Skip to content

Commit 0fdad49

Browse files
rgommersrgommers
authored
Merge pull request numpy#242 from bjnath/ecosystem_edit
DOC: Emphasize NumPy in Ecosystem openers
2 parents 8c7cdb5 + 62d6b30 commit 0fdad49

File tree

5 files changed

+85
-107
lines changed

5 files changed

+85
-107
lines changed

‎layouts/partials/array-libraries.html

Lines changed: 12 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -1,17 +1,16 @@
11
<!-- Array libraries Tab Content -->
22
<li class="array-libraries">
33
<p>
4-
Numpy array forms the core of the organically growing numeric
5-
Python <b>array library</b> ecosystem that now supports GPUs, sparse,
6-
distributed arrays and more.
7-
</p>
8-
<p>
9-
Several of these newer libraries such as CuPy, Sparse and Dask,
10-
implement the NumPy API adding support for modern user cases,
11-
newer hardware and higher scalability of array computing. Other
12-
array libraries such as Xarray, TensorLy consume NumPy API and
13-
build newer functionality on top of it, thus enhancing array
14-
computing in Python beyond Numpy capabilities.
4+
When libraries emerge to exploit new
5+
hardware technologies and architectures, they take
6+
NumPy as their starting point.
7+
<a href="https://cupy.chainer.org">CuPy</a>,
8+
<a href="https://sparse.pydata.org/en/latest/">Sparse</a>, and
9+
<a href="https://dask.org/">Dask</a>
10+
implement the NumPy API with support for modern user cases and
11+
scalable hardware;
12+
<a href="https://xarray.pydata.org/en/stable/index.html">Xarray</a> and
13+
<a href="http://tensorly.org/stable/home.html">Tensor.ly</a> add newer functionality.
1514
</p>
1615
<table>
1716
<tr class="highlight-th">
@@ -56,7 +55,7 @@
5655
astronomy, satellite imagery and mobile network modeling.</td>
5756
</tr>
5857
<tr>
59-
<td><img class="first-column-layout" src="images/content_images/arlib/CuPy.png" alt="CuPy"></td>
58+
<td><img class="first-column-layout" src="images/content_images/arlib/cupy.png" alt="CuPy"></td>
6059
<td class="full-center-text"><a href="https://cupy.chainer.org">CuPy</a></td>
6160
<td class="left-text">NumPy-compatible matrix library accelerated by CUDA used to implement Neural Networks
6261
for Deep Learning.</td>
@@ -82,7 +81,7 @@
8281
</tr>
8382
<tr>
8483
<td><img class="first-column-layout" src="images/content_images/arlib/xtensor.png" alt="xtensor"></td>
85-
<td class="full-center-text"><a href="" https://github.com/xtensor-stack/xtensor-python>xtensor </a> </td>
84+
<td class="full-center-text"><a href="https://github.com/xtensor-stack/xtensor-python">xtensor</a> </td>
8685
<td class="left-text">Multi-dimensional arrays with broadcasting and lazy computing for numerical
8786
analysis.</td>
8887
</tr>

‎layouts/partials/data-science.html

Lines changed: 40 additions & 23 deletions
Original file line numberDiff line numberDiff line change
@@ -8,27 +8,44 @@
88
</div>
99
<div>
1010
<p>
11-
Data Science makes it possible to analyze massive amounts of data
12-
and gain meaningful insights. A typical data science workflow involves
13-
various techniques and tools such as:
11+
NumPy lies at the core of a rich ecosystem of data science libraries.
12+
</p>
13+
<p>
14+
Data science is the analysis of massive amounts of data
15+
to gain insight. A typical workflow might be:
16+
1417
<ul class="content-tab">
15-
<li><b>Extract, Transform, Load (ETL):</b> Pandas, Beautiful Soup, Intake</li>
16-
<li><b>Explore:</b> Seaborn, Matplotlib</li>
17-
<li><b>Model:</b> Scikit-learn, SciPy, statsmodels</li>
18-
<li><b>Evaluate:</b> NumPy, TensorFlow </li>
18+
<li><b>Extract, Transform, Load (ETL):</b>
19+
<a href="https://pandas.pydata.org">Pandas</a>,
20+
<a href="https://www.crummy.com/software/BeautifulSoup/">Beautiful Soup</a>,
21+
<a href="https://intake.readthedocs.io/en/latest/"> Intake</a>
22+
</li>
23+
24+
<li><b>Explore:</b>
25+
<a href="https://seaborn.pydata.org"> Seaborn</a>,
26+
<a href="https://matplotlib.org">Matplotlib</a>,
27+
28+
</li>
29+
30+
<li><b>Model:</b>
31+
<a href="https://scikit-learn.org">scikit-learn</a>,
32+
<a href="https://www.scipy.org">SciPy</a>,
33+
<a href="https://www.statsmodels.org/stable/index.html"> statsmodels</a>.
34+
</li>
35+
36+
<li><b>Evaluate:</b>
37+
NumPy,
38+
<a href="https://www.tensorflow.org">TensorFlow</a>
39+
</li>
40+
1941
<li>
20-
<b>Presentation:</b>
42+
<b>Display:</b>
2143
<a href="./index.html/#tab-visual"> Data Visualization Tools</a>
2244
</li>
2345
</ul>
2446
</p>
2547
</div>
2648
</div>
27-
<p>
28-
Python has a rich ecosystem of libraries that enable Data Science
29-
workflows. <b> NumPy</b> is the foundation of almost all of these tools
30-
such as Pandas, Seaborn, Beautiful Soup and several others.
31-
</p>
3249
<div class="grid-container">
3350
<div>
3451
<p>
@@ -37,13 +54,13 @@
3754
data access and distribution, while
3855
<a href="https://www.crummy.com/software/BeautifulSoup/">Beautiful Soup</a>
3956
is widely used for web-scraping and gathering data sets.
40-
<a href="https://seaborn.pydata.org"> Seaborn</a> is well known for its
41-
<a href="https://towardsdatascience.com/how-to-perform-exploratory-data-analysis-with-seaborn-97e3413e841d">exploratory data analysis (EDA)</a>
42-
capabilities, <a href="https://scikit-learn.org">Scikit-learn</a> and
43-
<a href="https://www.scipy.org">Scipy</a> (statistical computing) serve some
57+
<a href="https://seaborn.pydata.org"> Seaborn</a> is well known for
58+
<a href="https://towardsdatascience.com/how-to-perform-exploratory-data-analysis-with-seaborn-97e3413e841d">exploratory data analysis (EDA)</a>;
59+
<a href="https://scikit-learn.org">scikit-learn</a> and
60+
<a href="https://www.scipy.org">SciPy</a> (statistical computing) serve some
4461
of the backbone processes required for machine learning (regression methods,
4562
classification, clustering, model validation and selection).
46-
Statistical data exploration, estimation of various statistical models
63+
Statistical data exploration, estimation of various statistical models,
4764
and conducting statistical tests are some of the functions offered by
4865
<a href="https://www.statsmodels.org/stable/index.html"> statsmodels</a>.
4966
</p>
@@ -53,11 +70,11 @@
5370
</div>
5471
</div>
5572
<p>
56-
Effective data analytics require deep knowledge of the data domain (e.g.,
57-
Retail, Healthcare, Marketing, Finance, Social Media, Automation, Sales, Travel,
58-
etc.) as well as other core disciplines of Data Science, Data Engineering and
59-
Data Visualization. Tools such as <a href="https://mlflow.org">MLFlow</a> address
60-
experiment hyper-parameter and result tracking needs, while
73+
Effective data analytics requires deep knowledge of the data domain (e.g.,
74+
retail, healthcare, marketing, finance, social media, automation, sales, travel,
75+
etc.) as well as other core disciplines of data science, data engineering, and
76+
data visualization. Tools such as <a href="https://mlflow.org">MLFlow</a> address
77+
experiment hyperparameter and result tracking needs, while
6178
<a href="https://dvc.org"> DVC</a> provides data version control for data science
6279
and machine learning workflows.
6380
</p>

‎layouts/partials/machine-learning.html

Lines changed: 20 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -13,29 +13,21 @@
1313
</div>
1414
<div>
1515
<p>
16-
<b>Machine learning</b> (ML) enables computers to learn using
17-
data, without having to be explicitly programmed.
18-
<b>NumPy</b> is the foundation of all data pre-processing
19-
that happens in the implementation of several ML Algorithms.
20-
</p>
21-
<p>
22-
Python’s rich machine language and deep learning ecosystem
23-
provides powerful tools such as
24-
<a href="https://scikit-learn.org/stable/">Scikit-learn</a>
25-
that is built on top of NumPy and
26-
<a href="https://www.scipy.org">SciPy</a> and offers data
27-
mining and analytics using classical ML algorithms.
28-
</p>
29-
<p>
30-
<a href="https://www.tensorflow.org">Tensorflow’s</a>
31-
deep learning capabilities help to define and run
32-
computations involving tensors that have broad
33-
applications in Speech and image recognition, Text-based
34-
applications, Time-Series analysis and Video Detection.
35-
<a href="https://pytorch.org">PyTorch </a> is another deep
36-
learning library that is very popular among researchers for
37-
computer vision and NLP applications. <a href="https://github.com/apache/incubator-mxnet">MXNet</a>
38-
is another AI package that provides blueprints and
16+
NumPy forms the basis of powerful machine learning libraries
17+
like
18+
<a href="https://scikit-learn.org">scikit-learn</a> and
19+
<a href="https://www.scipy.org">SciPy</a>.
20+
As machine learning grows, so does the
21+
list of libraries built on NumPy.
22+
<a href="https://www.tensorflow.org">TensorFlow’s</a>
23+
deep learning capabilities have broad
24+
applications &mdash; among them speech and image recognition, text-based
25+
applications, time-series analysis, and video detection.
26+
<a href="https://pytorch.org">PyTorch</a>, another deep
27+
learning library, is popular among researchers in
28+
computer vision and natural language processing.
29+
<a href="https://github.com/apache/incubator-mxnet">MXNet</a>
30+
is another AI package, providing blueprints and
3931
templates for deep learning.
4032
</p>
4133
</div>
@@ -44,15 +36,15 @@
4436
<div>
4537
<p>
4638
Statistical techniques called
47-
<a href="https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205">Ensemble</a>
39+
<a href="https://towardsdatascience.com/ensemble-methods-bagging-boosting-and-stacking-c9214a10a205">ensemble</a>
4840
methods such as binning,
49-
bagging, stacking and boosting are widely used in various ML
41+
bagging, stacking, and boosting are among the ML
5042
algorithms implemented by tools such as
5143
<a href="https://github.com/dmlc/xgboost">XGBoost</a>,
52-
<a href="https://lightgbm.readthedocs.io/en/latest/">LightGBM</a>,
53-
<a href="https://catboost.ai">CatBoost</a> - one of the
44+
<a href="https://lightgbm.readthedocs.io/en/latest/">LightGBM</a>, and
45+
<a href="https://catboost.ai">CatBoost</a> &mdash; one of the
5446
fastest inference engines.
55-
<a href="https://www.scikit-yb.org/en/latest/">Yellowbrick</a>,
47+
<a href="https://www.scikit-yb.org/en/latest/">Yellowbrick</a> and
5648
<a href="https://eli5.readthedocs.io/en/latest/">Eli5</a>
5749
offer machine learning visualizations.
5850
</p>

‎layouts/partials/scientific-domains.html

Lines changed: 7 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -1,14 +1,12 @@
11
<!-- Scientific Domains Tab Content -->
22
<li class="scientific-domains">
33
<p>
4-
Data acquisition (experimental, simulation), processing and
5-
visualization are the core data related tasks in almost all the
6-
scientific domains. Visualization of results through high quality
7-
figures make scientific reports and publications easy to
8-
understand. Python is easier to learn and computationally
9-
efficient for scientific computing. NumPy, SciPy and Matplotlib
10-
form the core Python packages that are used across various
11-
scientific domains.
4+
Nearly every scientist working in Python draws on the power of NumPy.
5+
</p>
6+
<p>
7+
NumPy brings the computational power of languages like C and Fortran
8+
to Python, a language much easier to learn and use. With this power
9+
comes simplicity: a solution in NumPy is often clear and elegant.
1210
</p>
1311
<!-- First Row -->
1412
<table>
@@ -99,7 +97,7 @@
9997
<td class="center-text"></td>
10098
<td class="center-text"></td>
10199
<td class="center-text"><a
102-
href="https://towardsdatascience.com/easy-steps-to-plot-geographic-data-on-a-map-python-11217859a2db">NumPy</a>
100+
href="https://towardsdatascience.com/easy-steps-to-plot-geographic-data-on-a-map-python-11217859a2db">NumPy</a>
103101
</td>
104102
<td class="center-text"></td>
105103
</tr>
@@ -122,9 +120,4 @@
122120
<td class="lastrow-center-text"></td>
123121
</tr>
124122
</table>
125-
<p>
126-
NumPy’s powerful array processing capabilities and elegant syntax
127-
helps to clearly and efficiently express computational algorithms
128-
in various scientific computing domains.
129-
</p>
130123
</li>

‎layouts/partials/visualization.html

Lines changed: 6 additions & 29 deletions
Original file line numberDiff line numberDiff line change
@@ -54,47 +54,24 @@
5454
</div>
5555
<div>
5656
<p>
57-
<a href="https://www.slideshare.net/Visage/data-visualization-101-how-to-design-chartsandgraphs">Data
58-
Visualization</a>
59-
exposes patterns, trends and correlations in textual-data, making it easier for humans to
60-
analyse and interpret large volumes of data.
61-
</p>
62-
<p>
63-
<a href="https://python-graph-gallery.com">Visualization elements</a>
64-
such as bar graphs, pie charts, line charts, maps, infographics, dashboards,
65-
geographic maps, heatmaps, and interactive images offer valuable
66-
insights for making data-driven decisions.
67-
</p>
68-
</div>
69-
<div>
70-
<p>
71-
NumPy is the key data transformation building block for the burgeoning
72-
<a href="https://pyviz.org/overviews/index.html">Python visualization landscape</a> comprising of
57+
NumPy is an essential component in the burgeoning
58+
<a href="https://pyviz.org/overviews/index.html">Python visualization landscape</a>, which includes
7359
<a href="https://matplotlib.org">Matplotlib</a>,
7460
<a href="https://seaborn.pydata.org">Seaborn</a>,
7561
<a href="https://plot.ly">Plotly</a>,
7662
<a href="https://altair-viz.github.io">Altair</a>,
7763
<a href="https://docs.bokeh.org/en/latest/">Bokeh</a>,
7864
<a href="http://holoviz.org">Holoviz</a>,
79-
<a href="http://vispy.org">Vispy</a> and
65+
<a href="http://vispy.org">Vispy</a>, and
8066
<a href="https://github.com/napari/napari">Napari</a>,
8167
to name a few.
8268
</p>
8369
<p>
84-
By performing parallel operations on large arrays, all at once, NumPy accelerates data-processing and
85-
visualization of large quantities of data, beyond Python's native performance levels, for data
86-
visualization at
87-
scale.
70+
NumPy's accelerated processing of large arrays allows researchers to visualize
71+
datasets far larger than native Python could handle.
8872
</p>
8973
</div>
9074
<div>
91-
<p>
92-
<a href="https://rougier.github.io/python-visualization-landscape/landscape-colors.png">
93-
<img src="images/content_images/vis-landscape.png"
94-
alt="Mindmap linking several concepts, such as Javascript, Matplotlib, d3js and OpenGL."
95-
align="left">
96-
</a>
97-
</p>
9875
</div>
9976
</div>
100-
</li>
77+
</li>

0 commit comments

Comments
 (0)