Coursera - Tensorflow
Adds a layer of neurons
What does Dense do? it...
Labelling the data
When I tell a computer what the data represents (i.e. this data is for walking, this data is for running), what is that process called?
(training_images, training_labels), (test_images, test_labels) = mnist.load_data()
how do you load mnist data into training and test sets in tf?
plt.imshow(training_images[i])
how do you show a picture of mnist using matplotlib?
stochastic gradient descent
in SGD optimizer, SGD stands for ...
Keras
is a high-level neural networks API, written in Python and capable of running on top of TensorFlow
optimizer
the ... thinks about how good or how badly the guess was done using the data from the loss function. Then the logic is that each guess should be better than the one before
loss function
the optimizer thinks about how good or how badly the guess was done using the data from the ... . Then the logic is that each guess should be better than the one before
guess
the optimizer thinks about how good or how badly the guess was done using the data from the loss function. Then the logic is that each ... should be better than the one before
Relu
... effectively means "If X>0 return X, else return 0" -- so what it does it it only passes values 0 or greater to the next layer in the network
keras
... is an API in TensorFlow and it makes it really easy to define neural networks
Softmax, coding
... takes a set of values, and effectively picks the biggest one, so, for example, if the output of the last layer looks like [0.1, 0.1, 0.05, 0.1, 9.5, 0.1, 0.05, 0.05, 0.05], it saves you from fishing through it looking for the biggest value, and turns it into [0,0,0,0,1,0,0,0,0] -- The goal is to save a lot of ...!
Flatten
... takes this 28 by 28 square and turns it into a simple linear array.
Sequential
...: That defines a SEQUENCE of layers in the neural network
converge
An iterative algorithm is said to ... when as the iterations proceed the output gets closer and closer to a minimum error
minimum error
An iterative algorithm is said to converge when as the iterations proceed the output gets closer and closer to a ...
linear array
Flatten takes this 28 by 28 square and turns it into a simple ... .
model = tf.keras.models.Sequential()
How do you create a deep learning model using Keras?
ok
How to create a conda environment: conda create -n EnvName pip python=3.6 (type ok)
epochs
If the ... equals 500 value means that it will go through the training loop 500 times
layer, single neuron
If there's only one dense in the code and there is only one unit in it, it means there's only one ... and it's a ... .
sequential
In Tensorflow and Keras, layers are defined in ... .
dense
In keras, you use the word ... to define a layer of connected neurons
shape of input
Inside the sequential and dense, you also define the ... to the neural network
activation function
Each layer of neurons need an ... to tell them what to do. There's lots of options
one byte, 784 bytes
Each pixel can be represented in values from zero to 255 and so it's only ... per pixel. With 28 by 28 pixels in an image, only ... are needed to store the entire image
Flatten
Remember earlier where our images were a square, when you printed them out? ... just takes that square and turns it into a 1 dimensional set.
flatten
The first layer is a ... layer with the input shaping 28 by 28.
ten neurons
The important things to look at are the first and the last layers. The last layer has ... in it because we have ten classes of clothing in the dataset. They should always match.
loss function, optimizer
The neural network has no idea of the relationship between X and Y, so it makes a guess. It will then use the training data to measure how good or how bad its guess was. The ... measures this and then gives the data to the ... which figures out the next guess
