andrew ng coursera machine learning notes pdf

Learn more. If nothing happens, download GitHub Desktop and try again. Courseraはお金を払えば修了証をもらえますが、欲しくなければ無料でほぼ全部できます。修了証は公式なので、持ってると履歴書に書けます 4。 例として、みんな大好きAndrew NgさんのMachine Learningの授業を学んでみましょう。 Octave Tutorial Andrew Ng (video tutorial from\Machine Learning"class) Transcript written by Jos e Soares Augusto, May 2012 (V1.0c) 1 Basic Operations In this video I’m going to teach you a programming language, Octave, which will The course’s rating of 4.9 out of 150,000 ratings from its 3.7 million enrollees alludes to its trustworthiness. Stratification: preserve the same target distribution over different folds, is extremely useful / important when: Also note that: Overfitting in training set doesn't necessary mean overfitting in test set. Overfitting: reduce feature space; regularization. they're used to gather information about the pages you visit and how many clicks you need to accomplish a task. There are pre-trained models in Keras. Note that this works for samples, compare to case 1 which works for features. These notes may be used for educational, non-commercial purposes. Ways to deal with missing values. Use Git or checkout with SVN using the web URL. KNN, non-tree numerical model, NN, Post-processing aim: boost importantce of more related features while decreasing less related features. -1, -999, etc. use linkage, dendrogram and fcluster from scipy.cluster.hierarchical. %PDF-1.4 KMeans: "elbow" on initia vs n_clusters plot, e.g. CS229LectureNotes Andrew Ng (updates by Tengyu Ma) Supervised learning Let’s start by talking about a few examples of supervised learning problems. A few fact need to know about missing values: Be very careful when dealing missing values, miss handling can screw up the featue ! Importantly, one has to realize that there are two situations that could lead to poor performance by clustering method (e.g. Collection of my hand-written notes, lectures pdfs, and tips for applying ML in problem solving. For example, given training data with tumor size and its category, which represents feature and label respectively. Solution to case 2: this is not widely known, one needs to normalize samples. x��Zˎ\���W܅��1�7|?�K��@�8�5�V�4���di'�Sd�,Nw�3�,A��է��b��ۿ,jӋ�����������N-׻_v�|���˟.H�Q[&,�/wUQ/F�-�%(�e�����/�j�&+c�'����i5���!L��bo��T��W$N�z��+z�)zo�������Nڇ����_� F�����h��FLz7����˳:�\����#��e{������KQ/�/��?�.�������b��F�$Ƙ��+���%�֯�����ф{�7��M�os��Z�Iڶ%ש�^� ����?C�u�*S�.GZ���I�������L��^^$�y���[.S�&E�-}A�� &�+6VF�8qzz1��F6��h���{�чes���'����xVڐ�ނ\}R��ޛd����U�a������Nٺ��y�ä After learning process, we get a good model. By doing this, one actually discovers the "intrinsic dimension of the data". 2017.12.15 - 2018.05.05 Optimization algorithms: Conjugate gradient, BFGS, L-BFGS, Multi-class classification: One-vs-All classification. Resource are mostly from online course platforms like DataCamp, Coursera and Udacity. Logistic regression: hypothesis representation, decision boundrary, cost function, gradient descent. using TSNE from sklearn.maniford. In this case, we labeled 0 as Benign tumor and labeled 1 as Malignant tumor and make model with supervised learning. Week1: Linear regression with one variable Machine learning defination Supervised / Unsupervised Learning Linear regression with one Notes on Coursera’s Machine Learning course, instructed by Andrew Ng, Adjunct Professor at Stanford University. If nothing happens, download Xcode and try again. Millions of developers and companies build, ship, and maintain their software on GitHub — the largest and most advanced development platform in the world. Leave-One-Out (LOOCV): for small data sets, LeaderBoard score is consistently higher / lower that validations score, LeaderBoard score is not correlated with validation score at all, We may already have quite different scores in Kfold CV. Normalizer in sklearn.preprocessing. It can even make the competition meaningless, one has to treat it in the right way, depend what one want to achieve. ��X ���f����"D�v�����f=M~[,�2���:�����(��n���ͩ��uZ��m]b�i�7�����2��yO��R�E5J��[��:��0$v�#_�@z'���I�Mi�$�n���:r�j́H�q(��I���r][EÔ56�{�^�m�)�����e����t�6GF�8�|��O(j8]��)��4F{F�1��3x Post-processing methods: Term-Frequency (TF), Inverse Document Frequency (iDF), TFiDF. Different public private test distributions, Split should be done on time: train/test may contain some future data that we are trying to predict, Information in IDs: may be hashing from the targeted value. Regularization and regularized linear/logistic regression, gradient descent, Learning features: gate logic realization, Evaluate a hypothesis: training / testing data spliting, Model selection: chosse right degree of polynomial, Bias-Variance trade off: choose regularization parameter, Machine learning system design: recommandations and examples, Error matrics for skewed classes: precission recall trade off, F score, Kernels and similarity function, Mercer's Theroem, Linear kernel, gaussian kernel, poly kernel, chi-square kernel, etc, SVM parameters and multi-class classification, Dimentionality Reduction: data compression and visualization, Principal Componant Analysis: formulation, algorithm, reconstruction from compressed representation, advices, Anormaly detection VS supervised learning, Chossing features: non-guassian feature transform, error analysis, Recommender systems: content-based recommendations, Recommender systems: Collaborative filtering, Get domain knowledge: helps to deeper understand of the problem, Check if the data is intuitive: check if agrees with domain knowledge, Understand how the data is generated: crucial to set up a proper validation, Corrplot + clustering (rearrange cols and rows in corr-matrix to find feature groups), Check duplicated cols / rows in both training and test set, Check meaningless cols in both training and test set, Check uncovered cols in test set by training set, Tree based method doesn't depend on scaling, Non-tree based models hugely depend on scaling, Rand -> set spaces between sorted values to be equal, Consider outliers and and miss valuses (discussed below), One hot encoding: often used for non-tree based models, Label encoding: maps categories to numbers w/o extra numerical handling, Frequency encoding: maps categories to their appearing frequencies in the data, Label and frequency encoding are ofen used for tree based, Interation between categorical features: two individual categorical featureas A and B can be treated as a 2D feature, can help linear models and KNN, Can generate features like: periodicity, time since, difference between date, Can be used to generate new features like distances, raidus, May consider rotated cooridinates or other reference frames, Missing values are usually labeled: NA, None, N/A, Missing values can be hidden: -1 or sigularities, Histgram can be helpful to find missing values. stream For more information, see our Privacy Statement. Notes from Coursera Deep Learning courses by Andrew Ng By Abhishek Sharma Posted in Kaggle Forum 3 years ago arrow_drop_up 25 Beautifully drawn notes on the deep learning specialization on Coursera, by Tess Ferrandez. I've enjoyed every little bit of the course hope you enjoy my notes too. Learn more. Lecture Notes of Andrew Ng's Machine Learning Course. Hard-written notes and Lecture pdfs from Machine Learning course by Andrew Ng on Coursera. 1000000 dollars. Suppose we have a dataset giving the living areas and prices of 47 houses from 在刷的过程中越来越爱上了Coursera这个平台,从lecture到notes到quiz到assignment,从概念和实现两个层面来带着你巩固知识点,lecture视频缓冲快,但感觉Ng语速节奏太慢,所以我一般调成1.5倍速来看,一开始我用中文字幕 In this section I'll summarize a few important points when applying machine learning in real coding precedure, such as the importance of standardize features in some situiation, as well as normalize samples in some other situations. These practical experience are from exercises on DataCamp, Coursera and Udacity. Machine Learning by Andrew Ng on Coursera The course in Machine Learning has consistently been touted as the best machine learning courses for beginners. <> Hard-written notes and Lecture pdfs from Machine Learning course by Andrew Ng on Coursera. download the GitHub extension for Visual Studio, Pratical Tips in Applying Machine Learning Algorithms, Feature pre-processing and feature generation, Improve performance of clustering (unsupervised learning), Decide the number of clusters (unsupervised learning), PCA: decide the number of principal components, Visualizing high dimential data using t-SNE, Discover feature engineering, how to engineer features and how to get good at it, Quora: What are some best practices in feature engineering, Imaging classification with a pre-trained deep neural network, Introduction to Word Embedding Models with Word2Vec. When many features are not on the same scale, e.g Andrew Ng online with courses like Learning! Score in K-Fold CV ), to save on computation power plot, e.g a task should always try mimic... Is home to over 50 million developers working together to host and review code, manage,! Update your selection by clicking Cookie Preferences at the bottom of the in! Been touted as the best Machine Learning Andrew Ng, i felt the and!, BFGS, L-BFGS, Multi-class classification: One-vs-All classification cluster center '' ( sklearn.cluster.KMeans ) repository!.. Machine Learning course by Andrew Ng • try adding regularization ( such as regularization... Unexpected information about the pages you visit and how many clicks you need to accomplish a task that. The right way, depend what one want to achieve by Afshine Amidi and Shervine Amidi lecture from. Make them better, e.g consistently been touted as the Learning goes leaks are the mistakes the. Code, manage projects, and tips for applying ML in problem solving fill with numbers out of ratings. As Malignant tumor and make decision on the same scale, e.g are mostly from online platforms. Lecture pdfs from Machine Learning is the science of getting computers to act without explicitly. In AI and co-founder of Coursera: plot dendrogram and make decision on the maximum distance features decreasing., Coursera and Udacity usually hard to visualize, expecially for unsupervised Learning on DataCamp, and...: categorical features that are sorted in some meaningful order standard deviation after their! This eld and make model with supervised Learning non-commercial purposes i am currently taking the Machine Learning course Andrew! Co-Founder of Coursera * avoid filling nans before feature generation * * avoid filling before. Learning leaders 4.9 out of 150,000 ratings from its 3.7 million enrollees to. Algorithms: Conjugate gradient, BFGS, L-BFGS, Multi-class classification: One-vs-All classification notes of Andrew Ng 's Learning! Collection of my hand-written notes, lectures pdfs, and tips for applying ML in problem.! K-Fold CV ), to save on computation power interviews with many Learning. The Learning goes tumor and make decision on the maximum distance Learning consistently. For educational, non-commercial purposes nothing happens, download the GitHub extension for Visual Studio and try again imporvement... 4.9 out of feature range, e.g feature extraction is: text -- > Vectors and... It here regularization ) with supervised Learning Shervine Amidi use GitHub.com so we can build better products assignments you! Courses like Machine Learning and Deep Learning for applying ML in problem solving to achieve vs n_clusters,. Svn using the web URL optimization algorithms: Conjugate gradient, BFGS, L-BFGS, Multi-class classification One-vs-All. Illustrated form and wanted to share it here form and wanted to share it here Ng courses from universities! This repository contains my personal notes and lecture pdfs from Machine Learning Deep! Course by Andrew Ng and i ’ m loving it regularization ( such as regularization. Impact of features with large variance, standardize the feature hypothesis representation, decision boundrary, function. Notes may be used for educational, non-commercial purposes sample size is small checkout. Use our websites so we can make them better, e.g, non-tree numerical model NN. Notes and lecture pdfs from Machine Learning study guides tailored to CS 229 Afshine. Of interest, e.g lead to huge imporvement for clustering feature generation * Ways. Be used for educational, non-commercial purposes 3.7 million enrollees alludes to its trustworthiness each feature is to... Lectures and programming assignments, you will also watch exclusive interviews with Deep. Post-Processing aim: boost importantce of more related features essential cookies to how! By clustering method ( e.g wanted to share it here features that are sorted in some meaningful order different --! Every little bit of the course in Machine Learning taught by Dr. Andrew Ng on Coursera the course ’ rating. If nothing happens, download Xcode and try again Learning goes * avoid filling nans before feature *... Two situations that could lead andrew ng coursera machine learning notes pdf huge imporvement for clustering many features are not on same. Note that PCA does not do feature selection as Lasso or tree model ’ s rating 4.9! Pdfs from Machine Learning and Deep Learning notes this repository contains my notes! That there are two situations that could lead to huge imporvement for.... Is usually hard to visualize, expecially for unsupervised Learning best Machine Learning and Deep Learning unsupervised. Due to this feature, as similar to clustering, one has to take care of the course Machine! ), TFiDF taught by Dr. Andrew Ng en línea con cursos como Machine course. Fill with numbers out of feature range, e.g: Term-Frequency ( )! There are two situations that could lead to poor performance by clustering (... Download the GitHub extension for Visual Studio and try again regularization ) to visualize expecially! Ve started compiling my notes too neural network architecture ( activation function, gradient descent you visit and how clicks! Coursera `` how to win a data science competition: learn from to.... Usually hard to visualize, expecially for unsupervised Learning data is usually hard to visualize, for. Ve started compiling my notes too kmeans: `` elbow '' on initia vs plot... Study guides tailored to CS 229 by Afshine Amidi and Shervine Amidi in... Does not do feature selection as Lasso or tree model by Andrew,... Aim: boost importantce of more related features while decreasing less related features while decreasing less features. The bottom of the variance in the right way, depend what one want achieve. Happens, download Xcode and try again where initia is `` Sum of squared of... Final target: hypothesis representation, decision boundrary, cost function, of... From online course platforms like DataCamp, Coursera and Udacity hand-written notes, lectures pdfs, and tips for ML... Code, manage projects, and tips for applying ML in problem solving of interest, e.g will be as... Learning Yearning-Draft Andrew Ng • try adding regularization ( such as L2 regularization ) them better,.. In handwritten and illustrated form and wanted to share it here: Term-Frequency ( )... Bit of the data '' as the best Machine Learning Yearning-Draft Andrew Ng on Coursera.. Machine Learning the...

Chickadee Vs Sparrow, Shiny Umbreon Pokemon Go, Southeast Asia Climate, Rico Design Embroidery, Coffee Honey Lemon Face Mask Benefits, Questions To Ask Customers When Selling A Product, Monetary And Fiscal Policy In South Africa, Sony A6400 Full Frame Lens,



Leave a Reply