Path: blob/master/C1 - Neural Networks and Deep Learning/Week 3/Planar data classification with one hidden layer/__pycache__/planar_utils.cpython-313.pyc
9597 views
�
u5h� � �H � S SK Jr S SKrS SKrS SKrS SKrS rS r S r
S r g)� Nc �, � USS S 24 R 5 S-
USS S 24 R 5 S- pCUSS S 24 R 5 S-
USS S 24 R 5 S- peSn[ R " [ R " X4U5 [ R " XVU5 5 u p�U " [ R
UR
5 U R
5 4 5 n
U
R UR 5 n
[ R " X�U
[ R R S9 [ R " S5 [ R " S5 [ R " USS S 24 USS S 24 U[ R R S9 g )Nr � g{�G�z�?)�cmap�x2�x1)�cr )�min�max�np�meshgrid�arange�c_�ravel�reshape�shape�plt�contourf�cm�Spectral�ylabel�xlabel�scatter) �model�X�y�x_min�x_max�y_min�y_max�h�xx�yy�Zs ڼ/Users/Teo/Documents/2.Queens/1.Masters/Coursera-Deep-Learning-Specialization/C1 - Neural Networks and Deep Learning/Week 3/Planar data classification with one hidden layer/planar_utils.py�plot_decision_boundaryr% s � ��Q��T�7�;�;�=�1�$�a��1��g�k�k�m�a�&7�5��Q��T�7�;�;�=�1�$�a��1��g�k�k�m�a�&7�5� �A�
�[�[����5��3�R�Y�Y�u�Q�5O�
P�F�B�
�b�e�e�B�H�H�J����
�*�+�,�A� � � �"�(�(��A��L�L������ � �1��J�J�t���J�J�t���K�K��!�Q�$���1�a�4��A�C�F�F�O�O�<� c �@ � SS[ R " U * 5 - - nU$ )zi
Compute the sigmoid of x
Arguments:
x -- A scalar or numpy array of any size.
Return:
s -- sigmoid(x)
r )r �exp)�x�ss r$ �sigmoidr+ s! � �
�1�R�V�V�Q�B�Z�<��A� �Hr& c �� � [ R R S5 Sn [ U S- 5 nSn[ R " X45 n[ R " U S4SS9nSn[ S5 H� n[ X- XS- - 5 n[ R " US- US- S- U5 [ R R U5 S- - nU[ R " SU- 5 - [ R R U5 S- - n [ R U [ R " U5 - U [ R " U5 - 4 X7' XdU' M� UR nUR nX44$ ) Nr i� � �uint8)�dtype� g�(\���@皙�����?) r �random�seed�int�zeros�range�linspace�randn�sinr �cos�T)
�m�N�Dr �Y�a�j�ix�t�rs
r$ �load_planar_datasetrE % s# � ��I�I�N�N�1�� �A� �A�a�C��A� �A�
���!���A�
���!�A��g�&�A� �A�
�1�X��
�1�3�q�A�#�w�
�� �K�K��$���!��T�z�!� ,�r�y�y���q�/A�#�/E� E��
�b�f�f�Q�q�S�k�M�B�I�I�O�O�A� .�s� 2� 2�����a����q� �k�1�R�V�V�A�Y�;�.�/����"�� �
���A� ���A� �4�Kr& c
� � Sn [ R R U SSS9n[ R R U SS9n[ R R U SSS S
9n[ R R S SU SSS S S 9n[ R R U S5 [ R R U S5 4nXX4U4$ )
N�� g �?g333333�?)� n_samples�factor�noiser1 )rH rJ � r- � )rH � random_state�
n_features�centersT)�mean�covrH rN � n_classes�shufflerM ) �sklearn�datasets� make_circles�
make_moons�
make_blobs�make_gaussian_quantilesr r2 �rand)r= �
noisy_circles� noisy_moons�blobs�gaussian_quantiles� no_structures r$ �load_extra_datasetsr` : s� � � �A��$�$�1�1�A�b�PR�1�S�M��"�"�-�-���-�D�K� � � � '� '�!�!�PQ�[\� '� ]�E� �)�)�A�A�t�QT�`a�no�{|� GK� Z^�A� _���9�9�>�>�!�Q�'���� � ��1�)=�=�L� �u�,� N�Nr& ) �matplotlib.pyplot�pyplotr �numpyr rT �sklearn.datasets�sklearn.linear_modelr% r+ rE r` � r& r$ �<module>rg s) �� � � � � �=�"
��*Or&