coding ...

owlbarn · Nov 22, 2023 · a8210cb · a8210cb
1 parent 4868377
commit a8210cb
Show file tree

Hide file tree

Showing 34 changed files with 34 additions and 33 deletions.
diff --git a/docs/cn/aeos.html b/docs/cn/aeos.html
@@ -8,7 +8,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="automatic-empirical-tuning">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="automatic-empirical-tuning">
 <h1>Automatic Empirical Tuning</h1>
 <p>The behaviours of a software system are controlled by many of its parameters. These parameters can significantly impact the performance of the software. Assigning optimal values to the parameters to achieve the best performance is one core task for software optimisation. This chapter reveals the technical details of the parameter tunning module in Owl.</p>
 <section class="level2" id="what-is-parameter-tuning">

diff --git a/docs/cn/algodiff.html b/docs/cn/algodiff.html
@@ -7,7 +7,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="algorithmic-differentiation">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="algorithmic-differentiation">
 <h1>Algorithmic Differentiation</h1>
 <p>In science and engineering it is often necessary to study the relationship between two or more quantities, where changing of one quantity leads to changes of others. For example, in describing the motion an object, we denote velocity <span class="math inline">\(v\)</span> of an object with the change of the distance regarding time:</p>
 <p><span class="math display">\[v = \lim_{\Delta~t}\frac{\Delta~s}{\Delta~t} = \frac{ds}{dt}.\]</span> {#eq:algodiff:def}</p>

diff --git a/docs/cn/architecture.html b/docs/cn/architecture.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="architecture-overview">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="architecture-overview">
 <h1>Architecture Overview</h1>
 <p>Owl is an emerging library developed in the OCaml language for scientific and engineering computing. It focuses on providing a comprehensive set of high-level numerical functions so that developers can quickly build up any data analytical applications. After over one-year intensive development and continuous optimisation, Owl has evolved into a powerful software system with competitive performance to mainstream numerical libraries. Meanwhile, Owl’s overall architecture remains simple and elegant. Its small code base can be easily managed by a small group of developers.</p>
 <p>In this chapter, we first present Owl’s design, core components, and its key functionality. We show that Owl benefits greatly from OCaml’s module system which not only allows us to write concise generic code with superior performance, but also leads to a very unique design to enable parallel and distributed computing. OCaml’s static type checking significantly reduces potential bugs and accelerates the development cycle. We also share the knowledge and lessons learnt from building up a full-fledge system for scientific computing with the functional programming community.</p>

diff --git a/docs/cn/backend.html b/docs/cn/backend.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="compiler-backends">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="compiler-backends">
 <h1>Compiler Backends</h1>
 <p>For a numerical library, it is always beneficial and a challenge to extend to multiple execution backends. We have seen how we support accelerators such as GPU by utilising symbolic representation and computation graph standard such as ONNX. In this chapter we introduce how Owl can be used on more edge-oriented backends, including JavaScript and MirageOS. We also introduce the <code>base</code> library in Owl, since this pure OCaml library is built to support these backends.</p>
 <section class="level2" id="base-library">

diff --git a/docs/cn/case-image-inception.html b/docs/cn/case-image-inception.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---image-recognition">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---image-recognition">
 <h1>Case - Image Recognition</h1>
 <p>How can a computer take an image and answer questions like “what is in this picture? A cat, dog, or something else?” In the recent few years the machine learning community has been making tremendous progress on tackling this difficult problem. In particular, Deep Neural Network (DNN) is able to achieve reasonable performance on visual recognition tasks – matching or even exceeding human performance in some domains.</p>
 <p>We have introduced the neural network module in the previous chapter. In this chapter, we will show one specific example that is built on the neural network module: using the InceptionV3 architecture to perform the image classification task.</p>

diff --git a/docs/cn/case-nst.html b/docs/cn/case-nst.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---neural-style-transfer">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---neural-style-transfer">
 <h1>Case - Neural Style Transfer</h1>
 <p>What is Neural Style Transfer (NST)? It is “the process of using DNN to migrate the semantic content of one image to different styles”, a pretty cool application of the deep neural network. The idea is actually very simple. As the <span data-cites="fig:case-nst:example_01" class="citation">[@fig:case-nst:example_01]</span> shows, this application takes two images A and B as input. Let’s say A is a normal street view, and B is “The Starry Night” of Vincent van Gogh. We then specify A as the “content image” and B as the “style image”. What a NST application can produce is a new street view image with the style of Van Gogh. If you want another style, just replace image B and run the application again. Impressionism, abstractionism, classical art, you name it. Or you can also apply the same “Starry Sky” style to any other images.</p>
 <figure>

diff --git a/docs/cn/case-obj-detect.html b/docs/cn/case-obj-detect.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---instance-segmentation">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---instance-segmentation">
 <h1>Case - Instance Segmentation</h1>
 <p>Computer vision is a field dealing with many different automated tasks whose goal are to give high-level descriptions of images and videos. It has been applied to a wide variety of domains ranging from highly technical ones, such as automatic tagging of satellite images, analysis of medical images etc., to more mundane applications, including categorising pictures in your phone, making your face into an emoji, etc. This field has seen tremendous progress since 2012, when <a href="https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks">A. Krizhevsky et al.</a> used the first deep learning approach to computer vision, crushing all their opponents in the <a href="http://image-net.org/challenges/LSVRC/2012/results.html">ImageNet challenge</a>. We have already discussed this image recognition task in the previous chapter.</p>
 <p>In this chapter, we are going to introduce another classical computer vision task: the <em>Instance Segmentation</em>, which is to give different labels for separate instances of objects in an image. We will discuss its connection with other similar applications, how the deep neural network is constructed in Owl, and how the network, loaded with pre-trained weights, can be used to process users’ input images. We have also included an online demo for this application. Let’s begin!</p>

diff --git a/docs/cn/case-recommender.html b/docs/cn/case-recommender.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---recommender-system">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="case---recommender-system">
 <h1>Case - Recommender System</h1>
 <section class="level2" id="introduction">
 <h2>Introduction</h2>

diff --git a/docs/cn/cgraph.html b/docs/cn/cgraph.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="computation-graph">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="computation-graph">
 <h1>Computation Graph</h1>
 <p>This chapter first gives a bird’s-eye-view on the computation graph in Owl. Then we will continue to cover the design and implementation details of the computation graph and how it is fitted into Owl’s functor stack, and its implications on the architecture of numerical systems.</p>
 <section class="level2" id="introduction">

diff --git a/docs/cn/constant.html b/docs/cn/constant.html
@@ -2,7 +2,7 @@
       window.dataLayer = window.dataLayer || [];
       function gtag(){dataLayer.push(arguments);}
       gtag('js', new Date());
-      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="constants-and-metric-system">
+      gtag('config', 'UA-123353217-1');</script></head><body><div class="title-bar"><div class="title"><h1>OCaml Scientific Computing Tutorials</h1><h5></h5><nav><a href="index.html">Home</a><a href="toc.html">Table of Contents</a><a href="https://ocaml.xyz/api/">API</a><a href="https://blog.ocaml.xyz">经典算法</a><a href="../index.html">英文版</a></nav></div></div><div class="wrap"><div class="left-column"><a class="to-chapter" href="toc.html"><small>Back</small><h5>Table of Contents</h5></a></div><article class="main-body"><section class="level1" id="constants-and-metric-system">
 <h1>Constants and Metric System</h1>
 <p>In many scientific computing problems, numbers are not abstract but reflect the realistic meanings. In other words, these numbers only make sense on top of a well-defined metric system.</p>
 <section class="level2" id="what-is-a-metric-system">