Simple Logistic Regression with Neural Networks Mindset

Question

Got this simple exercise where I have to build a NN with the help of Logistic Regression. My dataset is built in this way :

You are given a dataset ("data.h5") containing:

• a training set of m_train images labeled as cat (y=1) or non-cat (y=0)
• a test set of m_test images labeled as cat or non-cat
• each image is of shape (num_px, num_px, 3) where 3 is for the 3 channels (RGB). Thus, each image is square (height = num_px) and (width = num_px).

To show an image from the database the text gives me an example :

# Example of a picture# Examp 
index = 25
plt.imshow(train_set_x_orig[index])
print ("y = " + str(train_set_y[:,index]) + ", it's a '" + classes[np.squeeze(train_set_y[:,index])].decode("utf-8") +  "' picture.")

I have 2 questions :

1) I did not understand how this works : str(train_set_y[:,index])

2) The big problem is that , due to site problem , I can not download this database and in order to do the exercise I would like to understand how is it built. Can someone intuitively tell me how it could be structured?

Answer 1

I assume this code snippet is from the Coursera Deep Learning Course 1.

"train_set_y" is a vector of shape (1, 209) ie it has the labels 0 or 1 for all 209 training examples, "train_set_y[:,25]" gives an integer label 0 or 1 from the 25th position of the vector train_set_y. As we are concatenating the strings ("y = " + str(train_set_y[:,index])) . we need to convert it into string using str.

Check the lr_utils.py file in the notebook it will give you an clear idea how the dataset is loaded and transformed.

Below is the code snippet from the lr_utils.py file

def load_dataset():
   train_dataset = h5py.File('datasets/train_catvnoncat.h5', "r")
   train_set_x_orig = np.array(train_dataset["train_set_x"][:]) # your train set features
   train_set_y_orig = np.array(train_dataset["train_set_y"][:]) # your train set labels

   test_dataset = h5py.File('datasets/test_catvnoncat.h5', "r")
   test_set_x_orig = np.array(test_dataset["test_set_x"][:]) # your test set features
   test_set_y_orig = np.array(test_dataset["test_set_y"][:]) # your test set labels

   classes = np.array(test_dataset["list_classes"][:]) # the list of classes

   train_set_y_orig = train_set_y_orig.reshape((1, train_set_y_orig.shape[0]))
   test_set_y_orig = test_set_y_orig.reshape((1, test_set_y_orig.shape[0]))

   return train_set_x_orig, train_set_y_orig, test_set_x_orig, test_set_y_orig, classes

Answer 2

1. The dataset can be download at this location, (Thanks to Anderson!)

   • https://github.com/andersy005/deep-learning-specialization-coursera/tree/master/01-Neural-Networks-and-Deep-Learning/week2/Programming-Assignments/datasets

2. Then build @taurz 's lr_utils function, put into sys.path() any directory, BUT make sure you delete 'datasets/' from train_dataset = h5py.File('datasets/train_catvnoncat.h5', "r")

3. str(train_set_y[:,index]) is the label, >> train_set_y , you can see all the feature, train_set_y.shape = (1,209), train_set_y[:,25][0] = 1 , that means it's cat.