diff --git a/.github/workflows/gh-pages.yml b/.github/workflows/gh-pages.yml
index a7c9ae6b4f..e6aab7e52f 100644
--- a/.github/workflows/gh-pages.yml
+++ b/.github/workflows/gh-pages.yml
@@ -25,4 +25,4 @@ jobs:
           branch: main
           cname: numpy.org
           repo: numpy/numpy.github.com
-          hugoVersion: extended_0.104.3
+          hugoVersion: extended_0.115.4
diff --git a/config.yaml.in b/config.yaml.in
index 566fb0c9eb..11246456ab 100644
--- a/config.yaml.in
+++ b/config.yaml.in
@@ -27,21 +27,24 @@ languages:
     title: NumPy
     weight: 1
     contentDir: content/en
-    < content/en/config.yaml >
-    < content/en/tabcontents.yaml >
-  
+    include-files:
+      - content/en/config.yaml
+      - content/en/tabcontents.yaml
+
   # Portuguese
   pt:
     title: NumPy
     weight: 2
     contentDir: content/pt
-    < content/pt/config.yaml >
-    < content/pt/tabcontents.yaml >
+    include-files:
+      - content/pt/config.yaml
+      - content/pt/tabcontents.yaml
 
   # Japanese
   ja:
     title: NumPy
     weight: 3
     contentDir: content/ja
-    < content/ja/config.yaml >
-    < content/ja/tabcontents.yaml >
+    include-files:
+      - content/ja/config.yaml
+      - content/ja/tabcontents.yaml
diff --git a/content/en/config.yaml b/content/en/config.yaml
index d2e6e7a68d..0c0598b972 100644
--- a/content/en/config.yaml
+++ b/content/en/config.yaml
@@ -63,13 +63,13 @@ params:
       text: The core of NumPy is well-optimized C code. Enjoy the flexibility of Python with the speed of compiled code.
     - title: Easy to use
       text: NumPy's high level syntax makes it accessible and productive for programmers from any background or experience level.
-    
+
   tabs:
     title: ECOSYSTEM
-  section5: false
 
+  section5: false
 
-navbar:
+  navbar:
   - title: Install
     url: /install
   - title: Documentation
@@ -84,50 +84,50 @@ navbar:
     url: /news
   - title: Contribute
     url: /contribute
-footer:
-  logo: logo.svg
-  socialmediatitle: ""
-  socialmedia:
-  - link: https://github.com/numpy/numpy
-    icon: github
-  - link: https://www.youtube.com/channel/UCguIL9NZ7ybWK5WQ53qbHng
-    icon: youtube
-  - link: https://twitter.com/numpy_team
-    icon: twitter
-  quicklinks:
-    column1:
-      title: ""
-      links:
-        - text: Install
-          link: /install
-        - text: Documentation
-          link: https://numpy.org/doc/stable
-        - text: Learn
-          link: /learn
-        - text: Citing Numpy
-          link: /citing-numpy
-        - text: Roadmap
-          link: https://numpy.org/neps/roadmap.html
-    column2:
-      links:
-        - text: About us
-          link: /about
-        - text: Community
-          link: /community
-        - text: User surveys
-          link: /user-surveys
-        - text: Contribute
-          link: /contribute
-        - text: Code of conduct
-          link: /code-of-conduct
-    column3:
-      links:
-        - text: Get help
-          link: /gethelp
-        - text: Terms of use
-          link: /terms
-        - text: Privacy
-          link: /privacy
-        - text: Press kit
-          link: /press-kit
 
+  footer:
+    logo: logo.svg
+    socialmediatitle: ""
+    socialmedia:
+    - link: https://github.com/numpy/numpy
+      icon: github
+    - link: https://www.youtube.com/channel/UCguIL9NZ7ybWK5WQ53qbHng
+      icon: youtube
+    - link: https://twitter.com/numpy_team
+      icon: twitter
+    quicklinks:
+      column1:
+        title: ""
+        links:
+          - text: Install
+            link: /install
+          - text: Documentation
+            link: https://numpy.org/doc/stable
+          - text: Learn
+            link: /learn
+          - text: Citing Numpy
+            link: /citing-numpy
+          - text: Roadmap
+            link: https://numpy.org/neps/roadmap.html
+      column2:
+        links:
+          - text: About us
+            link: /about
+          - text: Community
+            link: /community
+          - text: User surveys
+            link: /user-surveys
+          - text: Contribute
+            link: /contribute
+          - text: Code of conduct
+            link: /code-of-conduct
+      column3:
+        links:
+          - text: Get help
+            link: /gethelp
+          - text: Terms of use
+            link: /terms
+          - text: Privacy
+            link: /privacy
+          - text: Press kit
+            link: /press-kit
diff --git a/content/en/tabcontents.yaml b/content/en/tabcontents.yaml
index 2b251ec15f..bfdaf7fa93 100644
--- a/content/en/tabcontents.yaml
+++ b/content/en/tabcontents.yaml
@@ -1,189 +1,190 @@
-machinelearning:
-  paras:
-  - para1: NumPy forms the basis of powerful machine learning libraries like [scikit-learn](https://scikit-learn.org) and [SciPy](https://www.scipy.org). As machine learning grows, so does the list of libraries built on NumPy. [TensorFlow’s](https://www.tensorflow.org) deep learning capabilities have broad applications &mdash; among them speech and image recognition, text-based applications, time-series analysis, and video detection. [PyTorch](https://pytorch.org), another deep learning library, is popular among researchers in computer vision and natural language processing. [MXNet](https://github.com/apache/incubator-mxnet) is another AI package, providing blueprints and templates for deep learning.
-    para2: Statistical techniques called [ensemble](https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205) methods such as binning, bagging, stacking, and boosting are among the ML algorithms implemented by tools such as [XGBoost](https://github.com/dmlc/xgboost), [LightGBM](https://lightgbm.readthedocs.io/en/latest/), and [CatBoost](https://catboost.ai) &mdash; one of the fastest inference engines. [Yellowbrick](https://www.scikit-yb.org/en/latest/) and [Eli5](https://eli5.readthedocs.io/en/latest/) offer machine learning visualizations.
+params:
+  machinelearning:
+    paras:
+    - para1: NumPy forms the basis of powerful machine learning libraries like [scikit-learn](https://scikit-learn.org) and [SciPy](https://www.scipy.org). As machine learning grows, so does the list of libraries built on NumPy. [TensorFlow’s](https://www.tensorflow.org) deep learning capabilities have broad applications &mdash; among them speech and image recognition, text-based applications, time-series analysis, and video detection. [PyTorch](https://pytorch.org), another deep learning library, is popular among researchers in computer vision and natural language processing. [MXNet](https://github.com/apache/incubator-mxnet) is another AI package, providing blueprints and templates for deep learning.
+      para2: Statistical techniques called [ensemble](https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205) methods such as binning, bagging, stacking, and boosting are among the ML algorithms implemented by tools such as [XGBoost](https://github.com/dmlc/xgboost), [LightGBM](https://lightgbm.readthedocs.io/en/latest/), and [CatBoost](https://catboost.ai) &mdash; one of the fastest inference engines. [Yellowbrick](https://www.scikit-yb.org/en/latest/) and [Eli5](https://eli5.readthedocs.io/en/latest/) offer machine learning visualizations.
 
-arraylibraries:
-  intro:
-  - text: NumPy's API is the starting point when libraries are written to exploit innovative hardware, create specialized array types, or add capabilities beyond what NumPy provides.
-  
-  headers:
-  - text: Array Library
-  - text: Capabilities & Application areas
+  arraylibraries:
+    intro:
+    - text: NumPy's API is the starting point when libraries are written to exploit innovative hardware, create specialized array types, or add capabilities beyond what NumPy provides.
 
-  libraries:
-  - title: Dask
-    text: Distributed arrays and advanced parallelism for analytics, enabling performance at scale.
-    img: /images/content_images/arlib/dask.png
-    alttext: Dask
-    url: https://dask.org/
-  - title: CuPy
-    text: NumPy-compatible array library for GPU-accelerated computing with Python.
-    img: /images/content_images/arlib/cupy.png
-    alttext: CuPy
-    url: https://cupy.chainer.org
-  - title: JAX
-    text: "Composable transformations of NumPy programs: differentiate, vectorize, just-in-time compilation to GPU/TPU."
-    img: /images/content_images/arlib/jax_logo_250px.png
-    alttext: JAX
-    url: https://github.com/google/jax
-  - title: Xarray
-    text: Labeled, indexed multi-dimensional arrays for advanced analytics and visualization
-    img: /images/content_images/arlib/xarray.png
-    alttext: xarray
-    url: https://xarray.pydata.org/en/stable/index.html
-  - title: Sparse
-    text: NumPy-compatible sparse array library that integrates with Dask and SciPy's sparse linear algebra.
-    img: /images/content_images/arlib/sparse.png
-    alttext: sparse
-    url: https://sparse.pydata.org/en/latest/
-  - title: PyTorch
-    text: Deep learning framework that accelerates the path from research prototyping to production deployment.
-    img: /images/content_images/arlib/pytorch-logo-dark.svg
-    alttext: PyTorch
-    url: https://pytorch.org/
-  - title: TensorFlow
-    text: An end-to-end platform for machine learning to easily build and deploy ML powered applications.
-    img: /images/content_images/arlib/tensorflow-logo.svg
-    alttext: TensorFlow
-    url: https://www.tensorflow.org
-  - title: MXNet
-    text: Deep learning framework suited for flexible research prototyping and production.
-    img: /images/content_images/arlib/mxnet_logo.png
-    alttext: MXNet
-    url: https://mxnet.apache.org/
-  - title: Arrow
-    text: A cross-language development platform for columnar in-memory data and analytics.
-    img: /images/content_images/arlib/arrow.png
-    alttext: arrow
-    url: https://github.com/apache/arrow
-  - title: xtensor
-    text: Multi-dimensional arrays with broadcasting and lazy computing for numerical analysis.
-    img: /images/content_images/arlib/xtensor.png
-    alttext: xtensor
-    url: https://github.com/xtensor-stack/xtensor-python
-  - title: Awkward Array
-    text: Manipulate JSON-like data with NumPy-like idioms.
-    img: /images/content_images/arlib/awkward.svg
-    alttext: awkward
-    url: https://awkward-array.org/
-  - title: uarray
-    text: Python backend system that decouples API from implementation; unumpy provides a NumPy API.
-    img: /images/content_images/arlib/uarray.png
-    alttext: uarray
-    url: https://uarray.org/en/latest/
-  - title: tensorly
-    text: Tensor learning, algebra and backends to seamlessly use NumPy, MXNet, PyTorch, TensorFlow or CuPy.
-    img: /images/content_images/arlib/tensorly.png
-    alttext: tensorly
-    url: http://tensorly.org/stable/home.html
-  
-scientificdomains:
-  intro:
-  - text: Nearly every scientist working in Python draws on the power of NumPy.
-  - text: "NumPy brings the computational power of languages like C and Fortran to Python, a language much easier to learn and use. With this power comes simplicity: a solution in NumPy is often clear and elegant."
+    headers:
+    - text: Array Library
+    - text: Capabilities & Application areas
 
-  librariesrow1:
-  - title: Quantum Computing
-    alttext: A computer chip.
-    img: /images/content_images/sc_dom_img/quantum_computing.svg
-  - title: Statistical Computing
-    alttext: A line graph with the line moving up.
-    img: /images/content_images/sc_dom_img/statistical_computing.svg
-  - title: Signal Processing
-    alttext: A bar chart with positive and negative values.
-    img: /images/content_images/sc_dom_img/signal_processing.svg
-  - title: Image Processing
-    alttext: An photograph of the mountains.
-    img: /images/content_images/sc_dom_img/image_processing.svg
-  - title: Graphs and Networks 
-    alttext: A simple graph.
-    img: /images/content_images/sc_dom_img/sd6.svg
-  - title: Astronomy Processes
-    alttext: A telescope.
-    img: /images/content_images/sc_dom_img/astronomy_processes.svg
-  - title: Cognitive Psychology
-    alttext: A human head with gears.
-    img: /images/content_images/sc_dom_img/cognitive_psychology.svg
+    libraries:
+    - title: Dask
+      text: Distributed arrays and advanced parallelism for analytics, enabling performance at scale.
+      img: /images/content_images/arlib/dask.png
+      alttext: Dask
+      url: https://dask.org/
+    - title: CuPy
+      text: NumPy-compatible array library for GPU-accelerated computing with Python.
+      img: /images/content_images/arlib/cupy.png
+      alttext: CuPy
+      url: https://cupy.chainer.org
+    - title: JAX
+      text: "Composable transformations of NumPy programs: differentiate, vectorize, just-in-time compilation to GPU/TPU."
+      img: /images/content_images/arlib/jax_logo_250px.png
+      alttext: JAX
+      url: https://github.com/google/jax
+    - title: Xarray
+      text: Labeled, indexed multi-dimensional arrays for advanced analytics and visualization
+      img: /images/content_images/arlib/xarray.png
+      alttext: xarray
+      url: https://xarray.pydata.org/en/stable/index.html
+    - title: Sparse
+      text: NumPy-compatible sparse array library that integrates with Dask and SciPy's sparse linear algebra.
+      img: /images/content_images/arlib/sparse.png
+      alttext: sparse
+      url: https://sparse.pydata.org/en/latest/
+    - title: PyTorch
+      text: Deep learning framework that accelerates the path from research prototyping to production deployment.
+      img: /images/content_images/arlib/pytorch-logo-dark.svg
+      alttext: PyTorch
+      url: https://pytorch.org/
+    - title: TensorFlow
+      text: An end-to-end platform for machine learning to easily build and deploy ML powered applications.
+      img: /images/content_images/arlib/tensorflow-logo.svg
+      alttext: TensorFlow
+      url: https://www.tensorflow.org
+    - title: MXNet
+      text: Deep learning framework suited for flexible research prototyping and production.
+      img: /images/content_images/arlib/mxnet_logo.png
+      alttext: MXNet
+      url: https://mxnet.apache.org/
+    - title: Arrow
+      text: A cross-language development platform for columnar in-memory data and analytics.
+      img: /images/content_images/arlib/arrow.png
+      alttext: arrow
+      url: https://github.com/apache/arrow
+    - title: xtensor
+      text: Multi-dimensional arrays with broadcasting and lazy computing for numerical analysis.
+      img: /images/content_images/arlib/xtensor.png
+      alttext: xtensor
+      url: https://github.com/xtensor-stack/xtensor-python
+    - title: Awkward Array
+      text: Manipulate JSON-like data with NumPy-like idioms.
+      img: /images/content_images/arlib/awkward.svg
+      alttext: awkward
+      url: https://awkward-array.org/
+    - title: uarray
+      text: Python backend system that decouples API from implementation; unumpy provides a NumPy API.
+      img: /images/content_images/arlib/uarray.png
+      alttext: uarray
+      url: https://uarray.org/en/latest/
+    - title: tensorly
+      text: Tensor learning, algebra and backends to seamlessly use NumPy, MXNet, PyTorch, TensorFlow or CuPy.
+      img: /images/content_images/arlib/tensorly.png
+      alttext: tensorly
+      url: http://tensorly.org/stable/home.html
 
-  librariesrow2:
-  - title: Bioinformatics
-    alttext: A strand of DNA.
-    img: /images/content_images/sc_dom_img/bioinformatics.svg
-  - title: Bayesian Inference
-    alttext: A graph with a bell-shaped curve.
-    img: /images/content_images/sc_dom_img/bayesian_inference.svg
-  - title: Mathematical Analysis
-    alttext: Four mathematical symbols.
-    img: /images/content_images/sc_dom_img/mathematical_analysis.svg
-  - title: Chemistry
-    alttext: A test tube.
-    img: /images/content_images/sc_dom_img/chemistry.svg
-  - title: Geoscience
-    alttext: The Earth.
-    img: /images/content_images/sc_dom_img/geoscience.svg
-  - title: Geographic Processing
-    alttext: A map.
-    img: /images/content_images/sc_dom_img/GIS.svg
-  - title: Architecture & Engineering
-    alttext: A microprocessor development board.
-    img: /images/content_images/sc_dom_img/robotics.svg
-  
-datascience:
+  scientificdomains:
+    intro:
+    - text: Nearly every scientist working in Python draws on the power of NumPy.
+    - text: "NumPy brings the computational power of languages like C and Fortran to Python, a language much easier to learn and use. With this power comes simplicity: a solution in NumPy is often clear and elegant."
 
-  intro: "NumPy lies at the core of a rich ecosystem of data science libraries. A typical exploratory data science workflow might look like:"
+    librariesrow1:
+    - title: Quantum Computing
+      alttext: A computer chip.
+      img: /images/content_images/sc_dom_img/quantum_computing.svg
+    - title: Statistical Computing
+      alttext: A line graph with the line moving up.
+      img: /images/content_images/sc_dom_img/statistical_computing.svg
+    - title: Signal Processing
+      alttext: A bar chart with positive and negative values.
+      img: /images/content_images/sc_dom_img/signal_processing.svg
+    - title: Image Processing
+      alttext: An photograph of the mountains.
+      img: /images/content_images/sc_dom_img/image_processing.svg
+    - title: Graphs and Networks
+      alttext: A simple graph.
+      img: /images/content_images/sc_dom_img/sd6.svg
+    - title: Astronomy Processes
+      alttext: A telescope.
+      img: /images/content_images/sc_dom_img/astronomy_processes.svg
+    - title: Cognitive Psychology
+      alttext: A human head with gears.
+      img: /images/content_images/sc_dom_img/cognitive_psychology.svg
 
-  image1:
-  - img: /images/content_images/ds-landscape.png
-    alttext: Diagram of Python Libraries. The five catagories are 'Extract, Transform, Load', 'Data Exploration', 'Data Modeling', 'Data Evaluation' and 'Data Presentation'.
-  
-  image2:
-  - img: /images/content_images/data-science.png
-    alttext: Diagram of three overlapping circles. The circles are labeled 'Mathematics', 'Computer Science' and 'Domain Expertise'. In the middle of the diagram, which has the three circles overlapping it, is an area labeled 'Data Science'.
+    librariesrow2:
+    - title: Bioinformatics
+      alttext: A strand of DNA.
+      img: /images/content_images/sc_dom_img/bioinformatics.svg
+    - title: Bayesian Inference
+      alttext: A graph with a bell-shaped curve.
+      img: /images/content_images/sc_dom_img/bayesian_inference.svg
+    - title: Mathematical Analysis
+      alttext: Four mathematical symbols.
+      img: /images/content_images/sc_dom_img/mathematical_analysis.svg
+    - title: Chemistry
+      alttext: A test tube.
+      img: /images/content_images/sc_dom_img/chemistry.svg
+    - title: Geoscience
+      alttext: The Earth.
+      img: /images/content_images/sc_dom_img/geoscience.svg
+    - title: Geographic Processing
+      alttext: A map.
+      img: /images/content_images/sc_dom_img/GIS.svg
+    - title: Architecture & Engineering
+      alttext: A microprocessor development board.
+      img: /images/content_images/sc_dom_img/robotics.svg
 
-  examples:
-  - text: "<b>Extract, Transform, Load: </b>[Pandas](https://pandas.pydata.org), [Intake](https://intake.readthedocs.io), [PyJanitor](https://pyjanitor.readthedocs.io/)"
-  - text: "<b>Exploratory analysis: </b>[Jupyter](https://jupyter.org), [Seaborn](https://seaborn.pydata.org), [Matplotlib](https://matplotlib.org), [Altair](https://altair-viz.github.io)"
-  - text: "<b>Model and evaluate: </b>[scikit-learn](https://scikit-learn.org), [statsmodels](https://www.statsmodels.org/stable/index.html), [PyMC3](https://docs.pymc.io), [spaCy](https://spacy.io)"
-  - text: "<b>Report in a dashboard: </b>[Dash](https://plotly.com/dash), [Panel](https://panel.holoviz.org), [Voila](https://github.com/voila-dashboards/voila)"
+  datascience:
 
-  content:
-  - text: For high data volumes, [Dask](https://dask.org) and [Ray](https://ray.io/) are designed to scale. Stable deployments rely on data versioning ([DVC](https://dvc.org)), experiment tracking ([MLFlow](https://mlflow.org)), and workflow automation ([Airflow](https://airflow.apache.org), [Dagster](https://dagster.io) and [Prefect](https://www.prefect.io)).
+    intro: "NumPy lies at the core of a rich ecosystem of data science libraries. A typical exploratory data science workflow might look like:"
 
-visualization:
-  images:
-  - url: https://www.fusioncharts.com/blog/best-python-data-visualization-libraries
-    img: /images/content_images/v_matplotlib.png
-    alttext: A streamplot made in matplotlib
-  - url: https://github.com/yhat/ggpy
-    img: /images/content_images/v_ggpy.png
-    alttext: A scatter-plot graph made in ggpy
-  - url: https://www.journaldev.com/19692/python-plotly-tutorial
-    img: /images/content_images/v_plotly.png
-    alttext:  A box-plot made in plotly
-  - url: https://altair-viz.github.io/gallery/streamgraph.html
-    img: /images/content_images/v_altair.png
-    alttext:  A streamgraph made in altair
-  - url: https://seaborn.pydata.org
-    img: /images/content_images/v_seaborn.png
-    alttext:  A pairplot of two types of graph, a plot-graph and a frequency graph made in seaborn"
-  - url: https://docs.pyvista.org/examples/index.html
-    img: /images/content_images/v_pyvista.png
-    alttext:  A 3D volume rendering made in PyVista.
-  - url: https://napari.org
-    img: /images/content_images/v_napari.png
-    alttext:  A multi-dimensionan image made in napari.
-  - url: https://vispy.org/gallery/index.html
-    img: /images/content_images/v_vispy.png
-    alttext:  A Voronoi diagram made in vispy.
-  
-  content:
-  - text: NumPy is an essential component in the burgeoning 
-          [Python visualization landscape](https://pyviz.org/overviews/index.html), 
-          which includes [Matplotlib](https://matplotlib.org), 
-          [Seaborn](https://seaborn.pydata.org), [Plotly](https://plot.ly), 
-          [Altair](https://altair-viz.github.io), [Bokeh](https://docs.bokeh.org/en/latest/), 
-          [Holoviz](https://holoviz.org), [Vispy](http://vispy.org), [Napari](https://github.com/napari/napari), 
-          and [PyVista](https://github.com/pyvista/pyvista), to name a few.
-  - text: NumPy's accelerated processing of large arrays allows researchers to visualize
-          datasets far larger than native Python could handle.
+    image1:
+    - img: /images/content_images/ds-landscape.png
+      alttext: Diagram of Python Libraries. The five catagories are 'Extract, Transform, Load', 'Data Exploration', 'Data Modeling', 'Data Evaluation' and 'Data Presentation'.
+
+    image2:
+    - img: /images/content_images/data-science.png
+      alttext: Diagram of three overlapping circles. The circles are labeled 'Mathematics', 'Computer Science' and 'Domain Expertise'. In the middle of the diagram, which has the three circles overlapping it, is an area labeled 'Data Science'.
+
+    examples:
+    - text: "<b>Extract, Transform, Load: </b>[Pandas](https://pandas.pydata.org), [Intake](https://intake.readthedocs.io), [PyJanitor](https://pyjanitor.readthedocs.io/)"
+    - text: "<b>Exploratory analysis: </b>[Jupyter](https://jupyter.org), [Seaborn](https://seaborn.pydata.org), [Matplotlib](https://matplotlib.org), [Altair](https://altair-viz.github.io)"
+    - text: "<b>Model and evaluate: </b>[scikit-learn](https://scikit-learn.org), [statsmodels](https://www.statsmodels.org/stable/index.html), [PyMC3](https://docs.pymc.io), [spaCy](https://spacy.io)"
+    - text: "<b>Report in a dashboard: </b>[Dash](https://plotly.com/dash), [Panel](https://panel.holoviz.org), [Voila](https://github.com/voila-dashboards/voila)"
+
+    content:
+    - text: For high data volumes, [Dask](https://dask.org) and [Ray](https://ray.io/) are designed to scale. Stable deployments rely on data versioning ([DVC](https://dvc.org)), experiment tracking ([MLFlow](https://mlflow.org)), and workflow automation ([Airflow](https://airflow.apache.org), [Dagster](https://dagster.io) and [Prefect](https://www.prefect.io)).
+
+  visualization:
+    images:
+    - url: https://www.fusioncharts.com/blog/best-python-data-visualization-libraries
+      img: /images/content_images/v_matplotlib.png
+      alttext: A streamplot made in matplotlib
+    - url: https://github.com/yhat/ggpy
+      img: /images/content_images/v_ggpy.png
+      alttext: A scatter-plot graph made in ggpy
+    - url: https://www.journaldev.com/19692/python-plotly-tutorial
+      img: /images/content_images/v_plotly.png
+      alttext:  A box-plot made in plotly
+    - url: https://altair-viz.github.io/gallery/streamgraph.html
+      img: /images/content_images/v_altair.png
+      alttext:  A streamgraph made in altair
+    - url: https://seaborn.pydata.org
+      img: /images/content_images/v_seaborn.png
+      alttext:  A pairplot of two types of graph, a plot-graph and a frequency graph made in seaborn"
+    - url: https://docs.pyvista.org/examples/index.html
+      img: /images/content_images/v_pyvista.png
+      alttext:  A 3D volume rendering made in PyVista.
+    - url: https://napari.org
+      img: /images/content_images/v_napari.png
+      alttext:  A multi-dimensionan image made in napari.
+    - url: https://vispy.org/gallery/index.html
+      img: /images/content_images/v_vispy.png
+      alttext:  A Voronoi diagram made in vispy.
+
+    content:
+    - text: NumPy is an essential component in the burgeoning
+            [Python visualization landscape](https://pyviz.org/overviews/index.html),
+            which includes [Matplotlib](https://matplotlib.org),
+            [Seaborn](https://seaborn.pydata.org), [Plotly](https://plot.ly),
+            [Altair](https://altair-viz.github.io), [Bokeh](https://docs.bokeh.org/en/latest/),
+            [Holoviz](https://holoviz.org), [Vispy](http://vispy.org), [Napari](https://github.com/napari/napari),
+            and [PyVista](https://github.com/pyvista/pyvista), to name a few.
+    - text: NumPy's accelerated processing of large arrays allows researchers to visualize
+            datasets far larger than native Python could handle.
diff --git a/content/ja/config.yaml b/content/ja/config.yaml
index 1007d30234..d76679188b 100644
--- a/content/ja/config.yaml
+++ b/content/ja/config.yaml
@@ -72,89 +72,89 @@ params:
   tabs:
     title: NumPyのエコシステム
   section5: false
-navbar:
-  - 
-    title: インストール
-    url: /ja/install
-  - 
-    title: ドキュメント
-    url: https://numpy.org/doc/stable
-  - 
-    title: 学び方
-    url: /ja/learn
-  - 
-    title: コミュニティ
-    url: /ja/community
-  - 
-    title: 私達について
-    url: /ja/about
-  - 
-    title: ニュース
-    url: /ja/news
-  - 
-    title: NumPyに貢献する
-    url: /ja/contribute
-footer:
-  logo: logo.svg
-  socialmediatitle: ""
-  socialmedia:
+  navbar:
     - 
-      link: https://github.com/numpy/numpy
-      icon: github
+      title: インストール
+      url: /ja/install
     - 
-      link: https://www.youtube.com/channel/UCguIL9NZ7ybWK5WQ53qbHng
-      icon: YouTube
+      title: ドキュメント
+      url: https://numpy.org/doc/stable
     - 
-      link: https://twitter.com/numpy_team
-      icon: twitter
-  quicklinks:
-    column1:
-      title: ""
-      links:
-        - 
-          text: インストール
-          link: /ja/install
-        - 
-          text: ドキュメント
-          link: https://numpy.org/doc/stable
-        - 
-          text: 学び方
-          link: /ja/learn
-        - 
-          text: 引用する
-          link: /ja/citing-numpy
-        - 
-          text: ロードマップ
-          link: https://numpy.org/neps/roadmap.html
-    column2:
-      links:
-        - 
-          text: 私達について
-          link: /ja/about
-        - 
-          text: コミュニティ
-          link: /ja/community
-        - 
-          text: ユーザーの調査
-          link: /ja/user-surveys
-        - 
-          text: NumPyに貢献する
-          link: /ja/contribute
-        - 
-          text: 行動規範
-          link: /ja/code-of-conduct
-    column3:
-      links:
-        - 
-          text: サポートを得る方法
-          link: /ja/gethelp
-        - 
-          text: 利用規約
-          link: /ja/terms
-        - 
-          text: プライバシーポリシー
-          link: /ja/privacy
-        - 
-          text: プレス用資料
-          link: /ja/press-kit
+      title: 学び方
+      url: /ja/learn
+    - 
+      title: コミュニティ
+      url: /ja/community
+    - 
+      title: 私達について
+      url: /ja/about
+    - 
+      title: ニュース
+      url: /ja/news
+    - 
+      title: NumPyに貢献する
+      url: /ja/contribute
+  footer:
+    logo: logo.svg
+    socialmediatitle: ""
+    socialmedia:
+      - 
+        link: https://github.com/numpy/numpy
+        icon: github
+      - 
+        link: https://www.youtube.com/channel/UCguIL9NZ7ybWK5WQ53qbHng
+        icon: YouTube
+      - 
+        link: https://twitter.com/numpy_team
+        icon: twitter
+    quicklinks:
+      column1:
+        title: ""
+        links:
+          - 
+            text: インストール
+            link: /ja/install
+          - 
+            text: ドキュメント
+            link: https://numpy.org/doc/stable
+          - 
+            text: 学び方
+            link: /ja/learn
+          - 
+            text: 引用する
+            link: /ja/citing-numpy
+          - 
+            text: ロードマップ
+            link: https://numpy.org/neps/roadmap.html
+      column2:
+        links:
+          - 
+            text: 私達について
+            link: /ja/about
+          - 
+            text: コミュニティ
+            link: /ja/community
+          - 
+            text: ユーザーの調査
+            link: /ja/user-surveys
+          - 
+            text: NumPyに貢献する
+            link: /ja/contribute
+          - 
+            text: 行動規範
+            link: /ja/code-of-conduct
+      column3:
+        links:
+          - 
+            text: サポートを得る方法
+            link: /ja/gethelp
+          - 
+            text: 利用規約
+            link: /ja/terms
+          - 
+            text: プライバシーポリシー
+            link: /ja/privacy
+          - 
+            text: プレス用資料
+            link: /ja/press-kit
 
diff --git a/content/ja/tabcontents.yaml b/content/ja/tabcontents.yaml
index 4c1b555bdd..c633eb5276 100644
--- a/content/ja/tabcontents.yaml
+++ b/content/ja/tabcontents.yaml
@@ -1,218 +1,219 @@
-machinelearning:
-  paras:
-    - 
-      para1: NumPyは、[scikit-learn](https://scikit-learn.org)や[SciPy](https://www.scipy.org)のような強力な機械学習ライブラリの基礎を形成しています。機械学習の技術分野が成長するにつれ、NumPyをベースにしたライブラリの数も増えています。[TensorFlow](https://www.tensorflow.org)の深層学習機能は、音声認識や画像認識、テキストベースのアプリケーション、時系列分析、動画検出など、幅広い応用用途があります。[PyTorch](https://pytorch.org)も、コンピュータビジョンや自然言語処理の研究者に人気のある深層学習ライブラリです。[MXNet](https://github.com/apache/incubator-mxnet)もAIパッケージの一つで、深層学習の設計図やテンプレート機能を提供しています。
-      para2: '[ensemble](https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205)法と呼ばれる統計的手法であるビンニング、バギング、スタッキングや、[XGBoost](https://github.com/dmlc/xgboost)、[LightGBM](https://lightgbm.readthedocs.io/en/latest/)、[CatBoost](https://catboost.ai)などのツールで実装されているブースティングなどは、機械学習アルゴリズムの一つであり、最速の推論エンジンの一つです。[Yellowbrick](https://www.scikit-yb.org/en/latest/)や[Eli5](https://eli5.readthedocs.io/en/latest/)は機械学習の可視化機能を提供しています。'
-arraylibraries:
-  intro:
-    - 
-      text: NumPyのAPIは、革新的なハードウェアを利用したり、特殊な配列タイプを作成したり、NumPyが提供する以上の機能を追加するためにライブラリを作成する際の基礎となります。
-  headers:
-    - 
-      text: 配列ライブラリ
-    - 
-      text: 機能と応用分野
-  libraries:
-    - 
-      title: Dask
-      text: 分析用の分散配列と高度な並列処理により、大規模な処理を可能にします。
-      img: /images/content_images/arlib/dask.png
-      alttext: Dask
-      url: https://dask.org/
-    - 
-      title: CuPy
-      text: Python を使用した GPUによる高速計算用のNumPy互換配列ライブラリ
-      img: /images/content_images/arlib/cupy.png
-      alttext: CuPy
-      url: https://cupy.chainer.org
-    - 
-      title: JAX
-      text: "NumPyコードの合成可能な変換ライブラリ:  微分、ベクトル化、GPU/TPUへのジャストインタイムコンパイル"
-      img: /images/content_images/arlib/jax_logo_250px.png
-      alttext: JAX
-      url: https://github.com/google/jax
-    - 
-      title: Xarray
-      text: 高度な分析と視覚化のためのラベルとインデックス付き多次元配列
-      img: /images/content_images/arlib/xarray.png
-      alttext: xarray
-      url: https://xarray.pydata.org/en/stable/index.html
-    - 
-      title: Sparse
-      text: Dask と SciPy の疎行列の線形代数ライブラリを統合した、Numpy 互換の疎行列ライブラリ
-      img: /images/content_images/arlib/sparse.png
-      alttext: sparse
-      url: https://sparse.pydata.org/en/latest/
-    - 
-      title: PyTorch
-      text: 研究用のプロトタイピングから本番運用への展開を加速させる、深層学習フレームワーク
-      img: /images/content_images/arlib/pytorch-logo-dark.svg
-      alttext: PyTorch
-      url: https://pytorch.org/
-    - 
-      title: TensorFlow
-      text: 機械学習を利用したアプリケーションを簡単に構築・展開するための、エンド・ツー・エンドの機械学習プラットフォーム
-      img: /images/content_images/arlib/tensorflow-logo.svg
-      alttext: TensorFlow
-      url: https://www.tensorflow.org
-    - 
-      title: MXNet
-      text: 柔軟や研究用のプロトタイピングから、実際の運用まで利用可能な深層学習フレームワーク
-      img: /images/content_images/arlib/mxnet_logo.png
-      alttext: MXNet
-      url: https://mxnet.apache.org/
-    - 
-      title: Arrow
-      text: 列型のインメモリーデータやその分析のための、複数の言語に対応した開発プラットフォーム
-      img: /images/content_images/arlib/arrow.png
-      alttext: arrow
-      url: https://github.com/apache/arrow
-    - 
-      title: xtensor
-      text: 数値解析のためのブロードキャスティングと遅延計算を備えた多次元配列
-      img: /images/content_images/arlib/xtensor.png
-      alttext: xtensor
-      url: https://github.com/xtensor-stack/xtensor-python
-    - 
-      title: XND
-      text: Numpy のような イディオムを使って JSON のようなデータを操作するライブラリ
-      img: /images/content_images/arlib/xnd.png
-      alttext: xnd
-      url: https://xnd.io
-    - 
-      title: uarray
-      text: APIを実装から切り離すPythonバックエンドシステム (unumpyはNumPy APIを提供しています)
-      img: /images/content_images/arlib/uarray.png
-      alttext: uarray
-      url: https://uarray.org/en/latest/
-    - 
-      title: tensorly
-      text: Numpy、MXNet、PyTorch、TensorFlowまたはCupyをシームレスに使用するための、テンソル学習、テンソル代数、およびそれらのテンソル計算のためのバックエンド
-      img: /images/content_images/arlib/tensorly.png
-      alttext: tensorly
-      url: http://tensorly.org/stable/home.html
-scientificdomains:
-  intro:
-    - 
-      text: Pythonを使って働くほとんどの科学者はNumPyの力を利用しています。
-    - 
-      text: "Numpy は、 C や Fortran のような言語の計算パフォーマンスを、Pythonにもたらします。 このパワーはNumPyのシンプルさから来ており、NumPyによるソリューションの多くは明確でエレガントになります。"
-  librariesrow1:
-    - 
-      title: 量子コンピューティング
-      alttext: コンピューターチップ
-      img: /images/content_images/sc_dom_img/quantum_computing.svg
-    - 
-      title: 統計コンピューティング
-      alttext: 線グラフで、グラフが上に移動します。
-      img: /images/content_images/sc_dom_img/statistical_computing.svg
-    - 
-      title: 信号処理
-      alttext: 正と負の値を持つ棒グラフ。
-      img: /images/content_images/sc_dom_img/signal_processing.svg
-    - 
-      title: 画像処理
-      alttext: 山々の写真
-      img: /images/content_images/sc_dom_img/image_processing.svg
-    - 
-      title: グラフとネットワーク
-      alttext: シンプルなグラフ
-      img: /images/content_images/sc_dom_img/sd6.svg
-    - 
-      title: 天文学における計算
-      alttext: 望遠鏡
-      img: /images/content_images/sc_dom_img/astronomy_processes.svg
-    - 
-      title: 認知心理学
-      alttext: ギアをつけた人間の頭部
-      img: /images/content_images/sc_dom_img/cognitive_psychology.svg
-  librariesrow2:
-    - 
-      title: 生命情報科学
-      alttext: DNAの鎖
-      img: /images/content_images/sc_dom_img/bioinformatics.svg
-    - 
-      title: ベイズ推論
-      alttext: 鐘形の曲線のグラフ
-      img: /images/content_images/sc_dom_img/bayesian_inference.svg
-    - 
-      title: 数学的分析
-      alttext: ４つの数学記号
-      img: /images/content_images/sc_dom_img/mathematical_analysis.svg
-    - 
-      title: 化学
-      alttext: 試験管
-      img: /images/content_images/sc_dom_img/chemistry.svg
-    - 
-      title: 地球科学
-      alttext: 地球
-      img: /images/content_images/sc_dom_img/geoscience.svg
-    - 
-      title: 地理情報処理
-      alttext: 地図
-      img: /images/content_images/sc_dom_img/GIS.svg
-    - 
-      title: アーキテクチャとエンジニアリング
-      alttext: マイクロプロセッサ開発ボード
-      img: /images/content_images/sc_dom_img/robotics.svg
-datascience:
-  intro: "Numpy は豊富なデータサイエンスライブラリのエコシステムの中核にあります。一般的なデータサイエンスのワークフローは次のようになります。"
-  image1:
-    - 
-      img: /images/content_images/ds-landscape.png
-      alttext: Python ライブラリの図 。5 つのカテゴリに分類され、「抽出、変換、読み込み」、「データ探索」、「モデリング」、「評価」、「可視化」です。
-  image2:
-    - 
-      img: /images/content_images/data-science.png
-      alttext: 三つの円が重なり合う図。円はそれぞれ「数学」、「コンピューターサイエンス」、「専門知識」でラベル付けされています。図の中心部には、三つの円が重なり合って形成されるエリアがあり、「データサイエンス」とラベル付けされています。
-  examples:
-    - 
-      text: "<b>抽出, 変換, 読み込み: </b>[Pandas](https://pandas.pydata.org), [Intake](https://intake.readthedocs.io), [PyJanitor](https://pyjanitor.readthedocs.io/)"
-    - 
-      text: "<b>探索的解析: </b>[Jupyter](https://jupyter.org), [Seaborn](https://seaborn.pydata.org), [Matplotlib](https://matplotlib.org), [Altair](https://altair-viz.github.io)"
-    - 
-      text: "<b>モデリングと評価: </b>[scikit-learn](https://scikit-learn.org), [statsmodels](https://www.statsmodels.org/stable/index.html), [PyMC3](https://docs.pymc.io), [spaCy](https://spacy.io)"
-    - 
-      text: "<b>ダッシュボードでのレポート: </b>[Dash](https://plotly.com/dash), [Panel](https://panel.holoviz.org), [Voila](https://github.com/voila-dashboards/voila)"
-  content:
-    - 
-      text: 大規模データに対して、[Dask](https://dask.org)と[Ray](https://ray.io/)はスケールすることを目指して設計されています。安定したデプロイメントに関しては、データのバージョニング（[DVC](https://dvc.org)）、実験の追跡（[MLFlow](https://mlflow.org)）、ワークフローの自動化（[Airflow](https://airflow.apache.org)および[Prefect](https://www.prefect.io)が重要ですが様々なNumPyベースのツールが提供されています。
-visualization:
-  images:
-    - 
-      url: https://www.fusioncharts.com/blog/best-python-data-visualization-libraries
-      img: /images/content_images/v_matplotlib.png
-      alttext: matplotlibで作られたストリームプロット
-    - 
-      url: https://github.com/yhat/ggpy
-      img: /images/content_images/v_ggpy.png
-      alttext: ggpyで作られた散布図グラフ
-    - 
-      url: https://www.journaldev.com/19692/python-plotly-tutorial
-      img: /images/content_images/v_plotly.png
-      alttext: plotyで作られた箱ひげ図
-    - 
-      url: https://alta-viz.github.io/gallery/streamgraph.html
-      img: /images/content_images/v_altair.png
-      alttext: altairで作られたストリームグラフ
-    - 
-      url: https://seaborn.pydata.org
-      img: /images/content_images/v_seaborn.png
-      alttext: 2種類のグラフによるペアプロット。seabornで作られたプロットと周波数グラフ"
-    - 
-      url: https://docs.pyvista.org/examples/index.html
-      img: /images/content_images/v_pyvista.png
-      alttext: PyVista製の3Dボリュームレンダリング
-    - 
-      url: https://napari.org
-      img: /images/content_images/v_napari.png
-      alttext: ナパリで作られた多次元画像
-    - 
-      url: https://vispy.org/gallery/index.html
-      img: /images/content_images/v_vispy.png
-      alttext: vispyで作られたボロノイ図
-  content:
-    - 
-      text: NumPyは、[Matplotlib](https://matplotlib.org)、[Seaborn](https://seaborn.pydata.org)、[Plotly](https://plot.ly)、[Altair](https://altair-viz.github.io)、[Bokeh](https://docs.bokeh.org/en/latest/)、[Holoviz](https://holoviz.org)、[Vispy](http://vispy.org)、[Napari](https://github.com/napari/napari)、[PyVista](https://github.com/pyvista/pyvista)などの、急成長している[Python visualization landscape](https://pyviz.org/overviews/index.html)に欠かせないコンポーネントです。
-    - 
-      text: NumPy の大規模配列の高速処理により、研究者はネイティブの Python が扱うことができるよりも、はるかに大きなデータセットを可視化することができます。
+params:
+  machinelearning:
+    paras:
+      - 
+        para1: NumPyは、[scikit-learn](https://scikit-learn.org)や[SciPy](https://www.scipy.org)のような強力な機械学習ライブラリの基礎を形成しています。機械学習の技術分野が成長するにつれ、NumPyをベースにしたライブラリの数も増えています。[TensorFlow](https://www.tensorflow.org)の深層学習機能は、音声認識や画像認識、テキストベースのアプリケーション、時系列分析、動画検出など、幅広い応用用途があります。[PyTorch](https://pytorch.org)も、コンピュータビジョンや自然言語処理の研究者に人気のある深層学習ライブラリです。[MXNet](https://github.com/apache/incubator-mxnet)もAIパッケージの一つで、深層学習の設計図やテンプレート機能を提供しています。
+        para2: '[ensemble](https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205)法と呼ばれる統計的手法であるビンニング、バギング、スタッキングや、[XGBoost](https://github.com/dmlc/xgboost)、[LightGBM](https://lightgbm.readthedocs.io/en/latest/)、[CatBoost](https://catboost.ai)などのツールで実装されているブースティングなどは、機械学習アルゴリズムの一つであり、最速の推論エンジンの一つです。[Yellowbrick](https://www.scikit-yb.org/en/latest/)や[Eli5](https://eli5.readthedocs.io/en/latest/)は機械学習の可視化機能を提供しています。'
+  arraylibraries:
+    intro:
+      - 
+        text: NumPyのAPIは、革新的なハードウェアを利用したり、特殊な配列タイプを作成したり、NumPyが提供する以上の機能を追加するためにライブラリを作成する際の基礎となります。
+    headers:
+      - 
+        text: 配列ライブラリ
+      - 
+        text: 機能と応用分野
+    libraries:
+      - 
+        title: Dask
+        text: 分析用の分散配列と高度な並列処理により、大規模な処理を可能にします。
+        img: /images/content_images/arlib/dask.png
+        alttext: Dask
+        url: https://dask.org/
+      - 
+        title: CuPy
+        text: Python を使用した GPUによる高速計算用のNumPy互換配列ライブラリ
+        img: /images/content_images/arlib/cupy.png
+        alttext: CuPy
+        url: https://cupy.chainer.org
+      - 
+        title: JAX
+        text: "NumPyコードの合成可能な変換ライブラリ:  微分、ベクトル化、GPU/TPUへのジャストインタイムコンパイル"
+        img: /images/content_images/arlib/jax_logo_250px.png
+        alttext: JAX
+        url: https://github.com/google/jax
+      - 
+        title: Xarray
+        text: 高度な分析と視覚化のためのラベルとインデックス付き多次元配列
+        img: /images/content_images/arlib/xarray.png
+        alttext: xarray
+        url: https://xarray.pydata.org/en/stable/index.html
+      - 
+        title: Sparse
+        text: Dask と SciPy の疎行列の線形代数ライブラリを統合した、Numpy 互換の疎行列ライブラリ
+        img: /images/content_images/arlib/sparse.png
+        alttext: sparse
+        url: https://sparse.pydata.org/en/latest/
+      - 
+        title: PyTorch
+        text: 研究用のプロトタイピングから本番運用への展開を加速させる、深層学習フレームワーク
+        img: /images/content_images/arlib/pytorch-logo-dark.svg
+        alttext: PyTorch
+        url: https://pytorch.org/
+      - 
+        title: TensorFlow
+        text: 機械学習を利用したアプリケーションを簡単に構築・展開するための、エンド・ツー・エンドの機械学習プラットフォーム
+        img: /images/content_images/arlib/tensorflow-logo.svg
+        alttext: TensorFlow
+        url: https://www.tensorflow.org
+      - 
+        title: MXNet
+        text: 柔軟や研究用のプロトタイピングから、実際の運用まで利用可能な深層学習フレームワーク
+        img: /images/content_images/arlib/mxnet_logo.png
+        alttext: MXNet
+        url: https://mxnet.apache.org/
+      - 
+        title: Arrow
+        text: 列型のインメモリーデータやその分析のための、複数の言語に対応した開発プラットフォーム
+        img: /images/content_images/arlib/arrow.png
+        alttext: arrow
+        url: https://github.com/apache/arrow
+      - 
+        title: xtensor
+        text: 数値解析のためのブロードキャスティングと遅延計算を備えた多次元配列
+        img: /images/content_images/arlib/xtensor.png
+        alttext: xtensor
+        url: https://github.com/xtensor-stack/xtensor-python
+      - 
+        title: XND
+        text: Numpy のような イディオムを使って JSON のようなデータを操作するライブラリ
+        img: /images/content_images/arlib/xnd.png
+        alttext: xnd
+        url: https://xnd.io
+      - 
+        title: uarray
+        text: APIを実装から切り離すPythonバックエンドシステム (unumpyはNumPy APIを提供しています)
+        img: /images/content_images/arlib/uarray.png
+        alttext: uarray
+        url: https://uarray.org/en/latest/
+      - 
+        title: tensorly
+        text: Numpy、MXNet、PyTorch、TensorFlowまたはCupyをシームレスに使用するための、テンソル学習、テンソル代数、およびそれらのテンソル計算のためのバックエンド
+        img: /images/content_images/arlib/tensorly.png
+        alttext: tensorly
+        url: http://tensorly.org/stable/home.html
+  scientificdomains:
+    intro:
+      - 
+        text: Pythonを使って働くほとんどの科学者はNumPyの力を利用しています。
+      - 
+        text: "Numpy は、 C や Fortran のような言語の計算パフォーマンスを、Pythonにもたらします。 このパワーはNumPyのシンプルさから来ており、NumPyによるソリューションの多くは明確でエレガントになります。"
+    librariesrow1:
+      - 
+        title: 量子コンピューティング
+        alttext: コンピューターチップ
+        img: /images/content_images/sc_dom_img/quantum_computing.svg
+      - 
+        title: 統計コンピューティング
+        alttext: 線グラフで、グラフが上に移動します。
+        img: /images/content_images/sc_dom_img/statistical_computing.svg
+      - 
+        title: 信号処理
+        alttext: 正と負の値を持つ棒グラフ。
+        img: /images/content_images/sc_dom_img/signal_processing.svg
+      - 
+        title: 画像処理
+        alttext: 山々の写真
+        img: /images/content_images/sc_dom_img/image_processing.svg
+      - 
+        title: グラフとネットワーク
+        alttext: シンプルなグラフ
+        img: /images/content_images/sc_dom_img/sd6.svg
+      - 
+        title: 天文学における計算
+        alttext: 望遠鏡
+        img: /images/content_images/sc_dom_img/astronomy_processes.svg
+      - 
+        title: 認知心理学
+        alttext: ギアをつけた人間の頭部
+        img: /images/content_images/sc_dom_img/cognitive_psychology.svg
+    librariesrow2:
+      - 
+        title: 生命情報科学
+        alttext: DNAの鎖
+        img: /images/content_images/sc_dom_img/bioinformatics.svg
+      - 
+        title: ベイズ推論
+        alttext: 鐘形の曲線のグラフ
+        img: /images/content_images/sc_dom_img/bayesian_inference.svg
+      - 
+        title: 数学的分析
+        alttext: ４つの数学記号
+        img: /images/content_images/sc_dom_img/mathematical_analysis.svg
+      - 
+        title: 化学
+        alttext: 試験管
+        img: /images/content_images/sc_dom_img/chemistry.svg
+      - 
+        title: 地球科学
+        alttext: 地球
+        img: /images/content_images/sc_dom_img/geoscience.svg
+      - 
+        title: 地理情報処理
+        alttext: 地図
+        img: /images/content_images/sc_dom_img/GIS.svg
+      - 
+        title: アーキテクチャとエンジニアリング
+        alttext: マイクロプロセッサ開発ボード
+        img: /images/content_images/sc_dom_img/robotics.svg
+  datascience:
+    intro: "Numpy は豊富なデータサイエンスライブラリのエコシステムの中核にあります。一般的なデータサイエンスのワークフローは次のようになります。"
+    image1:
+      - 
+        img: /images/content_images/ds-landscape.png
+        alttext: Python ライブラリの図 。5 つのカテゴリに分類され、「抽出、変換、読み込み」、「データ探索」、「モデリング」、「評価」、「可視化」です。
+    image2:
+      - 
+        img: /images/content_images/data-science.png
+        alttext: 三つの円が重なり合う図。円はそれぞれ「数学」、「コンピューターサイエンス」、「専門知識」でラベル付けされています。図の中心部には、三つの円が重なり合って形成されるエリアがあり、「データサイエンス」とラベル付けされています。
+    examples:
+      - 
+        text: "<b>抽出, 変換, 読み込み: </b>[Pandas](https://pandas.pydata.org), [Intake](https://intake.readthedocs.io), [PyJanitor](https://pyjanitor.readthedocs.io/)"
+      - 
+        text: "<b>探索的解析: </b>[Jupyter](https://jupyter.org), [Seaborn](https://seaborn.pydata.org), [Matplotlib](https://matplotlib.org), [Altair](https://altair-viz.github.io)"
+      - 
+        text: "<b>モデリングと評価: </b>[scikit-learn](https://scikit-learn.org), [statsmodels](https://www.statsmodels.org/stable/index.html), [PyMC3](https://docs.pymc.io), [spaCy](https://spacy.io)"
+      - 
+        text: "<b>ダッシュボードでのレポート: </b>[Dash](https://plotly.com/dash), [Panel](https://panel.holoviz.org), [Voila](https://github.com/voila-dashboards/voila)"
+    content:
+      - 
+        text: 大規模データに対して、[Dask](https://dask.org)と[Ray](https://ray.io/)はスケールすることを目指して設計されています。安定したデプロイメントに関しては、データのバージョニング（[DVC](https://dvc.org)）、実験の追跡（[MLFlow](https://mlflow.org)）、ワークフローの自動化（[Airflow](https://airflow.apache.org)および[Prefect](https://www.prefect.io)が重要ですが様々なNumPyベースのツールが提供されています。
+  visualization:
+    images:
+      - 
+        url: https://www.fusioncharts.com/blog/best-python-data-visualization-libraries
+        img: /images/content_images/v_matplotlib.png
+        alttext: matplotlibで作られたストリームプロット
+      - 
+        url: https://github.com/yhat/ggpy
+        img: /images/content_images/v_ggpy.png
+        alttext: ggpyで作られた散布図グラフ
+      - 
+        url: https://www.journaldev.com/19692/python-plotly-tutorial
+        img: /images/content_images/v_plotly.png
+        alttext: plotyで作られた箱ひげ図
+      - 
+        url: https://alta-viz.github.io/gallery/streamgraph.html
+        img: /images/content_images/v_altair.png
+        alttext: altairで作られたストリームグラフ
+      - 
+        url: https://seaborn.pydata.org
+        img: /images/content_images/v_seaborn.png
+        alttext: 2種類のグラフによるペアプロット。seabornで作られたプロットと周波数グラフ"
+      - 
+        url: https://docs.pyvista.org/examples/index.html
+        img: /images/content_images/v_pyvista.png
+        alttext: PyVista製の3Dボリュームレンダリング
+      - 
+        url: https://napari.org
+        img: /images/content_images/v_napari.png
+        alttext: ナパリで作られた多次元画像
+      - 
+        url: https://vispy.org/gallery/index.html
+        img: /images/content_images/v_vispy.png
+        alttext: vispyで作られたボロノイ図
+    content:
+      - 
+        text: NumPyは、[Matplotlib](https://matplotlib.org)、[Seaborn](https://seaborn.pydata.org)、[Plotly](https://plot.ly)、[Altair](https://altair-viz.github.io)、[Bokeh](https://docs.bokeh.org/en/latest/)、[Holoviz](https://holoviz.org)、[Vispy](http://vispy.org)、[Napari](https://github.com/napari/napari)、[PyVista](https://github.com/pyvista/pyvista)などの、急成長している[Python visualization landscape](https://pyviz.org/overviews/index.html)に欠かせないコンポーネントです。
+      - 
+        text: NumPy の大規模配列の高速処理により、研究者はネイティブの Python が扱うことができるよりも、はるかに大きなデータセットを可視化することができます。
diff --git a/content/pt/config.yaml b/content/pt/config.yaml
index a0b70147f7..202bc1bf35 100644
--- a/content/pt/config.yaml
+++ b/content/pt/config.yaml
@@ -72,89 +72,89 @@ params:
   tabs:
     title: ECOSSISTEMA
   section5: false
-navbar:
-  - 
-    title: Instalação
-    url: /pt/install
-  - 
-    title: Documentação
-    url: https://numpy.org/doc/stable
-  - 
-    title: Aprenda
-    url: /pt/learn
-  - 
-    title: Comunidade
-    url: /pt/community
-  - 
-    title: Sobre
-    url: /pt/about
-  - 
-    title: Notícias
-    url: /pt/news
-  - 
-    title: Contribuir
-    url: /contribute
-footer:
-  logo: logo.svg
-  socialmediatitle: ""
-  socialmedia:
+  navbar:
     - 
-      link: https://github.com/numpy/numpy
-      icon: github
+      title: Instalação
+      url: /pt/install
     - 
-      link: https://www.youtube.com/channel/UCguIL9NZ7ybWK5WQ53qbHng
-      icon: youtube
+      title: Documentação
+      url: https://numpy.org/doc/stable
     - 
-      link: https://twitter.com/numpy_team
-      icon: twitter
-  quicklinks:
-    column1:
-      title: ""
-      links:
-        - 
-          text: Instalação
-          link: /pt/install
-        - 
-          text: Documentação
-          link: https://numpy.org/doc/stable
-        - 
-          text: Aprenda
-          link: /pt/learn
-        - 
-          text: Citando o Numpy
-          link: /pt/citing-numpy
-        - 
-          text: Roadmap
-          link: https://numpy.org/neps/roadmap.html
-    column2:
-      links:
-        - 
-          text: Sobre
-          link: /pt/about
-        - 
-          text: Comunidade
-          link: /pt/community
-        - 
-          text: Pesquisas de usuário
-          link: /pt/user-surveys
-        - 
-          text: Contribuir
-          link: /pt/contribute
-        - 
-          text: Código de Conduta
-          link: /pt/code-of-conduct
-    column3:
-      links:
-        - 
-          text: Ajuda
-          link: /pt/gethelp
-        - 
-          text: Termos de uso (EN)
-          link: /pt/terms
-        - 
-          text: Privacidade
-          link: /pt/privacy
-        - 
-          text: Kit de imprensa
-          link: /pt/press-kit
+      title: Aprenda
+      url: /pt/learn
+    - 
+      title: Comunidade
+      url: /pt/community
+    - 
+      title: Sobre
+      url: /pt/about
+    - 
+      title: Notícias
+      url: /pt/news
+    - 
+      title: Contribuir
+      url: /contribute
+  footer:
+    logo: logo.svg
+    socialmediatitle: ""
+    socialmedia:
+      - 
+        link: https://github.com/numpy/numpy
+        icon: github
+      - 
+        link: https://www.youtube.com/channel/UCguIL9NZ7ybWK5WQ53qbHng
+        icon: youtube
+      - 
+        link: https://twitter.com/numpy_team
+        icon: twitter
+    quicklinks:
+      column1:
+        title: ""
+        links:
+          - 
+            text: Instalação
+            link: /pt/install
+          - 
+            text: Documentação
+            link: https://numpy.org/doc/stable
+          - 
+            text: Aprenda
+            link: /pt/learn
+          - 
+            text: Citando o Numpy
+            link: /pt/citing-numpy
+          - 
+            text: Roadmap
+            link: https://numpy.org/neps/roadmap.html
+      column2:
+        links:
+          - 
+            text: Sobre
+            link: /pt/about
+          - 
+            text: Comunidade
+            link: /pt/community
+          - 
+            text: Pesquisas de usuário
+            link: /pt/user-surveys
+          - 
+            text: Contribuir
+            link: /pt/contribute
+          - 
+            text: Código de Conduta
+            link: /pt/code-of-conduct
+      column3:
+        links:
+          - 
+            text: Ajuda
+            link: /pt/gethelp
+          - 
+            text: Termos de uso (EN)
+            link: /pt/terms
+          - 
+            text: Privacidade
+            link: /pt/privacy
+          - 
+            text: Kit de imprensa
+            link: /pt/press-kit
 
diff --git a/content/pt/tabcontents.yaml b/content/pt/tabcontents.yaml
index 270fac1e56..6bb9588f8a 100644
--- a/content/pt/tabcontents.yaml
+++ b/content/pt/tabcontents.yaml
@@ -1,218 +1,219 @@
-machinelearning:
-  paras:
-    - 
-      para1: O NumPy forma a base de bibliotecas de aprendizagem de máquina poderosas como [scikit-learn](https://scikit-learn.org) e [SciPy](https://www.scipy.org). À medida que a disciplina de aprendizagem de máquina cresce, a lista de bibliotecas construidas a partir do NumPy também cresce. As funcionalidades de deep learning do [TensorFlow](https://www.tensorflow.org) tem diversas aplicações &mdash; entre elas, reconhecimento de imagem e de fala, aplicações baseadas em texto, análise de séries temporais, e detecção de vídeo. O [PyTorch](https://pytorch.org), outra biblioteca de deep learning, é popular entre pesquisadores em visão computacional e processamento de linguagem natural. O [MXNet](https://github.com/apache/incubator-mxnet) é outro pacote de IA, que fornece templates e protótipos para deep learning.
-      para2: Técnicas estatísticas chamadas métodos de [ensemble](https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205) tais como binning, bagging, stacking, e boosting estão entre os algoritmos de ML implementados por ferramentas tais como [XGBoost](https://github.com/dmlc/xgboost), [LightGBM](https://lightgbm.readthedocs.io/en/latest/), e [CatBoost](https://catboost.ai) &mdash; um dos motores de inferência mais rápidos. [Yellowbrick](https://www.scikit-yb.org/en/latest/) e [Eli5](https://eli5.readthedocs.io/en/latest/) oferecem visualizações para aprendizagem de máquina.
-arraylibraries:
-  intro:
-    - 
-      text: A API do NumPy é o ponto de partida quando bibliotecas são escritas para explorar hardware inovador, criar tipos de arrays especializados, ou adicionar capacidades além do que o NumPy fornece.
-  headers:
-    - 
-      text: Biblioteca de Arrays
-    - 
-      text: Recursos e áreas de aplicação
-  libraries:
-    - 
-      title: Dask
-      text: Arrays distribuídas e paralelismo avançado para análise, permitindo desempenho em escala.
-      img: /images/content_images/arlib/dask.png
-      alttext: Dask
-      url: https://dask.org/
-    - 
-      title: CuPy
-      text: Biblioteca de matriz compatível com NumPy para computação acelerada pela GPU com Python.
-      img: /images/content_images/arlib/cupy.png
-      alttext: CuPy
-      url: https://cupy.chainer.org
-    - 
-      title: JAX
-      text: "Composable transformations of NumPy programs differentiate: vectorize, just-in-time compilation to GPU/TPU."
-      img: /images/content_images/arlib/jax_logo_250px.png
-      alttext: JAX
-      url: https://github.com/google/jax
-    - 
-      title: Xarray
-      text: Arrays multidimensionais rotuladas e indexadas para análise e visualização avançadas
-      img: /images/content_images/arlib/xarray.png
-      alttext: xarray
-      url: https://xarray.pydata.org/en/stable/index.html
-    - 
-      title: Sparse
-      text: Biblioteca de arrays compatíveis com o NumPy que pode ser integrada com Dask e álgebra linear esparsa da SciPy.
-      img: /images/content_images/arlib/sparse.png
-      alttext: sparse
-      url: https://sparse.pydata.org/en/latest/
-    - 
-      title: PyTorch
-      text: Framework de deep learning que acelera o caminho entre prototipação de pesquisa e colocação em produção.
-      img: /images/content_images/arlib/pytorch-logo-dark.svg
-      alttext: PyTorch
-      url: https://pytorch.org/
-    - 
-      title: TensorFlow
-      text: Uma plataforma completa para aprendizagem de máquina que permite construir e colocar em produção aplicações usando ML facilmente.
-      img: /images/content_images/arlib/tensorflow-logo.svg
-      alttext: TensorFlow
-      url: https://www.tensorflow.org
-    - 
-      title: MXNet
-      text: Framework de deep learning voltado para flexibilizar prototipação em pesquisa e produção.
-      img: /images/content_images/arlib/mxnet_logo.png
-      alttext: MXNet
-      url: https://mxnet.apache.org/
-    - 
-      title: Arrow
-      text: Uma plataforma de desenvolvimento multi-linguagens para dados e análise para dados armazenados em colunas na memória.
-      img: /images/content_images/arlib/arrow.png
-      alttext: arrow
-      url: https://github.com/apache/arrow
-    - 
-      title: xtensor
-      text: Arrays multidimensionais com broadcasting e avaliação preguiçosa (lazy computing) para análise numérica.
-      img: /images/content_images/arlib/xtensor.png
-      alttext: xtensor
-      url: https://github.com/xtensor-stack/xtensor-python
-    - 
-      title: XND
-      text: Develop libraries for array computing, recreating NumPy's foundational concepts.
-      img: /images/content_images/arlib/xnd.png
-      alttext: xnd
-      url: https://xnd.io
-    - 
-      title: uarray
-      text: Sistema de backend Python que dissocia a API da implementação; unumpy fornece uma API NumPy.
-      img: /images/content_images/arlib/uarray.png
-      alttext: uarray
-      url: https://uarray.org/en/latest/
-    - 
-      title: tensorly
-      text: Ferramentas para aprendizagem com tensores, algebra e backends para usar NumPy, MXNet, PyTorch, TensorFlow ou CuPy sem esforço.
-      img: /images/content_images/arlib/tensorly.png
-      alttext: tensorly
-      url: http://tensorly.org/stable/home.html
-scientificdomains:
-  intro:
-    - 
-      text: Quase todos os cientistas que trabalham em Python se baseiam na potência do NumPy.
-    - 
-      text: "NumPy traz o poder computacional de linguagens como C e Fortran para Python, uma linguagem muito mais fácil de aprender e usar. Com esse poder vem a simplicidade: uma solução no NumPy é frequentemente clara e elegante."
-  librariesrow1:
-    - 
-      title: Computação quântica
-      alttext: Um chip de computador.
-      img: /images/content_images/sc_dom_img/quantum_computing.svg
-    - 
-      title: Computação estatística
-      alttext: Um gráfico com uma linha em movimento para cima.
-      img: /images/content_images/sc_dom_img/statistical_computing.svg
-    - 
-      title: Processamento de sinais
-      alttext: Um gráfico de barras com valores positivos e negativos.
-      img: /images/content_images/sc_dom_img/signal_processing.svg
-    - 
-      title: Processamento de imagens
-      alttext: Uma fotografia das montanhas.
-      img: /images/content_images/sc_dom_img/image_processing.svg
-    - 
-      title: Gráficos e Redes
-      alttext: Um grafo simples.
-      img: /images/content_images/sc_dom_img/sd6.svg
-    - 
-      title: Processos de Astronomia
-      alttext: Um telescópio.
-      img: /images/content_images/sc_dom_img/astronomy_processes.svg
-    - 
-      title: Psicologia Cognitiva
-      alttext: Uma cabeça humana com engrenagens.
-      img: /images/content_images/sc_dom_img/cognitive_psychology.svg
-  librariesrow2:
-    - 
-      title: Bioinformática
-      alttext: Um pedaço de DNA.
-      img: /images/content_images/sc_dom_img/bioinformatics.svg
-    - 
-      title: Inferência Bayesiana
-      alttext: Um gráfico com uma curva em forma de sino.
-      img: /images/content_images/sc_dom_img/bayesian_inference.svg
-    - 
-      title: Análise Matemática
-      alttext: Quatro símbolos matemáticos.
-      img: /images/content_images/sc_dom_img/mathematical_analysis.svg
-    - 
-      title: Química
-      alttext: Um tubo de ensaio.
-      img: /images/content_images/sc_dom_img/chemistry.svg
-    - 
-      title: Geociências
-      alttext: A Terra.
-      img: /images/content_images/sc_dom_img/geoscience.svg
-    - 
-      title: Processamento Geográfico
-      alttext: Um mapa.
-      img: /images/content_images/sc_dom_img/GIS.svg
-    - 
-      title: Arquitetura e Engenharia
-      alttext: Uma placa de desenvolvimento de microprocessador.
-      img: /images/content_images/sc_dom_img/robotics.svg
-datascience:
-  intro: "NumPy está no centro de um rico ecossistema de bibliotecas de ciência de dados. Um fluxo de trabalho típico de ciência de dados exploratório pode parecer assim:"
-  image1:
-    - 
-      img: /images/content_images/ds-landscape.png
-      alttext: Diagrama de bibliotecas Python. As cinco categorias são 'Extrair, Transformar, Carregar', 'Exploração de Dados', 'Modelo de Dados', 'Avaliação de Dados' e 'Apresentação de Dados'.
-  image2:
-    - 
-      img: /images/content_images/data-science.png
-      alttext: Diagram of three overlapping circle. The circles labeled 'Mathematics', 'Computer Science' and 'Domain Expertise'. In the middle of the diagram, which has the three circles overlapping it, is an area labeled 'Data Science'.
-  examples:
-    - 
-      text: "<b>Extract, Transform, Load: </b>[Pandas](https://pandas.pydata.org),[ Intake](https://intake.readthedocs.io),[PyJanitor](https://pyjanitor.readthedocs.io/)"
-    - 
-      text: "<b>Exploratory analysis: </b>[Jupyter](https://jupyter.org),[Seaborn](https://seaborn.pydata.org),[ Matplotlib](https://matplotlib.org),[ Altair](https://altair-viz.github.io)"
-    - 
-      text: "<b>Model and evaluate: </b>[scikit-learn](https://scikit-learn.org),[ statsmodels](https://www.statsmodels.org/stable/index.html),[ PyMC3](https://docs.pymc.io),[ spaCy](https://spacy.io)"
-    - 
-      text: "<b>Report in a dashboard: </b>[Dash](https://plotly.com/dash),[ Panel](https://panel.holoviz.org),[ Voila](https://github.com/voila-dashboards/voila)"
-  content:
-    - 
-      text: For high data volumes, [Dask](https://dask.org) and[Ray](https://ray.io/) are designed to scale. Stabledeployments rely on data versioning ([DVC](https://dvc.org)),experiment tracking ([MLFlow](https://mlflow.org)), andworkflow automation ([Airflow](https://airflow.apache.org) and[Prefect](https://www.prefect.io)).
-visualization:
-  images:
-    - 
-      url: https://www.fusioncharts.com/blog/best-python-data-visualization-libraries
-      img: /images/content_images/v_matplotlib.png
-      alttext: Um streamplot feito em matplotlib
-    - 
-      url: https://github.com/yhat/ggpy
-      img: /images/content_images/v_ggpy.png
-      alttext: Um gráfico scatter-plot feito em ggpy
-    - 
-      url: https://www.journaldev.com/19692/python-plotly-tutorial
-      img: /images/content_images/v_plotly.png
-      alttext: Um box-plot feito no plotly
-    - 
-      url: https://altair-viz.github.io/gallery/streamgraph.html
-      img: /images/content_images/v_altair.png
-      alttext: Um gráfico streamgraph feito em altair
-    - 
-      url: https://seaborn.pydata.org
-      img: /images/content_images/v_seaborn.png
-      alttext: A plot duplo com dois tipos de gráficos, um plot-graph e um gráfico de frequência feitos no seaborn
-    - 
-      url: https://docs.pyvista.org/examples/index.html
-      img: /images/content_images/v_pyvista.png
-      alttext: Uma renderização de volume 3D feita no PyVista.
-    - 
-      url: https://napari.org
-      img: /images/content_images/v_napari.png
-      alttext: Uma imagem multidimensional, feita em napari.
-    - 
-      url: https://vispy.org/gallery/index.html
-      img: /images/content_images/v_vispy.png
-      alttext: Diagrama de Voronoi feito com vispy.
-  content:
-    - 
-      text: NumPy é um componente essencial no crescente [campo de visualização em Python](https://pyviz.org/overviews/index.html), que inclui [Matplotlib](https://matplotlib.org), [Seaborn](https://seaborn.pydata.org), [Plotly](https://plot.ly), [Altair](https://altair-viz.github.io), [Bokeh](https://docs.bokeh.org/en/latest/), [Holoviz](https://holoviz.org), [Vispy](http://vispy.org), [Napari](https://github.com/napari/napari), e [PyVista](https://github.com/pyvista/pyvista), para citar alguns.
-    - 
-      text: O processamento de grandes arrays acelerado pela NumPy permite que os pesquisadores visualizem conjuntos de dados muito maiores do que o Python nativo poderia permitir.
+params:
+  machinelearning:
+    paras:
+      - 
+        para1: O NumPy forma a base de bibliotecas de aprendizagem de máquina poderosas como [scikit-learn](https://scikit-learn.org) e [SciPy](https://www.scipy.org). À medida que a disciplina de aprendizagem de máquina cresce, a lista de bibliotecas construidas a partir do NumPy também cresce. As funcionalidades de deep learning do [TensorFlow](https://www.tensorflow.org) tem diversas aplicações &mdash; entre elas, reconhecimento de imagem e de fala, aplicações baseadas em texto, análise de séries temporais, e detecção de vídeo. O [PyTorch](https://pytorch.org), outra biblioteca de deep learning, é popular entre pesquisadores em visão computacional e processamento de linguagem natural. O [MXNet](https://github.com/apache/incubator-mxnet) é outro pacote de IA, que fornece templates e protótipos para deep learning.
+        para2: Técnicas estatísticas chamadas métodos de [ensemble](https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205) tais como binning, bagging, stacking, e boosting estão entre os algoritmos de ML implementados por ferramentas tais como [XGBoost](https://github.com/dmlc/xgboost), [LightGBM](https://lightgbm.readthedocs.io/en/latest/), e [CatBoost](https://catboost.ai) &mdash; um dos motores de inferência mais rápidos. [Yellowbrick](https://www.scikit-yb.org/en/latest/) e [Eli5](https://eli5.readthedocs.io/en/latest/) oferecem visualizações para aprendizagem de máquina.
+  arraylibraries:
+    intro:
+      - 
+        text: A API do NumPy é o ponto de partida quando bibliotecas são escritas para explorar hardware inovador, criar tipos de arrays especializados, ou adicionar capacidades além do que o NumPy fornece.
+    headers:
+      - 
+        text: Biblioteca de Arrays
+      - 
+        text: Recursos e áreas de aplicação
+    libraries:
+      - 
+        title: Dask
+        text: Arrays distribuídas e paralelismo avançado para análise, permitindo desempenho em escala.
+        img: /images/content_images/arlib/dask.png
+        alttext: Dask
+        url: https://dask.org/
+      - 
+        title: CuPy
+        text: Biblioteca de matriz compatível com NumPy para computação acelerada pela GPU com Python.
+        img: /images/content_images/arlib/cupy.png
+        alttext: CuPy
+        url: https://cupy.chainer.org
+      - 
+        title: JAX
+        text: "Composable transformations of NumPy programs differentiate: vectorize, just-in-time compilation to GPU/TPU."
+        img: /images/content_images/arlib/jax_logo_250px.png
+        alttext: JAX
+        url: https://github.com/google/jax
+      - 
+        title: Xarray
+        text: Arrays multidimensionais rotuladas e indexadas para análise e visualização avançadas
+        img: /images/content_images/arlib/xarray.png
+        alttext: xarray
+        url: https://xarray.pydata.org/en/stable/index.html
+      - 
+        title: Sparse
+        text: Biblioteca de arrays compatíveis com o NumPy que pode ser integrada com Dask e álgebra linear esparsa da SciPy.
+        img: /images/content_images/arlib/sparse.png
+        alttext: sparse
+        url: https://sparse.pydata.org/en/latest/
+      - 
+        title: PyTorch
+        text: Framework de deep learning que acelera o caminho entre prototipação de pesquisa e colocação em produção.
+        img: /images/content_images/arlib/pytorch-logo-dark.svg
+        alttext: PyTorch
+        url: https://pytorch.org/
+      - 
+        title: TensorFlow
+        text: Uma plataforma completa para aprendizagem de máquina que permite construir e colocar em produção aplicações usando ML facilmente.
+        img: /images/content_images/arlib/tensorflow-logo.svg
+        alttext: TensorFlow
+        url: https://www.tensorflow.org
+      - 
+        title: MXNet
+        text: Framework de deep learning voltado para flexibilizar prototipação em pesquisa e produção.
+        img: /images/content_images/arlib/mxnet_logo.png
+        alttext: MXNet
+        url: https://mxnet.apache.org/
+      - 
+        title: Arrow
+        text: Uma plataforma de desenvolvimento multi-linguagens para dados e análise para dados armazenados em colunas na memória.
+        img: /images/content_images/arlib/arrow.png
+        alttext: arrow
+        url: https://github.com/apache/arrow
+      - 
+        title: xtensor
+        text: Arrays multidimensionais com broadcasting e avaliação preguiçosa (lazy computing) para análise numérica.
+        img: /images/content_images/arlib/xtensor.png
+        alttext: xtensor
+        url: https://github.com/xtensor-stack/xtensor-python
+      - 
+        title: XND
+        text: Develop libraries for array computing, recreating NumPy's foundational concepts.
+        img: /images/content_images/arlib/xnd.png
+        alttext: xnd
+        url: https://xnd.io
+      - 
+        title: uarray
+        text: Sistema de backend Python que dissocia a API da implementação; unumpy fornece uma API NumPy.
+        img: /images/content_images/arlib/uarray.png
+        alttext: uarray
+        url: https://uarray.org/en/latest/
+      - 
+        title: tensorly
+        text: Ferramentas para aprendizagem com tensores, algebra e backends para usar NumPy, MXNet, PyTorch, TensorFlow ou CuPy sem esforço.
+        img: /images/content_images/arlib/tensorly.png
+        alttext: tensorly
+        url: http://tensorly.org/stable/home.html
+  scientificdomains:
+    intro:
+      - 
+        text: Quase todos os cientistas que trabalham em Python se baseiam na potência do NumPy.
+      - 
+        text: "NumPy traz o poder computacional de linguagens como C e Fortran para Python, uma linguagem muito mais fácil de aprender e usar. Com esse poder vem a simplicidade: uma solução no NumPy é frequentemente clara e elegante."
+    librariesrow1:
+      - 
+        title: Computação quântica
+        alttext: Um chip de computador.
+        img: /images/content_images/sc_dom_img/quantum_computing.svg
+      - 
+        title: Computação estatística
+        alttext: Um gráfico com uma linha em movimento para cima.
+        img: /images/content_images/sc_dom_img/statistical_computing.svg
+      - 
+        title: Processamento de sinais
+        alttext: Um gráfico de barras com valores positivos e negativos.
+        img: /images/content_images/sc_dom_img/signal_processing.svg
+      - 
+        title: Processamento de imagens
+        alttext: Uma fotografia das montanhas.
+        img: /images/content_images/sc_dom_img/image_processing.svg
+      - 
+        title: Gráficos e Redes
+        alttext: Um grafo simples.
+        img: /images/content_images/sc_dom_img/sd6.svg
+      - 
+        title: Processos de Astronomia
+        alttext: Um telescópio.
+        img: /images/content_images/sc_dom_img/astronomy_processes.svg
+      - 
+        title: Psicologia Cognitiva
+        alttext: Uma cabeça humana com engrenagens.
+        img: /images/content_images/sc_dom_img/cognitive_psychology.svg
+    librariesrow2:
+      - 
+        title: Bioinformática
+        alttext: Um pedaço de DNA.
+        img: /images/content_images/sc_dom_img/bioinformatics.svg
+      - 
+        title: Inferência Bayesiana
+        alttext: Um gráfico com uma curva em forma de sino.
+        img: /images/content_images/sc_dom_img/bayesian_inference.svg
+      - 
+        title: Análise Matemática
+        alttext: Quatro símbolos matemáticos.
+        img: /images/content_images/sc_dom_img/mathematical_analysis.svg
+      - 
+        title: Química
+        alttext: Um tubo de ensaio.
+        img: /images/content_images/sc_dom_img/chemistry.svg
+      - 
+        title: Geociências
+        alttext: A Terra.
+        img: /images/content_images/sc_dom_img/geoscience.svg
+      - 
+        title: Processamento Geográfico
+        alttext: Um mapa.
+        img: /images/content_images/sc_dom_img/GIS.svg
+      - 
+        title: Arquitetura e Engenharia
+        alttext: Uma placa de desenvolvimento de microprocessador.
+        img: /images/content_images/sc_dom_img/robotics.svg
+  datascience:
+    intro: "NumPy está no centro de um rico ecossistema de bibliotecas de ciência de dados. Um fluxo de trabalho típico de ciência de dados exploratório pode parecer assim:"
+    image1:
+      - 
+        img: /images/content_images/ds-landscape.png
+        alttext: Diagrama de bibliotecas Python. As cinco categorias são 'Extrair, Transformar, Carregar', 'Exploração de Dados', 'Modelo de Dados', 'Avaliação de Dados' e 'Apresentação de Dados'.
+    image2:
+      - 
+        img: /images/content_images/data-science.png
+        alttext: Diagram of three overlapping circle. The circles labeled 'Mathematics', 'Computer Science' and 'Domain Expertise'. In the middle of the diagram, which has the three circles overlapping it, is an area labeled 'Data Science'.
+    examples:
+      - 
+        text: "<b>Extract, Transform, Load: </b>[Pandas](https://pandas.pydata.org),[ Intake](https://intake.readthedocs.io),[PyJanitor](https://pyjanitor.readthedocs.io/)"
+      - 
+        text: "<b>Exploratory analysis: </b>[Jupyter](https://jupyter.org),[Seaborn](https://seaborn.pydata.org),[ Matplotlib](https://matplotlib.org),[ Altair](https://altair-viz.github.io)"
+      - 
+        text: "<b>Model and evaluate: </b>[scikit-learn](https://scikit-learn.org),[ statsmodels](https://www.statsmodels.org/stable/index.html),[ PyMC3](https://docs.pymc.io),[ spaCy](https://spacy.io)"
+      - 
+        text: "<b>Report in a dashboard: </b>[Dash](https://plotly.com/dash),[ Panel](https://panel.holoviz.org),[ Voila](https://github.com/voila-dashboards/voila)"
+    content:
+      - 
+        text: For high data volumes, [Dask](https://dask.org) and[Ray](https://ray.io/) are designed to scale. Stabledeployments rely on data versioning ([DVC](https://dvc.org)),experiment tracking ([MLFlow](https://mlflow.org)), andworkflow automation ([Airflow](https://airflow.apache.org) and[Prefect](https://www.prefect.io)).
+  visualization:
+    images:
+      - 
+        url: https://www.fusioncharts.com/blog/best-python-data-visualization-libraries
+        img: /images/content_images/v_matplotlib.png
+        alttext: Um streamplot feito em matplotlib
+      - 
+        url: https://github.com/yhat/ggpy
+        img: /images/content_images/v_ggpy.png
+        alttext: Um gráfico scatter-plot feito em ggpy
+      - 
+        url: https://www.journaldev.com/19692/python-plotly-tutorial
+        img: /images/content_images/v_plotly.png
+        alttext: Um box-plot feito no plotly
+      - 
+        url: https://altair-viz.github.io/gallery/streamgraph.html
+        img: /images/content_images/v_altair.png
+        alttext: Um gráfico streamgraph feito em altair
+      - 
+        url: https://seaborn.pydata.org
+        img: /images/content_images/v_seaborn.png
+        alttext: A plot duplo com dois tipos de gráficos, um plot-graph e um gráfico de frequência feitos no seaborn
+      - 
+        url: https://docs.pyvista.org/examples/index.html
+        img: /images/content_images/v_pyvista.png
+        alttext: Uma renderização de volume 3D feita no PyVista.
+      - 
+        url: https://napari.org
+        img: /images/content_images/v_napari.png
+        alttext: Uma imagem multidimensional, feita em napari.
+      - 
+        url: https://vispy.org/gallery/index.html
+        img: /images/content_images/v_vispy.png
+        alttext: Diagrama de Voronoi feito com vispy.
+    content:
+      - 
+        text: NumPy é um componente essencial no crescente [campo de visualização em Python](https://pyviz.org/overviews/index.html), que inclui [Matplotlib](https://matplotlib.org), [Seaborn](https://seaborn.pydata.org), [Plotly](https://plot.ly), [Altair](https://altair-viz.github.io), [Bokeh](https://docs.bokeh.org/en/latest/), [Holoviz](https://holoviz.org), [Vispy](http://vispy.org), [Napari](https://github.com/napari/napari), e [PyVista](https://github.com/pyvista/pyvista), para citar alguns.
+      - 
+        text: O processamento de grandes arrays acelerado pela NumPy permite que os pesquisadores visualizem conjuntos de dados muito maiores do que o Python nativo poderia permitir.
diff --git a/gen_config.py b/gen_config.py
index 0f5a4636b9..4f0ccef5dd 100644
--- a/gen_config.py
+++ b/gen_config.py
@@ -1,23 +1,45 @@
 import os
 import re
+import yaml
 
 
-with open('config.yaml.in', 'r', encoding='utf-8') as templ:
-    lines = templ.readlines()
+config = yaml.load(
+    open("config.yaml.in", "r", encoding="utf-8"), Loader=yaml.SafeLoader
+)
 
-pattern = re.compile('< content\/\w\w\/\w*.yaml >')
-with open('config.yaml', 'w', encoding='utf-8') as f:
-    for line in lines:
-        match = pattern.search(line)
-        if match:
-            with open(match.group()[2:-2], 'r', encoding='utf-8') as f2:
-                for f2_line in f2.readlines():
-                    # indent to get correct yaml formatting
-                    f.write('    ' + f2_line)
-        elif line.startswith('disableLanguages'):
-            if os.environ.get('NUMPYORG_WITH_TRANSLATIONS'):
-                line = "#" + line
 
-            f.write(line)
+def merge_dicts(d1, d2):
+    for key, value in d2.items():
+        if key in d1:
+            if isinstance(value, list):
+                d1[key].extend(value)
+            elif isinstance(value, dict):
+                merge_dicts(d1[key], value)
         else:
-            f.write(line)
+            d1[key] = value
+
+    return d1
+
+
+def include_files(d):
+    external = {}
+    for key, val in d.items():
+        if isinstance(val, dict):
+            d[key] = include_files(val)
+        elif key == "include-files":
+            for otherfile in val:
+                external_data = yaml.load(
+                    open(otherfile, "r", encoding="utf-8"), Loader=yaml.SafeLoader
+                )
+                external = merge_dicts(external, external_data)
+
+    d.pop("include-files", None)
+    return {**d, **external}
+
+
+config = include_files(config)
+if os.environ.get("NUMPYORG_WITH_TRANSLATIONS"):
+    del config["disableLanguages"]
+
+
+yaml.dump(config, open('config.yaml', 'w', encoding='utf-8'), sort_keys=False)
diff --git a/netlify.toml b/netlify.toml
index 35942b0ce4..9fea4e7e7e 100644
--- a/netlify.toml
+++ b/netlify.toml
@@ -1,9 +1,9 @@
 # Settings in the [build] context are global and are applied to all contexts
 # unless otherwise overridden by more specific contexts.
 [build.environment]
-  PYTHON_VERSION = "3.8"
+  PYTHON_VERSION = "3.8" # netlify currently only support 2.7 and 3.8
   # Same Hugo version as in .github/workflows/gh-pages.yml
-  HUGO_VERSION = "0.104.3"
+  HUGO_VERSION = "0.115.4"
 
 # Here is another way to define context specific environment variables.
 [context.deploy-preview.environment]