y size changed from 121 to 11

This commit is contained in:
lkrsnik 2017-07-02 11:49:41 +02:00
parent 3e76883ce6
commit f061573c0b
2 changed files with 102 additions and 22 deletions

View File

@ -2,6 +2,7 @@
<project version="4"> <project version="4">
<component name="ChangeListManager"> <component name="ChangeListManager">
<list default="true" id="8a8ba9af-e1a4-433a-9968-475192610776" name="Default" comment=""> <list default="true" id="8a8ba9af-e1a4-433a-9968-475192610776" name="Default" comment="">
<change type="MODIFICATION" beforePath="$PROJECT_DIR$/.idea/workspace.xml" afterPath="$PROJECT_DIR$/.idea/workspace.xml" />
<change type="MODIFICATION" beforePath="$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary/character_based_ffnn_keras.ipynb" afterPath="$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary/character_based_ffnn_keras.ipynb" /> <change type="MODIFICATION" beforePath="$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary/character_based_ffnn_keras.ipynb" afterPath="$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary/character_based_ffnn_keras.ipynb" />
<change type="MODIFICATION" beforePath="$PROJECT_DIR$/prepare_data.py" afterPath="$PROJECT_DIR$/prepare_data.py" /> <change type="MODIFICATION" beforePath="$PROJECT_DIR$/prepare_data.py" afterPath="$PROJECT_DIR$/prepare_data.py" />
</list> </list>
@ -33,8 +34,8 @@
<file leaf-file-name="prepare_data.py" pinned="false" current-in-tab="true"> <file leaf-file-name="prepare_data.py" pinned="false" current-in-tab="true">
<entry file="file://$PROJECT_DIR$/prepare_data.py"> <entry file="file://$PROJECT_DIR$/prepare_data.py">
<provider selected="true" editor-type-id="text-editor"> <provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="532"> <state relative-caret-position="781">
<caret line="499" column="40" lean-forward="false" selection-start-line="499" selection-start-column="40" selection-end-line="499" selection-end-column="40" /> <caret line="318" column="23" lean-forward="false" selection-start-line="318" selection-start-column="23" selection-end-line="318" selection-end-column="23" />
<folding> <folding>
<element signature="e#24#63#0" expanded="true" /> <element signature="e#24#63#0" expanded="true" />
</folding> </folding>
@ -148,6 +149,7 @@
<option value="$PROJECT_DIR$/theano_tutorial/tutorial_loop.py" /> <option value="$PROJECT_DIR$/theano_tutorial/tutorial_loop.py" />
<option value="$PROJECT_DIR$/theano_tutorial/logistic_regression.py" /> <option value="$PROJECT_DIR$/theano_tutorial/logistic_regression.py" />
<option value="$PROJECT_DIR$/theano_tutorial/logistic_regression_loop.py" /> <option value="$PROJECT_DIR$/theano_tutorial/logistic_regression_loop.py" />
<option value="$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary/cnn_test_on_other_attributes.ipynb" />
<option value="$PROJECT_DIR$/prepare_data.py" /> <option value="$PROJECT_DIR$/prepare_data.py" />
</list> </list>
</option> </option>
@ -173,6 +175,7 @@
<foldersAlwaysOnTop value="true" /> <foldersAlwaysOnTop value="true" />
</navigator> </navigator>
<panes> <panes>
<pane id="Scratches" />
<pane id="ProjectPane"> <pane id="ProjectPane">
<subPane> <subPane>
<PATH> <PATH>
@ -185,9 +188,40 @@
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" /> <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
</PATH_ELEMENT> </PATH_ELEMENT>
</PATH> </PATH>
<PATH>
<PATH_ELEMENT>
<option name="myItemId" value="accetuation" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.ProjectViewProjectNode" />
</PATH_ELEMENT>
<PATH_ELEMENT>
<option name="myItemId" value="accetuation" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
</PATH_ELEMENT>
<PATH_ELEMENT>
<option name="myItemId" value="cnn" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
</PATH_ELEMENT>
</PATH>
<PATH>
<PATH_ELEMENT>
<option name="myItemId" value="accetuation" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.ProjectViewProjectNode" />
</PATH_ELEMENT>
<PATH_ELEMENT>
<option name="myItemId" value="accetuation" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
</PATH_ELEMENT>
<PATH_ELEMENT>
<option name="myItemId" value="cnn" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
</PATH_ELEMENT>
<PATH_ELEMENT>
<option name="myItemId" value="word_accetuation" />
<option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
</PATH_ELEMENT>
</PATH>
</subPane> </subPane>
</pane> </pane>
<pane id="Scratches" />
<pane id="Scope" /> <pane id="Scope" />
</panes> </panes>
</component> </component>
@ -197,6 +231,7 @@
</component> </component>
<component name="RecentsManager"> <component name="RecentsManager">
<key name="CopyFile.RECENT_KEYS"> <key name="CopyFile.RECENT_KEYS">
<recent name="$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary" />
<recent name="$PROJECT_DIR$/theano_tutorial" /> <recent name="$PROJECT_DIR$/theano_tutorial" />
</key> </key>
</component> </component>
@ -505,16 +540,6 @@
<watches-manager /> <watches-manager />
</component> </component>
<component name="editorHistoryManager"> <component name="editorHistoryManager">
<entry file="file://$PROJECT_DIR$/theano_tutorial/tutorial_conditions.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="18">
<caret line="1" column="31" lean-forward="false" selection-start-line="1" selection-start-column="31" selection-end-line="1" selection-end-column="31" />
<folding>
<element signature="e#55#85#0" expanded="true" />
</folding>
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/theanoTest.py"> <entry file="file://$PROJECT_DIR$/theanoTest.py">
<provider selected="true" editor-type-id="text-editor"> <provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="0"> <state relative-caret-position="0">
@ -919,6 +944,16 @@
</state> </state>
</provider> </provider>
</entry> </entry>
<entry file="file://$PROJECT_DIR$/theano_tutorial/logistic_regression.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="162">
<caret line="11" column="0" lean-forward="false" selection-start-line="11" selection-start-column="0" selection-end-line="11" selection-end-column="0" />
<folding>
<element signature="e#0#12#0" expanded="true" />
</folding>
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/theano_tutorial/logistic_regression_loop.py"> <entry file="file://$PROJECT_DIR$/theano_tutorial/logistic_regression_loop.py">
<provider selected="true" editor-type-id="text-editor"> <provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="828"> <state relative-caret-position="828">
@ -965,20 +1000,17 @@
</state> </state>
</provider> </provider>
</entry> </entry>
<entry file="file://$PROJECT_DIR$/theano_tutorial/logistic_regression.py"> <entry file="file://$PROJECT_DIR$/cnn/word_accetuation/cnn_dictionary/cnn_test_on_other_attributes.ipynb">
<provider selected="true" editor-type-id="text-editor"> <provider selected="true" editor-type-id="ipnb-editor">
<state relative-caret-position="162"> <state>
<caret line="11" column="0" lean-forward="false" selection-start-line="11" selection-start-column="0" selection-end-line="11" selection-end-column="0" /> <selected id="-1" />
<folding>
<element signature="e#0#12#0" expanded="true" />
</folding>
</state> </state>
</provider> </provider>
</entry> </entry>
<entry file="file://$PROJECT_DIR$/prepare_data.py"> <entry file="file://$PROJECT_DIR$/prepare_data.py">
<provider selected="true" editor-type-id="text-editor"> <provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="532"> <state relative-caret-position="781">
<caret line="499" column="40" lean-forward="false" selection-start-line="499" selection-start-column="40" selection-end-line="499" selection-end-column="40" /> <caret line="318" column="23" lean-forward="false" selection-start-line="318" selection-start-column="23" selection-end-line="318" selection-end-column="23" />
<folding> <folding>
<element signature="e#24#63#0" expanded="true" /> <element signature="e#24#63#0" expanded="true" />
</folding> </folding>

View File

@ -279,7 +279,55 @@ def generate_full_matrix_inputs():
return X_train, X_other_features_train, y_train, X_validate, X_other_features_validate, y_validate return X_train, X_other_features_train, y_train, X_validate, X_other_features_validate, y_validate
# Generate each y as an array of 11 numbers (with possible values between 0 and 1)
def generate_X_and_y(dictionary, max_word, max_num_vowels, content, vowels, accetuated_vowels, feature_dictionary): def generate_X_and_y(dictionary, max_word, max_num_vowels, content, vowels, accetuated_vowels, feature_dictionary):
y = np.zeros((len(content), max_num_vowels))
X = np.zeros((len(content), max_word, len(dictionary)))
print('CREATING OTHER FEATURES...')
X_other_features = create_X_features(content, feature_dictionary)
print('OTHER FEATURES CREATED!')
i = 0
for el in content:
j = 0
for c in list(el[0]):
index = 0
for d in dictionary:
if c == d:
X[i][j][index] = 1
break
index += 1
j += 1
j = 0
word_accetuations = []
num_vowels = 0
for c in list(el[3]):
index = 0
if is_vowel(el[3], j, vowels):
num_vowels += 1
for d in accetuated_vowels:
if c == d:
word_accetuations.append(num_vowels)
break
index += 1
j += 1
if len(word_accetuations) > 0:
y_value = 1/len(word_accetuations)
for el in word_accetuations:
y[i][el] = y_value
else:
y[i][0] = 1
# y[i][generate_presentable_y(word_accetuations, list(el[3]), max_num_vowels)] = 1
i += 1
print('SHUFFELING INPUTS...')
X, y, X_other_features = shuffle_inputs(X, y, X_pure=X_other_features)
print('INPUTS SHUFFELED!')
return X, X_other_features, y
# Generate each y as an array of 121 numbers (with one 1 per line and the rest zeros)
def generate_X_and_y_one_classification(dictionary, max_word, max_num_vowels, content, vowels, accetuated_vowels, feature_dictionary):
y = np.zeros((len(content), max_num_vowels * max_num_vowels )) y = np.zeros((len(content), max_num_vowels * max_num_vowels ))
X = np.zeros((len(content), max_word, len(dictionary))) X = np.zeros((len(content), max_word, len(dictionary)))
print('CREATING OTHER FEATURES...') print('CREATING OTHER FEATURES...')