I have the following method get_train_instances this takes a matrix (train_mat), which is a one-hot matrix and consists of users and items, it also takes a number of negative ones (num_negatives), for example 4.
I would like to understand this method in more detail and am therefore looking for an explanation of why this method is run. Are there books, writings, papers, newspaper articles about which I could learn more about why this method should be carried out and what exactly it does?
def get_train_instances(train_mat, num_negatives):
user_input, item_input, labels = [], [], []
num_user, num_item = train_mat.shape
for (u, i) in train_mat.keys():
user_input.append(u)
item_input.append(i)
labels.append(1)
# negative instances
for t in range(num_negatives):
j = np.random.randint(num_item)
while (u, j) in train_mat.keys():
j = np.random.randint(num_item)
user_input.append(u)
item_input.append(j)
labels.append(0)
return user_input, item_input, labels
This method is used in a deep learning neural network. I understood this to mean that this method is used to create training samples so that the model can then improve. However, I am not sure and am therefore looking for a good explanation with a recommendation of a book, paper, newspaper article, ... about which I can learn more about why and how to proceed.
question from:https://stackoverflow.com/questions/65625792/why-is-the-method-for-creating-training-instances-there
