|
|
@@ -9,8 +9,6 @@
|
|
|
<meta name="viewport" content="width=device-width, initial-scale=1">
|
|
|
<meta name="theme-color" content="#ffffff" />
|
|
|
|
|
|
- <base href="">
|
|
|
-
|
|
|
<link rel="stylesheet" href="book.css">
|
|
|
<link href="https://fonts.googleapis.com/css?family=Open+Sans:300italic,400italic,600italic,700italic,800italic,400,300,600,700,800" rel="stylesheet" type="text/css">
|
|
|
<link href="https://fonts.googleapis.com/css?family=Source+Code+Pro:500" rel="stylesheet" type="text/css">
|
|
|
@@ -18,7 +16,7 @@
|
|
|
<link rel="shortcut icon" href="favicon.png">
|
|
|
|
|
|
<!-- Font Awesome -->
|
|
|
- <link rel="stylesheet" href="_FontAwesome/css/font-awesome.css">
|
|
|
+ <link rel="stylesheet" href="FontAwesome/css/font-awesome.css">
|
|
|
|
|
|
<link rel="stylesheet" href="highlight.css">
|
|
|
<link rel="stylesheet" href="tomorrow-night.css">
|
|
|
@@ -136,24 +134,24 @@
|
|
|
|
|
|
<div id="content" class="content">
|
|
|
<main>
|
|
|
- <a class="header" href="print.html#qcgpu" id="qcgpu"><h1>QCGPU</h1></a>
|
|
|
+ <a class="header" href="#qcgpu" id="qcgpu"><h1>QCGPU</h1></a>
|
|
|
<p><strong>QCGPU</strong> is a high performance, hardware accelerated quantum computer simulator written with <a href="https://rust-lang.org/">Rust</a> and <a href="https://www.khronos.org/opencl/">OpenCL</a>.</p>
|
|
|
<p>QCGPU is free and open source. The source code is available on <a href="https://github.com/qcgpu/qcgpu-rust">GitHub</a>. Please report any places where the documentation is unclear, or where examples could be added or improved by creating an <a href="https://github.com/qcgpu/qcgpu-rust/issues">issue</a>.</p>
|
|
|
-<a class="header" href="print.html#help-wanted" id="help-wanted"><h2>Help Wanted</h2></a>
|
|
|
+<a class="header" href="#help-wanted" id="help-wanted"><h2>Help Wanted</h2></a>
|
|
|
<p>Please request any functionality you would like, need or appreciate by creating an <a href="https://github.com/qcgpu/qcgpu-rust/issues">issue</a>. If you would like to add functionality or fix issues, please create a <a href="https://github.com/qcgpu/qcgpu-rust/pulls">pull request</a> and I will get back to you (usually within the day).</p>
|
|
|
<p>Any other feedback, suggestions or even tiny bits of criticism are welcome. When in doubt, file an <a href="https://github.com/qcgpu/qcgpu-rust/issues">issue</a>!</p>
|
|
|
-<a class="header" href="print.html#api-docs" id="api-docs"><h2>API Docs</h2></a>
|
|
|
+<a class="header" href="#api-docs" id="api-docs"><h2>API Docs</h2></a>
|
|
|
<p>Alongside this guide, you may also want the <a href="https://qcgpu.github.com/qcgpu/documentation">documentation</a>.</p>
|
|
|
-<a class="header" href="print.html#license" id="license"><h2>License</h2></a>
|
|
|
+<a class="header" href="#license" id="license"><h2>License</h2></a>
|
|
|
<p>QCGPU is licensed under the <a href="https://github.com/qcgpu/qcgpu-rust/blob/master/LICENSE">MIT</a> License.</p>
|
|
|
-<a class="header" href="print.html#getting-started" id="getting-started"><h1>Getting Started</h1></a>
|
|
|
-<a class="header" href="print.html#installing-the-requirements" id="installing-the-requirements"><h2>Installing The Requirements</h2></a>
|
|
|
+<a class="header" href="#getting-started" id="getting-started"><h1>Getting Started</h1></a>
|
|
|
+<a class="header" href="#installing-the-requirements" id="installing-the-requirements"><h2>Installing The Requirements</h2></a>
|
|
|
<p>To use QCGPU, you will need Rust and OpenCL installed.</p>
|
|
|
<p>The setup process for OpenCL will be different for every device. All apple devices (MacOS / OSX) will have OpenCL already installed.
|
|
|
For some hints on how to install on linux, look at the <a href="https://github.com/QCGPU/qcgpu-rust/blob/master/EC2-install.md">AWS EC2 Install Instructions</a>.
|
|
|
There is also a good chance that it is installed already. Check that <code>clinfo</code> for some other diagnostic command will run.</p>
|
|
|
<p>Rust is very easy to install. Check out <a href="https://www.rustup.rs">rustup</a>.</p>
|
|
|
-<a class="header" href="print.html#adding-the-dependency" id="adding-the-dependency"><h2>Adding The Dependency</h2></a>
|
|
|
+<a class="header" href="#adding-the-dependency" id="adding-the-dependency"><h2>Adding The Dependency</h2></a>
|
|
|
<p>To use the library with rust, you must add the following to your <code>cargo.toml</code> file:</p>
|
|
|
<pre><code class="language-toml">[dependencies]
|
|
|
qcgpu = "0.1"
|
|
|
@@ -162,10 +160,10 @@ qcgpu = "0.1"
|
|
|
<pre><code>extern crate qcgpu;
|
|
|
</code></pre>
|
|
|
<p>to <code>lib.rs</code> or <code>main.rs</code>, depending on if you are writing an executable or a library.</p>
|
|
|
-<a class="header" href="print.html#user-guide" id="user-guide"><h1>User Guide</h1></a>
|
|
|
+<a class="header" href="#user-guide" id="user-guide"><h1>User Guide</h1></a>
|
|
|
<p>This chapter covers the usage of the QCGPU library. It will also contain some information about the mathematics that each of the functions represents.</p>
|
|
|
<p>All aspects of the library are covered by this chapter, whereas complete examples (in the form of algorithm implementations) are available in the <a href="../algorithms/algorithms.html">Algorithms Chapter</a></p>
|
|
|
-<a class="header" href="print.html#quantum-registers" id="quantum-registers"><h1>Quantum Registers</h1></a>
|
|
|
+<a class="header" href="#quantum-registers" id="quantum-registers"><h1>Quantum Registers</h1></a>
|
|
|
<p>All of the simulation is done through quantum registers. QCGPU provides a struct as a register, but that contains fields to do with the OpenCL buffers and related items, so the creation of registers should be done through the provided methods.</p>
|
|
|
<p>The library is optimized for complex superpositions, so the registers are all dense. This means that the number of qubits you can initialize is directly related to the capacity / available memory of the device.</p>
|
|
|
<p>The register struct is called <code>State</code> and is available through <code>qcgpu::State</code>.</p>
|
|
|
@@ -195,9 +193,9 @@ let mut register = State::from_bit_string("|0100>", 0);
|
|
|
</code></pre></pre>
|
|
|
<p>The second argument is the same as before. The register that is outputed from this method is equivilent to the state</p>
|
|
|
<p>\[ \lvert 0100 \rangle\]</p>
|
|
|
-<a class="header" href="print.html#quantum-gates" id="quantum-gates"><h1>Quantum Gates</h1></a>
|
|
|
+<a class="header" href="#quantum-gates" id="quantum-gates"><h1>Quantum Gates</h1></a>
|
|
|
<p>Gates are used to manipulate quantum registers and to implement quantum algorithms.</p>
|
|
|
-<a class="header" href="print.html#built-in-gates" id="built-in-gates"><h2>Built In Gates</h2></a>
|
|
|
+<a class="header" href="#built-in-gates" id="built-in-gates"><h2>Built In Gates</h2></a>
|
|
|
<p>There are a number of gates built in to QCGPU. They can all be applied the same way:</p>
|
|
|
<pre><pre class="playpen"><code class="language-rust">
|
|
|
# #![allow(unused_variables)]
|
|
|
@@ -240,7 +238,7 @@ use qcgpu::State;
|
|
|
let mut state = State::new(5, 0);
|
|
|
state.apply_controlled_gate(x(), 0, 1);
|
|
|
#}</code></pre></pre>
|
|
|
-<a class="header" href="print.html#user-defined-gates" id="user-defined-gates"><h2>User Defined Gates</h2></a>
|
|
|
+<a class="header" href="#user-defined-gates" id="user-defined-gates"><h2>User Defined Gates</h2></a>
|
|
|
<p>Gates in QCGPU are represented by the <code>Gate</code> struct, available through <code>qcgpu::Gate</code>.</p>
|
|
|
<p>It is defined as follows:</p>
|
|
|
<pre><pre class="playpen"><code class="language-rust">
|
|
|
@@ -279,9 +277,9 @@ let x = Gate {
|
|
|
let mut state = State::new(1, 0);
|
|
|
state.apply_gate(x, 0);
|
|
|
#}</code></pre></pre>
|
|
|
-<a class="header" href="print.html#quantum-operations" id="quantum-operations"><h1>Quantum Operations</h1></a>
|
|
|
+<a class="header" href="#quantum-operations" id="quantum-operations"><h1>Quantum Operations</h1></a>
|
|
|
<p>There are a number of operations you can preform on quantum registers with QCGPU.</p>
|
|
|
-<a class="header" href="print.html#measurement" id="measurement"><h2>Measurement</h2></a>
|
|
|
+<a class="header" href="#measurement" id="measurement"><h2>Measurement</h2></a>
|
|
|
<p>You can measure the register in two ways. You can either do a single measurement and return an integer with the measured value or you can measure multiple times and return a <code>HashMap<String, i32></code>, with the key being the bitstring, and the value being the number of times it was measured.</p>
|
|
|
<p>The measurements don't collapse the state.</p>
|
|
|
<p>They are used as follows:</p>
|
|
|
@@ -303,7 +301,7 @@ use qcgpu::State;
|
|
|
let mut state = State::new(5, 0);
|
|
|
state.measure_first(3, 1000); // Measures the first 3 qubits 1000 times, returns a HashMap<String, i32>
|
|
|
#}</code></pre></pre>
|
|
|
-<a class="header" href="print.html#probability" id="probability"><h2>Probability</h2></a>
|
|
|
+<a class="header" href="#probability" id="probability"><h2>Probability</h2></a>
|
|
|
<p>QCGPU provides another method for getting the probability of each outcome.</p>
|
|
|
<p>The probability is calculated for a state \(\lvert \psi \rangle = \sum_{j = 0}^{2^n - 1} \alpha_j \lvert j \rangle\),</p>
|
|
|
<p>\[P(j) = |\alpha_j|^2\]</p>
|
|
|
@@ -317,7 +315,7 @@ let mut state = State::new(1, 0);
|
|
|
state.h(0);
|
|
|
state.get_probabilities(); // [0.5, 0.5]
|
|
|
#}</code></pre></pre>
|
|
|
-<a class="header" href="print.html#examples" id="examples"><h1>Examples</h1></a>
|
|
|
+<a class="header" href="#examples" id="examples"><h1>Examples</h1></a>
|
|
|
<p>Here is a complete example of using QCGPU to simulate the bell state.</p>
|
|
|
<pre><pre class="playpen"><code class="language-rust">extern crate qcgpu;
|
|
|
|
|
|
@@ -341,7 +339,7 @@ fn main() {
|
|
|
<pre><code class="language-shell">Measured: {"00": 516, "11": 484}
|
|
|
</code></pre>
|
|
|
<p>For more, non trivial examples, see the <a href="../algorithms/algorithms.html">Algorithms</a></p>
|
|
|
-<a class="header" href="print.html#decoherence" id="decoherence"><h1>Decoherence</h1></a>
|
|
|
+<a class="header" href="#decoherence" id="decoherence"><h1>Decoherence</h1></a>
|
|
|
<p>QCGPU provides a way to easily simulate the effects of decoherence on the quantum computer. The effects are simulated by a random gate, corresponding to a random rotation around the \(z\) axis of the bloch sphere.</p>
|
|
|
<p>The angle of the rotation is a normal distrobuted value, with the varience as the strength factor <code>d</code>.</p>
|
|
|
<p>To avoid performance costs when not being used, decoherence can be enabled via a feature.</p>
|
|
|
@@ -366,10 +364,10 @@ register.set_decoherence(0.4);
|
|
|
#fn main() {
|
|
|
register.decohere();
|
|
|
#}</code></pre></pre>
|
|
|
-<a class="header" href="print.html#algorithms" id="algorithms"><h1>Algorithms</h1></a>
|
|
|
+<a class="header" href="#algorithms" id="algorithms"><h1>Algorithms</h1></a>
|
|
|
<p>The following chapter details some implementations of non trivial quantum algorithms. They are based on the source code from <a href="http://www.libquantum.de/">libquantum</a> along with the lecture notes from <a href="https://cs.uwaterloo.ca/%7Ewatrous/LectureNotes.html">the University of Waterloo</a>.</p>
|
|
|
<p>Note that some implementations may not yet be complete.</p>
|
|
|
-<a class="header" href="print.html#bernstein-vazirani-algorithm" id="bernstein-vazirani-algorithm"><h1>Bernstein-Vazirani Algorithm</h1></a>
|
|
|
+<a class="header" href="#bernstein-vazirani-algorithm" id="bernstein-vazirani-algorithm"><h1>Bernstein-Vazirani Algorithm</h1></a>
|
|
|
<p>This algorithm finds a hidden integer \(a \in { 0, 1}^n\)from
|
|
|
an oracle \(f_a\)which returns a bit \(a \cdot x \equiv \sum_i a_i x_i \mod 2\)
|
|
|
for an input \(x \in {0,1}^n\).</p>
|
|
|
@@ -380,7 +378,7 @@ oracle with the inputs \(x = 1,2,\dots,2^i,2^{n-1}\), where each
|
|
|
query reveals the \(i\)th bit of \(a\) (\(a_i\)).
|
|
|
This is the optimal classical solution, and is O(n). Using a quantum oracle and the
|
|
|
Bernstein-Vazirani algorithm, \(a\) can be found with just one query to the oracle.</p>
|
|
|
-<a class="header" href="print.html#the-algorithm" id="the-algorithm"><h2>The Algorithm</h2></a>
|
|
|
+<a class="header" href="#the-algorithm" id="the-algorithm"><h2>The Algorithm</h2></a>
|
|
|
<ol>
|
|
|
<li>Initialize \(n\) qubits in the state \(\lvert 0, \dots, 0\rangle\).</li>
|
|
|
<li>Apply the Hadamard gate \(H\) to each qubit.</li>
|
|
|
@@ -423,7 +421,7 @@ fn main() {
|
|
|
// Measurement Results: {"0000000001100101": 1000}
|
|
|
}
|
|
|
</code></pre></pre>
|
|
|
-<a class="header" href="print.html#deutsch-jozsa-algorithm" id="deutsch-jozsa-algorithm"><h1>Deutsch-Jozsa Algorithm</h1></a>
|
|
|
+<a class="header" href="#deutsch-jozsa-algorithm" id="deutsch-jozsa-algorithm"><h1>Deutsch-Jozsa Algorithm</h1></a>
|
|
|
<p>This algorithm was the first to show that quantum computers could have a speedup over classical computers.</p>
|
|
|
<p>Consider a function \(f(x)\) which takes an input of an \(n\)-bit string \(x\) and returns 0 or 1.</p>
|
|
|
<p>Suppose that \(f(x)\) is either a <strong>constant</strong> function that has the same value \(c \in {0, 1}, \forall x\), or a <strong>balanced</strong> function, where the value is 0 for half of the inputs, and 1 for the other half.</p>
|
|
|
@@ -431,7 +429,7 @@ fn main() {
|
|
|
<p>Using classical computing, in the worst case, this requires \(2^{n-1}+1\) function evaluations.
|
|
|
Using quantum computing, this can be done with just one function evaluation.</p>
|
|
|
<p>The function \(f\), to be used in a quantum computer, must be specified by an oracle circuit \(U_{f}\) such that \(U_{f} \lvert x \rangle = (-1)^{f(x)}\lvert x \rangle\).</p>
|
|
|
-<a class="header" href="print.html#the-algorithm-1" id="the-algorithm-1"><h2>The Algorithm</h2></a>
|
|
|
+<a class="header" href="#the-algorithm-1" id="the-algorithm-1"><h2>The Algorithm</h2></a>
|
|
|
<ol>
|
|
|
<li>Initialize \(n\) qubits in the state \(\lvert 0, \dots, 0\rangle\).</li>
|
|
|
<li>Apply the Hadamard gate \(H\) to each qubit.</li>
|
|
|
@@ -490,7 +488,7 @@ fn main() {
|
|
|
);
|
|
|
}
|
|
|
</code></pre></pre>
|
|
|
-<a class="header" href="print.html#grovers-algorithm" id="grovers-algorithm"><h1>Grovers Algorithm</h1></a>
|
|
|
+<a class="header" href="#grovers-algorithm" id="grovers-algorithm"><h1>Grovers Algorithm</h1></a>
|
|
|
<p>Given an unstructured set \(N = {a_1, a_2,\dots,a_n}\), find
|
|
|
a given element \(a_i \in N\).</p>
|
|
|
<p>This implementation looks for a given number \(target\) in the set \({0,1,\dots, \text{regwidth}}\).</p>
|
|
|
@@ -614,10 +612,10 @@ fn get_bit(number: i32, n: i32) -> i32 {
|
|
|
0
|
|
|
}
|
|
|
</code></pre></pre>
|
|
|
-<a class="header" href="print.html#shors-algorithm" id="shors-algorithm"><h1>Shors Algorithm</h1></a>
|
|
|
+<a class="header" href="#shors-algorithm" id="shors-algorithm"><h1>Shors Algorithm</h1></a>
|
|
|
<p>This algorithm finds the prime factors (\)u\) and \(v\)) of an odd, composite integer \(n\),
|
|
|
that is not a prime power.</p>
|
|
|
-<a class="header" href="print.html#the-algorithm-2" id="the-algorithm-2"><h2>The Algorithm</h2></a>
|
|
|
+<a class="header" href="#the-algorithm-2" id="the-algorithm-2"><h2>The Algorithm</h2></a>
|
|
|
<p>(pseudo code)</p>
|
|
|
<pre><code class="language-pseudo">Repeat
|
|
|
Randomly choose \\(a \in \{ 2, \dots, n - 1 \}\\)
|
|
|
@@ -678,7 +676,7 @@ fn main() {
|
|
|
}
|
|
|
}
|
|
|
</code></pre></pre>
|
|
|
-<a class="header" href="print.html#order-finding" id="order-finding"><h2>Order Finding</h2></a>
|
|
|
+<a class="header" href="#order-finding" id="order-finding"><h2>Order Finding</h2></a>
|
|
|
<p>Given a positive integer \(n \geq 2\) and an element \(a \in \mathbb{Z}_n^* \),
|
|
|
Find the order of \(a\) in \(\mathbb{Z}_n^*\).</p>
|
|
|
<p>Order finding is the only quantum part in shors algorithm.</p>
|
|
|
@@ -700,7 +698,7 @@ fn find_order(a: i32, n: i32) -> i32 {
|
|
|
unimplemented!()
|
|
|
}
|
|
|
#}</code></pre></pre>
|
|
|
-<a class="header" href="print.html#super-dense-coding" id="super-dense-coding"><h1>Super Dense Coding</h1></a>
|
|
|
+<a class="header" href="#super-dense-coding" id="super-dense-coding"><h1>Super Dense Coding</h1></a>
|
|
|
<p>if Alice and Bob share a pair of entangled qubits, then Alice can encode two classical bits into her one entangled qubit,
|
|
|
send it to Bob, and Bob can decode it with the help of his entangled qubit.</p>
|
|
|
<pre><pre class="playpen"><code class="language-rust">extern crate qcgpu;
|
|
|
@@ -780,17 +778,12 @@ fn main() {
|
|
|
|
|
|
|
|
|
|
|
|
- <script type="text/javascript">
|
|
|
- document.addEventListener('DOMContentLoaded', function() {
|
|
|
- window.print();
|
|
|
- })
|
|
|
- </script>
|
|
|
-
|
|
|
-
|
|
|
-
|
|
|
|
|
|
|
|
|
<script src="searchindex.js" type="text/javascript" charset="utf-8"></script>
|
|
|
+ <script>
|
|
|
+ var path_to_root = "";
|
|
|
+ </script>
|
|
|
|
|
|
|
|
|
<script src="elasticlunr.min.js" type="text/javascript" charset="utf-8"></script>
|
|
|
@@ -805,5 +798,17 @@ fn main() {
|
|
|
<!-- Custom JS scripts -->
|
|
|
|
|
|
|
|
|
+
|
|
|
+
|
|
|
+ <script type="text/javascript">
|
|
|
+ window.addEventListener('load', function() {
|
|
|
+ MathJax.Hub.Register.StartupHook('End', function() {
|
|
|
+ window.setTimeout(window.print, 100);
|
|
|
+ });
|
|
|
+ });
|
|
|
+ </script>
|
|
|
+
|
|
|
+
|
|
|
+
|
|
|
</body>
|
|
|
</html>
|
Deployment Bot (from Travis CI)
