mirror of
https://github.com/rasbt/python_reference.git
synced 2024-10-05 20:49:27 +00:00
note about numpy in-place ops
This commit is contained in:
rasbt
parent
b75d40b8cb
commit
018a0b281e
BIN
Images/literature.png
Normal file
BIN
Images/literature.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 6.5 KiB |
BIN
Images/literature_small.png
Normal file
BIN
Images/literature_small.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 4.4 KiB |
|
|
@ -235,6 +235,8 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
<br>
|
||||
<br>
|
||||
|
||||
|
||||
|
||||
<table cellpadding="2" cellspacing="0" style="page-break-before: always">
|
||||
<col width="383">
|
||||
<col width="387">
|
||||
|
|
@ -243,7 +245,7 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
<col width="511">
|
||||
<col width="690">
|
||||
<tr>
|
||||
<td height="57" bgcolor="#0066cc" style="border: none; padding: 0in">
|
||||
<td height="59" bgcolor="#0066cc" style="border: none; padding: 0in">
|
||||
<p align="center"><font color="#ffffff"><font face="Arial"><font size="5" style="font-size: 18pt"><b>Task</b></font></font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#0066cc" style="border: none; padding: 0in">
|
||||
|
|
@ -264,9 +266,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td colspan="6" height="70" bgcolor="#ffffff" style="border: none; padding: 0in" id="creating">
|
||||
<td colspan="6" height="70" bgcolor="#ffffff" style="border: none; padding: 0in">
|
||||
<p align="center"><font face="Arial"><font size="4" style="font-size: 14pt"><b>CREATING
|
||||
MATRICES </b></font></font><br>[<a href='#cheat_overview'>back to cheat sheet overview</a>]</p>
|
||||
MATRICES</b></font></font></p>
|
||||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
|
|
@ -313,10 +315,15 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
3</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
a = np.array([[1],[2],[3]])<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
a<br>array([[1],<br> [2],<br>
|
||||
[3]])</font></font></p>
|
||||
<p align="left"><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font>
|
||||
<font face="Andale Mono"><font size="1" style="font-size: 8pt">a</font></font><font color="#2fff12">
|
||||
</font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">=
|
||||
np.array([1,2,3]).reshape(1,3)</font></font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><br></font></font><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font>
|
||||
<font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">b.shape<br>(1,
|
||||
3)</font></font></font></p>
|
||||
<p align="left"><br>
|
||||
</p>
|
||||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#0000ff"><font size="2" style="font-size: 10pt"><b>R></b></font></font>
|
||||
|
|
@ -345,6 +352,14 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
b = np.array([1,2,3])<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
b<br>array([1, 2, 3])</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt">#
|
||||
note that numpy doesn't have <br># explicit “row-vectors”, but
|
||||
1-D <br># arrays</font></font></p>
|
||||
<p style="margin-bottom: 0in"><font face="Liberation Serif, serif"><font size="3" style="font-size: 12pt"><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font><font color="#2fff12">
|
||||
</font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">b.shape</font></font></font></font></font></p>
|
||||
<p style="margin-bottom: 0in"><font face="Andale Mono"><font size="1" style="font-size: 8pt">(3,)</font></font></p>
|
||||
<p align="left"><br>
|
||||
</p>
|
||||
</td>
|
||||
<td bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#0000ff"><font size="2" style="font-size: 10pt"><b>R></b></font></font>
|
||||
|
|
@ -434,7 +449,7 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
np.ones([3,2])<br>array([[ 1., 1.],<br>
|
||||
np.ones((3,2))<br>array([[ 1., 1.],<br>
|
||||
[ 1., 1.],<br> [ 1., 1.]])</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
|
|
@ -519,9 +534,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td colspan="6" height="128" bgcolor="#ffffff" style="border: none; padding: 0in" id='accessing'>
|
||||
<td colspan="6" height="128" bgcolor="#ffffff" style="border: none; padding: 0in">
|
||||
<p align="center"><font face="Arial"><font size="4" style="font-size: 14pt"><b>ACCESSING
|
||||
MATRIX ELEMENTS</b></font></font><br>[<a href='#cheat_overview'>back to cheat sheet overview</a>]</p>
|
||||
MATRIX ELEMENTS</b></font></font></p>
|
||||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
|
|
@ -714,9 +729,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td colspan="6" height="128" bgcolor="#ffffff" style="border: none; padding: 0in" id="manipulating">
|
||||
<td colspan="6" height="128" bgcolor="#ffffff" style="border: none; padding: 0in">
|
||||
<p align="center"><font face="Arial"><font size="4" style="font-size: 14pt"><b>MANIPULATING
|
||||
SHAPE AND DIMENSIONS</b></font></font><br>[<a href='#cheat_overview'>back to cheat sheet overview</a>]</p>
|
||||
SHAPE AND DIMENSIONS</b></font></font></p>
|
||||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
|
|
@ -772,10 +787,10 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
A = np.array([[1,2,3],[4,5,6],[7,8,9]])<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A<br>array([[1, 2, 3],<br> [4, 5, 9],<br>
|
||||
[7, 8, 9]])<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
total_elements = A.shape[0] * A.shape[1]<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
B = A.reshape(1, total_elements) <br><br># or
|
||||
A.reshape(1,9)<br># Alternative: A.shape = (1,9) <br>#
|
||||
to change the array in place<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
total_elements = np.prod(A.shape)</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
B = A.reshape(1, total_elements) <br><br># alternative
|
||||
shortcut:<br># A.reshape(1,-1)<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
B<br>array([[1, 2, 3, 4, 5, 6, 7, 8, 9]])</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
|
|
@ -885,9 +900,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td colspan="6" height="114" bgcolor="#ffffff" style="border: none; padding: 0in" id='basic_ops'>
|
||||
<td colspan="6" height="114" bgcolor="#ffffff" style="border: none; padding: 0in">
|
||||
<p align="center"><font face="Arial"><font size="4" style="font-size: 14pt"><b>BASIC
|
||||
MATRIX OPERATIONS</b></font></font><br>[<a href='#cheat_overview'>back to cheat sheet overview</a>]</p>
|
||||
MATRIX OPERATIONS</b></font></font></p>
|
||||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
|
|
@ -912,6 +927,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
A + 2<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A - 2<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A / 2</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt">#
|
||||
Note that NumPy was optimized for<br># in-place assignments<br>#
|
||||
e.g., A += A instead of <br># A = A + A</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#0000ff"><font size="2" style="font-size: 10pt"><b>R></b></font></font>
|
||||
|
|
@ -1025,6 +1043,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
A + A<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A - A<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A / A</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt">#
|
||||
Note that NumPy was optimized for<br># in-place assignments<br>#
|
||||
e.g., A += A instead of <br># A = A + A</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#0000ff"><font size="2" style="font-size: 10pt"><b>R></b></font></font>
|
||||
|
|
@ -1232,9 +1253,9 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td colspan="6" height="202" bgcolor="#ffffff" style="border: none; padding: 0in" id='advanced_ops'>
|
||||
<td colspan="6" height="202" bgcolor="#ffffff" style="border: none; padding: 0in">
|
||||
<p align="center"><font face="Arial"><font size="4" style="font-size: 14pt"><b>ADVANCED
|
||||
MATRIX OPERATIONS</b></font></font><br>[<a href='#cheat_overview'>back to cheat sheet overview</a>]</p>
|
||||
MATRIX OPERATIONS</b></font></font></p>
|
||||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
|
|
@ -1328,7 +1349,7 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
<td height="539" bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: none; border-right: 1px solid #000000; padding-top: 0in; padding-bottom: 0in; padding-left: 0in; padding-right: 0.02in">
|
||||
<td height="541" bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: none; border-right: 1px solid #000000; padding-top: 0in; padding-bottom: 0in; padding-left: 0in; padding-right: 0.02in">
|
||||
<p align="left"><font face="Arial"><font size="2" style="font-size: 10pt"><b>Generating
|
||||
a Gaussian dataset:<br><br>creating random vectors from the
|
||||
multivariate normal<br>distribution given mean and covariance
|
||||
|
|
@ -1395,6 +1416,8 @@ Bezanson, J., Karpinski, S., Shah, V.B. and Edelman, A. (2012), “Julia: A fast
|
|||
</td>
|
||||
</tr>
|
||||
</table>
|
||||
|
||||
|
||||
<br>
|
||||
|
||||
<br><br><font size=0.6>
|
||||
|
|
|
|||
|
|
@ -5,7 +5,7 @@
|
|||
<title></title>
|
||||
<meta name="generator" content="LibreOffice 4.2.4.2 (MacOSX)">
|
||||
<meta name="created" content="0;0">
|
||||
<meta name="changed" content="20140616;230046743144000">
|
||||
<meta name="changed" content="20140622;123733939958000">
|
||||
<style type="text/css">
|
||||
<!--
|
||||
td p { color: #000000; font-family: "Verdana"; font-size: 12pt }
|
||||
|
|
@ -22,7 +22,7 @@
|
|||
<col width="511">
|
||||
<col width="690">
|
||||
<tr>
|
||||
<td height="58" bgcolor="#0066cc" style="border: none; padding: 0in">
|
||||
<td height="59" bgcolor="#0066cc" style="border: none; padding: 0in">
|
||||
<p align="center"><font color="#ffffff"><font face="Arial"><font size="5" style="font-size: 18pt"><b>Task</b></font></font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#0066cc" style="border: none; padding: 0in">
|
||||
|
|
@ -92,12 +92,12 @@
|
|||
3</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font><font face="Andale Mono"><font size="1" style="font-size: 8pt">
|
||||
</font></font><font face="Andale Mono"><font size="1" style="font-size: 8pt">a</font></font><font color="#2fff12"><font face="AndaleMono, serif"><font size="4" style="font-size: 14pt">
|
||||
</font></font></font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">=
|
||||
<p align="left"><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font>
|
||||
<font face="Andale Mono"><font size="1" style="font-size: 8pt">a</font></font><font color="#2fff12">
|
||||
</font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">=
|
||||
np.array([1,2,3]).reshape(1,3)</font></font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><br></font></font><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font><font face="Andale Mono"><font size="1" style="font-size: 8pt">
|
||||
</font></font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">b.shape<br>(1,
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><br></font></font><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font>
|
||||
<font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">b.shape<br>(1,
|
||||
3)</font></font></font></p>
|
||||
<p align="left"><br>
|
||||
</p>
|
||||
|
|
@ -132,10 +132,10 @@
|
|||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt">#
|
||||
note that numpy doesn't have <br># explicit “row-vectors”, but
|
||||
1-D <br># arrays</font></font></p>
|
||||
<p style="margin-bottom: 0in"><font face="Liberation Serif, serif"><font size="3" style="font-size: 12pt"><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font><font color="#2fff12"><font face="AndaleMono, serif"><font size="4" style="font-size: 14pt"><b>
|
||||
</b></font></font></font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">b.shape</font></font></font></font></font></p>
|
||||
<p style="margin-bottom: 0in"><font face="Liberation Serif, serif"><font size="3" style="font-size: 12pt"><font color="#ff420e"><font face="Andale Mono"><font size="2" style="font-size: 10pt"><b>P></b></font></font></font><font color="#2fff12">
|
||||
</font><font color="#000000"><font face="Andale Mono"><font size="1" style="font-size: 8pt">b.shape</font></font></font></font></font></p>
|
||||
<p style="margin-bottom: 0in"><font face="Andale Mono"><font size="1" style="font-size: 8pt">(3,)</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><br></font></font><br>
|
||||
<p align="left"><br>
|
||||
</p>
|
||||
</td>
|
||||
<td bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
|
|
@ -704,6 +704,9 @@
|
|||
A + 2<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A - 2<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A / 2</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt">#
|
||||
Note that NumPy was optimized for<br># in-place assignments<br>#
|
||||
e.g., A += A instead of <br># A = A + A</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#cfe7f5" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#0000ff"><font size="2" style="font-size: 10pt"><b>R></b></font></font>
|
||||
|
|
@ -817,6 +820,9 @@
|
|||
A + A<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A - A<br><br><font color="#ff420e"><font size="2" style="font-size: 10pt"><b>P></b></font></font>
|
||||
A / A</font></font></p>
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt">#
|
||||
Note that NumPy was optimized for<br># in-place assignments<br>#
|
||||
e.g., A += A instead of <br># A = A + A</font></font></p>
|
||||
</td>
|
||||
<td bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: 1px dotted #000000; border-right: 1px dotted #000000; padding: 0in 0.02in">
|
||||
<p align="left"><font face="Andale Mono"><font size="1" style="font-size: 8pt"><font color="#0000ff"><font size="2" style="font-size: 10pt"><b>R></b></font></font>
|
||||
|
|
@ -1120,7 +1126,7 @@
|
|||
</td>
|
||||
</tr>
|
||||
<tr valign="top">
|
||||
<td height="540" bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: none; border-right: 1px solid #000000; padding-top: 0in; padding-bottom: 0in; padding-left: 0in; padding-right: 0.02in">
|
||||
<td height="541" bgcolor="#ffffff" style="border-top: none; border-bottom: none; border-left: none; border-right: 1px solid #000000; padding-top: 0in; padding-bottom: 0in; padding-left: 0in; padding-right: 0.02in">
|
||||
<p align="left"><font face="Arial"><font size="2" style="font-size: 10pt"><b>Generating
|
||||
a Gaussian dataset:<br><br>creating random vectors from the
|
||||
multivariate normal<br>distribution given mean and covariance
|
||||
|
|
|
|||
Loading…
Reference in New Issue
Block a user