Data Science Programming All-in-One For Dummies

1 - Title Page [Seite 3]
2 - Copyright Page [Seite 4]
3 - Table of Contents [Seite 7]
4 - Introduction [Seite 21]
4.1 - About This Book [Seite 21]
4.2 - Foolish Assumptions [Seite 23]
4.3 - Icons Used in This Book [Seite 24]
4.4 - Beyond the Book [Seite 24]
4.5 - Where to Go from Here [Seite 25]
5 - Book 1 Defining Data Science [Seite 27]
5.1 - Chapter 1 Considering the History and Uses of Data Science [Seite 29]
5.1.1 - Considering the Elements of Data Science [Seite 30]
5.1.1.1 - Considering the emergence of data science [Seite 30]
5.1.1.2 - Outlining the core competencies of a data scientist [Seite 31]
5.1.1.3 - Linking data science, big data, and AI [Seite 32]
5.1.1.4 - Understanding the role of programming [Seite 32]
5.1.2 - Defining the Role of Data in the World [Seite 33]
5.1.2.1 - Enticing people to buy products [Seite 33]
5.1.2.2 - Keeping people safer [Seite 34]
5.1.2.3 - Creating new technologies [Seite 35]
5.1.2.4 - Performing analysis for research [Seite 36]
5.1.2.5 - Providing art and entertainment [Seite 37]
5.1.2.6 - Making life more interesting in other ways [Seite 38]
5.1.3 - Creating the Data Science Pipeline [Seite 38]
5.1.3.1 - Preparing the data [Seite 38]
5.1.3.2 - Performing exploratory data analysis [Seite 38]
5.1.3.3 - Learning from data [Seite 39]
5.1.3.4 - Visualizing [Seite 39]
5.1.3.5 - Obtaining insights and data products [Seite 39]
5.1.4 - Comparing Different Languages Used for Data Science [Seite 40]
5.1.4.1 - Obtaining an overview of data science languages [Seite 40]
5.1.4.2 - Defining the pros and cons of using Python [Seite 42]
5.1.4.3 - Defining the pros and cons of using R [Seite 43]
5.1.5 - Learning to Perform Data Science Tasks Fast [Seite 45]
5.1.5.1 - Loading data [Seite 46]
5.1.5.2 - Training a model [Seite 46]
5.1.5.3 - Viewing a result [Seite 46]
5.2 - Chapter 2 Placing Data Science within the Realm of AI [Seite 49]
5.2.1 - Seeing the Data to Data Science Relationship [Seite 50]
5.2.1.1 - Considering the data architecture [Seite 50]
5.2.1.2 - Acquiring data from various sources [Seite 51]
5.2.1.3 - Performing data analysis [Seite 52]
5.2.1.4 - Archiving the data [Seite 53]
5.2.2 - Defining the Levels of AI [Seite 53]
5.2.2.1 - Beginning with AI [Seite 54]
5.2.2.2 - Advancing to machine learning [Seite 59]
5.2.2.3 - Getting detailed with deep learning [Seite 63]
5.2.3 - Creating a Pipeline from Data to AI [Seite 67]
5.2.3.1 - Considering the desired output [Seite 67]
5.2.3.2 - Defining a data architecture [Seite 67]
5.2.3.3 - Combining various data sources [Seite 67]
5.2.3.4 - Checking for errors and fixing them [Seite 68]
5.2.3.5 - Performing the analysis [Seite 68]
5.2.3.6 - Validating the result [Seite 69]
5.2.3.7 - Enhancing application performance [Seite 69]
5.3 - Chapter 3 Creating a Data Science Lab of Your Own [Seite 71]
5.3.1 - Considering the Analysis Platform Options [Seite 72]
5.3.1.1 - Using a desktop system [Seite 73]
5.3.1.2 - Working with an online IDE [Seite 73]
5.3.1.3 - Considering the need for a GPU [Seite 74]
5.3.2 - Choosing a Development Language [Seite 76]
5.3.3 - Obtaining and Using Python [Seite 78]
5.3.3.1 - Working with Python in this book [Seite 78]
5.3.3.2 - Obtaining and installing Anaconda for Python [Seite 79]
5.3.3.3 - Defining a Python code repository [Seite 84]
5.3.3.4 - Working with Python using Google Colaboratory [Seite 89]
5.3.3.5 - Defining the limits of using Azure Notebooks with Python and R [Seite 91]
5.3.4 - Obtaining and Using R [Seite 92]
5.3.4.1 - Obtaining and installing Anaconda for R [Seite 92]
5.3.4.2 - Starting the R environment [Seite 93]
5.3.4.3 - Defining an R code repository [Seite 95]
5.3.5 - Presenting Frameworks [Seite 96]
5.3.5.1 - Defining the differences [Seite 96]
5.3.5.2 - Explaining the popularity of frameworks [Seite 97]
5.3.5.3 - Choosing a particular library [Seite 99]
5.3.6 - Accessing the Downloadable Code [Seite 100]
5.4 - Chapter 4 Considering Additional Packages and Libraries You Might Want [Seite 101]
5.4.1 - Considering the Uses for Third-Party Code [Seite 102]
5.4.2 - Obtaining Useful Python Packages [Seite 103]
5.4.2.1 - Accessing scientific tools using SciPy [Seite 104]
5.4.2.2 - Performing fundamental scientific computing using NumPy [Seite 105]
5.4.2.3 - Performing data analysis using pandas [Seite 105]
5.4.2.4 - Implementing machine learning using Scikit-learn [Seite 106]
5.4.2.5 - Going for deep learning with Keras and TensorFlow [Seite 106]
5.4.2.6 - Plotting the data using matplotlib [Seite 107]
5.4.2.7 - Creating graphs with NetworkX [Seite 108]
5.4.2.8 - Parsing HTML documents using Beautiful Soup [Seite 108]
5.4.3 - Locating Useful R Libraries [Seite 109]
5.4.3.1 - Using your Python code in R with reticulate [Seite 109]
5.4.3.2 - Conducting advanced training using caret [Seite 110]
5.4.3.3 - Performing machine learning tasks using mlr [Seite 110]
5.4.3.4 - Visualizing data using ggplot2 [Seite 111]
5.4.3.5 - Enhancing ggplot2 using esquisse [Seite 111]
5.4.3.6 - Creating graphs with igraph [Seite 111]
5.4.3.7 - Parsing HTML documents using rvest [Seite 112]
5.4.3.8 - Wrangling dates using lubridate [Seite 112]
5.4.3.9 - Making big data simpler using dplyr and purrr [Seite 113]
5.5 - Chapter 5 Leveraging a Deep Learning Framework [Seite 115]
5.5.1 - Understanding Deep Learning Framework Usage [Seite 116]
5.5.2 - Working with Low-End Frameworks [Seite 117]
5.5.2.1 - Chainer [Seite 117]
5.5.2.2 - PyTorch [Seite 118]
5.5.2.3 - MXNet [Seite 118]
5.5.2.4 - Microsoft Cognitive Toolkit/CNTK [Seite 119]
5.5.3 - Understanding TensorFlow [Seite 120]
5.5.3.1 - Grasping why TensorFlow is so good [Seite 121]
5.5.3.2 - Making TensorFlow easier by using TFLearn [Seite 122]
5.5.3.3 - Using Keras as the best simplifier [Seite 122]
5.5.3.4 - Getting your copy of TensorFlow and Keras [Seite 123]
5.5.3.5 - Fixing the C++ build tools error in Windows [Seite 126]
5.5.3.6 - Accessing your new environment in Notebook [Seite 128]
6 - Book 2 Interacting with Data Storage [Seite 129]
6.1 - Chapter 1 Manipulating Raw Data [Seite 131]
6.1.1 - Defining the Data Sources [Seite 132]
6.1.1.1 - Obtaining data locally [Seite 132]
6.1.1.2 - Using online data sources [Seite 137]
6.1.1.3 - Employing dynamic data sources [Seite 141]
6.1.1.4 - Considering other kinds of data sources [Seite 143]
6.1.2 - Considering the Data Forms [Seite 144]
6.1.2.1 - Working with pure text [Seite 144]
6.1.2.2 - Accessing formatted text [Seite 145]
6.1.2.3 - Deciphering binary data [Seite 146]
6.1.3 - Understanding the Need for Data Reliability [Seite 148]
6.2 - Chapter 2 Using Functional Programming Techniques [Seite 151]
6.2.1 - Defining Functional Programming [Seite 152]
6.2.1.1 - Differences with other programming paradigms [Seite 152]
6.2.1.2 - Understanding its goals [Seite 153]
6.2.2 - Understanding Pure and Impure Languages [Seite 154]
6.2.2.1 - Using the pure approach [Seite 154]
6.2.2.2 - Using the impure approach [Seite 154]
6.2.3 - Comparing the Functional Paradigm [Seite 155]
6.2.3.1 - Imperative [Seite 155]
6.2.3.2 - Procedural [Seite 156]
6.2.3.3 - Object-oriented [Seite 156]
6.2.3.4 - Declarative [Seite 156]
6.2.4 - Using Python for Functional Programming Needs [Seite 157]
6.2.5 - Understanding How Functional Data Works [Seite 158]
6.2.5.1 - Working with immutable data [Seite 159]
6.2.5.2 - Considering the role of state [Seite 159]
6.2.5.3 - Eliminating side effects [Seite 160]
6.2.5.4 - Passing by reference versus by value [Seite 160]
6.2.6 - Working with Lists and Strings [Seite 162]
6.2.6.1 - Creating lists [Seite 164]
6.2.6.2 - Evaluating lists [Seite 164]
6.2.6.3 - Performing common list manipulations [Seite 166]
6.2.6.4 - Understanding the Dict and Set alternatives [Seite 167]
6.2.6.5 - Considering the use of strings [Seite 168]
6.2.7 - Employing Pattern Matching [Seite 170]
6.2.7.1 - Looking for patterns in data [Seite 170]
6.2.7.2 - Understanding regular expressions [Seite 172]
6.2.7.3 - Using pattern matching in analysis [Seite 175]
6.2.7.4 - Working with pattern matching [Seite 176]
6.2.8 - Working with Recursion [Seite 179]
6.2.8.1 - Performing tasks more than once [Seite 179]
6.2.8.2 - Understanding recursion [Seite 181]
6.2.8.3 - Using recursion on lists [Seite 182]
6.2.8.4 - Considering advanced recursive tasks [Seite 183]
6.2.8.5 - Passing functions instead of variables [Seite 184]
6.2.9 - Performing Functional Data Manipulation [Seite 185]
6.2.9.1 - Slicing and dicing [Seite 186]
6.2.9.2 - Mapping your data [Seite 187]
6.2.9.3 - Filtering data [Seite 188]
6.2.9.4 - Organizing data [Seite 189]
6.3 - Chapter 3 Working with Scalars, Vectors, and Matrices [Seite 191]
6.3.1 - Considering the Data Forms [Seite 192]
6.3.2 - Defining Data Type through Scalars [Seite 193]
6.3.3 - Creating Organized Data with Vectors [Seite 194]
6.3.3.1 - Defining a vector [Seite 195]
6.3.3.2 - Creating vectors of a specific type [Seite 195]
6.3.3.3 - Performing math on vectors [Seite 196]
6.3.3.4 - Performing logical and comparison tasks on vectors [Seite 196]
6.3.3.5 - Multiplying vectors [Seite 197]
6.3.4 - Creating and Using Matrices [Seite 198]
6.3.4.1 - Creating a matrix [Seite 198]
6.3.4.2 - Creating matrices of a specific type [Seite 199]
6.3.4.3 - Using the matrix class [Seite 201]
6.3.4.4 - Performing matrix multiplication [Seite 201]
6.3.4.5 - Executing advanced matrix operations [Seite 203]
6.3.5 - Extending Analysis to Tensors [Seite 205]
6.3.6 - Using Vectorization Effectively [Seite 206]
6.3.7 - Selecting and Shaping Data [Seite 207]
6.3.7.1 - Slicing rows [Seite 208]
6.3.7.2 - Slicing columns [Seite 208]
6.3.7.3 - Dicing [Seite 209]
6.3.7.4 - Concatenating [Seite 209]
6.3.7.5 - Aggregating [Seite 214]
6.3.8 - Working with Trees [Seite 215]
6.3.8.1 - Understanding the basics of trees [Seite 215]
6.3.8.2 - Building a tree [Seite 216]
6.3.9 - Representing Relations in a Graph [Seite 218]
6.3.9.1 - Going beyond trees [Seite 218]
6.3.9.2 - Arranging graphs [Seite 219]
6.4 - Chapter 4 Accessing Data in Files [Seite 221]
6.4.1 - Understanding Flat File Data Sources [Seite 222]
6.4.2 - Working with Positional Data Files [Seite 223]
6.4.3 - Accessing Data in CSV Files [Seite 225]
6.4.3.1 - Working with a simple CSV file [Seite 225]
6.4.3.2 - Making use of header information [Seite 228]
6.4.4 - Moving On to XML Files [Seite 229]
6.4.4.1 - Working with a simple XML file [Seite 229]
6.4.4.2 - Parsing XML [Seite 231]
6.4.4.3 - Using XPath for data extraction [Seite 232]
6.4.5 - Considering Other Flat-File Data Sources [Seite 234]
6.4.6 - Working with Nontext Data [Seite 235]
6.4.7 - Downloading Online Datasets [Seite 238]
6.4.7.1 - Working with package datasets [Seite 238]
6.4.7.2 - Using public domain datasets [Seite 239]
6.5 - Chapter 5 Working with a Relational DBMS [Seite 243]
6.5.1 - Considering RDBMS Issues [Seite 244]
6.5.1.1 - Defining the use of tables [Seite 245]
6.5.1.2 - Understanding keys and indexes [Seite 246]
6.5.1.3 - Using local versus online databases [Seite 247]
6.5.1.4 - Working in read-only mode [Seite 248]
6.5.2 - Accessing the RDBMS Data [Seite 248]
6.5.2.1 - Using the SQL language [Seite 249]
6.5.2.2 - Relying on scripts [Seite 251]
6.5.2.3 - Relying on views [Seite 251]
6.5.2.4 - Relying on functions [Seite 252]
6.5.3 - Creating a Dataset [Seite 253]
6.5.3.1 - Combining data from multiple tables [Seite 253]
6.5.3.2 - Ensuring data completeness [Seite 254]
6.5.3.3 - Slicing and dicing the data as needed [Seite 254]
6.5.4 - Mixing RDBMS Products [Seite 254]
6.6 - Chapter 6 Working with a NoSQL DMBS [Seite 257]
6.6.1 - Considering the Ramifications of Hierarchical Data [Seite 258]
6.6.1.1 - Understanding hierarchical organization [Seite 258]
6.6.1.2 - Developing strategies for freeform data [Seite 259]
6.6.1.3 - Performing an analysis [Seite 260]
6.6.1.4 - Working around dangling data [Seite 261]
6.6.2 - Accessing the Data [Seite 263]
6.6.2.1 - Creating a picture of the data form [Seite 263]
6.6.2.2 - Employing the correct transiting strategy [Seite 264]
6.6.2.3 - Ordering the data [Seite 267]
6.6.3 - Interacting with Data from NoSQL Databases [Seite 268]
6.6.4 - Working with Dictionaries [Seite 269]
6.6.5 - Developing Datasets from Hierarchical Data [Seite 270]
6.6.6 - Processing Hierarchical Data into Other Forms [Seite 271]
7 - Book 3 Manipulating Data Using Basic Algorithms [Seite 273]
7.1 - Chapter 1 Working with Linear Regression [Seite 275]
7.1.1 - Considering the History of Linear Regression [Seite 276]
7.1.2 - Combining Variables [Seite 277]
7.1.2.1 - Working through simple linear regression [Seite 277]
7.1.2.2 - Advancing to multiple linear regression [Seite 280]
7.1.2.3 - Considering which question to ask [Seite 282]
7.1.2.4 - Reducing independent variable complexity [Seite 283]
7.1.3 - Manipulating Categorical Variables [Seite 285]
7.1.3.1 - Creating categorical variables [Seite 286]
7.1.3.2 - Renaming levels [Seite 287]
7.1.3.3 - Combining levels [Seite 288]
7.1.4 - Using Linear Regression to Guess Numbers [Seite 289]
7.1.4.1 - Defining the family of linear models [Seite 290]
7.1.4.2 - Using more variables in a larger dataset [Seite 291]
7.1.4.3 - Understanding variable transformations [Seite 294]
7.1.4.4 - Doing variable transformations [Seite 295]
7.1.4.5 - Creating interactions between variables [Seite 297]
7.1.4.6 - Understanding limitations and problems [Seite 302]
7.1.5 - Learning One Example at a Time [Seite 303]
7.1.5.1 - Using Gradient Descent [Seite 303]
7.1.5.2 - Implementing Stochastic Gradient Descent [Seite 303]
7.1.5.3 - Considering the effects of regularization [Seite 307]
7.2 - Chapter 2 Moving Forward with Logistic Regression [Seite 309]
7.2.1 - Considering the History of Logistic Regression [Seite 310]
7.2.2 - Differentiating between Linear and Logistic Regression [Seite 311]
7.2.2.1 - Considering the model [Seite 311]
7.2.2.2 - Defining the logistic function [Seite 312]
7.2.2.3 - Understanding the problems that logistic regression solves [Seite 314]
7.2.2.4 - Fitting the curve [Seite 315]
7.2.2.5 - Considering a pass/fail example [Seite 316]
7.2.3 - Using Logistic Regression to Guess Classes [Seite 317]
7.2.3.1 - Applying logistic regression [Seite 317]
7.2.3.2 - Considering when classes are more [Seite 318]
7.2.3.3 - Defining logistic regression performance [Seite 320]
7.2.4 - Switching to Probabilities [Seite 321]
7.2.4.1 - Specifying a binary response [Seite 321]
7.2.4.2 - Transforming numeric estimates into probabilities [Seite 322]
7.2.5 - Working through Multiclass Regression [Seite 325]
7.2.5.1 - Understanding multiclass regression [Seite 325]
7.2.5.2 - Developing a multiclass regression implementation [Seite 326]
7.3 - Chapter 3 Predicting Outcomes Using Bayes [Seite 329]
7.3.1 - Understanding Bayes' Theorem [Seite 330]
7.3.1.1 - Delving into Bayes history [Seite 330]
7.3.1.2 - Considering the basic theorem [Seite 332]
7.3.2 - Using Naïve Bayes for Predictions [Seite 333]
7.3.2.1 - Finding out that Naïve Bayes isn't so naïve [Seite 334]
7.3.2.2 - Predicting text classifications [Seite 335]
7.3.2.3 - Getting an overview of Bayesian inference [Seite 338]
7.3.3 - Working with Networked Bayes [Seite 344]
7.3.3.1 - Considering the network types and uses [Seite 344]
7.3.3.2 - Understanding Directed Acyclic Graphs (DAGs) [Seite 347]
7.3.3.3 - Employing networked Bayes in predictions [Seite 348]
7.3.3.4 - Deciding between automated and guided learning [Seite 352]
7.3.4 - Considering the Use of Bayesian Linear Regression [Seite 352]
7.3.5 - Considering the Use of Bayesian Logistic Regression [Seite 353]
7.4 - Chapter 4 Learning with K-Nearest Neighbors [Seite 355]
7.4.1 - Considering the History of K-Nearest Neighbors [Seite 356]
7.4.2 - Learning Lazily with K-Nearest Neighbors [Seite 357]
7.4.2.1 - Understanding the basis of KNN [Seite 357]
7.4.2.2 - Predicting after observing neighbors [Seite 358]
7.4.2.3 - Choosing the k parameter wisely [Seite 361]
7.4.3 - Leveraging the Correct k Parameter [Seite 362]
7.4.3.1 - Understanding the k parameter [Seite 362]
7.4.3.2 - Experimenting with a flexible algorithm [Seite 363]
7.4.4 - Implementing KNN Regression [Seite 365]
7.4.5 - Implementing KNN Classification [Seite 367]
8 - Book 4 Performing Advanced Data Manipulation [Seite 371]
8.1 - Chapter 1 Leveraging Ensembles of Learners [Seite 373]
8.1.1 - Leveraging Decision Trees [Seite 374]
8.1.1.1 - Growing a forest of trees [Seite 376]
8.1.1.2 - Seeing Random Forests in action [Seite 378]
8.1.1.3 - Understanding the importance measures [Seite 380]
8.1.1.4 - Configuring your system for importance measures with Python [Seite 381]
8.1.1.5 - Seeing importance measures in action [Seite 381]
8.1.2 - Working with Almost Random Guesses [Seite 384]
8.1.2.1 - Understanding the premise [Seite 385]
8.1.2.2 - Bagging predictors with AdaBoost [Seite 386]
8.1.3 - Meeting Again with Gradient Descent [Seite 389]
8.1.3.1 - Understanding the GBM difference [Seite 389]
8.1.3.2 - Seeing GBM in action [Seite 391]
8.1.4 - Averaging Different Predictors [Seite 392]
8.2 - Chapter 2 Building Deep Learning Models [Seite 393]
8.2.1 - Discovering the Incredible Perceptron [Seite 394]
8.2.1.1 - Understanding perceptron functionality [Seite 395]
8.2.1.2 - Touching the nonseparability limit [Seite 396]
8.2.2 - Hitting Complexity with Neural Networks [Seite 398]
8.2.2.1 - Considering the neuron [Seite 399]
8.2.2.2 - Pushing data with feed-forward [Seite 401]
8.2.2.3 - Defining hidden layers [Seite 403]
8.2.2.4 - Executing operations [Seite 404]
8.2.2.5 - Considering the details of data movement through the neural network [Seite 406]
8.2.2.6 - Using backpropagation to adjust learning [Seite 407]
8.2.3 - Understanding More about Neural Networks [Seite 410]
8.2.3.1 - Getting an overview of the neural network process [Seite 411]
8.2.3.2 - Defining the basic architecture [Seite 411]
8.2.3.3 - Documenting the essential modules [Seite 413]
8.2.3.4 - Solving a simple problem [Seite 416]
8.2.4 - Looking Under the Hood of Neural Networks [Seite 419]
8.2.4.1 - Choosing the right activation function [Seite 419]
8.2.4.2 - Relying on a smart optimizer [Seite 421]
8.2.4.3 - Setting a working learning rate [Seite 422]
8.2.5 - Explaining Deep Learning Differences with Other Forms of AI [Seite 422]
8.2.5.1 - Adding more layers [Seite 423]
8.2.5.2 - Changing the activations [Seite 425]
8.2.5.3 - Adding regularization by dropout [Seite 426]
8.2.5.4 - Using online learning [Seite 427]
8.2.5.5 - Transferring learning [Seite 427]
8.2.5.6 - Learning end to end [Seite 428]
8.3 - Chapter 3 Recognizing Images with CNNs [Seite 429]
8.3.1 - Beginning with Simple Image Recognition [Seite 430]
8.3.1.1 - Considering the ramifications of sight [Seite 430]
8.3.1.2 - Working with a set of images [Seite 431]
8.3.1.3 - Extracting visual features [Seite 437]
8.3.1.4 - Recognizing faces using Eigenfaces [Seite 439]
8.3.1.5 - Classifying images [Seite 443]
8.3.2 - Understanding CNN Image Basics [Seite 447]
8.3.3 - Moving to CNNs with Character Recognition [Seite 449]
8.3.3.1 - Accessing the dataset [Seite 450]
8.3.3.2 - Reshaping the dataset [Seite 451]
8.3.3.3 - Encoding the categories [Seite 452]
8.3.3.4 - Defining the model [Seite 452]
8.3.3.5 - Using the model [Seite 453]
8.3.4 - Explaining How Convolutions Work [Seite 455]
8.3.4.1 - Understanding convolutions [Seite 455]
8.3.4.2 - Simplifying the use of pooling [Seite 459]
8.3.4.3 - Describing the LeNet architecture [Seite 460]
8.3.5 - Detecting Edges and Shapes from Images [Seite 466]
8.3.5.1 - Visualizing convolutions [Seite 467]
8.3.5.2 - Unveiling successful architectures [Seite 469]
8.3.5.3 - Discussing transfer learning [Seite 470]
8.4 - Chapter 4 Processing Text and Other Sequences [Seite 473]
8.4.1 - Introducing Natural Language Processing [Seite 474]
8.4.1.1 - Defining the human perspective as it relates to data science [Seite 474]
8.4.1.2 - Considering the computer perspective as it relates to data science [Seite 475]
8.4.2 - Understanding How Machines Read [Seite 476]
8.4.2.1 - Creating a corpus [Seite 477]
8.4.2.2 - Performing feature extraction [Seite 477]
8.4.2.3 - Understanding the BoW [Seite 478]
8.4.2.4 - Processing and enhancing text [Seite 479]
8.4.2.5 - Maintaining order using n-grams [Seite 481]
8.4.2.6 - Stemming and removing stop words [Seite 482]
8.4.2.7 - Scraping textual datasets from the web [Seite 485]
8.4.2.8 - Handling problems with raw text [Seite 490]
8.4.2.9 - Storing processed text data in sparse matrices [Seite 493]
8.4.3 - Understanding Semantics Using Word Embeddings [Seite 498]
8.4.4 - Using Scoring and Classification [Seite 502]
8.4.4.1 - Performing classification tasks [Seite 502]
8.4.4.2 - Analyzing reviews from e-commerce [Seite 505]
9 - Book 5 Performing Data-Related Tasks [Seite 511]
9.1 - Chapter 1 Making Recommendations [Seite 513]
9.1.1 - Realizing the Recommendation Revolution [Seite 514]
9.1.2 - Downloading Rating Data [Seite 515]
9.1.2.1 - Navigating through anonymous web data [Seite 516]
9.1.2.2 - Encountering the limits of rating data [Seite 519]
9.1.3 - Leveraging SVD [Seite 526]
9.1.3.1 - Considering the origins of SVD [Seite 526]
9.1.3.2 - Understanding the SVD connection [Seite 528]
9.2 - Chapter 2 Performing Complex Classifications [Seite 529]
9.2.1 - Using Image Classification Challenges [Seite 530]
9.2.1.1 - Delving into ImageNet and Coco [Seite 531]
9.2.1.2 - Learning the magic of data augmentation [Seite 533]
9.2.2 - Distinguishing Traffic Signs [Seite 536]
9.2.2.1 - Preparing the image data [Seite 537]
9.2.2.2 - Running a classification task [Seite 540]
9.3 - Chapter 3 Identifying Objects [Seite 545]
9.3.1 - Distinguishing Classification Tasks [Seite 546]
9.3.1.1 - Understanding the problem [Seite 546]
9.3.1.2 - Performing localization [Seite 547]
9.3.1.3 - Classifying multiple objects [Seite 548]
9.3.1.4 - Annotating multiple objects in images [Seite 549]
9.3.1.5 - Segmenting images [Seite 550]
9.3.2 - Perceiving Objects in Their Surroundings [Seite 551]
9.3.2.1 - Considering vision needs in self-driving cars [Seite 551]
9.3.2.2 - Discovering how RetinaNet works [Seite 552]
9.3.2.3 - Using the Keras-RetinaNet code [Seite 554]
9.3.3 - Overcoming Adversarial Attacks on Deep Learning Applications [Seite 558]
9.3.3.1 - Tricking pixels [Seite 559]
9.3.3.2 - Hacking with stickers and other artifacts [Seite 561]
9.4 - Chapter 4 Analyzing Music and Video [Seite 563]
9.4.1 - Learning to Imitate Art and Life [Seite 564]
9.4.1.1 - Transferring an artistic style [Seite 565]
9.4.1.2 - Reducing the problem to statistics [Seite 566]
9.4.1.3 - Understanding that deep learning doesn't create [Seite 568]
9.4.2 - Mimicking an Artist [Seite 568]
9.4.2.1 - Defining a new piece based on a single artist [Seite 569]
9.4.2.2 - Combining styles to create new art [Seite 570]
9.4.2.3 - Visualizing how neural networks dream [Seite 571]
9.4.2.4 - Using a network to compose music [Seite 571]
9.4.2.5 - Other creative avenues [Seite 572]
9.4.3 - Moving toward GANs [Seite 573]
9.4.3.1 - Finding the key in the competition [Seite 574]
9.4.3.2 - Considering a growing field [Seite 576]
9.5 - Chapter 5 Considering Other Task Types [Seite 579]
9.5.1 - Processing Language in Texts [Seite 580]
9.5.1.1 - Considering the processing methodologies [Seite 580]
9.5.1.2 - Defining understanding as tokenization [Seite 581]
9.5.1.3 - Putting all the documents into a bag [Seite 582]
9.5.1.4 - Using AI for sentiment analysis [Seite 586]
9.5.2 - Processing Time Series [Seite 594]
9.5.2.1 - Defining sequences of events [Seite 594]
9.5.2.2 - Performing a prediction using LSTM [Seite 595]
9.6 - Chapter 6 Developing Impressive Charts and Plots [Seite 599]
9.6.1 - Starting a Graph, Chart, or Plot [Seite 600]
9.6.1.1 - Understanding the differences between graphs, charts, and plots [Seite 600]
9.6.1.2 - Considering the graph, chart, and plot types [Seite 602]
9.6.1.3 - Defining the plot [Seite 603]
9.6.1.4 - Drawing multiple lines [Seite 604]
9.6.1.5 - Drawing multiple plots [Seite 604]
9.6.1.6 - Saving your work [Seite 606]
9.6.2 - Setting the Axis, Ticks, and Grids [Seite 607]
9.6.2.1 - Getting the axis [Seite 607]
9.6.2.2 - Formatting the ticks [Seite 610]
9.6.2.3 - Adding grids [Seite 610]
9.6.3 - Defining the Line Appearance [Seite 611]
9.6.3.1 - Working with line styles [Seite 612]
9.6.3.2 - Adding markers [Seite 613]
9.6.4 - Using Labels, Annotations, and Legends [Seite 614]
9.6.4.1 - Adding labels [Seite 615]
9.6.4.2 - Annotating the chart [Seite 616]
9.6.4.3 - Creating a legend [Seite 618]
9.6.5 - Creating Scatterplots [Seite 619]
9.6.5.1 - Depicting groups [Seite 619]
9.6.5.2 - Showing correlations [Seite 620]
9.6.6 - Plotting Time Series [Seite 623]
9.6.6.1 - Representing time on axes [Seite 624]
9.6.6.2 - Plotting trends over time [Seite 625]
9.6.7 - Plotting Geographical Data [Seite 628]
9.6.7.1 - Getting the toolkit [Seite 628]
9.6.7.2 - Drawing the map [Seite 629]
9.6.7.3 - Plotting the data [Seite 633]
9.6.8 - Visualizing Graphs [Seite 635]
9.6.8.1 - Understanding the adjacency matrix [Seite 635]
9.6.8.2 - Using NetworkX basics [Seite 635]
10 - Book 6 Diagnosing and Fixing Errors [Seite 639]
10.1 - Chapter 1 Locating Errors in Your Data [Seite 641]
10.1.1 - Considering the Types of Data Errors [Seite 642]
10.1.2 - Obtaining the Required Data [Seite 644]
10.1.2.1 - Considering the data sources [Seite 644]
10.1.2.2 - Obtaining reliable data [Seite 645]
10.1.2.3 - Making human input more reliable [Seite 646]
10.1.2.4 - Using automated data collection [Seite 648]
10.1.3 - Validating Your Data [Seite 649]
10.1.3.1 - Figuring out what's in your data [Seite 649]
10.1.3.2 - Removing duplicates [Seite 651]
10.1.3.3 - Creating a data map and a data plan [Seite 652]
10.1.4 - Manicuring the Data [Seite 654]
10.1.4.1 - Dealing with missing data [Seite 654]
10.1.4.2 - Considering data misalignments [Seite 659]
10.1.4.3 - Separating out useful data [Seite 660]
10.1.5 - Dealing with Dates in Your Data [Seite 660]
10.1.5.1 - Formatting date and time values [Seite 661]
10.1.5.2 - Using the right time transformation [Seite 661]
10.2 - Chapter 2 Considering Outrageous Outcomes [Seite 663]
10.2.1 - Deciding What Outrageous Means [Seite 664]
10.2.2 - Considering the Five Mistruths in Data [Seite 665]
10.2.2.1 - Commission [Seite 665]
10.2.2.2 - Omission [Seite 666]
10.2.2.3 - Perspective [Seite 666]
10.2.2.4 - Bias [Seite 667]
10.2.2.5 - Frame-of-reference [Seite 668]
10.2.3 - Considering Detection of Outliers [Seite 669]
10.2.3.1 - Understanding outlier basics [Seite 669]
10.2.3.2 - Finding more things that can go wrong [Seite 671]
10.2.3.3 - Understanding anomalies and novel data [Seite 671]
10.2.4 - Examining a Simple Univariate Method [Seite 673]
10.2.4.1 - Using the pandas package [Seite 673]
10.2.4.2 - Leveraging the Gaussian distribution [Seite 675]
10.2.4.3 - Making assumptions and checking out [Seite 676]
10.2.5 - Developing a Multivariate Approach [Seite 677]
10.2.5.1 - Using principle component analysis [Seite 678]
10.2.5.2 - Using cluster analysis [Seite 679]
10.2.5.3 - Automating outliers detection with Isolation Forests [Seite 681]
10.3 - Chapter 3 Dealing with Model Overfitting and Underfitting [Seite 683]
10.3.1 - Understanding the Causes [Seite 684]
10.3.1.1 - Considering the problem [Seite 684]
10.3.1.2 - Looking at underfitting [Seite 685]
10.3.1.3 - Looking at overfitting [Seite 686]
10.3.1.4 - Plotting learning curves for insights [Seite 688]
10.3.2 - Determining the Sources of Overfitting and Underfitting [Seite 690]
10.3.2.1 - Understanding bias and variance [Seite 691]
10.3.2.2 - Having insufficient data [Seite 691]
10.3.2.3 - Being fooled by data leakage [Seite 692]
10.3.3 - Guessing the Right Features [Seite 692]
10.3.3.1 - Selecting variables like a pro [Seite 693]
10.3.3.2 - Using nonlinear transformations [Seite 696]
10.3.3.3 - Regularizing linear models [Seite 704]
10.4 - Chapter 4 Obtaining the Correct Output Presentation [Seite 709]
10.4.1 - Considering the Meaning of Correct [Seite 710]
10.4.2 - Determining a Presentation Type [Seite 711]
10.4.2.1 - Considering the audience [Seite 711]
10.4.2.2 - Defining a depth of detail [Seite 712]
10.4.2.3 - Ensuring that the data is consistent with audience needs [Seite 713]
10.4.2.4 - Understanding timeliness [Seite 713]
10.4.3 - Choosing the Right Graph [Seite 714]
10.4.3.1 - Telling a story with your graphs [Seite 714]
10.4.3.2 - Showing parts of a whole with pie charts [Seite 714]
10.4.3.3 - Creating comparisons with bar charts [Seite 715]
10.4.3.4 - Showing distributions using histograms [Seite 717]
10.4.3.5 - Depicting groups using boxplots [Seite 719]
10.4.3.6 - Defining a data flow using line graphs [Seite 720]
10.4.3.7 - Seeing data patterns using scatterplots [Seite 721]
10.4.4 - Working with External Data [Seite 722]
10.4.4.1 - Embedding plots and other images [Seite 723]
10.4.4.2 - Loading examples from online sites [Seite 723]
10.4.4.3 - Obtaining online graphics and multimedia [Seite 724]
10.5 - Chapter 5 Developing Consistent Strategies [Seite 727]
10.5.1 - Standardizing Data Collection Techniques [Seite 727]
10.5.2 - Using Reliable Sources [Seite 729]
10.5.3 - Verifying Dynamic Data Sources [Seite 731]
10.5.3.1 - Considering the problem [Seite 732]
10.5.3.2 - Analyzing streams with the right recipe [Seite 734]
10.5.4 - Looking for New Data Collection Trends [Seite 735]
10.5.5 - Weeding Old Data [Seite 736]
10.5.6 - Considering the Need for Randomness [Seite 737]
10.5.6.1 - Considering why randomization is needed [Seite 738]
10.5.6.2 - Understanding how probability works [Seite 738]
11 - Index [Seite 741]
12 - EULA [Seite 771]