課程大綱 Syllabus |
學生學習目標 Learning Objectives |
單元學習活動 Learning Activities |
學習成效評量 Evaluation |
備註 Notes |
序 No. | 單元主題 Unit topic |
內容綱要 Content summary |
1 | 機器學習簡介 |
1. 何謂機器學習
2. 機器學習之分類
3. 機器學習之應用
4. 介紹機器學習 vs. 深度學習 |
1. 認識本課程將討論之各項主題,並簡要的複習相關數學
2. 了解機器學習與深度學習的差異 |
上機實習 講授 實作 媒體教學
|
上機測驗
|
|
2 | Python, Colab and Snap!AI overview |
1. Snap!AI 功能介紹
2. Python constructs
3. Google Colab for running Python and ML programs |
1. 培養學生能夠使用Snap!AI工具的能力
2. 培養學生能夠使用Google Colab工具的能力 |
上機實習 講授 實作
|
上機測驗
|
|
3 | Digit Recognition I |
1. MNIST dataset overview
2. Training, Validation and Testing
3. Dense Neural Network (Deep learning) |
1. 了解 MNIST 資料集
2. 了解機器學習Training, Validation and Testing流程
3. 培養機器學習Dense Neural Network for solving Digit Recognition problem |
上機實習 講授 實作
|
上機測驗
|
|
4 | Digit Recognition I I |
1. MNIST dataset overview
2. Training, Validation and Testing
3. Dense Neural Network (Deep learning) |
1. 了解 MNIST 資料集
2. 了解機器學習Training, Validation and Testing流程
3. 培養機器學習Dense Neural Network for solving Digit Recognition problem |
上機實習 講授 實作
|
上機測驗
|
|
5 | Machine Learning Overview |
1. Neural Network Model
2. Activation functions
3. Hyperparameters Turning |
1. 了解 Neural Network Model
2. 了解 Activation functions
3. 了解 Hyperparameters Turning |
上機實習 講授 實作
|
作業
|
|
6 | Convolutional Neural Network I |
1. Convolutional operations and Convolutional layers
2. Max pooling operations and MaxPooling layers
3. Flatten layers
4. Dense layers |
1. 理解Convolutional layers, MaxPooling layers, Flatten layers and Dense layers
2. 具有實作CNN機器學習的能力 |
上機實習 講授 實作
|
上機測驗
|
|
7 | Convolutional Neural Network II |
1. Convolutional operations and Convolutional layers
2. Max pooling operations and MaxPooling layers
3. Flatten layers
4. Dense layers |
1. 理解Convolutional layers, MaxPooling layers, Flatten layers and Dense layers
2. 具有實作CNN機器學習的能力 |
上機實習 講授 實作
|
上機測驗
|
|
8 | Midterm |
None |
None |
|
期中考
|
|
9 | Convolutional Neural Network III |
1. Convolutional operations and Convolutional layers
2. Max pooling operations and MaxPooling layers
3. Flatten layers
4. Dense layers |
1. 理解Convolutional layers, MaxPooling layers, Flatten layers and Dense layers
2. 具有實作CNN機器學習的能力 |
上機實習 講授 實作
|
上機測驗
|
|
10 | Convolutional Neural Network IV |
1. Convolutional operations and Convolutional layers
2. Max pooling operations and MaxPooling layers
3. Flatten layers
4. Dense layers |
1. 理解Convolutional layers, MaxPooling layers, Flatten layers and Dense layers
2. 具有實作CNN機器學習的能力 |
上機實習 講授 實作
|
作業
|
|
11 | Save and Load Keras Model |
1.Machine learning Keras whole model: the architecture and the weights of the model
2. Save Model APIs
3. Load Model APIs
3. C4.5
4. Occam's razor |
1.理解 Machine learning Keras whole model: the architecture and the weights of the model
2. 具有實作機器學習Save Model APIs的能力
3. 具有實作機器學習Load Model APIs的能力 |
上機實習 講授 實作
|
|
|
12 | Keras pre-built models |
1. Models for image classification with weights trained on ImageNet
2. Fine tune pre-built models |
1. 了解 pre-built Models for image classification with weights trained on ImageNet
2. 了解 Fine tune pre-built models |
上機實習 講授 實作
|
|
|
13 | Recurrent Neural Networks I |
1. SimpleRNN cell
2. Long Short Term Memory (LSTM) and
3. Gated Recurrent Unit (GRU) |
1. 了解 SimpleRNN cell, Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU)
2. 培養具有實作RNN機器學習的能力 |
上機實習 講授 實作
|
上機測驗
|
|
14 | Recurrent Neural Networks II |
1. SimpleRNN cell
2. Long Short Term Memory (LSTM) and
3. Gated Recurrent Unit (GRU) |
1. 了解 SimpleRNN cell, Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU)
2. 培養具有實作RNN機器學習的能力 |
上機實習 講授 實作
|
作業
|
|
15 | Recurrent Neural Networks III |
1. SimpleRNN cell
2. Long Short Term Memory (LSTM) and
3. Gated Recurrent Unit (GRU) |
1. 了解 SimpleRNN cell, Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU)
2. 培養具有實作RNN機器學習的能力 |
上機實習 講授 實作
|
上機測驗
|
|
16 | 期末考 Final Exam |
期末考 Final Exam |
期末考 Final Exam |
|
期末考
|
|
17 | 專題報告 I |
機器學習專題設計與報告 |
1. 機器學習可以解決的問題痛點的重要性, 價值主張與目標客群的描述詳細程度.
2. 機器學習實作 |
討論 設計研究 實作
|
彈性教學
|
|
18 | 專題報告 II |
找出機器學習可以解決的問題, 用機器學習的方法來解決這個痛點. |
1. 機器學習可以解決的問題痛點的重要性, 價值主張與目標客群的描述詳細程度. 2. 機器學習實作 |
實作
|
彈性教學
|
|