Machine Learning

在這裡下載程式： NCUAI_20260424.zip

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

參考資料 :

李宏毅老師的課程 ML Lecture 10: Convolutional Neural Network (1:19:28)

https://www.youtube.com/watch?v=FrKWiRv254g&list=PLJV_el3uVTsPy9oCRY30oBPNLCo89yu49&index=19

https://www.youtube.com/playlist?list=PLJV_el3uVTsPy9oCRY30oBPNLCo89yu49

https://github.com/Sakura-gh/ML-notes

2月26日 (週四) : 3PM ~ 5PM

(1) ML-01: 人工智慧概論 02, 原理 (49:09)

https://www.youtube.com/watch?v=3_NQFxuJ4ow&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=1

(2) ML-02: (2026-02-23 版本) Install Anaconda, Tensorflow (Keras), Jupyter Notebook, Spyder. (15:25)

https://www.youtube.com/watch?v=wdfYvwaEAI8&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=2

(3) ML-03: How to run Python Programs in Jupyter Notebook and Spyder (37:03)

https://www.youtube.com/watch?v=cplezXd-W14&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=3

解決 Anaconda 閃退的方法

https://cloud.tencent.com/developer/article/2186579

彭彭的課程: Matplotlib 簡介、安裝、快速開始 - Python 資料視覺化教學課程

https://www.youtube.com/watch?v=MceOR4Kvv9I&list=PL-g0fdC5RMbqDdag2l_F3ejf4xQ_QjGbq

Python 基礎教學影片 (莫烦)

https://www.youtube.com/watch?v=1PC3etgLwVc&list=PLXO45tsB95cIRP5gCi8AlYwQ1uFO2aQBw

Python 書籍 :

一步到位！Python 程式設計-最強入門教科書 / 作者：陳惠貞 / 出版社：旗標

3月5日 (週四) : 3PM ~ 4PM

import warnings

warnings.filterwarnings('ignore')

3月5日 (週四) : 4PM ~ 5PM

YOLO 原創者的演講 : How computers learn to recognize objects instantly | Joseph Redmon (7:37)

https://www.youtube.com/watch?v=Cgxsv1riJhI

當一個天才程式設計師說「我不幹了」：YOLO 背後那些你不知道的故事

https://hackmd.io/@BASHCAT/r1mtdXcFge

YOLO-v7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Authors: Chien-Yao Wang, Alexey Bochkovskiy, Hong-Yuan Mark Liao. arXiv:2207.02696 (2022)

https://arxiv.org/abs/2207.02696

Github:

https://github.com/WongKinYiu/yolov7

(8) ML-08: How to install YOLO-v7 (Python = 3.8) (17:09)

https://www.youtube.com/watch?v=19NXLqns1pw&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=8

pip install -r requirements.txt

3月12日 (週四) : 3PM ~ 4PM

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\Keras_01_Iris.ipynb

投影片在: 人工智慧概論\1_PDF_Lecture_Note\AI_06_Keras_Iris.pdf

投影片在: 人工智慧概論\1_PDF_Lecture_Note\AI_01_什麼是人工智慧.pdf

(9) ML-09: 三種鳶尾花, 資料處理 (22:37)

https://www.youtube.com/watch?v=vllpguGZwUY&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=9

(10) ML-10: 三種鳶尾花, One-Hot encoding and Multi-Layer Perceptron (25:43)

https://www.youtube.com/watch?v=if-5pfb6tfo&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=10

3月12日 (週四) : 4PM ~ 5PM

(11) ML-11: 過度擬合和交叉驗證 (21:37)

https://www.youtube.com/watch?v=dtMkhsiVNFA&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=11

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\iris_cross_validation.ipynb

投影片在: 人工智慧概論\1_PDF_Lecture_Note\AI_07_OverFitting_CrossValidation.pdf

(12) ML-12: 三種鳶尾花, Save and Load, Model and Weights (15:00)

https://www.youtube.com/watch?v=lexI-3wehyc&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=12

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\Keras_02_iris_save_model_weights

投影片在: 人工智慧概論\1_PDF_Lecture_Note\AI_09_Save_Load_Model_Parameters.pdf

KERAS 使用網頁 :

Keras: API - Applications

https://keras.io/api/applications/

https://keras.io/getting_started/intro_to_keras_for_engineers/

Keras: Code examples

https://keras.io/examples/

3月19日 (週四) : 3PM ~ 5PM

(13) Regression

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\iris-randomforest

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\iris-svm

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\iris-knn

程式在: 人工智慧概論\3_Keras_experiment\Keras_1_Iris\code\iris-GradientBoostingClassfier

PyInvest - Video

https://www.youtube.com/watch?v=DeFIGJu6tbY&list=PLy7MS-q4l3xDYoR8MACYA3YyidUbEiz6j&index=1

程式檔案在此:

程式在: NCUAI_20230903\NCUAI_20230903\Python_20210720\ml_toturial-master

https://github.com/pyinvest/ml_toturial

參考資料: ML-13: PyInvest and GitHub (18:01)

https://www.youtube.com/watch?v=n6eHypiVNmc&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=13

(14) Scikit-Learn ： https://scikit-learn.org/stable/

ML-14: How to use Scikit-Learn (58:38), starting from 14:00

https://www.youtube.com/watch?v=L3hx8bsC8iE&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=14

@ 47:33, How to save and load model using: (1) joblib; (2) pickle

程式在: NCUAI_20230903\NCUAI_20230903\Python_20210720\Sklearn_03_IrisKneighborSaveModel_001

參考資料 :

sklearn.metrics: Metrics

from sklearn.metrics import classification_report, accuracy_score, precision_score, recall_score, f1_score, roc_auc_score

https://scikit-learn.org/stable/modules/classes.html#sklearn-metrics-metrics

F-1 Score for Multi-Class Classification

https://www.baeldung.com/cs/multi-class-f1-score

Sensitivity, Specificity, Accuracy, Associated Confidence Interval and ROC Analysis

https://www.lexjansen.com/nesug/nesug10/hl/hl07.pdf

Advanced Time Series Analysis: Decomposition

https://www.kaggle.com/code/bextuychiev/advanced-time-series-analysis-decomposition/notebook

Precision, Recall & F-Measure (13:41)

https://www.youtube.com/watch?v=j-EB6RqqjGI

ROC and AUC, Clearly Explained! (16:16)

https://www.youtube.com/watch?v=4jRBRDbJemM

3月26日 (週四) : 3PM ~ 4PM

乳癌資料和 EDA (Exploratory Data Analysis)

(15) ML-15: 乳癌資料, Random Forest (35:05)

https://www.youtube.com/watch?v=eurwX_f9r08&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=15

程式在: 人工智慧概論\3_Keras_experiment\Keras_2_Cancer\code\Keras_02_Breast_Cancer.ipynb

投影片在: 人工智慧概論\1_PDF_Lecture_Note\AI_10_Keras_Cancer.pdf

(16) ML-16: Kaggle (11:16)

https://www.youtube.com/watch?v=mrkksUOW-BQ&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=16

* Kaggle Courses:

https://www.kaggle.com/learn

作業 #2, Kaggle 競賽 (20%) (5月1日, 23:59 前上傳到 Kaggle 網站)

2026-GS-4519-C-course

https://www.kaggle.com/t/7f2deb756a694e7fa07c9f6aba168710

---------------------------------------

Example : GS-3073-AI-course, Regression or classification?

https://www.kaggle.com/competitions/gs-3073-ai-course/overview

https://www.kaggle.com/competitions/gs-3073-ai-course/code

https://www.kaggle.com/competitions/gs-3073-ai-course/leaderboard

3月26日 (週四) : 4PM ~ 5PM

在這裡下載說明檔及程式： (1) Copilot_explains_RAG_20260315, (2) LLM_20260321.7z

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

LLM_01_OpenAI.ipynb

LLM_02_OpenAI_ENV.ipynb

LLM_03_GoogleGemini.ipynb

LLM_04_Ollama.ipynb

Chat with OpenAI, Gemini, and Ollama

(17) ML-17: How to install LLM environment (Python = 3.10) (7:23)

https://www.youtube.com/watch?v=hN5iZLpLDBI&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=17

pip install -r requirements.txt

(18) ML-18: How to install Ollama (7:30)

https://www.youtube.com/watch?v=6T5Exvcl8Z4&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=18

requirements.txt for (17) ML-17: How to install LLM environment (Python = 3.10)

# Python: 3.10.20

langchain==1.2.15

langchain-openai==1.1.15

langchain-core==1.3.0

langchain-classic==1.0.4

langchain-community==0.4.1

langchain-experimental==0.4.1

langchain-huggingface==1.2.2

langchain-text-splitters== 1.1.2

langchain-google-genai==4.2.2

sentence-transformers==5.4.1

chardet==7.4.3

ipywidgets==8.1.8

faiss-cpu==1.13.2

python-dotenv==1.2.2

rich==14.2.0

TAME、Breeze、TAIDE 都屬於「開源（Open Source）」的台灣本土 LLM

1. TAME（台大語音實驗室）— ✔ 開源、免費、Apache-2.0 授權

2. Breeze（台灣 AI Labs Breeze LLM）— ✔ 開源、免費（多為 Apache-2.0）

3. TAIDE（國科會 TAIDE 計畫）— ✔ 開源、免費，但授權較特殊

-------------------------------

模型 LangChain 支援方式最簡單的使用方式

TAME ✔ HF Pipeline → LangChain HuggingFacePipeline(...)

Breeze ✔ Ollama → LangChain Ollama(model="jcai/breeze-7b...")

TAIDE ✔ Ollama → LangChain Ollama(model="jcai/llama3-taide...")

4月2日 (週四) : 放假

4月9日 (週四) : 3PM ~ 5PM

(19) ML-19: 如何使用 CoLab (32:20) (只看初級班, 0:00~12:44)

https://www.youtube.com/watch?v=YbmEyJpx7xk&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=31

在這裡下載程式： LLM_20260407.7z

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

LLM_05_OpenAI_Stream.ipynb

LLM_06_Prompt_Template.ipynb

LLM_07_Chat_Prompt_Template.ipynb

LLM_08_InMemoryChatMessageHistory.ipynb

Prompt 共享平台:

(1) 英文 : LangChain Prompt Hub 是一個用於上傳、瀏覽、拉取和管理提示詞（prompts）的平台。它的目標是創建一個分享和發現提示的場所，使開發者更容易發現新用例和精煉提示。

LangChain Hub / Explore and contribute prompts to the community hub

https://smith.langchain.com/hub

(2) 中文 : ChatGPT指令大全：最全中文指令庫

提問（Prompt）撰寫的CLAR原則與LACES問題模型

https://github.com/AIzhibei/chatgpt-prompts-wiki-zh

4月16日 (週四) : 3PM ~ 5PM

在這裡下載程式： LLM_20260416.7z

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

LLM_09_HuggingFace_FAISS.ipynb

LLM_10_RAG.ipynb

LLM_11_Agent_Python.ipynb

LLM_12_Agent_Python.py

LLM_13_RAG_Python.py

prompt = PromptTemplate(

input_variables=["context", "question"],

template = """

You are a precise assistant. Use ONLY the context below.

Context:

{context}

Question:

{question}

Answer clearly and cite the context.

"""

)

4月23日 (週四) :

虛擬環境 (venv)

Python 虛擬環境的創建（venv）

https://www.runoob.com/python3/python-venv.html

How to use venv to create virtual environments - python

https://www.youtube.com/watch?v=LzaBf2QjR8s

【Python環境安裝】Venv+VSCode - Windows版

https://www.youtube.com/watch?v=czeEyrCm-bQ

在這裡下載程式： LLM_20260422.7z

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

LLM_14_OneFunction.ipynb

LLM_15_ThreeFunctions.ipynb

LLM_16_Streamlit_demo_app.py

LLM_17_Streamlit_Invoke.py

(20) ML-20: How to install Streamlit (Python = 3.10) (7:33)

https://www.youtube.com/watch?v=6mZ01CX1DD4&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=34

4月30日 (週四) : 3PM ~ 4PM

在這裡下載程式：人工智慧概論.7z

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

But what is a neural network? 究竟神經網路是什麼？ (19:13)

https://www.youtube.com/watch?v=aircAruvnKk

手寫數字辨識 (MNIST)

MNIST, Multi-Layer Perceptron (MLP) and Save the best model

Create an environment with python 3.13

then in anaconda, install keras, tensorflow, pandas, matplotlib, scikit learn¶

程式在: 人工智慧概論\3_Keras_experiment\Keras_4_MNIST\code\Keras_MNIST_MLP_Save_Best.ipynb

4月30日 (週四) : 4PM ~ 5PM

Convolutional Neural Networks from Scratch | In Depth

https://www.youtube.com/watch?v=jDe5BAsT2-Y

李宏毅老師的課程 ML Lecture 10: Convolutional Neural Network (1:19:28)

https://www.youtube.com/watch?v=FrKWiRv254g&list=PLJV_el3uVTsPy9oCRY30oBPNLCo89yu49&index=19

李宏毅老師的課程 ML Lecture 7: Backpropagation

https://www.youtube.com/watch?v=ibJpTrp5mcE&list=PLJV_el3uVTsPy9oCRY30oBPNLCo89yu49&index=12

程式在: 人工智慧概論\3_Keras_experiment\Keras_4_MNIST\code\Keras_MNIST_CNN_step_1.ipynb

程式在: 人工智慧概論\3_Keras_experiment\Keras_4_MNIST\code\Keras_MNIST_CNN_step_2.ipynb

參考資料 :

(20) ML-20: MNIST, Multi-Layer Perceptron from Kaggle (37:57)

程式在: NCUAI_20260424\Python_20210803\mnist-deep-neural-network-with-keras.ipynb

程式在: NCUAI_20260424\Python_20210803\mnist-deep-neural-network-with-keras-revision.ipynb

https://www.youtube.com/watch?v=ItFJFOO5l4w&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=19

(21) ML-21: MNIST digits classification dataset and Simple Conv-Net (21:40)

程式在: NCUAI_20260424\Python_20210803\Load_MNIST_DataSet_001.py

程式在: NCUAI_20260424\Python_20210803\Keras_06_SimpleConvMNIST_001.py

https://www.youtube.com/watch?v=I3kLFT3FdXs&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=20

https://keras.io/api/datasets/mnist/

https://keras.io/examples/vision/mnist_convnet/

5月7日 (週四) : 3 PM ~ 4 PM

VGG16 Neural Network Visualization (7:35)

https://www.youtube.com/watch?v=RNnKtNrsrmg

(22) ML-22: CIFAR-10 image classification with CNN (CPU vs GPU) (37:10)

程式在: NCUAI_20260424\Python_20210803\cifar-10-image-classification-with-cnn-Revision.ipynb

https://www.youtube.com/watch?v=xbVI2xXhnmc&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=21

https://www.kaggle.com/bhuvanchennoju/cifar-10-image-classification-with-cnn

5月7日 (週四) : 4 PM ~ 5 PM

程式在: NCUAI_20230903\Python_20210810\TransferLearning_ShoulderImplants\code

(26) ML-26: Transfer Learning for Analyzing Medical Images (33:45)

https://www.youtube.com/watch?v=3x2YGfnaoXI&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=26

(a) Download the paper: Classifying shoulder implants in X-ray images using deep learning

https://www.sciencedirect.com/science/article/pii/S2001037019304179

(b) Download the dataset: 這裡可以下載 UCI 的597張照片

Shoulder Implant X-Ray Manufacturer Classification Data Set

https://archive.ics.uci.edu/dataset/517/shoulder+implant+x+ray+manufacturer+classification

(c) Display the images: 這裡是顯示照片的 PYTHON 程式

UCI Shoulder Implant X-Ray Manufacturer

https://www.kaggle.com/redwankarimsony/uci-shoulder-implant-x-ray-manufacturer/code

(d) CMU 的網友, 將程式寫出來了, 提供大家參考:

(註解: 作者將全部的資料, 都拿去training, 再拿去prediction, 所以準確率接近100%)

https://github.com/mannylazalde/ShoulderImplantClassification

5月14日 (週四) : 3 PM ~ 5 PM

分組報告 (10%)

各組繳交期末專題的題目, 每組報告3分鐘, 討論3分鐘.

請說明 : (1)題目重要性, (2)資料來源, (3)EDA

5月21日 (週四) : 3 PM ~ 4 PM

道瓊指數, EDA 和 LSTM (Long and Short Term Memory)：

程式在: NCUAI_20230903\人工智慧概論\3_Keras_experiment\Keras_3_Stock\code\Keras_03_Stock.ipynb

投影片在: NCUAI_20230903\人工智慧概論\1_PDF_Lecture_Note\AI_11_Keras_Stock.pdf

(24) ML-24: 道瓊指數, 資料處理 (EDA) (16:58)

https://www.youtube.com/watch?v=u4b5fyg4X2M&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=17

(25) ML-25: 道瓊指數, LSTM程式 (21:07)

https://www.youtube.com/watch?v=LBD6jQydzKM&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=18

5月21日 (週四) : 4 PM ~ 5 PM

1D-CNN

5月28日 (週四) : 3 PM ~ 5 PM

矩陣: Inverse and Pseudo Inverse

https://www.youtube.com/watch?v=-t1BnsvctRQ&list=PLYgGtiVoYLPcLDi-vW02DnSqmD6e6m6lL&index=8

矩陣: 曲線擬合, 國小近視比例

https://www.youtube.com/watch?v=zGChVR9vaWU&list=PLYgGtiVoYLPcLDi-vW02DnSqmD6e6m6lL&index=9

矩陣: XW=S

https://www.youtube.com/watch?v=pS48hZzUnUI&list=PLYgGtiVoYLPcLDi-vW02DnSqmD6e6m6lL&index=10

最佳化: 梯度下降

https://www.youtube.com/watch?v=rAaiM6AtIVI&list=PLYgGtiVoYLPcLDi-vW02DnSqmD6e6m6lL&index=11

最佳化: 約束函數, 曲線擬合, 國小近視比例

https://www.youtube.com/watch?v=nKLEKppcyeI&list=PLYgGtiVoYLPcLDi-vW02DnSqmD6e6m6lL&index=12

(27) ML-27: 人工智慧概論 10, 最佳化, 梯度下降 (39:08)

https://www.youtube.com/watch?v=WJ5LM27LpDw&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=27

(28) ML-28: 人工智慧概論 11, 最佳化, 約束函數, 曲線擬合, 國小近視比例 (29:13)

https://www.youtube.com/watch?v=OdYJ9gdXVWU&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=28

------------------------------

https://www.youtube.com/watch?v=ibJpTrp5mcE&list=PLJV_el3uVTsPy9oCRY30oBPNLCo89yu49&index=12

6月4日 (週四) : 3 PM ~ 5 PM

在此下載投影片和 Python 檔案： AI_04_05_Explainable_AI_CAM_GradCAM

https://drive.google.com/drive/u/2/folders/1o7u9yEnRE4FmPDbUwwGV2bu0lX_lDHeI

(29) 用可解釋的人工智慧做影像辨識：(1) CAM (31:24)

https://www.youtube.com/watch?v=zhujHoUZfKA&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=29

(30) 用可解釋的人工智慧做影像辨識：(2) Grad-CAM (25:14)

https://www.youtube.com/watch?v=vmt5miwXqTI&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=30

6月11日 (週四) : 3 PM ~ 5 PM

OpenPose:

Realtime Multi-Person 2D Human Pose Estimation using Part Affinity Fields, CVPR 2017 Oral

https://www.youtube.com/watch?v=pW6nZXeWlGM

CMU-Perceptual-Computing-Lab/openpose

https://github.com/CMU-Perceptual-Computing-Lab/openpose

Manipulating Facial Features With OpenCV and Dlib

https://towardsdatascience.com/manipulating-facial-features-with-opencv-and-dlib-14029f136a3d

spare ---

Co-Lab 課堂實作: Run the following three programs on Co-Lab

程式在: NCUAI_20230903\Python_20210810\Keras_10_ResNet50_001.ipynb

(21) ML-23: Classify ImageNet classes with ResNet50 (13:02)

https://www.youtube.com/watch?v=it4fjd_2hUU&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=23

https://keras.io/api/applications/#usage-examples-for-image-classification-models

程式在: NCUAI_20230903\Python_20210810\autoencoder.ipynb

(22) ML-24: Convolutional autoencoder for image denoising (19:00)

https://www.youtube.com/watch?v=PL043XfWRac&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=24

https://keras.io/examples/vision/autoencoder/

程式在: NCUAI_20230903\Python_20210810\3D_image_classification.ipynb

(23) ML-25: 3D image classification from CT scans (15:46)

https://www.youtube.com/watch?v=wor-nXI3OQE&list=PLYgGtiVoYLPdHNmslnjXuF9gjCIj3g-F3&index=25

https://keras.io/examples/vision/3D_image_classification/

On denoising autoencoders trained to minimise binary cross-entropy

https://arxiv.org/abs/1708.08487

spare ---

瑕疵檢測 (Anomaly Detection)

在這裡下載程式： PaDiM_20241218.zip

https://drive.google.com/drive/folders/1zDCFwfkl0lgDG8x7L3q425MxMmUcqG0k?usp=sharing

PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization

https://arxiv.org/abs/2011.08785

xiahaifeng1995/PaDiM-Anomaly-Detection-Localization-master

https://github.com/xiahaifeng1995/PaDiM-Anomaly-Detection-Localization-master

-------------------------------------------------------------

PaDiM 原理与代码解析

https://blog.csdn.net/ooooocj/article/details/127601035

PaDiM : A machine learning model for detecting defective products without retraining

https://medium.com/axinc-ai/padim-a-machine-learning-model-for-detecting-defective-products-without-retraining-5daa6f203377

modification in main_01.py:

replace: def parse_args():

add: debug_a = 'Y'

add: class_name = 'bottle', print(class.name)

comment out: the program in class_name loop

modification in mvtec.py:

replace: dataset_path='./dataset_mvtec'

replace: Image.LANCZOS

--------------------------

modification in main_02.py:

un-comment: the program in class_name loop

re-arrange: move the functions before the main program

--------------------------

modification in main_03.py:

add comments and print the values of layer.shape and B, C, H, W

參考資料 :

Stanford 李飛飛的CNN課程,

http://cs231n.stanford.edu/

Lecture 1 | Introduction to Convolutional Neural Networks for Visual Recognition

https://www.youtube.com/watch?v=vT1JzLTH4G4&list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv&index=1

Alex J. Smola of Berkeley and Amazon, 談理論的教科書

University of California - Berkeley 的免費AI課程, YouTube, 中英文投影片, 及PDF課本

https://courses.d2l.ai/berkeley-stat-157/syllabus.html

英文教科書 : https://d2l.ai/d2l-en.pdf

中文教科書 : https://zh-v2.d2l.ai/d2l-zh.pdf

Day 25 / DL x RL / Hello Reinforcement Learning —— CartPole

https://ithelp.ithome.com.tw/articles/10251599

Deep Reinforcement Learning Hands-On, 2nd Edition

book ==> https://www.amazon.com/Deep-Reinforcement-Learning-Hands-optimization/dp/1838826998

code ==> https://github.com/PacktPublishing/Deep-Reinforcement-Learning-Hands-On-Second-Edition

Reinforcement Learning — Implement TicTacToe

https://towardsdatascience.com/reinforcement-learning-implement-tictactoe-189582bea542

參考資料 :

One-Dimensional CNN, Introduction to 1D Convolutional Neural Networks in Keras for Time Sequences

https://blog.goodaudience.com/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences-3a7ff801a2cf

https://github.com/xitu/gold-miner/blob/master/TODO1/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences.md

Feature selection: Lasso

Lasso Regression with Python

https://www.kirenz.com/post/2019-08-12-python-lasso-regression-auto/

处理不均衡数据 (深度学习)! Dealing with imbalanced data (deep learning) (3:18)

https://www.youtube.com/watch?v=doXeC9_vMhg

Handling Imbalanced Datasets in Deep Learning ==> Class weight, over and under sampling

https://towardsdatascience.com/handling-imbalanced-datasets-in-deep-learning-f48407a0e758

Deep learning unbalanced training data? Solve it like this. ==> Image augmentation

https://towardsdatascience.com/deep-learning-unbalanced-training-data-solve-it-like-this-6c528e9efea6

How to Configure Image Data Augmentation in Keras ==> Image augmentation

https://machinelearningmastery.com/how-to-configure-image-data-augmentation-when-training-deep-learning-neural-networks/

Image data preprocessing

https://keras.io/api/preprocessing/image/

ImageDataGenerator的使用

https://www.itread01.com/content/1546542668.html

Machine learning dataset :

Kaggle

https://www.kaggle.com/datasets

University of California at Irvine

https://archive.ics.uci.edu/

Paper with code

https://paperswithcode.com/datasets

PhysioNet

https://physionet.org/about/database/

Train YOLOv8 :

YOLOV8: How to Train a Custom YOLOv8 Object Detector | Detect, Segment and Classify Objects

https://www.youtube.com/watch?v=ZUhRZ9UTkIM

Train Yolov8 object detection on a custom dataset

https://www.youtube.com/watch?v=m9fH9OWn8YM

期末報告案例

https://drive.google.com/drive/folders/1mBJMGQ9sNjU3xp5vV_UYdAJNs-IaF1Vw?usp=sharing

2020年第一學期的期末報告 :

(1) 使用YOLO3, 進行硬幣種類辨識

(2) 使用YOLO3,分辨蘋果的品種

(3) 使用YOLO3, 進行手勢辨識達成指定命令

(4) 非監督式新聞主題名稱生成

(5) 使用YOLO3, 進行猜拳競賽

(6) 語音記帳機器人

(7) 使用YOLO3, 辨識手部動作

(8) 使用YOLO3, 辨識臉部表情

(9) Openpose辨識姿勢 - 提升用戶專注力

(10) COVID 19 ~ 利用CNN做肺炎檢測

(11) 運用 RNN 預測比特幣收盤價

(12) 使用音樂訊號, 判斷歌曲的演唱者

(13) 人工智慧辨識塗鴉

下載檔案: 01_20210113_人工智慧導論(B班)_期末報告.zip

https://drive.google.com/drive/folders/1mBJMGQ9sNjU3xp5vV_UYdAJNs-IaF1Vw?usp=sharing

2021年第一學期的期末報告 :

(1) 野生貓科動物辨識

(2) Keras 網頁介紹

(3) 音樂產生器

(4) 透過GMCNN進行圖像修復

(5) 變臉是替身攻擊

(6) 雨量分析與預測

(7) 分析網路評論

(8) 深度學習音樂產生器

(9) 預測學生課業表現

(10) Scikit-Learn 網頁介紹

(11) GitHub 網站介紹

(12) Blazepose架構概論

(13) 臉部辨識

(14) 即時語音模仿

(15) 車牌辨識

下載檔案: 02_20220112_機器學習概論(B班)_期末報告.zip

https://drive.google.com/drive/folders/1mBJMGQ9sNjU3xp5vV_UYdAJNs-IaF1Vw?usp=sharing

2022年第一學期的期末報告 :

(1) Forest Fire

(2) Skin Segmentation

(3) Bike Sharing Dataset

(4) 葡萄乾的分類

(5) 問答系統

(6) 使用YOLO辨識撲克牌

(7) 預測避孕法

(8) Bank Marketing

(9) 具有武器辨識

(10) 銷售額預測

(11) 機械學習K組種族分析

(12) 口音辨識

(13) 如何用大型自然語言模型

(14) 使用YOLO計算人數

(15) 判別蘑菇是否能夠食用

(16) 預測比特幣收盤價

下載檔案: 03_20221226_機器學習概論(B班)_期末報告.zip

https://drive.google.com/drive/folders/1mBJMGQ9sNjU3xp5vV_UYdAJNs-IaF1Vw?usp=sharing

2023年第一學期的期末報告 :

(1) 使用YOLOV8_進行車牌辨識

(2) 心臟病發分析與預測

(3) 電池壽命預測

(4) 利用CNN實現人種辨識

(5) 利用CNN做垃圾分類

(6) 辨識是否戴口罩

(7) Company Bankruptcy Prediction

(8) 宵夜要吃什麼

(9) 恆星種類的分辨

(10) RESNET垃圾自動分類系統

(11) 糖尿病的特徵預測

(12) 辨識狗是否跳上沙發及未來發展

(13) 黑白猜

(14) 使用YOLO8進行英文字母辨識

(15) 基於CNN之手勢辨識

(16) 使用YOLOv8進行工地檢測

下載檔案: 04_20231226_機器學習概論(B班)_期末報告.zip

https://drive.google.com/drive/folders/1mBJMGQ9sNjU3xp5vV_UYdAJNs-IaF1Vw?usp=sharing

2024年第一學期的期末報告 :

(1) 太陽能發電預測

(2) 自動垃圾分類

(3) 手勢人機互動系統

(4) 這天氣該穿甚麼好呢

(5) Cleaned vs Dirty

(6) 驗證碼辨識

(7) 使用CNN進行剪刀石頭布之手勢辨識

(8) 手機使用與使用者行為分類

(9) 結合D-Fire資料與深度學習技術提升火災預測能力

(10) A Simple Hawk-eye system implementation

(11) 表情識別

(12) Fine-Tuning SAM for Breast Tumor Segmentation Using Ultrasound Images

下載檔案: 05_20241225_機器學習概論(B班)_期末報告.zip

https://drive.google.com/drive/folders/1mBJMGQ9sNjU3xp5vV_UYdAJNs-IaF1Vw?usp=sharing

Google Sites

Report abuse

Machine Learning

參考資料 :

李宏毅老師的課程 ML Lecture 10: Convolutional Neural Network (1:19:28)

推薦 ==> 台灣大學李宏毅老師的 Machine Learning 影片

推薦 ==> 有人幫李宏毅老師的影片, 寫出完整的筆記

2月26日 (週四) : 3PM ~ 5PM

(1) ML-01: 人工智慧概論 02, 原理 (49:09)

(2) ML-02: (2026-02-23 版本) Install Anaconda, Tensorflow (Keras), Jupyter Notebook, Spyder. (15:25)

(3) ML-03: How to run Python Programs in Jupyter Notebook and Spyder (37:03)

解決 Anaconda 閃退的方法

3月5日 (週四) : 3PM ~ 4PM

(4) ML-04: How to use Numpy (29:09)

(5) ML-05: How to use Pandas (27:19)

(6) ML-06: How to use debugger in Spyder (13:41)

(7) ML-07: How to use Matplotlib (44:37)

import warnings

warnings.filterwarnings('ignore')

3月5日 (週四) : 4PM ~ 5PM

YOLO 原創者的演講 : How computers learn to recognize objects instantly | Joseph Redmon (7:37)

YOLO-v7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Authors: Chien-Yao Wang, Alexey Bochkovskiy, Hong-Yuan Mark Liao. arXiv:2207.02696 (2022)

(8) ML-08: How to install YOLO-v7 (Python = 3.8) (17:09)

3月12日 (週四) : 3PM ~ 4PM

(9) ML-09: 三種鳶尾花, 資料處理 (22:37)

(10) ML-10: 三種鳶尾花, One-Hot encoding and Multi-Layer Perceptron (25:43)

3月12日 (週四) : 4PM ~ 5PM

(11) ML-11: 過度擬合和交叉驗證 (21:37)

(12) ML-12: 三種鳶尾花, Save and Load, Model and Weights (15:00)

KERAS 使用網頁 :

Keras: API - Applications

Keras: Code examples

3月19日 (週四) : 3PM ~ 5PM

(13) Regression

PyInvest - Video

(14) Scikit-Learn ： https://scikit-learn.org/stable/

ML-14: How to use Scikit-Learn (58:38), starting from 14:00

參考資料 :

sklearn.metrics: Metrics

F-1 Score for Multi-Class Classification

Sensitivity, Specificity, Accuracy, Associated Confidence Interval and ROC Analysis

Advanced Time Series Analysis: Decomposition

Precision, Recall & F-Measure (13:41)

ROC and AUC, Clearly Explained! (16:16)

3月26日 (週四) : 3PM ~ 4PM

乳癌資料和 EDA (Exploratory Data Analysis)

(15) ML-15: 乳癌資料, Random Forest (35:05)

(16) ML-16: Kaggle (11:16)

* Kaggle Courses:

作業 #2, Kaggle 競賽 (20%) (5月1日, 23:59 前上傳到 Kaggle 網站)

Example : GS-3073-AI-course, Regression or classification?

3月26日 (週四) : 4PM ~ 5PM

(17) ML-17: How to install LLM environment (Python = 3.10) (7:23)

(18) ML-18: How to install Ollama (7:30)

4月2日 (週四) : 放假

4月9日 (週四) : 3PM ~ 5PM

(19) ML-19: 如何使用 CoLab (32:20) (只看初級班, 0:00~12:44)

4月16日 (週四) : 3PM ~ 5PM

4月23日 (週四) :

(20) ML-20: How to install Streamlit (Python = 3.10) (7:33)

4月30日 (週四) : 3PM ~ 4PM

But what is a neural network? 究竟神經網路是什麼？ (19:13)

手寫數字辨識 (MNIST)

4月30日 (週四) : 4PM ~ 5PM

李宏毅老師的課程 ML Lecture 10: Convolutional Neural Network (1:19:28)

參考資料 :

(20) ML-20: MNIST, Multi-Layer Perceptron from Kaggle (37:57)

(21) ML-21: MNIST digits classification dataset and Simple Conv-Net (21:40)

5月7日 (週四) : 3 PM ~ 4 PM

VGG16 Neural Network Visualization (7:35)

(22) ML-22: CIFAR-10 image classification with CNN (CPU vs GPU) (37:10)

5月7日 (週四) : 4 PM ~ 5 PM

(26) ML-26: Transfer Learning for Analyzing Medical Images (33:45)

(a) Download the paper: Classifying shoulder implants in X-ray images using deep learning

(b) Download the dataset: 這裡可以下載 UCI 的597張照片

Shoulder Implant X-Ray Manufacturer Classification Data Set

(c) Display the images: 這裡是顯示照片的 PYTHON 程式

UCI Shoulder Implant X-Ray Manufacturer

(d) CMU 的網友, 將程式寫出來了, 提供大家參考:

(註解: 作者將全部的資料, 都拿去training, 再拿去prediction, 所以準確率接近100%)

5月14日 (週四) : 3 PM ~ 5 PM