This is my repository for my works on Deep Learning Specialization course - I am going to make it more fun.
Note: Course label is Course1
for tracking all changes for this week
- Week 1: Intorductions
- Week 2: Logistic Regression as a Neural Network and Python Basics with Numpy(utils dir). This week label is
Course1-Week2
for tracking the changes- loss function implementations
L1
,L2
methods added. (utils/loss.py
) - image utils
image2vector
,display_image
,flatten_X
, andstandardize_dataset
methods added. (utils/image.py
) - activation functions
sigmoid
,softmax
, andsigmoid_derivative
methods added.(utils/activation.py
) - base utils,
initialize_with_zeros
andplot_cost_function
added. (utils/base.py
) - dataloader utils for loading different datasets added.(
utils/dataloader.py
) - normalization methods
normalize_image
andnormalize_rows
methods added.(utils/normalization.py
) - LogisticRegression model with Neural Network mindset a directory with its own files added.
- Dataset
dataset/catvnotcat
were used for this week added to dataset dir.
- loss function implementations
- Week 3: Planar data classification with one hidden layer directory
nn.py
consist of nn model for classification of planner data, with anNN
classtest_nn.py
contains unit testsutil.py
for visualizationnn.ipynb
is a jupyter notebook file for demo of classification
- Week 4: Buidling DNN-Step by Step for CatvsNot Classification a dictionary with its own files added:
activations.py
contains activations and their backward calculations for DNNtest_dnn.py
contains unit testsDNN.ipynb
is a jupyter notebook file for demo of dnn classificationdnn.py
is a deep neural network implementations. it containsDNN
class as a base class andTwoLayerModel
,LLayerModel
classes as a child class ofDNN
.
Extra and useful links: