heatwave_outputs.to_csv('heatwave_evaluation_results.csv') ``` + +## Using command line initialization: + +```shell +$ ewb --default +``` +**Note**: this will run every event type, case, target source, and metric for the individual event type as they become available for GFS initialized FourCastNetv2. It is expected a full evaluation will take some time, even on a large VM. From 35e3ee34e3d7c1f8e0ad53024061a088f7ae70a5 Mon Sep 17 00:00:00 2001 From: Taylor Mandelbaum Date: Tue, 27 Jan 2026 15:42:32 -0500 Subject: [PATCH 5/6] Change AMS talk references from names to numbers --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index e2cec9c6..fbd9816d 100644 --- a/README.md +++ b/README.md @@ -22,8 +22,8 @@ EWB has cases broken down by multiple event types within `src/extremeweatherbenc # EWB paper and talks -* AMS 2025 talk: [Amy](https://ams.confex.com/ams/105ANNUAL/meetingapp.cgi/Paper/451220) -* AMS 2026 talks: [Amy](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477140), [Taylor](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477141) +* AMS 2025 talk: [1](https://ams.confex.com/ams/105ANNUAL/meetingapp.cgi/Paper/451220) +* AMS 2026 talks: [1](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477140), [2](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477141) * EWB paper is in preparation and will be submitted in late 2025 # How do I suggest new data, metrics, or otherwise get involved? From 19e98376cce5807cf46ed60d7350ccd9ec48a133 Mon Sep 17 00:00:00 2001 From: Taylor Mandelbaum Date: Tue, 27 Jan 2026 15:43:09 -0500 Subject: [PATCH 6/6] Revise EWB paper submission timeline in README --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index fbd9816d..c39e4803 100644 --- a/README.md +++ b/README.md @@ -24,7 +24,7 @@ EWB has cases broken down by multiple event types within `src/extremeweatherbenc * AMS 2025 talk: [1](https://ams.confex.com/ams/105ANNUAL/meetingapp.cgi/Paper/451220) * AMS 2026 talks: [1](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477140), [2](https://ams.confex.com/ams/106ANNUAL/meetingapp.cgi/Paper/477141) -* EWB paper is in preparation and will be submitted in late 2025 +* EWB paper is in preparation to be released soon # How do I suggest new data, metrics, or otherwise get involved?