[Scipy-svn] r2176 - in trunk/Lib/sandbox/ann: . doc
scipy-svn at scipy.org
scipy-svn at scipy.org
Tue Aug 22 10:03:05 EDT 2006
Author: fred.mailhot
Date: 2006-08-22 09:03:02 -0500 (Tue, 22 Aug 2006)
New Revision: 2176
Added:
trunk/Lib/sandbox/ann/doc/
trunk/Lib/sandbox/ann/doc/mlp.html
trunk/Lib/sandbox/ann/doc/rbf.html
trunk/Lib/sandbox/ann/doc/srn.html
Log:
HTML files for {mlp/rbf/srn}.py generated by pydoc uploaded.
Added: trunk/Lib/sandbox/ann/doc/mlp.html
===================================================================
--- trunk/Lib/sandbox/ann/doc/mlp.html 2006-08-22 13:46:37 UTC (rev 2175)
+++ trunk/Lib/sandbox/ann/doc/mlp.html 2006-08-22 14:03:02 UTC (rev 2176)
@@ -0,0 +1,93 @@
+
+<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
+<html><head><title>Python: module mlp</title>
+</head><body bgcolor="#f0f0f8">
+
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="heading">
+<tr bgcolor="#7799ee">
+<td valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"> <br><big><big><strong>mlp</strong></big></big></font></td
+><td align=right valign=bottom
+><font color="#ffffff" face="helvetica, arial"><a href=".">index</a><br><a href="file:/home/tilde/programming/SoC/scipy/Lib/sandbox/ann/mlp.py">/home/tilde/programming/SoC/scipy/Lib/sandbox/ann/mlp.py</a></font></td></tr></table>
+ <p><tt># <a href="#mlp">mlp</a>.py<br>
+# by: Fred Mailhot<br>
+# last mod: 2006-08-19</tt></p>
+<p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#aa55cc">
+<td colspan=3 valign=bottom> <br>
+<font color="#fffff" face="helvetica, arial"><big><strong>Modules</strong></big></font></td></tr>
+
+<tr><td bgcolor="#aa55cc"><tt> </tt></td><td> </td>
+<td width="100%"><table width="100%" summary="list"><tr><td width="25%" valign=top><a href="numpy.html">numpy</a><br>
+</td><td width="25%" valign=top></td><td width="25%" valign=top></td><td width="25%" valign=top></td></tr></table></td></tr></table><p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#ee77aa">
+<td colspan=3 valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"><big><strong>Classes</strong></big></font></td></tr>
+
+<tr><td bgcolor="#ee77aa"><tt> </tt></td><td> </td>
+<td width="100%"><dl>
+<dt><font face="helvetica, arial"><a href="mlp.html#mlp">mlp</a>
+</font></dt></dl>
+ <p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#ffc8d8">
+<td colspan=3 valign=bottom> <br>
+<font color="#000000" face="helvetica, arial"><a name="mlp">class <strong>mlp</strong></a></font></td></tr>
+
+<tr bgcolor="#ffc8d8"><td rowspan=2><tt> </tt></td>
+<td colspan=2><tt>Class to define, train and test a multilayer perceptron.<br> </tt></td></tr>
+<tr><td> </td>
+<td width="100%">Methods defined here:<br>
+<dl><dt><a name="mlp-__init__"><strong>__init__</strong></a>(self, ni, nh, no, f<font color="#909090">='linear'</font>, w<font color="#909090">=None</font>)</dt><dd><tt>Set up instance of <a href="#mlp">mlp</a>. Initial weights are drawn from a <br>
+zero-mean Gaussian w/ variance is scaled by fan-in.<br>
+Input:<br>
+ ni - <int> # of inputs<br>
+ nh - <int> # of hidden units<br>
+ no - <int> # of outputs<br>
+ f - <str> output activation fxn<br>
+ w - <array of float> vector of initial weights</tt></dd></dl>
+
+<dl><dt><a name="mlp-errfxn"><strong>errfxn</strong></a>(self, w, x, t)</dt><dd><tt>Return vector of squared-errors for the leastsq optimizer</tt></dd></dl>
+
+<dl><dt><a name="mlp-fwd_all"><strong>fwd_all</strong></a>(self, x, w<font color="#909090">=None</font>)</dt><dd><tt>Propagate values forward through the net. <br>
+Input:<br>
+ x - array (size>1) of input patterns<br>
+ w - optional 1-d vector of weights <br>
+Returns:<br>
+ y - array of outputs for all input patterns</tt></dd></dl>
+
+<dl><dt><a name="mlp-pack"><strong>pack</strong></a>(self)</dt><dd><tt>Compile weight matrices w1,b1,w2,b2 from net into a<br>
+single vector, suitable for optimization routines.</tt></dd></dl>
+
+<dl><dt><a name="mlp-test_all"><strong>test_all</strong></a>(self, x, t)</dt><dd><tt>Test network on an array (size>1) of patterns<br>
+Input:<br>
+ x - array of input data<br>
+ t - array of targets<br>
+Returns:<br>
+ sum-squared-error over all data</tt></dd></dl>
+
+<dl><dt><a name="mlp-train"><strong>train</strong></a>(self, x, t)</dt><dd><tt>Train network using scipy's leastsq optimizer<br>
+Input:<br>
+ x - array of input data <br>
+ t - array of targets<br>
+ <br>
+ N.B. x and t comprise the *entire* collection of training data<br>
+ <br>
+Returns:<br>
+ post-optimization weight vector</tt></dd></dl>
+
+<dl><dt><a name="mlp-unpack"><strong>unpack</strong></a>(self)</dt><dd><tt>Decompose 1-d vector of weights w into appropriate weight <br>
+matrices (w1,b1,w2,b2) and reinsert them into net</tt></dd></dl>
+
+</td></tr></table></td></tr></table><p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#eeaa77">
+<td colspan=3 valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"><big><strong>Functions</strong></big></font></td></tr>
+
+<tr><td bgcolor="#eeaa77"><tt> </tt></td><td> </td>
+<td width="100%"><dl><dt><a name="-main"><strong>main</strong></a>()</dt><dd><tt>Build/train/test MLP</tt></dd></dl>
+</td></tr></table>
+</body></html>
\ No newline at end of file
Added: trunk/Lib/sandbox/ann/doc/rbf.html
===================================================================
--- trunk/Lib/sandbox/ann/doc/rbf.html 2006-08-22 13:46:37 UTC (rev 2175)
+++ trunk/Lib/sandbox/ann/doc/rbf.html 2006-08-22 14:03:02 UTC (rev 2176)
@@ -0,0 +1,86 @@
+
+<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
+<html><head><title>Python: module rbf</title>
+</head><body bgcolor="#f0f0f8">
+
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="heading">
+<tr bgcolor="#7799ee">
+<td valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"> <br><big><big><strong>rbf</strong></big></big></font></td
+><td align=right valign=bottom
+><font color="#ffffff" face="helvetica, arial"><a href=".">index</a><br><a href="file:/home/tilde/programming/SoC/scipy/Lib/sandbox/ann/rbf.py">/home/tilde/programming/SoC/scipy/Lib/sandbox/ann/rbf.py</a></font></td></tr></table>
+ <p><tt># rbf2.py<br>
+# tilde<br>
+# 2006/08/20</tt></p>
+<p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#aa55cc">
+<td colspan=3 valign=bottom> <br>
+<font color="#fffff" face="helvetica, arial"><big><strong>Modules</strong></big></font></td></tr>
+
+<tr><td bgcolor="#aa55cc"><tt> </tt></td><td> </td>
+<td width="100%"><table width="100%" summary="list"><tr><td width="25%" valign=top><a href="numpy.html">numpy</a><br>
+</td><td width="25%" valign=top><a href="random.html">random</a><br>
+</td><td width="25%" valign=top></td><td width="25%" valign=top></td></tr></table></td></tr></table><p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#ee77aa">
+<td colspan=3 valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"><big><strong>Classes</strong></big></font></td></tr>
+
+<tr><td bgcolor="#ee77aa"><tt> </tt></td><td> </td>
+<td width="100%"><dl>
+<dt><font face="helvetica, arial"><a href="rbf.html#rbf">rbf</a>
+</font></dt></dl>
+ <p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#ffc8d8">
+<td colspan=3 valign=bottom> <br>
+<font color="#000000" face="helvetica, arial"><a name="rbf">class <strong>rbf</strong></a></font></td></tr>
+
+<tr bgcolor="#ffc8d8"><td rowspan=2><tt> </tt></td>
+<td colspan=2><tt>Class to define/train/test a radial basis function network<br> </tt></td></tr>
+<tr><td> </td>
+<td width="100%">Methods defined here:<br>
+<dl><dt><a name="rbf-__init__"><strong>__init__</strong></a>(self, ni, no, f<font color="#909090">='linear'</font>)</dt><dd><tt>Set up instance of RBF net. N.B. RBF centers and variance are selected at training time <br>
+Input:<br>
+ ni - <int> # of inputs<br>
+ no - <int> # of outputs<br>
+ f - <str> output activation fxn</tt></dd></dl>
+
+<dl><dt><a name="rbf-err_fxn"><strong>err_fxn</strong></a>(self, w, X, Y)</dt><dd><tt>Return vector of squared-errors for the leastsq optimizer</tt></dd></dl>
+
+<dl><dt><a name="rbf-fwd_all"><strong>fwd_all</strong></a>(self, X, w<font color="#909090">=None</font>)</dt><dd><tt>Propagate values forward through the net.<br>
+Inputs:<br>
+ inputs - vector of input values<br>
+ w - packed array of weights<br>
+Returns:<br>
+ array of outputs for all input patterns</tt></dd></dl>
+
+<dl><dt><a name="rbf-pack"><strong>pack</strong></a>(self)</dt><dd><tt>Compile weight matrices w,b from net into a<br>
+single vector, suitable for optimization routines.</tt></dd></dl>
+
+<dl><dt><a name="rbf-test_all"><strong>test_all</strong></a>(self, X, Y)</dt><dd><tt>Test network on an array (size>1) of patterns<br>
+Input:<br>
+ x - array of input data<br>
+ t - array of targets<br>
+Returns:<br>
+ sum-squared-error over all data</tt></dd></dl>
+
+<dl><dt><a name="rbf-train"><strong>train</strong></a>(self, X, Y)</dt><dd><tt>Train RBF network:<br>
+(i) select fixed centers randomly from input data (10%)<br>
+(ii) set fixed variance from max dist between centers<br>
+(iii) learn output weights using scipy's leastsq optimizer</tt></dd></dl>
+
+<dl><dt><a name="rbf-unpack"><strong>unpack</strong></a>(self)</dt><dd><tt>Decompose 1-d vector of weights w into appropriate weight<br>
+matrices (self.{w/b}) and reinsert them into net</tt></dd></dl>
+
+</td></tr></table></td></tr></table><p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#eeaa77">
+<td colspan=3 valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"><big><strong>Functions</strong></big></font></td></tr>
+
+<tr><td bgcolor="#eeaa77"><tt> </tt></td><td> </td>
+<td width="100%"><dl><dt><a name="-main"><strong>main</strong></a>()</dt><dd><tt>Build/train/test RBF net</tt></dd></dl>
+</td></tr></table>
+</body></html>
\ No newline at end of file
Added: trunk/Lib/sandbox/ann/doc/srn.html
===================================================================
--- trunk/Lib/sandbox/ann/doc/srn.html 2006-08-22 13:46:37 UTC (rev 2175)
+++ trunk/Lib/sandbox/ann/doc/srn.html 2006-08-22 14:03:02 UTC (rev 2176)
@@ -0,0 +1,90 @@
+
+<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
+<html><head><title>Python: module srn</title>
+</head><body bgcolor="#f0f0f8">
+
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="heading">
+<tr bgcolor="#7799ee">
+<td valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"> <br><big><big><strong>srn</strong></big></big></font></td
+><td align=right valign=bottom
+><font color="#ffffff" face="helvetica, arial"><a href=".">index</a><br><a href="file:/home/tilde/programming/SoC/scipy/Lib/sandbox/ann/srn.py">/home/tilde/programming/SoC/scipy/Lib/sandbox/ann/srn.py</a></font></td></tr></table>
+ <p><tt># <a href="#srn">srn</a>.py<br>
+# by: Fred Mailhot<br>
+# last mod: 2006-08-18</tt></p>
+<p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#aa55cc">
+<td colspan=3 valign=bottom> <br>
+<font color="#fffff" face="helvetica, arial"><big><strong>Modules</strong></big></font></td></tr>
+
+<tr><td bgcolor="#aa55cc"><tt> </tt></td><td> </td>
+<td width="100%"><table width="100%" summary="list"><tr><td width="25%" valign=top><a href="numpy.html">numpy</a><br>
+</td><td width="25%" valign=top></td><td width="25%" valign=top></td><td width="25%" valign=top></td></tr></table></td></tr></table><p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#ee77aa">
+<td colspan=3 valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"><big><strong>Classes</strong></big></font></td></tr>
+
+<tr><td bgcolor="#ee77aa"><tt> </tt></td><td> </td>
+<td width="100%"><dl>
+<dt><font face="helvetica, arial"><a href="srn.html#srn">srn</a>
+</font></dt></dl>
+ <p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#ffc8d8">
+<td colspan=3 valign=bottom> <br>
+<font color="#000000" face="helvetica, arial"><a name="srn">class <strong>srn</strong></a></font></td></tr>
+
+<tr bgcolor="#ffc8d8"><td rowspan=2><tt> </tt></td>
+<td colspan=2><tt>Class to define, train and test a simple recurrent network<br> </tt></td></tr>
+<tr><td> </td>
+<td width="100%">Methods defined here:<br>
+<dl><dt><a name="srn-__init__"><strong>__init__</strong></a>(self, ni, nh, no, f<font color="#909090">='linear'</font>, w<font color="#909090">=None</font>)</dt><dd><tt>Set up instance of <a href="#srn">srn</a>. Initial weights are drawn from a <br>
+zero-mean Gaussian w/ variance is scaled by fan-in.<br>
+Input:<br>
+ ni - <int> # of inputs<br>
+ nh - <int> # of hidden & context units<br>
+ no - <int> # of outputs<br>
+ f - <str> output activation fxn<br>
+ w - <array dtype=Float> weight vector</tt></dd></dl>
+
+<dl><dt><a name="srn-errfxn"><strong>errfxn</strong></a>(self, w, x, t)</dt><dd><tt>Return vector of squared-errors for the leastsq optimizer</tt></dd></dl>
+
+<dl><dt><a name="srn-fwd_all"><strong>fwd_all</strong></a>(self, x, w<font color="#909090">=None</font>)</dt><dd><tt>Propagate values forward through the net. <br>
+Input:<br>
+ x - matrix of all input patterns<br>
+ w - 1-d vector of weights<br>
+Returns:<br>
+ y - matrix of all outputs</tt></dd></dl>
+
+<dl><dt><a name="srn-pack"><strong>pack</strong></a>(self)</dt><dd><tt>Compile weight matrices w1,b1,wc,w2,b2 from net into a<br>
+single vector, suitable for optimization routines.</tt></dd></dl>
+
+<dl><dt><a name="srn-test_all"><strong>test_all</strong></a>(self, x, t)</dt><dd><tt>Test network on an array (size>1) of patterns<br>
+Input:<br>
+ x - array of input data<br>
+ t - array of targets<br>
+Returns:<br>
+ sum-squared-error over all data</tt></dd></dl>
+
+<dl><dt><a name="srn-train"><strong>train</strong></a>(self, x, t)</dt><dd><tt>Train a multilayer perceptron using scipy's leastsq optimizer<br>
+Input:<br>
+ x - matrix of input data<br>
+ t - matrix of target outputs<br>
+Returns:<br>
+ post-optimization weight vector</tt></dd></dl>
+
+<dl><dt><a name="srn-unpack"><strong>unpack</strong></a>(self)</dt><dd><tt>Decompose 1-d vector of weights w into appropriate weight <br>
+matrices (w1,b1,w2,b2) and reinsert them into net</tt></dd></dl>
+
+</td></tr></table></td></tr></table><p>
+<table width="100%" cellspacing=0 cellpadding=2 border=0 summary="section">
+<tr bgcolor="#eeaa77">
+<td colspan=3 valign=bottom> <br>
+<font color="#ffffff" face="helvetica, arial"><big><strong>Functions</strong></big></font></td></tr>
+
+<tr><td bgcolor="#eeaa77"><tt> </tt></td><td> </td>
+<td width="100%"><dl><dt><a name="-main"><strong>main</strong></a>()</dt><dd><tt>Set up a 1-2-1 SRN to solve the temporal-XOR problem from Elman 1990.</tt></dd></dl>
+</td></tr></table>
+</body></html>
\ No newline at end of file
More information about the Scipy-svn
mailing list