diff --git a/layouts/partials/data-science.html b/layouts/partials/data-science.html
index af6bf16569..c3993f55d8 100644
--- a/layouts/partials/data-science.html
+++ b/layouts/partials/data-science.html
@@ -9,38 +9,34 @@
NumPy lies at the core of a rich ecosystem of data science libraries.
-
-
- Data science is the analysis of massive amounts of data
- to gain insight. A typical workflow might be:
+ A typical exploratory data science workflow might look like:
- - Extract, Transform, Load (ETL):
+
- Extract, Transform, Load:
Pandas,
- Beautiful Soup,
- Intake
+ Intake,
+ PyJanitor
- - Explore:
+
- Exploratory analysis:
+ Jupyter,
Seaborn,
- Matplotlib,
+ Matplotlib,
+ Altair
- - Model:
+
- Model and evaluate:
scikit-learn,
- SciPy,
- statsmodels.
+ statsmodels,
+ PyMC3,
+ spaCy
- - Evaluate:
- NumPy,
- TensorFlow
-
-
- -
- Display:
- Data Visualization Tools
+
- Report in a dashboard:
+ Dash,
+ Panel,
+ Voila
@@ -48,34 +44,17 @@
-
- Pandas helps in data discovery and handling,
- Intake helps with
- data access and distribution, while
- Beautiful Soup
- is widely used for web-scraping and gathering data sets.
- Seaborn is well known for
- exploratory data analysis (EDA);
- scikit-learn and
- SciPy (statistical computing) serve some
- of the backbone processes required for machine learning (regression methods,
- classification, clustering, model validation and selection).
- Statistical data exploration, estimation of various statistical models,
- and conducting statistical tests are some of the functions offered by
- statsmodels.
+
+ For high data volumes, Dask and
+ Ray are designed to scale. Stable production
+ environments rely on data versioning (DVC),
+ experiment tracking (MLFlow), and
+ workflow automation (Airflow and
+ Prefect).
-
- Effective data analytics requires deep knowledge of the data domain (e.g.,
- retail, healthcare, marketing, finance, social media, automation, sales, travel,
- etc.) as well as other core disciplines of data science, data engineering, and
- data visualization. Tools such as MLFlow address
- experiment hyperparameter and result tracking needs, while
- DVC provides data version control for data science
- and machine learning workflows.
-
