Mercurial > pylearn
annotate linear_regression.py @ 284:8e923cb2e8fc
renamed file
author | Frederic Bastien <bastienf@iro.umontreal.ca> |
---|---|
date | Fri, 06 Jun 2008 13:52:37 -0400 |
parents | f6505ec32dc3 |
children | c9a89be5cb0a |
rev | line source |
---|---|
132
f6505ec32dc3
Updated documentation slightly
Joseph Turian <turian@gmail.com>
parents:
118
diff
changeset
|
1 """ |
f6505ec32dc3
Updated documentation slightly
Joseph Turian <turian@gmail.com>
parents:
118
diff
changeset
|
2 Implementation of linear regression, with or without L2 regularization. |
f6505ec32dc3
Updated documentation slightly
Joseph Turian <turian@gmail.com>
parents:
118
diff
changeset
|
3 This is one of the simplest example of L{learner}, and illustrates |
f6505ec32dc3
Updated documentation slightly
Joseph Turian <turian@gmail.com>
parents:
118
diff
changeset
|
4 the use of theano. |
f6505ec32dc3
Updated documentation slightly
Joseph Turian <turian@gmail.com>
parents:
118
diff
changeset
|
5 """ |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
6 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
7 from learner import * |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
8 from theano import tensor as t |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
9 from theano.scalar import as_scalar |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
10 |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
11 class LinearRegression(MinibatchUpdatesTLearner): |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
12 """ |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
13 Implement linear regression, with or without L2 regularization |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
14 (the former is called Ridge Regression and the latter Ordinary Least Squares). |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
15 |
92
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
16 The predictor parameters are obtained analytically from the training set. |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
17 Training can proceed sequentially (with multiple calls to update with |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
18 different disjoint subsets of the training sets). After each call to |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
19 update the predictor is ready to be used (and optimized for the union |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
20 of all the training sets passed to update since construction or since |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
21 the last call to forget). |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
22 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
23 For each (input[t],output[t]) pair in a minibatch,:: |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
24 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
25 output_t = b + W * input_t |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
26 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
27 where b and W are obtained by minimizing:: |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
28 |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
29 L2_regularizer sum_{ij} W_{ij}^2 + sum_t ||output_t - target_t||^2 |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
30 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
31 Let X be the whole training set inputs matrix (one input example per row), |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
32 with the first column full of 1's, and Let Y the whole training set |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
33 targets matrix (one example's target vector per row). |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
34 Let theta = the matrix with b in its first column and W in the others, |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
35 then each theta[:,i] is the solution of the linear system:: |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
36 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
37 XtX * theta[:,i] = XtY[:,i] |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
38 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
39 where XtX is a (n_inputs+1)x(n_inputs+1) matrix containing X'*X |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
40 plus L2_regularizer on the diagonal except at (0,0), |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
41 and XtY is a (n_inputs+1)*n_outputs matrix containing X'*Y. |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
42 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
43 The fields and attributes expected and produced by use and update are the following: |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
44 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
45 - Input and output fields (example-wise quantities): |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
46 |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
47 - 'input' (always expected by use and update as an input_dataset field) |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
48 - 'target' (optionally expected by use and update as an input_dataset field) |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
49 - 'output' (optionally produced by use as an output dataset field) |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
50 - 'squared_error' (optionally produced by use as an output dataset field, needs 'target') = example-wise squared error |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
51 |
92
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
52 - optional attributes (optionally expected as input_dataset attributes) |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
53 (warning, this may be dangerous, the 'use' method will use those provided in the |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
54 input_dataset rather than those learned during 'update'; currently no support |
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
55 for providing these to update): |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
56 |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
57 - 'L2_regularizer' |
92
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
58 - 'b' |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
59 - 'W' |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
60 - 'parameters' = [b, W] |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
61 - 'regularization_term' |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
62 - 'XtX' |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
63 - 'XtY' |
107
c4916445e025
Comments from Pascal V.
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
92
diff
changeset
|
64 |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
65 """ |
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
66 |
92
c4726e19b8ec
Finished first draft of TLearner
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
78
diff
changeset
|
67 def attributeNames(self): |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
68 return ["L2_regularizer","parameters","b","W","regularization_term","XtX","XtY"] |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
69 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
70 def useInputAttributes(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
71 return ["b","W"] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
72 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
73 def useOutputAttributes(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
74 return [] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
75 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
76 def updateInputAttributes(self): |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
77 return ["L2_regularizer","XtX","XtY"] |
77
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
78 |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
79 def updateMinibatchInputFields(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
80 return ["input","target"] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
81 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
82 def updateMinibatchInputAttributes(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
83 return ["XtX","XtY"] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
84 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
85 def updateMinibatchOutputAttributes(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
86 return ["new_XtX","new_XtY"] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
87 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
88 def updateEndInputAttributes(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
89 return ["theta","XtX","XtY"] |
77
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
90 |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
91 def updateEndOutputAttributes(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
92 return ["new_theta","b","W","regularization_term"] # CHECK: WILL b AND W CONTAIN OLD OR NEW THETA? @todo i.e. order of computation = ? |
77
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
93 |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
94 def parameterAttributes(self): |
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
95 return ["b","W"] |
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
96 |
78 | 97 def defaultOutputFields(self, input_fields): |
77
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
98 output_fields = ["output"] |
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
99 if "target" in input_fields: |
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
100 output_fields.append("squared_error") |
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
101 return output_fields |
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
102 |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
103 def __init__(self): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
104 self._input = t.matrix('input') # n_examples x n_inputs |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
105 self._target = t.matrix('target') # n_examples x n_outputs |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
106 self._L2_regularizer = as_scalar(0.,'L2_regularizer') |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
107 self._theta = t.matrix('theta') |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
108 self._W = self._theta[:,1:] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
109 self._b = self._theta[:,0] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
110 self._XtX = t.matrix('XtX') |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
111 self._XtY = t.matrix('XtY') |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
112 self._extended_input = t.prepend_one_to_each_row(self._input) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
113 self._output = t.dot(self._input,self._W.T) + self._b # (n_examples , n_outputs) matrix |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
114 self._squared_error = t.sum_within_rows(t.sqr(self._output-self._target)) # (n_examples ) vector |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
115 self._regularizer = self._L2_regularizer * t.dot(self._W,self._W) |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
116 self._new_XtX = add_inplace(self._XtX,t.dot(self._extended_input.T,self._extended_input)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
117 self._new_XtY = add_inplace(self._XtY,t.dot(self._extended_input.T,self._target)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
118 self._new_theta = t.solve_inplace(self._theta,self._XtX,self._XtY) |
77
1e2bb5bad636
toying with different ways to implement learners
bengioy@bengiomac.local
parents:
75
diff
changeset
|
119 |
118
d0a1bd0378c6
Finished draft of OneHiddenLayerNNetClassifier to debut learner.py
Yoshua Bengio <bengioy@iro.umontreal.ca>
parents:
111
diff
changeset
|
120 MinibatchUpdatesTLearner.__init__(self) |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
121 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
122 def allocate(self,minibatch): |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
123 minibatch_n_inputs = minibatch["input"].shape[1] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
124 minibatch_n_outputs = minibatch["target"].shape[1] |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
125 if not self._n_inputs: |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
126 self._n_inputs = minibatch_n_inputs |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
127 self._n_outputs = minibatch_n_outputs |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
128 self.XtX = numpy.zeros((1+self._n_inputs,1+self._n_inputs)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
129 self.XtY = numpy.zeros((1+self._n_inputs,self._n_outputs)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
130 self.theta = numpy.zeros((self._n_outputs,1+self._n_inputs)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
131 self.forget() |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
132 elif self._n_inputs!=minibatch_n_inputs or self._n_outputs!=minibatch_n_outputs: |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
133 # if the input or target changes dimension on the fly, we resize and forget everything |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
134 self.forget() |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
135 |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
136 def forget(self): |
110
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
137 if self._n_inputs and self._n_outputs: |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
138 self.XtX.resize((1+self.n_inputs,1+self.n_inputs)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
139 self.XtY.resize((1+self.n_inputs,self.n_outputs)) |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
140 self.XtX.data[:,:]=0 |
8fa1ef2411a0
Worked on OneShotTLearner and implementation of LinearRegression
bengioy@bengiomac.local
parents:
107
diff
changeset
|
141 self.XtY.data[:,:]=0 |
111
88257dfedf8c
Added another work in progress, for mlp's
bengioy@bengiomac.local
parents:
110
diff
changeset
|
142 numpy.diag(self.XtX.data)[1:]=self.L2_regularizer |
75
90e4c0784d6e
Added draft of LinearRegression learner
bengioy@bengiomac.local
parents:
diff
changeset
|
143 |