DeepLearning.AI TensorFlow Developer 筆記

TensorFlow Developer 筆記

Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning

Week 1 A New Programming Paradigm

Before you begin: TensorFlow 2.0 and this course

Introduction: A conversation with Andrew Ng

A primer in machine learning

The ‘Hello World’ of neural networks

import tensorflow as tf
from tensorflow import keras
import numpy as np

model = tf.keras.Sequential([keras.layers.Dense(units=1, input_shape=[1])])

model.compile(optimizer='sgd', loss='mean_squared_error')

xs = np.array([-1.0,  0.0, 1.0, 2.0, 3.0, 4.0], dtype=float)
ys = np.array([-3.0, -1.0, 1.0, 3.0, 5.0, 7.0], dtype=float)

model.fit(xs, ys, epochs=500)

print(model.predict([10.0]))

From rules to data

Working through ‘Hello World’ in TensorFlow and Python

Try it for yourself

GitHub Course 1 - Part 2 - Lesson 2 - Notebook.ipynb
Colab Course 1 - Part 2 - Lesson 2 - Notebook.ipynb

Week 1 Quiz (升級後提交)

Weekly Exercise - Your First Neural Network

Introduction to Google Colaboratory

Frequently Asked Questions

Get started with Google Colaboratory (Coding TensorFlow) - YouTube

Exercise 1 (Housing Prices) (升級後提交)

编程作业: Exercise 1 (Housing Prices) (升級後提交)

Week 1 Resources

Optional: Ungraded Google Colaboratory environment

Exercise 1 (Housing Prices)

Colab Exercise_1_House_Prices_Question.ipynb

Week 2 Introduction to Computer Vision

A Conversation with Andrew Ng

An Introduction to computer vision

Exploring how to use data

GitHub Fashion-MNIST

Writing code to load training data

The structure of Fashion MNIST data

Responsible AI practices

Coding a Computer Vision Neural Network

See how it’s done

Walk through a Notebook for computer vision

Get hands-on with computer vision

Colab Course 1 - Part 4 - Lesson 2 - Notebook.ipynb

import tensorflow as tf
print(tf.__version__)

mnist = tf.keras.datasets.fashion_mnist

(training_images, training_labels), (test_images, test_labels) = mnist.load_data()

import numpy as np
np.set_printoptions(linewidth=200)
import matplotlib.pyplot as plt
#Display image as grayscale using matplotlib
#https://stackoverflow.com/questions/3823752/display-image-as-grayscale-using-matplotlib
#plt.imshow(training_images[0])
plt.imshow(training_images[0], cmap='gray', vmin=0, vmax=255)
print(training_labels[0])
print(training_images[0])

training_images  = training_images / 255.0
test_images = test_images / 255.0

model = tf.keras.models.Sequential([tf.keras.layers.Flatten(), 
                                    tf.keras.layers.Dense(128, activation=tf.nn.relu), 
                                    tf.keras.layers.Dense(10, activation=tf.nn.softmax)])

model.compile(optimizer = tf.optimizers.Adam(),
              loss = 'sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(training_images, training_labels, epochs=5)

model.evaluate(test_images, test_labels)

classifications = model.predict(test_images)

print(classifications[0])

print(test_labels[0])

Using Callbacks to control training

See how to implement Callbacks

Colab Course 1 - Part 4 - Lesson 4 - Notebook.ipynb

import tensorflow as tf

class myCallback(tf.keras.callbacks.Callback):
  def on_epoch_end(self, epoch, logs={}):
    if(logs.get('accuracy')>0.6):
      print("\nReached 60% accuracy so cancelling training!")
      self.model.stop_training = True

mnist = tf.keras.datasets.fashion_mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

callbacks = myCallback()

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  tf.keras.layers.Dense(512, activation=tf.nn.relu),
  tf.keras.layers.Dense(10, activation=tf.nn.softmax)
])
model.compile(optimizer=tf.optimizers.Adam(),
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=10, callbacks=[callbacks])

Walk through a notebook with Callbacks

Week 2 Quiz (升級後提交)

Exercise 2 (Handwriting Recognition) (购买订阅以解锁此项目。)

编程作业: Exercise 2 (升級後提交)

Weekly Exercise - Implement a Deep Neural Network to recognize handwritten digits

Week 2 Resources

Beyond Hello, World - A Computer Vision Example

Exploring Callbacks

Exercise 2 - Handwriting Recognition - Answer

Optional: Ungraded Google Colaboratory environment

Exercise 2 (Handwriting Recognition)

Colab Exercise2-Question.ipynb

Week 3 Enhancing Vision with Convolutional Neural Networks

A conversation with Andrew Ng

What are convolutions and pooling?

Coding convolutions and pooling layers

tf.keras.layers.Conv2D
tf.keras.layers.MaxPool2D

Implementing convolutional layers

64 個 3*3 的 filter
- beyond the scope of this class
- they aren’t random. They start with a set of known good filters in a similar way to the pattern fitting that you saw earlier
relu 表示負值會被丟棄
1 表示顏色為灰階
For more details on convolutions and how they work, there’s a great set of resources here.