Deep Learning with Python Develop Deep Learning

Models on Theano and TensorFLow Using Keras

Jason Brownlee download

https://textbookfull.com/product/deep-learning-with-python-
develop-deep-learning-models-on-theano-and-tensorflow-using-
keras-jason-brownlee/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Deep Learning for Natural Language Processing Develop

Deep Learning Models for Natural Language in Python
Jason Brownlee

https://textbookfull.com/product/deep-learning-for-natural-
language-processing-develop-deep-learning-models-for-natural-
language-in-python-jason-brownlee/

Deep Learning Projects Using TensorFlow 2: Neural

Network Development with Python and Keras 1st Edition
Vinita Silaparasetty

https://textbookfull.com/product/deep-learning-projects-using-
tensorflow-2-neural-network-development-with-python-and-
keras-1st-edition-vinita-silaparasetty/

Deep Learning with Applications Using Python Chatbots

and Face, Object, and Speech Recognition With
TensorFlow and Keras Springerlink (Online Service)

https://textbookfull.com/product/deep-learning-with-applications-
using-python-chatbots-and-face-object-and-speech-recognition-
with-tensorflow-and-keras-springerlink-online-service/

Beginning Anomaly Detection Using Python-Based Deep

Learning: With Keras and PyTorch Sridhar Alla

https://textbookfull.com/product/beginning-anomaly-detection-
using-python-based-deep-learning-with-keras-and-pytorch-sridhar-
alla/
Computer Vision Using Deep Learning Neural Network
Architectures with Python and Keras 1st Edition Vaibhav
Verdhan

https://textbookfull.com/product/computer-vision-using-deep-
learning-neural-network-architectures-with-python-and-keras-1st-
edition-vaibhav-verdhan/

Deep Learning for Time Series Forecasting Predict the

Future with MLPs CNNs and LSTMs in Python Jason
Brownlee

https://textbookfull.com/product/deep-learning-for-time-series-
forecasting-predict-the-future-with-mlps-cnns-and-lstms-in-
python-jason-brownlee/

Reinforcement Learning: With Open AI, TensorFlow and

Keras Using Python 1st Edition Abhishek Nandy

https://textbookfull.com/product/reinforcement-learning-with-
open-ai-tensorflow-and-keras-using-python-1st-edition-abhishek-
nandy/

Deep Learning with Python Learn Best Practices of Deep

Learning Models with PyTorch 2nd Edition Nikhil Ketkar
Jojo Moolayil

https://textbookfull.com/product/deep-learning-with-python-learn-
best-practices-of-deep-learning-models-with-pytorch-2nd-edition-
nikhil-ketkar-jojo-moolayil/

Deep Learning with Python Learn Best Practices of Deep

Learning Models with PyTorch 2nd Edition Nikhil Ketkar
Jojo Moolayil

https://textbookfull.com/product/deep-learning-with-python-learn-
best-practices-of-deep-learning-models-with-pytorch-2nd-edition-
nikhil-ketkar-jojo-moolayil-2/

   

��������������������������������
����������������������������
�����

��������������
Jason Brownlee

Deep Learning With Python

Develop Deep Learning Models On Theano And TensorFlow Using
Keras
 i

Deep Learning With Python

Copyright 2016 Jason Brownlee. All Rights Reserved.

Edition: v1.7
Contents

Preface iii

I Introduction 1
1 Welcome 2
1.1 Deep Learning The Wrong Way . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Deep Learning With Python . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Book Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Requirements For This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 Your Outcomes From Reading This Book . . . . . . . . . . . . . . . . . . . . . . 7
1.6 What This Book is Not . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

II Background 10
2 Introduction to Theano 11
2.1 What is Theano? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 How to Install Theano . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Simple Theano Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.4 Extensions and Wrappers for Theano . . . . . . . . . . . . . . . . . . . . . . . . 13
2.5 More Theano Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3 Introduction to TensorFlow 15
3.1 What is TensorFlow? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 How to Install TensorFlow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3 Your First Examples in TensorFlow . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.4 Simple TensorFlow Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.5 More Deep Learning Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

4 Introduction to Keras 19
4.1 What is Keras? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2 How to Install Keras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.3 Theano and TensorFlow Backends for Keras . . . . . . . . . . . . . . . . . . . . 20

ii
 iii

4.4 Build Deep Learning Models with Keras . . . . . . . . . . . . . . . . . . . . . . 21

4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

5 Project: Develop Large Models on GPUs Cheaply In the Cloud 23

5.1 Project Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.2 Setup Your AWS Account . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.3 Launch Your Server Instance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.4 Login, Configure and Run . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.5 Build and Run Models on AWS . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5.6 Close Your EC2 Instance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.7 Tips and Tricks for Using Keras on AWS . . . . . . . . . . . . . . . . . . . . . . 34
5.8 More Resources For Deep Learning on AWS . . . . . . . . . . . . . . . . . . . . 34
5.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

III Multilayer Perceptrons 36

6 Crash Course In Multilayer Perceptrons 37
6.1 Crash Course Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6.2 Multilayer Perceptrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
6.3 Neurons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
6.4 Networks of Neurons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
6.5 Training Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

7 Develop Your First Neural Network With Keras 43

7.1 Tutorial Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
7.2 Pima Indians Onset of Diabetes Dataset . . . . . . . . . . . . . . . . . . . . . . 44
7.3 Load Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
7.4 Define Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
7.5 Compile Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
7.6 Fit Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
7.7 Evaluate Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
7.8 Tie It All Together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
7.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

8 Evaluate The Performance of Deep Learning Models 51

8.1 Empirically Evaluate Network Configurations . . . . . . . . . . . . . . . . . . . 51
8.2 Data Splitting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
8.3 Manual k-Fold Cross Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
8.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

9 Use Keras Models With Scikit-Learn For General Machine Learning 57

9.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
9.2 Evaluate Models with Cross Validation . . . . . . . . . . . . . . . . . . . . . . . 58
9.3 Grid Search Deep Learning Model Parameters . . . . . . . . . . . . . . . . . . . 59
9.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
 iv

10 Project: Multiclass Classification Of Flower Species 62

10.1 Iris Flowers Classification Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . 62
10.2 Import Classes and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
10.3 Initialize Random Number Generator . . . . . . . . . . . . . . . . . . . . . . . . 63
10.4 Load The Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
10.5 Encode The Output Variable . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
10.6 Define The Neural Network Model . . . . . . . . . . . . . . . . . . . . . . . . . 65
10.7 Evaluate The Model with k-Fold Cross Validation . . . . . . . . . . . . . . . . . 65
10.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

11 Project: Binary Classification Of Sonar Returns 68

11.1 Sonar Object Classification Dataset . . . . . . . . . . . . . . . . . . . . . . . . . 68
11.2 Baseline Neural Network Model Performance . . . . . . . . . . . . . . . . . . . . 69
11.3 Improve Performance With Data Preparation . . . . . . . . . . . . . . . . . . . 71
11.4 Tuning Layers and Neurons in The Model . . . . . . . . . . . . . . . . . . . . . 73
11.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

12 Project: Regression Of Boston House Prices 77

12.1 Boston House Price Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
12.2 Develop a Baseline Neural Network Model . . . . . . . . . . . . . . . . . . . . . 78
12.3 Lift Performance By Standardizing The Dataset . . . . . . . . . . . . . . . . . . 81
12.4 Tune The Neural Network Topology . . . . . . . . . . . . . . . . . . . . . . . . . 82
12.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

IV Advanced Multilayer Perceptrons and Keras 86

13 Save Your Models For Later With Serialization 87
13.1 Tutorial Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
13.2 Save Your Neural Network Model to JSON . . . . . . . . . . . . . . . . . . . . . 88
13.3 Save Your Neural Network Model to YAML . . . . . . . . . . . . . . . . . . . . 90
13.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

14 Keep The Best Models During Training With Checkpointing 93

14.1 Checkpointing Neural Network Models . . . . . . . . . . . . . . . . . . . . . . . 93
14.2 Checkpoint Neural Network Model Improvements . . . . . . . . . . . . . . . . . 94
14.3 Checkpoint Best Neural Network Model Only . . . . . . . . . . . . . . . . . . . 95
14.4 Loading a Saved Neural Network Model . . . . . . . . . . . . . . . . . . . . . . 96
14.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

15 Understand Model Behavior During Training By Plotting History 98

15.1 Access Model Training History in Keras . . . . . . . . . . . . . . . . . . . . . . 98
15.2 Visualize Model Training History in Keras . . . . . . . . . . . . . . . . . . . . . 99
15.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
 v

16 Reduce Overfitting With Dropout Regularization 102

16.1 Dropout Regularization For Neural Networks . . . . . . . . . . . . . . . . . . . . 102
16.2 Dropout Regularization in Keras . . . . . . . . . . . . . . . . . . . . . . . . . . 103
16.3 Using Dropout on the Visible Layer . . . . . . . . . . . . . . . . . . . . . . . . . 104
16.4 Using Dropout on Hidden Layers . . . . . . . . . . . . . . . . . . . . . . . . . . 105
16.5 Tips For Using Dropout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
16.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

17 Lift Performance With Learning Rate Schedules 108

17.1 Learning Rate Schedule For Training Models . . . . . . . . . . . . . . . . . . . . 108
17.2 Ionosphere Classification Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . 109
17.3 Time-Based Learning Rate Schedule . . . . . . . . . . . . . . . . . . . . . . . . 109
17.4 Drop-Based Learning Rate Schedule . . . . . . . . . . . . . . . . . . . . . . . . . 112
17.5 Tips for Using Learning Rate Schedules . . . . . . . . . . . . . . . . . . . . . . . 114
17.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

V Convolutional Neural Networks 115

18 Crash Course In Convolutional Neural Networks 116
18.1 The Case for Convolutional Neural Networks . . . . . . . . . . . . . . . . . . . . 116
18.2 Building Blocks of Convolutional Neural Networks . . . . . . . . . . . . . . . . . 117
18.3 Convolutional Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
18.4 Pooling Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
18.5 Fully Connected Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
18.6 Worked Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
18.7 Convolutional Neural Networks Best Practices . . . . . . . . . . . . . . . . . . . 119
18.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120

19 Project: Handwritten Digit Recognition 121

19.1 Handwritten Digit Recognition Dataset . . . . . . . . . . . . . . . . . . . . . . . 121
19.2 Loading the MNIST dataset in Keras . . . . . . . . . . . . . . . . . . . . . . . . 122
19.3 Baseline Model with Multilayer Perceptrons . . . . . . . . . . . . . . . . . . . . 123
19.4 Simple Convolutional Neural Network for MNIST . . . . . . . . . . . . . . . . . 127
19.5 Larger Convolutional Neural Network for MNIST . . . . . . . . . . . . . . . . . 131
19.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

20 Improve Model Performance With Image Augmentation 135

20.1 Keras Image Augmentation API . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
20.2 Point of Comparison for Image Augmentation . . . . . . . . . . . . . . . . . . . 136
20.3 Feature Standardization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
20.4 ZCA Whitening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
20.5 Random Rotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
20.6 Random Shifts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
20.7 Random Flips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
20.8 Saving Augmented Images to File . . . . . . . . . . . . . . . . . . . . . . . . . . 145
20.9 Tips For Augmenting Image Data with Keras . . . . . . . . . . . . . . . . . . . 147
 vi

20.10Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

21 Project Object Recognition in Photographs 148

21.1 Photograph Object Recognition Dataset . . . . . . . . . . . . . . . . . . . . . . 148
21.2 Loading The CIFAR-10 Dataset in Keras . . . . . . . . . . . . . . . . . . . . . . 149
21.3 Simple CNN for CIFAR-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
21.4 Larger CNN for CIFAR-10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
21.5 Extensions To Improve Model Performance . . . . . . . . . . . . . . . . . . . . . 157
21.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

22 Project: Predict Sentiment From Movie Reviews 159

22.1 Movie Review Sentiment Classification Dataset . . . . . . . . . . . . . . . . . . 159
22.2 Load the IMDB Dataset With Keras . . . . . . . . . . . . . . . . . . . . . . . . 160
22.3 Word Embeddings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
22.4 Simple Multilayer Perceptron Model . . . . . . . . . . . . . . . . . . . . . . . . 163
22.5 One-Dimensional Convolutional Neural Network . . . . . . . . . . . . . . . . . . 165
22.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

VI Recurrent Neural Networks 169

23 Crash Course In Recurrent Neural Networks 170
23.1 Support For Sequences in Neural Networks . . . . . . . . . . . . . . . . . . . . . 170
23.2 Recurrent Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
23.3 Long Short-Term Memory Networks . . . . . . . . . . . . . . . . . . . . . . . . . 172
23.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

24 Time Series Prediction with Multilayer Perceptrons 174

24.1 Problem Description: Time Series Prediction . . . . . . . . . . . . . . . . . . . . 174
24.2 Multilayer Perceptron Regression . . . . . . . . . . . . . . . . . . . . . . . . . . 176
24.3 Multilayer Perceptron Using the Window Method . . . . . . . . . . . . . . . . . 181
24.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

25 Time Series Prediction with LSTM Recurrent Neural Networks 185

25.1 LSTM Network For Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
25.2 LSTM For Regression Using the Window Method . . . . . . . . . . . . . . . . . 189
25.3 LSTM For Regression with Time Steps . . . . . . . . . . . . . . . . . . . . . . . 191
25.4 LSTM With Memory Between Batches . . . . . . . . . . . . . . . . . . . . . . . 194
25.5 Stacked LSTMs With Memory Between Batches . . . . . . . . . . . . . . . . . . 197
25.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200

26 Project: Sequence Classification of Movie Reviews 201

26.1 Simple LSTM for Sequence Classification . . . . . . . . . . . . . . . . . . . . . . 201
26.2 LSTM For Sequence Classification With Dropout . . . . . . . . . . . . . . . . . 203
26.3 LSTM and CNN For Sequence Classification . . . . . . . . . . . . . . . . . . . . 206
26.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
 vii

27 Understanding Stateful LSTM Recurrent Neural Networks 209

27.1 Problem Description: Learn the Alphabet . . . . . . . . . . . . . . . . . . . . . 209
27.2 LSTM for Learning One-Char to One-Char Mapping . . . . . . . . . . . . . . . 211
27.3 LSTM for a Feature Window to One-Char Mapping . . . . . . . . . . . . . . . . 214
27.4 LSTM for a Time Step Window to One-Char Mapping . . . . . . . . . . . . . . 216
27.5 LSTM State Maintained Between Samples Within A Batch . . . . . . . . . . . . 218
27.6 Stateful LSTM for a One-Char to One-Char Mapping . . . . . . . . . . . . . . . 221
27.7 LSTM with Variable Length Input to One-Char Output . . . . . . . . . . . . . . 224
27.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

28 Project: Text Generation With Alice in Wonderland 228

28.1 Problem Description: Text Generation . . . . . . . . . . . . . . . . . . . . . . . 228
28.2 Develop a Small LSTM Recurrent Neural Network . . . . . . . . . . . . . . . . . 229
28.3 Generating Text with an LSTM Network . . . . . . . . . . . . . . . . . . . . . . 234
28.4 Larger LSTM Recurrent Neural Network . . . . . . . . . . . . . . . . . . . . . . 237
28.5 Extension Ideas to Improve the Model . . . . . . . . . . . . . . . . . . . . . . . 240
28.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

VII Conclusions 242

29 How Far You Have Come 243

30 Getting More Help 244

30.1 Artificial Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
30.2 Deep Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
30.3 Python Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
30.4 Keras Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Preface

Deep learning is a fascinating field. Artificial neural networks have been around for a long time,
but something special has happened in recent years. The mixture of new faster hardware, new
techniques and highly optimized open source libraries allow very large networks to be created
with frightening ease.
This new wave of much larger and much deeper neural networks are also impressively skillful
on a range of problems. I have watched over recent years as they tackle and handily become
state-of-the-art across a range of difficult problem domains. Not least object recognition, speech
recognition, sentiment classification, translation and more.
When a technique comes a long that does so well on such a broad set of problems, you have
to pay attention. The problem is where do you start with deep learning? I created this book
because I thought that there was no gentle way for Python machine learning practitioners to
quickly get started developing deep learning models.
In developing the lessons in this book, I chose the best of breed Python deep learning library
called Keras that abstracted away all of the complexity, ruthlessly leaving you an API containing
only what you need to know to efficiently develop and evaluate neural network models.
This is the guide that I wish I had when I started apply deep learning to machine learning
problems. I hope that you find it useful on your own projects and have as much fun applying
deep learning as I did in creating this book for you.

Jason Brownlee
Melbourne, Australia
2016

viii
 Part I

Introduction

1
Chapter 1

Welcome

Welcome to Deep Learning With Python. This book is your guide to deep learning in Python.
You will discover the Keras Python library for deep learning and how to use it to develop and
evaluate deep learning models. In this book you will discover the techniques, recipes and skills
in deep learning that you can then bring to your own machine learning projects.
Deep learning does have a lot of fascinating math under the covers, but you do not need
to know it to be able to pick it up as a tool and wield it on important projects and deliver
real value. From the applied perspective, deep learning is quite a shallow field and a motivated
developer can quickly pick it up and start making very real and impactful contributions. This is
my goal for you and this book is your ticket to that outcome.

1.1 Deep Learning The Wrong Way

If you ask a deep learning practitioner how to get started with neural networks and deep learning,
what do they say? They say things like

You must have a strong foundation in linear algebra.

You must have a deep knowledge of traditional neural network techniques.

You really must know about probability and statistics.

You should really have a deep knowledge of machine learning.

You probably need to be a PhD in computer science.

You probably need 10 years of experience as a machine learning developer.

You can see that the “common sense” advice means that it is not until after you have
completed years of study and experience that you are ready to actually start developing and
evaluating machine learning model for your machine learning projects.
I think this advice is dead wrong.

2
1.2. Deep Learning With Python 3

1.2 Deep Learning With Python

The approach taken with this book and with all of Machine Learning Mastery is to flip the
traditional approach. If you are interested in deep learning, start by developing and evaluating
deep learning models. Then if you discover you really like it or have a knack for it, later you
can step deeper and deeper into the background and theory, as you need it in order to serve
you in developing better and more valuable results. This book is your ticket to jumping in and
making a ruckus with deep learning.
I have used many of the top deep learning platforms and libraries and I chose what I think
is the best-of-breed platform for getting started and very quickly developing powerful and even
state-of-the-art deep learning models in the Keras deep learning library for Python. Unlike R,
Python is a fully featured programming language allowing you to use the same libraries and
code for model development as you can use in production. Unlike Java, Python has the SciPy
stack for scientific computing and scikit-learn which is a professional grade machine library.
There are two top numerical platforms for developing deep learning models, they are Theano
developed by the University of Montreal and TensorFlow developed at Google. Both were
developed for use in Python and both can be leveraged by the super simple to use Keras library.
Keras wraps the numerical computing complexity of Theano and TensorFlow providing a concise
API that we will use to develop our own neural network and deep learning models.
You will develop your own and perhaps your first neural network and deep learning models
while working through this book, and you will have the skills to bring this amazing new
technology to your own projects. It is going to be a fun journey and I can’t wait to start.

1.3 Book Organization

This book is broken down into three parts.

Lessons where you learn about specific features of neural network models and or how to
use specific aspects of the Keras API.

Projects where you will pull together multiple lessons into an end-to-end project and
deliver a result, providing a template your your own projects.

Recipes where you can copy and paste the standalone code into your own project,
including all of the code presented in this book.

1.3.1 Lessons and Projects

Lessons are discrete and are focused on one topic, designed for you to complete in one sitting.
You can take as long as you need, from 20 minutes if you are racing through, to hours if you
want to experiment with the code or ideas and improve upon the presented results. Your lessons
are divided into five parts:

Background.

Multilayer Perceptrons.

Advanced Multilayer Perceptrons and Keras.

1.3. Book Organization 4

Convolutional Neural Networks.

Recurrent Neural Networks.

1.3.2 Part 2: Background

In this part you will learn about the Theano, TensorFlow and Keras libraries that lay the
foundation for your deep learning journey and about how you can leverage very cheap Amazon
Web Service computing in order to develop and evaluate your own large models in the cloud.
This part of the book includes the following lessons:
Introduction to the Theano Numerical Library.
Introduction to the TensorFlow Numerical Library.
Introduction to the Keras Deep Learning Library.
The lessons will introduce you to the important foundational libraries that you need to
install and use on your workstation. This is taken one step further in a project that shows how
you can cheaply harness GPU cloud computing to develop and evaluate very large deep learning
models.
Project: Develop Large Models on GPUs Cheaply In the Cloud.
At the end of this part you will be ready to start developing models in Keras on your
workstation or in the cloud.

1.3.3 Part 3: Multilayer Perceptrons

In this part you will learn about feedforward neural networks that may be deep or not and how
to expertly develop your own networks and evaluate them efficiently using Keras. This part of
the book includes the following lessons:
Crash Course In Multilayer Perceptrons.
Develop Your First Neural Network With Keras.
Evaluate The Performance of Deep Learning Models.
Use Keras Models With Scikit-Learn For General Machine Learning.
These important lessons are tied together with three foundation projects. These projects
demonstrate how you can quickly and efficiently develop neural network models for tabular
data and provide project templates that you can use on your own regression and classification
machine learning problems. These projects include:
Project: Multiclass Classification Problem.
Project: Binary Classification Problem.
Project: Regression Problem.
At the end of this part you will be ready to discover the finer points of deep learning using
the Keras API.
1.3. Book Organization 5

1.3.4 Part 4: Advanced Multilayer Perceptrons

In this part you will learn about some of the more finer points of the Keras library and API for
practical machine learning projects and some of the more important developments in applied
neural networks that you need to know in order to deliver world class results. This part of the
book includes the following lessons:

Save Your Models For Later With Network Serialization.

Keep The Best Models During Training With Checkpointing.

Understand Model Behavior During Training By Plotting History.

Reduce Overfitting With Dropout Regularization.

Lift Performance With Learning Rate Schedules.

At the end of this part you will know how to confidently wield Keras on your own machine
learning projects with a focus of the finer points of investigating model performance, persisting
models for later use and gaining lifts in performance over baseline models.

1.3.5 Part 5: Convolutional Neural Networks

In this part you will receive a crash course in the dominant model for computer vision machine
learning problems and some natural language problems and how you can best exploit the
capabilities of the Keras API for your own projects. This part of the book includes the following
lessons:

Crash Course In Convolutional Neural Networks.

Improve Model Performance With Image Augmentation.

The best way to learn about this impressive type of neural network model is to apply it.
You will work through three larger projects and apply CNN to image data for object recognition
and text data for sentiment classification.

Project: Handwritten Digit Recognition.

Project: Object Recognition in Photographs.

Project: Movie Review Sentiment Classification.

After completing the lessons and projects in this part you will have the skills and the
confidence of complete and working templates and recipes to tackle your own deep learning
projects using convolutional neural networks.
1.4. Requirements For This Book 6

1.3.6 Part 6: Recurrent Neural Networks

In this part you will receive a crash course in the dominant model for data with a sequence or
time component and how you can best exploit the capabilities of the Keras API for your own
projects. This part of the book includes the following lessons:

Crash Course In Recurrent Neural Networks.

Multilayer Perceptron Models for Time Series Problems.

LSTM Models for Time Series Problems.

Understanding State in LSTM Models for Sequence Prediction.

The best way to learn about this complex type of neural network model is to apply it.
You will work through two larger projects and apply RNN to sequence classification and text
generation.

Project: Sequence Classification of Movie Reviews.

Project: Text Generation With Alice in Wonderland.

After completing the lessons and projects in this part you will have the skills and the
confidence of complete and working templates and recipes to tackle your own deep learning
projects using recurrent neural networks.

1.3.7 Conclusions
The book concludes with some resources that you can use to learn more information about a
specific topic or find help if you need it as you start to develop and evaluate your own deep
learning models.

1.3.8 Recipes
Building up a catalog of code recipes is an important part of your deep learning journey. Each
time you learn about a new technique or new problem type, you should write up a short code
recipe that demonstrates it. This will give you a starting point to use on your next deep learning
or machine learning project.
As part of this book you will receive a catalog of deep learning recipes. This includes recipes
for all of the lessons presented in this book, as well as the complete code for all of the projects.
You are strongly encouraged to add to and build upon this catalog of recipes as you expand
your use and knowledge of deep learning in Python.

1.4 Requirements For This Book

1.4.1 Python and SciPy
You do not need to be a Python expert, but it would be helpful if you knew how to install and
setup Python and SciPy. The lessons and projects assume that you have a Python and SciPy
1.5. Your Outcomes From Reading This Book 7

environment available. This may be on your workstation or laptop, it may be in a VM or a

Docker instance that you run, or it may be a server instance that you can configure in the cloud
as taught in Part II of this book.
Technical Requirements: The technical requirements for the code and tutorials in this
book are as follows:

Python version 2 or 3 installed. This book was developed using Python version 2.7.11.

SciPy and NumPy installed. This book was developed with SciPy version 0.17.0 and
NumPy version 1.11.0.

Matplotlib installed. This book was developed with Matplotlib version 1.5.1.

Pandas installed. This book was developed with Pandas version 0.18.0.

scikit-learn installed. This book was developed with scikit-learn 0.17.1.

You do not need to match the version exactly, but if you are having problems running a
specific code example, please ensure that you update to the same or higher version as the library
specified. You will be guided as to how to install the deep learning libraries Theano, TensorFlow
and Keras in Part II of the book.

1.4.2 Machine Learning

You do not need to be a machine learning expert, but it would be helpful if you knew how to
navigate a small machine learning problem using scikit-learn. Basic concepts like cross validation
and one hot encoding used in lessons and projects are described, but only briefly. There are
resources to go into these topics in more detail at the end of the book, but some knowledge of
these areas might make things easier for you.

1.4.3 Deep Learning

You do not need to know the math and theory of deep learning algorithms, but it would be
helpful to have some basic idea of the field. You will get a crash course in neural network
terminology and models, but we will not go into much detail. Again, there will be resources for
more information at the end of the book, but it might be helpful if you can start with some
idea about neural networks.
Note: All tutorials can be completed on standard workstation hardware with a CPU. A
GPU is not required. Some tutorials later in the book can be sped up significantly by running
on the GPU and a suggestion is provided to consider using GPU hardware at the beginning of
those sections. You can access GPU hardware easily and cheaply in the cloud and a step-by-step
procedure is taught on how to do this in Chapter 5.

1.5 Your Outcomes From Reading This Book

This book will lead you from being a developer who is interested in deep learning with Python
to a developer who has the resources and capabilities to work through a new dataset end-to-end
using Python and develop accurate deep learning models. Specifically, you will know:
1.6. What This Book is Not 8

How to develop and evaluate neural network models end-to-end.

How to use more advanced techniques required for developing state-of-the-art deep learning
models.

How to build larger models for image and text data.

How to use advanced image augmentation techniques in order to lift model performance.

How to get help with deep learning in Python.

From here you can start to dive into the specifics of the functions, techniques and algorithms
used with the goal of learning how to use them better in order to deliver more accurate predictive
models, more reliably in less time. There are a few ways you can read this book. You can dip
into the lessons and projects as your need or interests motivate you. Alternatively, you can
work through the book end-to-end and take advantage of how the lessons and projects build in
complexity and range. I recommend the latter approach.
To get the very most from this book, I recommend taking each lesson and project and build
upon them. Attempt to improve the results, apply the method to a similar but di↵erent problem,
and so on. Write up what you tried or learned and share it on your blog, social media or send
me an email at [email protected]. This book is really what you make of it
and by putting in a little extra, you can quickly become a true force in applied deep learning.

1.6 What This Book is Not

This book solves a specific problem of getting you, a developer, up to speed applying deep
learning to your own machine learning projects in Python. As such, this book was not intended
to be everything to everyone and it is very important to calibrate your expectations. Specifically:

This is not a deep learning textbook. We will not be getting into the basic theory
of artificial neural networks or deep learning algorithms. You are also expected to have
some familiarity with machine learning basics, or be able to pick them up yourself.

This is not an algorithm book. We will not be working through the details of how
specific deep learning algorithms work. You are expected to have some basic knowledge of
deep learning algorithms or how to pick up this knowledge yourself.

This is not a Python programming book. We will not be spending a lot of time on
Python syntax and programming (e.g. basic programming tasks in Python). You are
expected to already be familiar with Python or a developer who can pick up a new C-like
language relatively quickly.

You can still get a lot out of this book if you are weak in one or two of these areas, but you
may struggle picking up the language or require some more explanation of the techniques. If
this is the case, see the Getting More Help chapter at the end of the book and seek out a good
companion reference text.
1.7. Summary 9

1.7 Summary
It is a special time right now. The tools for applied deep learning have never been so good.
The pace of change with neural networks and deep learning feels like it has never been so fast,
spurred by the amazing results that the methods are showing in such a broad range of fields.
This is the start of your journey into deep learning and I am excited for you. Take your time,
have fun and I’m so excited to see where you can take this amazing new technology.

1.7.1 Next
Let’s dive in. Next up is Part II where you will take a whirlwind tour of the foundation libraries
for deep learning in Python, namely the numerical libraries Theano and TensorFlow and the
library you will be using throughout this book called Keras.
 Part II

Background

10
Chapter 2

Introduction to Theano

Theano is a Python library for fast numerical computation that can be run on the CPU or GPU.
It is a key foundational library for deep learning in Python that you can use directly to create
deep learning models. After completing this lesson, you will know:

About the Theano library for Python.

How a very simple symbolic expression can be defined, compiled and calculated.

Where you can learn more about Theano.

Let’s get started.

2.1 What is Theano?

Theano is an open source project released under the BSD license and was developed by the LISA
(now MILA1 ) group at the University of Montreal, Quebec, Canada (home of Yoshua Bengio).
It is named after a Greek mathematician. At it’s heart Theano is a compiler for mathematical
expressions in Python. It knows how to take your structures and turn them into very efficient
code that uses NumPy, efficient native libraries like BLAS and native code to run as fast as
possible on CPUs or GPUs.
It uses a host of clever code optimizations to squeeze as much performance as possible from
your hardware. If you are into the nitty-gritty of mathematical optimizations in code, check out
this interesting list2 . The actual syntax of Theano expressions is symbolic, which can be o↵
putting to beginners. Specifically, expression are defined in the abstract sense, compiled and
later actually used to make calculations.
Theano was specifically designed to handle the types of computation required for large
neural network algorithms used in deep learning. It was one of the first libraries of its kind
(development started in 2007) and is considered an industry standard for deep learning research
and development.
1
http://mila.umontreal.ca/
2
http://deeplearning.net/software/theano/optimizations.html#optimizations

11
2.2. How to Install Theano 12

2.2 How to Install Theano

Theano provides extensive installation instructions for the major operating systems: Windows,
OS X and Linux. Read the Installing Theano guide for your platform3 . Theano assumes a
working Python 2 or Python 3 environment with SciPy. There are ways to make the installation
easier, such as using Anaconda4 to quickly setup Python and SciPy on your machine as well
as using Docker images. With a working Python and SciPy environment, it is relatively
straightforward to install Theano using pip, for example:
sudo pip install Theano

Listing 2.1: Install Theano with pip.

New releases of Theano may be announced and you will want to update to get any bug fixes
and efficiency improvements. You can upgrade Theano using pip as follows:
sudo pip install --upgrade --no-deps theano

Listing 2.2: Upgrade Theano with pip.

You may want to use the bleeding edge version of Theano checked directly out of GitHub.
This may be required for some wrapper libraries that make use of bleeding edge API changes.
You can install Theano directly from a GitHub checkout as follows:
sudo pip install --upgrade --no-deps git+git://github.com/Theano/Theano.git

Listing 2.3: Upgrade Theano with pip from GitHub.

You are now ready to run Theano on your CPU, which is just fine for the development of
small models. Large models may run slowly on the CPU. If you have a Nvidia GPU, you may
want to look into configuring Theano to use your GPU. There is a wealth of documentation of
the Theano homepage for further configuring the library.

Theano v0.8.2is the latest at the time of writing and is used in this book.

2.3 Simple Theano Example

In this section we demonstrate a simple Python script that gives you a flavor of Theano. In this
example we define two symbolic floating point variables a and b. We define an expression that
uses these variables (c = a + b). We then compile this symbolic expression into a function using
Theano that we can use later. Finally, we use our compiled expression by plugging in some real
values and performing the calculation using efficient compiled Theano code under the covers.
# Example of Theano library
import theano
from theano import tensor
# declare two symbolic floating-point scalars
a = tensor.dscalar()
b = tensor.dscalar()
3
http://deeplearning.net/software/theano/install.html
4
https://www.continuum.io/downloads
2.4. Extensions and Wrappers for Theano 13

# create a simple symbolic expression

c = a + b
# convert the expression into a callable object that takes (a,b) and computes c
f = theano.function([a,b], c)
# bind 1.5 to a , 2.5 to b , and evaluate c
result = f(1.5, 2.5)
print(result)

Listing 2.4: Example of Symbolic Arithmetic with Theano.

Running the example prints the output 4, which matches our expectation that 1.5 + 2.5 = 4.0.
This is a useful example as it gives you a flavor for how a symbolic expression can be defined,
compiled and used. Although we have only performed a basic introduction of adding 2 and 2,
you can see how pre-defining computation to be compiled for efficiency may be scaled up to
large vector and matrix operations required for deep learning.

2.4 Extensions and Wrappers for Theano

If you are new to deep learning you do not have to use Theano directly. In fact, you are highly
encouraged to use one of many popular Python projects that make Theano a lot easier to use
for deep learning. These projects provide data structures and behaviors in Python, specifically
designed to quickly and reliably create deep learning models whilst ensuring that fast and
efficient models are created and executed by Theano under the covers. The amount of Theano
syntax exposed by the libraries varies.
Keras is a wrapper library that hides Theano completely and provides a very simple API to
work with to create deep learning models. It hides Theano so well, that it can in fact run as a
wrapper for another popular foundation framework called TensorFlow (discussed next).

2.5 More Theano Resources

Looking for some more resources on Theano? Take a look at some of the following.

Theano Official Homepage

http://deeplearning.net/software/theano/

Theano GitHub Repository

https://github.com/Theano/Theano/

Theano: A CPU and GPU Math Compiler in Python (2010)

http://www.iro.umontreal.ca/~lisa/pointeurs/theano_scipy2010.pdf

List of Libraries Built on Theano

https://github.com/Theano/Theano/wiki/Related-projects

List of Theano configuration options

http://deeplearning.net/software/theano/library/config.html
2.6. Summary 14

2.6 Summary
In this lesson you discovered the Theano Python library for efficient numerical computation.
You learned:

Theano is a foundation library used for deep learning research and development.

Deep learning models can be developed directly in Theano if desired.

The development and evaluation of deep learning models is easier with wrapper libraries
like Keras.

2.6.1 Next
You now know about the Theano library for numerical computation in Python. In the next
lesson you will discover the TensorFlow library released by Google that attempts to o↵er the
same capabilities.
Chapter 3

Introduction to TensorFlow

TensorFlow is a Python library for fast numerical computing created and released by Google.
It is a foundation library that can be used to create deep learning models directly or by using
wrapper libraries that simplify the process built on top of TensorFlow. After completing this
lesson you will know:

About the TensorFlow library for Python.

How to define, compile and evaluate a simple symbolic expression in TensorFlow.

Where to go to get more information on the Library.

Let’s get started.

Note: TensorFlow is not easily supported on Windows at the time of writing. It may be
possible to get TensorFlow working on windows with Docker. TensorFlow is not required to
complete the rest of this book, and if you are on the Windows platform you can skip this lesson.

3.1 What is TensorFlow?

TensorFlow is an open source library for fast numerical computing. It was created and is
maintained by Google and released under the Apache 2.0 open source license. The API is
nominally for the Python programming language, although there is access to the underlying
C++ API. Unlike other numerical libraries intended for use in Deep Learning like Theano,
TensorFlow was designed for use both in research and development and in production systems,
not least RankBrain in Google search1 and the fun DeepDream project2 . It can run on single
CPU systems, GPUs as well as mobile devices and large scale distributed systems of hundreds
of machines.

3.2 How to Install TensorFlow

Installation of TensorFlow is straightforward if you already have a Python SciPy environment.
TensorFlow works with Python 2.7 and Python 3.3+. With a working Python and SciPy
1
https://en.wikipedia.org/wiki/RankBrain
2
https://en.wikipedia.org/wiki/DeepDream

15
3.3. Your First Examples in TensorFlow 16

environment, it is relatively straightforward to install TensorFlow using pip There are a number
of di↵erent distributions of TensorFlow, customized for di↵erent environments, therefore to
install TensorFlow you can follow the Download and Setup instructions3 on the TensorFlow
website. , for example:

TensorFlow v0.10.0is the latest at the time of writing and is used in this book.

3.3 Your First Examples in TensorFlow

Computation is described in terms of data flow and operations in the structure of a directed
graph.

Nodes: Nodes perform computation and have zero or more inputs and outputs. Data that
moves between nodes are known as tensors, which are multi-dimensional arrays of real
values.

Edges: The graph defines the flow of data, branching, looping and updates to state.
Special edges can be used to synchronize behavior within the graph, for example waiting
for computation on a number of inputs to complete.

Operation: An operation is a named abstract computation which can take input attributes
and produce output attributes. For example, you could define an add or multiply operation.

3.4 Simple TensorFlow Example

In this section we demonstrate a simple Python script that gives you a flavor of TensorFlow. In
this example we define two symbolic floating point variables a and b. We define an expression
that uses these variables (c = a + b). This is the same example used in the previous chapter that
introduced Theano. We then compile this symbolic expression into a function using TensorFlow
that we can use later. Finally, we use our complied expression by plugging in some real values
and performing the calculation using efficient compiled TensorFlow code under the covers.
# Example of TensorFlow library
import tensorflow as tf
# declare two symbolic floating-point scalars
a = tf.placeholder(tf.float32)
b = tf.placeholder(tf.float32)
# create a simple symbolic expression using the add function
add = tf.add(a, b)
# bind 1.5 to a , 2.5 to b , and evaluate c
sess = tf.Session()
binding = {a: 1.5, b: 2.5}
c = sess.run(add, feed_dict=binding)
print(c)

Listing 3.1: Example of Symbolic Arithmetic with TensorFlow.

3
https://www.tensorflow.org/versions/r0.9/get_started/os_setup.html
3.5. More Deep Learning Models 17

Running the example prints the output 4, which matches our expectation that 1.5 + 2.5 = 4.0.
This is a useful example as it gives you a flavor for how a symbolic expression can be defined,
compiled and used. Although we have only performed a basic introduction of adding 2 and 2,
you can see how pre-defining computation to be compiled for efficiency may be scaled up to
large vector and matrix operations required for deep learning.

3.5 More Deep Learning Models

Your TensorFlow installation comes with a number of Deep Learning models that you can use
and experiment with directly. Firstly, you need to find out where TensorFlow was installed on
your system. For example, you can use the following Python script:
python -c import os; import inspect; import tensorflow;
print(os.path.dirname(inspect.getfile(tensorflow)))

Listing 3.2: Print Install Directory for TensorFlow.

Change to this directory and take note of the models/ subdirectory. Included are a number
of deep learning models with tutorial-like comments, such as:

Multi-threaded word2vec mini-batched skip-gram model.

Multi-threaded word2vec unbatched skip-gram model.

CNN for the CIFAR-10 network.

Simple, end-to-end, LeNet-5-like convolutional MNIST model example.

Sequence-to-sequence model with an attention mechanism.

Also check the examples directory as it contains an example using the MNIST dataset.
There is also an excellent list of tutorials on the main TensorFlow website4 . They show how
to use di↵erent network types, di↵erent datasets and how to use the framework in various
di↵erent ways. Finally, there is the TensorFlow playground5 where you can experiment with
small networks right in your web browser.

3.6 Summary
In this lesson you discovered the TensorFlow Python library for deep learning. You learned:

TensorFlow is another efficient numerical library like Theano.

Like Theano, deep learning models can be developed directly in TensorFlow if desired.

Also like Theano, TensorFlow may be better leveraged by a wrapper library that abstracts
the complexity and lower level details.
4
https://www.tensorflow.org/versions/r0.9/tutorials/
5
http://playground.tensorflow.org/
3.6. Summary 18

3.6.1 Next
You now know about the Theano and TensorFlow libraries for efficient numerical computation
in Python. In the next lesson you will discover the Keras library that wraps both libraries and
gives you a clean and simple API for developing and evaluating deep learning models.
Chapter 4

Introduction to Keras

Two of the top numerical platforms in Python that provide the basis for deep learning research
and development are Theano and TensorFlow. Both are very powerful libraries, but both can
be difficult to use directly for creating deep learning models. In this lesson you will discover
the Keras Python library that provides a clean and convenient way to create a range of deep
learning models on top of Theano or TensorFlow. After completing this lesson you will know:

About the Keras Python library for deep learning.

How to configure Keras for Theano or TensorFlow.

The standard idiom for creating models with Keras.

Let’s get started.

4.1 What is Keras?

Keras is a minimalist Python library for deep learning that can run on top of Theano or
TensorFlow. It was developed to make developing deep learning models as fast and easy as
possible for research and development. It runs on Python 2.7 or 3.5 and can seamlessly execute
on GPUs and CPUs given the underlying frameworks. It is released under the permissive MIT
license. Keras was developed and maintained by François Chollet, a Google engineer using four
guiding principles:

Modularity: A model can be understood as a sequence or a graph alone. All the concerns
of a deep learning model are discrete components that can be combined in arbitrary ways.

Minimalism: The library provides just enough to achieve an outcome, no frills and
maximizing readability.

Extensibility: New components are intentionally easy to add and use within the frame-
work, intended for developers to trial and explore new ideas.

Python: No separate model files with custom file formats. Everything is native Python.

19
Other documents randomly have
different content
 Hec cum superba uerterit uices dextra.

Whan Fortune with a proud right hand hath torned hir

chaunginge stoundes, she fareth lyk the maneres of the boilinge
Eurype. Glosa. Eurype is an arm of the see that ebbeth and
floweth; and som-tyme the streem is on o syde, and som-tyme on
the other. Text. She, cruel Fortune, casteth adoun kinges 5
that whylom weren y-drad; and she, deceivable, enhaunseth up
the humble chere of him that is discomfited. Ne she neither
hereth ne rekketh of wrecchede wepinges; and she is so hard
that she laugheth and scorneth the wepinges of hem, the whiche
she hath maked wepe with hir free wille. Thus she pleyeth, 10
and thus she proeueth hir strengthes; and sheweth a greet wonder
to alle hir servauntes, yif that a wight is seyn weleful, and over-throwe
in an houre.

Me. I. 3. C. A. Eurippe (twice); Ed. Eurype. 5. C. the; A. that. 6. C. whilom; A.

somtyme. // C. enhanseth; A. enhaunseth. 7. C. vmble; A. humble. // C.
descounfited; A. discomfited. // C. Ne; A. and. 9. C. lyssheth; A. lauȝeth; Ed.
laugheth (Lat. ridet.) 11. A. preueth. // A. strengthe (Lat. uires). // C. A.
grete. 12. C. whiht; A. wyȝt.

Prose II.

Vellem autem pauca tecum.

Certes, I wolde pleten with thee a fewe thinges, usinge the

wordes of Fortune; tak hede now thy-self, yif that she axeth
right. "O thou man, wher-fore makest thou me gilty by thyne
every-dayes pleyninges? What wrong have I don thee? What
goodes have I bireft thee that weren thyne? Stryf or plete 5
with me, bifore what Iuge that thou wolt, of the possessioun
of richesses or of dignitees. And yif thou mayst shewen me
that ever any mortal man hath received any of tho thinges to
ben hise in propre, than wol I graunte frely that alle thilke
thinges weren thyne whiche that thou axest. Whan that nature 10
broughte thee forth out of thy moder wombe, I receyved thee
naked and nedy of alle thinges, and I norisshede thee with my
richesses, and was redy and ententif through my favour to
susteyne thee; and that maketh thee now inpacient ayeins me;
and I envirounde thee with alle the aboundance and shyninge 15
of alle goodes that ben in my right. Now it lyketh me to
with-drawen my hand; thou hast had grace as he that hath
used of foreine goodes: thou hast no right to pleyne thee, as
though thou haddest outrely for-lorn alle thy thinges. Why
pleynest thou thanne? I have done thee no wrong. Richesses, 20
honours, and swiche other thinges ben of my right. My servauntes
knowen me for hir lady; they comen with me, and departen
whan I wende. I dar wel affermen hardily, that yif tho thinges,
of which thou pleynest that thou hast forlorn, hadde ben thyne,
thou ne haddest not lorn hem. Shal I thanne only ben defended 25
to usen my right?

Certes, it is leveful to the hevene to make clere dayes, and,

after that, to coveren tho same dayes with derke nightes. The
yeer hath eek leve to apparailen the visage of the erthe, now
with floures and now with fruit, and to confounden hem som-tyme 30
with reynes and with coldes. The see hath eek his right
to ben som-tyme calme and blaundishing with smothe water,
and som-tyme to ben horrible with wawes and with tempestes.
But the covetise of men, that may nat ben stanched, shal it
binde me to ben stedefast, sin that stedefastnesse is uncouth 35
to my maneres? Swich is my strengthe, and this pley I pleye
continuely. I torne the whirlinge wheel with the torning cercle;
I am glad to chaungen the lowest to the heyest, and the heyest
to the lowest. Worth up, if thou wolt, so it be by this lawe,
that thou ne holde nat that I do thee wronge thogh thou 40
descende adoun, whan the resoun of my pley axeth it.

Wistest thou nat how Cresus, the king of Lydiens, of whiche

king Cyrus was ful sore agast a litel biforn, that this rewliche
Cresus was caught of Cyrus and lad to the fyr to ben brent,
but that a rayn descendede doun fro hevene that rescowede 45
him? And is it out of thy minde how that Paulus, consul of
Rome, whan he hadde taken the king of Perciens, weep pitously
for the captivitee of the self kinge? What other thing biwailen
the cryinges of tragedies but only the dedes of Fortune, that
with an unwar stroke overtorneth realmes of grete nobley? 50
Glose. Tragedie is to seyn, a ditee of a prosperitee for a tyme,
that endeth in wrecchednesse.

Lernedest nat thou in Greke, whan thou were yonge, that

in the entree, or in the celere, of Iupiter, ther ben couched two
tonnes; that on is ful of good, that other is ful of harm? What 55
right hast thou to pleyne, yif thou hast taken more plentevously
of the goode syde, that is to seyn, of my richesses and prosperites;
and what eek if I ne be nat al departed fro thee? What eek
yif my mutabilitee yiveth thee rightful cause of hope to han yit
beter thinges? Natheles dismaye thee nat in thy thought; and 60
thou that art put in the comune realme of alle, ne desyre nat to
liven by thyn only propre right.

Pr. II. 3. C. makes; A. makest. 4. A. wronges (Lat. iniuriam). 5. C. pleten; A.

plete (Lat. contende). 8. C. reseyued. // C. tho; A. these. 9. C. thykke; A.
thilke. 11. C. browht; A. brouȝt. // C. resseyued. 12. A. al thing. // C.
noryssede; A. norysshed. 13. C. fauor; A. fauour. 19. A. vtterly lorn. 20. C.
pleynes. 25. C. I shal; A. Shal I. // C. deffendyd. 28. C. coeueryn; A. keuere
(better coveren). // C. dirk; A. derke. 29. C. apayrelyn; A. apparaile. 30. C.
frut; A. fruyt. 32. C. kalm; A. calme. // C. blawndyssynge; A. blaundyshing.
33. C. om. 2nd with. 35. C. stidefast; A. stedfast. So stide(sted-)fastnesse. 41.
C. dessende. // A. doun. // A. om. the. 42. C. wistesthow; A. Wost thou (Lat.
Nesciebas). // A. om. the. 44. C. kawth; A. cauȝt. 45. C. dessendede; A.
descended. 48. C. kapteuite; A. captiuitee. // C. thinge; A. thinges. 49. C.
cryenges; A. criinges. 50. A. the realmes; C. om. the. // C. noblye; A. nobley.
54. A. seler. // C. cowched; A. couched (Lat. iacere). 56. C. hasthow. 57. A.
rycchesse. 58. A. om. be and al. 59. C. yeueth; A. ȝiueth. 60. A. desmaye. 61.
A. om. the.

Metre II.

Si quantas rapidis flatibus incitus.

Though Plentee, that is goddesse of richesses, hielde adoun

with ful horn, and withdraweth nat hir hand, as many richesses
as the see torneth upward sandes whan it is moeved with
ravisshinge blastes, or elles as many richesses as ther shynen
brighte sterres on hevene on the sterry nightes; yit, for al 5
that, mankinde nolde not cese to wepe wrecchede pleyntes.
And al be it so that god receyveth gladly hir preyers, and
yiveth them (as fool-large) moche gold, and aparaileth coveitous
men with noble or clere honours: yit semeth hem haven y-geten
no-thing, but alwey hir cruel ravyne, devouringe al that they 10
han geten, sheweth other gapinges; that is to seyn, gapen and
desyren yit after mo richesses. What brydles mighten withholden,
to any certein ende, the desordenee covetise of men, whan,
ever the rather that it fleteth in large yiftes, the more ay brenneth
in hem the thurst of havinge? Certes he that, quakinge and 15
dredful, weneth him-selven nedy, he ne liveth never-more riche."

Me. II. 1. A. rycche. // Both hielde; Ed. hylde. 2. A. recches(!). 4. C.

rauyssynge. // A. rycches. 5. A. nyȝt (Lat. noctibus). 6. C. plentes; A.
pleyntes. 7. C. resseyueth. // C. preyres; A. prayers. 8. C. A. yeueth. // A. ful
(for fool). 9. A. folk (for men). 10. C. thinge; A. thing. // C. crewel. 12. A.
rycchesse. 15. A. threst. 16. C. leueth; A. lyueth. // A. -mo.

Prose III.

Hiis igitur si pro se tecum Fortuna loqueretur.

Therfor, yif that Fortune spake with thee for hir-self in this
manere, for-sothe thou ne haddest nat what thou mightest answere.
And, if thou hast any-thing wherwith, thou mayest rightfully defenden
thy compleint, it behoveth thee to shewen it; and I wol
yeven thee space to tellen it.' 5

'Certeynly,' quod I thanne, 'thise beth faire thinges, and

enointed with hony swetenesse of rethorike and musike; and
only whyl they ben herd they ben delicious. But to wrecches is
a depper felinge of harm; this is to seyn, that wrecches felen the
harmes that they suffren more grevously than the remedies or the 10
delites of thise wordes mowen gladen or comforten hem; so that,
whan thise thinges stinten for to soune in eres, the sorwe that is
inset greveth the thought.'
'Right so is it,' quod she. 'For thise ne ben yit none remedies
of thy maladye; but they ben a maner norisshinges of thy sorwe, 15
yit rebel ayein thy curacioun. For whan that tyme is, I shal
moeve swiche thinges that percen hem-self depe. But natheles,
that thou shalt not wilne to leten thy-self a wrecche, hast thou
foryeten the noumber and the manere of thy welefulnesse? I
holde me stille, how that the soverayne men of the citee token 20
thee in cure and kepinge, whan thou were orphelin of fader and
moder, and were chosen in affinitee of princes of the citee; and
thou bigunne rather to be leef and dere than forto ben a neighbour;
the whiche thing is the most precious kinde of any propinquitee
or alyaunce that may ben. Who is it that ne seide tho 25
that thou were right weleful, with so grete a nobleye of thy fadres-in-lawe,
and with the chastitee of thy wyf, and with the oportunitee
and noblesse of thy masculin children, that is to seyn, thy sones?
And over al this—me list to passen the comune thinges—how
thou haddest in thy youthe dignitees that weren werned to olde 30
men. But it delyteth me to comen now to the singuler uphepinge
of thy welefulnesse. Yif any fruit of mortal thinges may han any
weighte or prys of welefulnesse, mightest thou ever foryeten, for
any charge of harm that mighte bifalle, the remembraunce of
thilke day that thou saye thy two sones maked conseileres, and 35
y-lad to-gedere fro thyn house under so greet assemblee of
senatoures and under the blythenesse of poeple; and whan thou
saye hem set in the court in here chayeres of dignitees? Thou,
rethorien or pronouncere of kinges preysinges, deservedest glorie
of wit and of eloquence, whan thou, sittinge bitwene thy two sones, 40
conseileres, in the place that highte Circo, fulfuldest the abydinge
of the multitude of poeple that was sprad abouten thee, with so large
preysinge and laude, as men singen in victories. Tho yave thou
wordes to Fortune, as I trowe, that is to seyn, tho feffedest thou
Fortune with glosinge wordes and deceivedest hir, whan she acoyede 45
thee and norisshede thee as hir owne delyces. Thou bere away of
Fortune a yifte, that is to seyn, swiche guerdoun, that she never yaf
to privee man. Wilt thou therfor leye a rekeninge with Fortune?
She hath now twinkled first upon thee with a wikkede eye. Yif
thou considere the noumbre and the manere of thy blisses and 50
of thy sorwes, thou mayst nat forsaken that thou art yit blisful.
For if thou therfor wenest thy-self nat weleful, for thinges that
tho semeden ioyful ben passed, ther nis nat why thou sholdest wene
thy-self a wrecche; for thinges that semen now sorye passen also.

Art thou now comen first, a sodein gest, in-to the shadwe or 55
tabernacle of this lyf; or trowest thou that any stedefastnesse be
in mannes thinges, whan ofte a swift houre dissolveth the same
man; that is to seyn, whan the soule departeth fro the body? For,
al-though that selde is ther any feith that fortunous thinges wolen
dwellen, yit natheles the laste day of a mannes lyf is a manere 60
deeth to Fortune, and also to thilke that hath dwelt. And therfor,
what, wenestow, thar [thee] recche, yif thou forlete hir in deyinge,
or elles that she, Fortune, forlete thee in fleeinge awey?

Pr. III. 2. A. om. nat. 4. A. tellen (for defenden). 6. C. bet (for beth); A. ben.
8. C. delysyos; A. deliciouse. 15. C. maledye. // C. noryssynges; A.
norissinges. // C. sorwes; A. sorwe (Lat. doloris). 17. C. swych; A. swiche. 20.
C. souerane; A. souerayn. 23. C. begunne; A. bygunne. 24. C. neysshebour;
A. neyȝbour. // C. presyous. 26. A. om. tho that. // A. nere (for were). // C.
fadyris. 27. C. castete; A. chastite. 29. C. lyste; A. lyst. // C. the; A. of. 30. A.
thought (for youthe); Ed. youthe. 32. C. wel-; A. wele-. // C. frute; A. fruyt.
36. C. A semble; A. Ed. assemble. 37. C. peeple; A. poeple. 39. C. des-; A.
de-. 40. C. bitwyen; A. bytwix; Ed. bytwene. 41. C. hihte; A. hyȝt. // C. A. Ed.
all insert and before fulfuldest; I omit it, because it obscures the sense. 42. A.
om. the and so. 44. C. to; A. of. 45. So Ed.; C. A. desseiuedest. 46. C.
noryssede; A. norsshed; Ed. norisshed. // A. hast had (for bere away). // C.
bar. 47. C. A. gerdoun; Ed. guerdon. 48. C. lye; A. leye; Ed. laye (Lat. ponere).
49. C. om. a. 50. C. blysse (wrongly); A. Ed. blisses. 51. C. art; A. Ed. nart. //
C. blysse-; A. blys-. 53. C. the; A. tho (Lat. tunc). 57. C. dyssoluede; A. Ed.
dissolueth. 59. C. al that thowgh; A. Ed. although that. // Ed. selde; C. ȝelde
(= zelde); A. yelde (= ȝelde); Lat. rara. // C. fortune; A. Ed. fortunous. 62: C.
weenestow; A. wenest thou. // C. dar; A. thar. // I supply thee. // C. recke; A.
recche.

Metre III.

Cum polo Phebus roseis quadrigis.

Whan Phebus, the sonne, biginneth to spreden his cleernesse

with rosene chariettes, thanne the sterre, y-dimmed, paleth hir
whyte cheres, by the flambes of the sonne that overcometh the
sterre-light. This is to seyn, whan the sonne is risen, the dey-sterre
wexeth pale, and leseth hir light for the grete brightnesse of the 5
sonne.

Whan the wode wexeth rody of rosene floures, in the first somer
sesoun, thorugh the brethe of the winde Zephirus that wexeth
warm, yif the cloudy wind Auster blowe felliche, than goth awey
the fairenesse of thornes. 10

Ofte the see is cleer and calm withoute moevinge flodes; and
ofte the horrible wind Aquilon moeveth boilinge tempestes and
over-whelveth the see.

Yif the forme of this worlde is so selde stable, and yif it turneth
by so many entrechaunginges, wolt thou thanne trusten in the 15
tomblinge fortunes of men? Wolt thou trowen on flittinge goodes?
It is certein and establisshed by lawe perdurable, that no-thing that
is engendred nis stedefast ne stable.'

Me. III. 1. C. hyr; A. Ed. his. 2. C. palyt. 3. A. flamus. 7. C. rosyn; A. rosene.

9. C. A. wynde. 10. C. thornesse. 11. C. floedes. 13. Ed. -whelueth; C. -
welueeth; A. -whelweth. 14. Ed. selde; C. ȝeelde (= zeelde); A. om. (Lat.
rara). 15. C. wolthow; A. Ed. wilt thou. 16. C. towmblynge; Ed. tomblyng; A.
trublynge (Lat. caducis). // C. wolthow; A. Ed. wilt thou. // C. Ed. on; A. in. //
C. flettynge; A. flittyng. 17. C. is it; A. It is. // C. A. establyssed; Ed.
establysshed. // C. thinge; A. thing. 18. C. estable; A. stable.

Prose IV.

Tunc ego, uera, inquam, commemoras.

Thanne seide I thus: 'O norice of alle vertues, thou seist ful
sooth; ne I ne may nat forsake the right swifte cours of my
prosperitee; that is to seyn, that prosperitee ne be comen to me
wonder swiftly and sone. But this is a thing that greetly smerteth
me whan it remembreth me. For in alle adversitee of fortune, 5
the most unsely kinde of contrarious fortune is to han ben
weleful.'

'But that thou,' quod she, 'abyest thus the torment of thy
false opinioun, that mayst thou nat rightfully blamen ne aretten
to thinges: as who seith, for thou hast yit many habundaunces of 10
thinges.

Text. For al be it so that the ydel name of aventurous

welefulnesse moeveth thee now, it is leveful that thou rekne with
me of how manye grete thinges thou hast yit plentee. And
therfor, yif that thilke thing that thou haddest for most precious 15
in al thy richesse of fortune be kept to thee yit, by the grace of
god, unwemmed and undefouled, mayst thou thanne pleyne
rightfully upon the meschef of Fortune, sin thou hast yit thy
beste thinges? Certes, yit liveth in good point thilke precious
honour of mankinde, Symacus, thy wyves fader, which that is 20
a man maked alle of sapience and of vertu; the whiche man
thou woldest byen redely with the prys of thyn owne lyf. He
biwayleth the wronges that men don to thee, and nat for him-self;
for he liveth in sikernesse of any sentences put ayeins him. And
yit liveth thy wyf, that is atempre of wit, and passinge other 25
wimmen in clennesse of chastetee; and for I wol closen shortely
hir bountees, she is lyk to hir fader. I telle thee wel, that she
liveth looth of this lyf, and kepeth to thee only hir goost; and is
al maat and overcomen by wepinge and sorwe for desyr of thee,
in the whiche thing only I moot graunten that thy welefulnesse is 30
amenused. What shal I seyn eek of thy two sones, conseilours,
of whiche, as of children of hir age, ther shyneth the lyknesse of
the wit of hir fader or of hir elder fader? And sin the sovereyn
cure of alle mortel folk is to saven hir owen lyves, O how weleful
art thou, yif thou knowe thy goodes! For yit ben ther 35
thinges dwelled to thee-ward, that no man douteth that they ne
ben more dereworthe to thee than thyn owen lyf. And for-thy
drye thy teres, for yit nis nat everich fortune al hateful to thee-ward,
ne over greet tempest hath nat yit fallen upon thee, whan
that thyn ancres cleven faste, that neither wolen suffren the 40
counfort of this tyme present ne the hope of tyme cominge to
passen ne to faylen.'

'And I preye,' quod I, 'that faste moten they halden; for

whyles that they halden, how-so-ever that thinges ben, I shal wel
fleten forth and escapen; but thou mayst wel seen how grete 45
aparayles and aray that me lakketh, that ben passed away fro
me.'

'I have som-what avaunsed and forthered thee,' quod she, 'yif
that thou anoye nat or forthinke nat of al thy fortune: as who
seith, I have som-what comforted thee, so that thou tempest thee nat 50
thus with al thy fortune, sin thou hast yit thy beste thinges. But
I may nat suffren thy delices, that pleynest so wepinge and
anguissous, for that ther lakketh som-what to thy welefulnesse.
For what man is so sad or of so parfit welefulnesse, that he ne
stryveth and pleyneth on som halve ayen the qualitee of his 55
estat? For-why ful anguissous thing is the condicioun of mannes
goodes; for either it cometh nat al-togider to a wight, or elles it
last nat perpetuel. For sum man hath grete richesses, but he is
ashamed of his ungentel linage; and som is renowned of noblesse
of kinrede, but he is enclosed in so grete anguisshe of nede 60
of thinges, that him were lever that he were unknowe. And
som man haboundeth both in richesse and noblesse, but yit he
bewaileth his chaste lyf, for he ne hath no wyf. And som man is
wel and selily y-maried, but he hath no children, and norissheth
his richesses to the eyres of strange folkes. And som man is 65
gladed with children, but he wepeth ful sory for the trespas of
his sone or of his doughter. And for this ther ne acordeth no
wight lightly to the condicioun of his fortune; for alwey to every
man ther is in som-what that, unassayed, he ne wot nat; or elles
he dredeth that he hath assayed. And adde this also, that every 70
weleful man hath a ful delicat felinge; so that, but-yif alle thinges
bifalle at his owne wil, for he is impacient, or is nat used to han
non adversitee, anon he is throwen adoun for every litel thing.
And ful litel thinges ben tho that withdrawen the somme or the
perfeccioun of blisfulnesse fro hem that ben most fortunat. How 75
many men, trowest thou, wolden demen hem-self to ben almost in
hevene, yif they mighten atayne to the leest party of the remnaunt
of thy fortune? This same place that thou clepest exil, is
contree to hem that enhabiten heer, and forthy nothing [is]
wrecched but whan thou wenest it: as who seith, thou thy-self, ne 80
no wight elles, nis a wrecche, but whan he weneth him-self a wrecche
by reputacioun of his corage. And ayeinward, alle fortune is blisful
to a man by the agreabletee or by the egalitee of him that
suffreth it.

What man is that, that is so weleful, that nolde changen his 85

estat whan he hath lost pacience? The swetnesse of mannes
welefulnesse is sprayned with many biternesses; the whiche welefulnesse,
al-though it seme swete and ioyful to hem that useth it,
yit may it nat ben with-holden that it ne goth away whan it wole.
Thanne is it wel sene, how wrecched is the blisfulnesse of mortal 90
thinges, that neither it dureth perpetuel with hem that every
fortune receiven agreablely or egaly, ne it delyteth nat in al to
hem that ben anguissous. O ye mortal folk, what seke ye thanne
blisfulnesse out of your-self, whiche that is put in your-self?
Errour and folye confoundeth yow. 95

I shal shewe thee shortely the poynt of sovereyne blisfulnesse.

Is ther any-thing more precious to thee than thy-self? Thou
wolt answere, "nay." Thanne, yif it so be that thou art mighty
over thy-self, that is to seyn, by tranquillitee of thy sowle, than hast
thou thing in thy power that thou noldest never lesen, ne Fortune 100
ne may nat beneme it thee. And that thou mayst knowe that
blisfulnesse ne may nat standen in thinges that ben fortunous
and temporel, now understonde and gader it to-gidere thus:
Yif blisfulnesse be the sovereyn good of nature that liveth by
resoun, ne thilke thing nis nat sovereyn good that may be taken 105
awey in any wyse, (for more worthy thing and more digne is
thilke thing that may nat ben taken awey); than sheweth it wel,
that the unstablenesse of fortune may nat atayne to receiven
verray blisfulnesse. And yit more-over: what man that this
toumbling welefulnesse ledeth, either he woot that it is chaungeable, 110
or elles he woot it nat. And yif he woot it nat, what blisful
fortune may ther be in the blindnesse of ignorance? And yif he
woot that it is chaungeable, he moot alwey ben adrad that he ne
lese that thing that he ne doubteth nat but that he may lesen it;
as who seith, he mot ben alwey agast, lest he lese that he wot wel he 115
may lese it. For which, the continuel dreed that he hath ne
suffreth him nat to ben weleful. Or yif he lese it, he weneth to
be dispysed and forleten. Certes eek, that is a ful litel good that
is born with evene herte whan it is lost; that is to seyn, that men
do no more fors of the lost than of the havinge. And for as moche 120
as thou thy-self art he, to whom it hath ben shewed and proved
by ful manye demonstraciouns, as I wot wel, that the sowles of
men ne mowe nat deyen in no wyse; and eek sin it is cleer and
certein, that fortunous welefulnesse endeth by the deeth of the
body; it may nat ben douted that, yif that deeth may take awey 125
blisfulnesse, that alle the kinde of mortal thinges ne descendeth
in-to wrecchednesse by the ende of the deeth. And sin we knowen
wel, that many a man hath sought the fruit of blisfulnesse nat
only with suffringe of deeth, but eek with suffringe of peynes and
tormentes; how mighte than this present lyf maken men blisful, 130
sin that, whan thilke selve lyf is ended, it ne maketh folk no
wrecches?

Pr. IV. 1. C. vertuus; A. vertues. 4. C. om. a. 6. C. vnȝely (= vnzely); A. Ed.

vnsely. 8. A. abaist (!). // C. tormentz; A. tourment (Lat. supplicium). 10. C. -
daunce; A. Ed. -daunces. 13. C. leefful; A. leueful. 15. C. thinge; A. thing. 19.
C. leueth; A. lyueth. 21. C. om. 2nd of. 24. C. leueth; A. liueth. 29. C. maad;
A. maat; Ed. mate. 30. C. thinge; A. thing. 31. C. amenyssed; A. Ed.
amenused. 32. C. lyke-; A. lyk-. 33. A. Ed. eldefadir. 35. A. But (for For). 36.
So C. Ed.; A. dwellyng. // A. -wardes. 40. A. cliue. 42. A. fallen. 43. A. holden.
44. C. A. halden. 45. C. mayste. 49. A. forthenke. 52. C. delites (?); A. Ed.
delices (Lat. delicias). 55. C. Ed. and; A. or. 57. A. om. nat. 58. A. lasteth. //
A. perpetuely. // A. rycchesse. 59. A. renomed. 60. anguisshe of] A. angre for.
63. Ed. chaste; C. caste; A. chast. 64. C. zelyly; A. Ed. selily. // C. hat. // C.
noriseth; A. norissheth. 66. C. A. sory; Ed. sore. 69. A. is in mest som-what.
71. A. wel (for ful). 72. Ed. is; C. A. om. 77. A. remenaunt. 79. I supply is;
Lat. nihil est miserum. 80. C. ho; A. who. 81. A. no (for a). 83. C. egreablete;
A. agreablete. 86. C. what (!); A. whan. // C. lost; A. lorn. 87. C. sprayngd (!);
A. y-spranid; Ed. spraynte. // C. beter-; A. bitter-. // C. weche. 89. C. wan. //
C. woole; A. wol. 92. C. resseyuen; A. receyuen. 100, 106. C. thinge; A. thing.
101. A. bynyme. 102. A. om. ne. 107. C. take; A. taken. 108. C. resseyuen; A.
receyue. 110. A. om. it. 115. C. list; A. lest. 116. A. om. it. 118. A. forleten
hit. 120. C. A. lost; Ed. losse. // C. meche (for moche). 126. C. dessendeth; A.
descendith. 128. C. frut; A. fruit.

Metre IV.

Quisquis uolet perennem Cautus ponere sedem.

What maner man, stable and war, that wole founden him
a perdurable sete, and ne wole nat ben cast down with the loude
blastes of the wind Eurus; and wole despyse the see, manasinge
with flodes; lat him eschewen to bilde on the cop of the mountaigne
or in the moiste sandes. For the felle wind Auster 5
tormenteth the cop of the mountaigne with all his strengthes;
and the lause sandes refusen to beren the hevy wighte.

And forthy, if thou wolt fleen the perilous aventure, that is to

seyn, of the worlde; have minde certeinly to ficchen thyn hous of
a merye site in a lowe stoon. For al-though the wind, troubling 10
the see, thondre with over-throwinges, thou that art put in quiete,
and weleful by strengthe of thy palis, shalt leden a cleer age,
scorninge the woodnesses and the ires of the eyr.

Me. IV. 1. C. waar. 7. Ed. lose; A. lowe see(!); (Lat. solutae). // A. weyȝte. 10.
C. lowh; A. Ed. lowe. 12. C. A. palys (Lat. ualli).

Prose V.

Set cum rationum iam in te.

But for as moche as the norisshinges of my resouns descenden

now in-to thee, I trowe it were tyme to usen a litel strenger
medicynes. Now understond heer, al were it so that the yiftes of
Fortune ne were nat brutel ne transitorie, what is ther in hem
that may be thyn in any tyme, or elles that it nis foul, yif that it 5
be considered and loked perfitly? Richesses, ben they precious
by the nature of hem-self, or elles by the nature of thee? What is
most worth of richesses? Is it nat gold or might of moneye
assembled? Certes, thilke gold and thilke moneye shyneth and
yeveth betere renoun to hem that despenden it thanne to thilke 10
folk that mokeren it; for avarice maketh alwey mokereres to ben
hated, and largesse maketh folk cleer of renoun. For sin that
swich thing as is transferred fram o man to another ne may nat
dwellen with no man; certes, thanne is thilke moneye precious
whan it is translated into other folk and stenteth to ben had, by 15
usage of large yevinge of him that hath yeven it. And also: yif
that al the moneye that is over-al in the worlde were gadered
toward o man, it sholde maken alle other men to ben nedy as of that.
And certes a voys al hool, that is to seyn, with-oute amenusinge,
fulfilleth to-gidere the hering of moche folk; but certes, youre 20
richesses ne mowen nat passen in-to moche folke with-oute
amenusinge. And whan they ben apassed, nedes they maken
hem pore that for-gon the richesses.

O! streite and nedy clepe I this richesse, sin that many folk
ne may nat han it al, ne al may it nat comen to o man with-outen 25
povertee of alle other folk! And the shyninge of gemmes, that
I clepe precious stones, draweth it nat the eyen of folk to hem-ward,
that is to seyn, for the beautee? But certes, yif ther were
beautee or bountee in the shyninge of stones, thilke cleernesse is
of the stones hem-self, and nat of men; for whiche I wondre 30
gretly that men mervailen on swiche thinges. For-why, what
thing is it, that yif it wanteth moeving and Ioynture of sowle and
body, that by right mighte semen a fair creature to him that hath
a sowle of resoun? For al be it so that gemmes drawen to hem-self
a litel of the laste beautee of the world, through the entente of 35
hir creatour and through the distinccioun of hem-self; yit, for as
mochel as they ben put under youre excellence, they ne han nat
deserved by no wey that ye sholden mervailen on hem. And
the beautee of feldes, delyteth it nat mochel un-to yow?'

Boece. 'Why sholde it nat delyten us, sin that it is a right fair 40
porcioun of the right faire werke, that is to seyn, of this world?
And right so ben we gladed som-tyme of the face of the see
whan it is cleer; and also mervailen we on the hevene and on the
sterres, and on the sonne and on the mone.'

Philosophye. 'Aperteneth,' quod she, 'any of thilke thinges to 45

thee? Why darst thou glorifyen thee in the shyninge of any
swiche thinges? Art thou distingwed and embelised by the
springinge floures of the first somer sesoun, or swelleth thy
plentee in the fruites of somer? Why art thou ravisshed with
ydel Ioyes? Why embracest thou straunge goodes as they weren 50
thyne? Fortune ne shal never maken that swiche thinges ben
thyne, that nature of thinges hath maked foreine fro thee. Sooth
is that, with-outen doute, the frutes of the erthe owen to ben to
the norissinge of bestes. And yif thou wolt fulfille thy nede after
that it suffyseth to nature, than is it no nede that thou seke after 55
the superfluitee of fortune. For with ful fewe things and with ful
litel thinges nature halt hir apayed; and yif thou wolt achoken
the fulfillinge of nature with superfluitees, certes, thilke thinges
that thou wolt thresten or pouren in-to nature shullen ben unioyful
to thee, or elles anoyous. Wenest thou eek that it be a fair 60
thing to shyne with dyverse clothinge? Of whiche clothinge yif
the beautee be agreeable to loken up-on, I wol mervailen on the
nature of the matere of thilke clothes, or elles on the werkman
that wroughte hem. But also a long route of meynee, maketh
that a blisful man? The whiche servants, yif they ben vicious of 65
condiciouns, it is a great charge and a distruccioun to the hous,
and a greet enemy to the lord him-self. And yif they ben goode
men, how shal straunge or foreine goodnesse ben put in the
noumbre of thy richesse? So that, by all these forseide thinges,
it is clearly y-shewed, that never oon of thilke thinges that thou 70
acountedest for thyne goodes nas nat thy good. In the whiche
thinges, yif ther be no beautee to ben desyred, why sholdest thou
ben sory yif thou lese hem, or why sholdest thou reioysen thee
to holden hem? For yif they ben faire of hir owne kinde, what
aperteneth that to thee? For al so wel sholden they han ben 75
faire by hem-selve, though they weren departed fram alle thyne
richesses. Forwhy faire ne precious ne weren they nat, for that
they comen among thy richesses; but, for they semeden faire and
precious, ther-for thou haddest lever rekne hem amonges thy
richesses. 80

But what desirest thou of Fortune with so grete a noise, and

with so grete a fare? I trowe thou seke to dryve awey nede with
habundaunce of thinges; but certes, it torneth to you al in the
contrarie. Forwhy certes, it nedeth of ful manye helpinges to
kepen the diversitee of precious ostelments. And sooth it is, 85
that of manye thinges han they nede that manye thinges han; and
ayeinward, of litel nedeth hem that mesuren hir fille after the nede
of kinde, and nat after the outrage of coveityse. Is it thanne so,
that ye men ne han no proper good y-set in you, for which
ye moten seken outward youre goodes in foreine and subgit 90
thinges? So is thanne the condicioun of thinges torned up-so-down,
that a man, that is a devyne beest by merite of his resoun,
thinketh that him-self nis neither faire ne noble, but-yif it be
thorugh possessioun of ostelments that ne han no sowles. And
certes, al other thinges ben apayed of hir owne beautee; but ye 95
men, that ben semblable to god by your resonable thought,
desiren to aparailen your excellent kinde of the lowest thinges;
ne ye understonden nat how greet a wrong ye don to your
creatour. For he wolde that mankinde were most worthy and
noble of any othre erthely thinges; and ye threste adoun your 100
dignitees benethe the lowest thinges. For yif that al the good of
every thinge be more precious than is thilke thing whos that
the good is: sin ye demen that the fouleste thinges ben youre
goodes, thanne submitten ye and putten your-selven under tho
fouleste thinges by your estimacioun; and certes, this tydeth nat 105
with-oute youre desertes. For certes, swiche is the condicioun of
alle mankinde, that only whan it hath knowinge of it-selve, than
passeth it in noblesse alle other thinges; and whan it forleteth the
knowinge of it-self, than is it brought binethen alle beestes. For-why
al other livinge beestes han of kinde to knowe nat hem-self; 110
but whan that men leten the knowinge of hemself, it cometh hem
of vice. But how brode sheweth the errour and the folye of yow
men, that wenen that any thing may ben aparailed with straunge
aparailements! But for sothe that may nat ben doon. For yif
a wight shyneth with thinges that ben put to him, as thus, if 115
thilke thinges shynen with which a man is aparailed, certes, thilke
thinges ben comended and preysed with which he is aparailed;
but natheles, the thing that is covered and wrapped under that
dwelleth in his filthe.

And I denye that thilke thing be good that anoyeth him that 120
hath it. Gabbe I of this?. Thou wolt seye "nay." Certes,
richesses han anoyed ful ofte hem that han tho richesses; sin that
every wikked shrewe, (and for his wikkednesse the more gredy
after other folkes richesses, wher-so ever it be in any place, be it
gold or precious stones), weneth him only most worthy that hath 125
hem. Thou thanne, that so bisy dredest now the swerd and now
the spere, yif thou haddest entred in the path of this lyf a voide
wayferinge man, than woldest thou singe beforn the theef; as
who seith, a pore man, that berth no richesse on him by the weye,
may boldely singe biforn theves, for he hath nat wherof to ben 130
robbed. O precious and right cleer is the blisfulnesse of mortal
richesses, that, whan thou hast geten it, than hast thou lorn thy
sikernesse!

Pr. V. 1. C. A. noryssinges; Ed. norisshynges. // C. dess-; A. desc-. 6. A.

Richesse. 8. A. worthi. // A. rycchesse. // C. om. it. 15. C. stenteth; A.
stynteth. 19. A. al hool; Ed. al hole; C. om.; (Lat. tota). 21. A. rycchesse. 24.
A. thise rycchesses. 25. A. om. 1st ne. 27. A. in-to. 28. C. beautes; A. Ed.
beaute. // C. But; A. For. 29. A. om. the. 31. C. gretely; A. gretly. 32. C.
Ioyngture; A. ioynture. 33. C. myht; A. myȝt. 35. C. last; A. laste. 36. C. om.
and. 38. C. A. desserued. // A. shullen. 41. C. ryhte; A ryȝt. 46. C. darsthow;
A. darst thou. 47. C. Arthow; A. Art thou. 49. A. om. the. // C. fructes; A.
fruytes. // C. arthow. // C. rauyssed; A. rauyshed. 52. A. om. hath. // A. Syche
(!). 53. A. on (for 2nd to). 59. C. shollen; A. shullen. 60. C. anoyos; A. anoies;
Ed. anoyous. 64. C. wrowht; A. wrouȝt. 70. oon] A. none. 71. A.
accoumptedest. 75. A. as (for al-so). 77, 78, 80. A. rycchesse. 90. A.
outwardes. 98. A. ne ye ne, &c. 100. A. Ed. erthely; C. wordly. 103. C. tho; A.
the. // C. A. foulest. 104. A. summytten. // C. the; A. tho. 106. A. desert. 110.
A. om. livinge. // C. hym-; A. hem-. 111. C. om. that. 119. So A.; C. felthe.
122. A. rycchesse (thrice). // C. tho; A. the. 125. C. A. Ed. and weneth; but
and must be omitted (see Latin text). // C. hat. 126. A. om. 2nd now. 128. A.
wayfaryng. 132. A. rycchesse.

Metre V.

Felix nimium prior etas.

Blisful was the first age of men! They helden hem apayed
with the metes that the trewe feldes broughten forth. They
ne distroyede nor deceivede nat hem-self with outrage. They
weren wont lightly to slaken hir hunger at even with acornes
of okes. They ne coude nat medly the yifte of Bachus to the 5
cleer hony; that is to seyn, they coude make no piment nor clarree;
ne they coude nat medle the brighte fleeses of the contree of
Seriens with the venim of Tyrie; this is to seyn, they coude nat
deyen whyte fleeses of Serien contree with the blode of a maner
shelfisshe that men finden in Tyrie, with whiche blood men deyen 10
purpur. They slepen hoolsom slepes up-on the gras, and
dronken of the renninge wateres; and layen under the shadwes
of the heye pyn-trees. Ne no gest ne straungere ne carf yit
the heye see with ores or with shippes; ne they ne hadde seyn
yit none newe strondes, to leden marchaundyse in-to dyverse 15
contrees. Tho weren the cruel clariouns ful hust and ful stille,
ne blood y-shad by egre hate ne hadde nat deyed yit armures.
For wher-to or which woodnesse of enemys wolde first moeven
armes, whan they seyen cruel woundes, ne none medes be of
blood y-shad? 20

I wolde that oure tymes sholde torne ayein to the olde

maneres! But the anguissous love of havinge brenneth in folk
more cruely than the fyr of the mountaigne Ethna, that ay brenneth.
Allas! what was he that first dalf up the gobetes or the weightes
of gold covered under erthe, and the precious stones that wolden 25
han ben hid? He dalf up precious perils. That is to seyn, that
he that hem first up dalf, he dalf up a precious peril; for-why for
the preciousnesse of swiche thinge, hath many man ben in peril.

Me. V. 2. Ed. feldes; C. feeldes; A. erthes. 3. C. desseyuyd; A. desceyued. 4.

C. accornes; A. acornes. 6. C. nor; Ed. or; A. of. 7. C. fleezes; A. flies; Ed.
fleces. 8. A. siriens (Lat. Serum). 9. C. flezes; A. flies; Ed. fleces. // C. syryen;
A. sirien; Ed. Syrien. 10. C. shylle-; A. Ed. shel-. 13. A. om. 3rd ne. // C.
karue; A. karf; Ed. carfe. 16. C. crwel (and so again below). // C. Ed. hust; A.
whist. 17. A. y-shed. // A. armurers (!). 18. C. wer to. 19. C. say; A. seien. 22.
C. angwissos; A. anguissous. 23. C. om. 2nd the. // A. Ed. of Ethna; C. om.
of. // A. euer (for ay). 27. C. om. 2nd he. 28. A. om. thinge. // A. ben; C. be.

Prose VI.

Quid autem de dignitatibus.

But what shal I seye of dignitees and of powers, the whiche

ye men, that neither knowen verray dignitee ne verray power,
areysen hem as heye as the hevene? The whiche dignitees and
powers, yif they comen to any wikked man, they don as grete
damages and destrucciouns as doth the flaumbe of the mountaigne 5
Ethna, whan the flaumbe walweth up; ne no deluge ne doth so
cruel harmes. Certes, thee remembreth wel, as I trowe, that
thilke dignitee that men clepen the imperie of consulers, the
whiche that whylom was biginninge of fredom, youre eldres
coveiteden to han don away that dignitee, for the pryde of the 10
consulers. And right for the same pryde your eldres, biforn that
tyme, hadden don awey, out of the citee of Rome, the kinges
name; that is to seyn, they nolde han no lenger no king. But
now, yif so be that dignitees and powers be yeven to goode men,
the whiche thing is ful selde, what agreable thing is ther in tho 15
dignitees or powers but only the goodnesse of folkes that usen
hem? And therfor it is thus, that honour ne comth nat to vertu
for cause of dignitee, but ayeinward honour comth to dignitee for
cause of vertu. But whiche is thilke youre dereworthe power,
that is so cleer and so requerable? O ye ertheliche bestes, 20
considere ye nat over which thinge that it semeth that ye han
power? Now yif thou saye a mous amonges other mys, that
chalaunged to him-self-ward right and power over alle other mys,
how greet scorn woldest thou han of it! Glosa. So fareth it by
men; the body hath power over the body. For yif thou loke wel 25
up-on the body of a wight, what thing shall thou finde more
freele than is mankinde; the whiche men wel ofte ben slayn with
bytinge of smale flyes, or elles with the entringe of crepinge
wormes in-to the privetees of mannes body? But wher shal man
finden any man that may exercen or haunten any right up-on 30
another man, but only up-on his body, or elles up-on thinges
that ben lowere than the body, the whiche I clepe fortunous
possessiouns? Mayst thou ever have any comaundement over
a free corage? Mayst thou remuen fro the estat of his propre
reste a thought that is clyvinge to-gidere in him-self by stedefast 35
resoun? As whylom a tyraunt wende to confounde a free man
of corage, and wende to constreyne him by torment, to maken
him discoveren and acusen folk that wisten of a coniuracioun,
which I clepe a confederacie, that was cast ayeins this tyraunt;
but this free man boot of his owne tonge and caste it in the 40
visage of thilke wode tyraunt; so that the torments that this
tyraunt wende to han maked matere of crueltee, this wyse man
maked it matere of vertu.

But what thing is it that a man may don to another man, that
he ne may receyven the same thing of othre folk in him-self: 45
or thus, what may a man don to folk, that folk ne may don him the
same? I have herd told of Busirides, that was wont to sleen his
gestes that herberweden in his hous; and he was sleyn him-self
of Ercules that was his gest. Regulus hadde taken in bataile
many men of Affrike and cast hem in-to feteres; but sone after 50
he moste yeve his handes to ben bounde with the cheynes of
hem that he hadde whylom overcomen. Wenest thou thanne
that he be mighty, that hath no power to don a thing, that othre
ne may don in him that he doth in othre? And yit more-over,
yif it so were that thise dignitees or poweres hadden any propre 55
or natural goodnesse in hem-self, never nolden they comen to
shrewes. For contrarious thinges ne ben nat wont to ben
y-felawshiped to-gidere. Nature refuseth that contrarious thinges
ben y-ioigned. And so, as I am in certein that right wikked folk
han dignitees ofte tyme, than sheweth it wel that dignitees and 60
powers ne ben nat goode of hir owne kinde; sin that they suffren
hem-self to cleven or ioinen hem to shrewes. And certes, the
same thing may I most digneliche iugen and seyn of alle the
yiftes of fortune that most plentevously comen to shrewes; of
the whiche yiftes, I trowe that it oughte ben considered, that no 65
man douteth that he nis strong in whom he seeth strengthe; and
in whom that swiftnesse is, sooth it is that he is swift. Also
musike maketh musiciens, and phisike maketh phisiciens, and
rethorike rethoriens. For-why the nature of every thing maketh
his propretee, ne it is nat entremedled with the effects of the 70
contrarious thinges; and, as of wil, it chaseth out thinges that
ben to it contrarie. But certes, richesse may not restreyne
avarice unstaunched; ne power ne maketh nat a man mighty
over him-self, whiche that vicious lustes holden destreyned with
cheynes that ne mowen nat be unbounden. And dignitees that 75
ben yeven to shrewede folk nat only ne maketh hem nat digne,
but it sheweth rather al openly that they ben unworthy and
undigne. And why is it thus? Certes, for ye han Ioye to clepen
thinges with false names that beren hem alle in the contrarie;
the whiche names ben ful ofte reproeved by the effecte of the 80
same thinges; so that thise ilke richesses ne oughten nat by
right to ben cleped richesses; ne swich power ne oughte nat
ben cleped power; ne swich dignitee ne oughte nat ben cleped
dignitee.

And at the laste, I may conclude the same thing of alle the 85
yiftes of Fortune, in which ther nis nothing to ben desired, ne
that hath in him-self naturel bountee, as it is ful wel y-sene. For
neither they ne ioignen hem nat alwey to goode men, ne maken
hem alwey goode to whom that they ben y-ioigned.

Pr. VI. 1. A. seyne. 2. A. om. ye. 5. C. flawmbe; A. flamme (twice). 6. A. ins.

wit (!) bef. walweth. 7. C. crwel. // C. remenbryth. 8. A. thilke; C. thikke. // A.
emperie; C. Imperiye. 11. A. conseilers. 13. A. kyng; C. kynge. 15. Ed. selde;
C. A. zelde. // C. A. Ed. thinges; read thing (Lat. quid placet). 19. A. om.
thilke. 22. C. musȝ; A. myse; Ed. myce. 23. C. mysȝ; A. myse; Ed. myce. 26.
C. shalthow. 27. A. mannes kynde. // A. whiche ben ful ofte slayn. 29. A.
mennes bodyes. 33. C. Maysthow. 34. C. Maysthow remwen. 35. A. cleuyng.
// C. stidefast; A. stedfast. 40. Ed. caste; C. A. cast. 42. C. crwelte. 45. C.
resseyuen; A. receyue. 48. A. herburghden. 52. C. om. he. // C. whylom; A.
somtyme. // C. weenesthow. 53. C. thinge; A. thing. 54. A. om. 1st in. // A. to
(for 2nd in). 63. Ed. I (after may); C. A. omit. 67. C. om. it. 68. So A.; C.
musuciens, phisissiens. 70. A. effectis; C. effect. // A. om. the. 72. C. A. to it
ben. 73. A. om. 2nd ne. 81, 82. A. rycchesse (twice). 82, 83. A. whiche (for
swich; twice). 87. C. I-seene; A. sene.

Metre VI.

Nouimus quantas dederit ruinas.

We han wel knowen how many grete harmes and destrucciouns

weren don by the emperor Nero. He leet brenne the citee of
Rome, and made sleen the senatoures. And he, cruel, whylom
slew his brother; and he was maked moist with the blood of
his moder; that is to seyn, he leet sleen and slitten the body of 5
his moder, to seen wher he was conceived; and he loked on every
halve up-on her colde dede body, ne no tere ne wette his face, but
he was so hard-herted that he mighte ben domes-man or Iuge of
hir dede beautee. And natheles, yit governede this Nero by
ceptre alle the poeples that Phebus the sonne may seen, cominge 10
from his outereste arysinge til he hyde his bemes under the
wawes; that is to seyn, he governed alle the poeples by ceptre imperial
that the sonne goth aboute, from est to west. And eek this
Nero governed by ceptre alle the poeples that ben under the
colde sterres that highten "septem triones"; this is to seyn, he 15
governede alle the poeples that ben under the party of the north.
And eek Nero governed alle the poeples that the violent wind
Nothus scorkleth, and baketh the brenning sandes by his drye
hete; that is to seyn, alle the poeples in the south. But yit ne
mighte nat al his hye power torne the woodnesse of this wikked 20
Nero. Allas! it is a grevous fortune, as ofte as wikked swerd
is ioigned to cruel venim; that is to seyn, venimous crueltee to
lordshippe.'

Me. VI. 2. C. let; A. letee (!). 3. C. crwel. // C. whylom; A. somtyme. 5. C.

lette (wrongly); A. let. 6. C. conseyued; A. conceiued. 7. A. half. // C. wecte;
A. wette. 9. A. ȝitte neuertheles. 11. A. hidde. 12. C. sceptre; A. ceptre. 15. C.
vii. tyryones (sic); A. the seuene triones; Ed. the Septentrions. 16. A. parties.
18. C. Ed. scorklith; A. scorchith. 19-21. A. om. But yit ... Nero; Ed. retains it,
omitting hye. // For Allas ... it is, A. has—But ne how greuous fortune is; C.
om. a bef. greuous, but Ed. retains it. C. repeats it is. 22. C. crwel; crwelte.

Prose VII.

Tum ego, scis, inquam.

Thanne seyde I thus: 'Thou wost wel thy-self that the coveitise
of mortal thinges ne hadde never lordshipe of me; but
I have wel desired matere of thinges to done, as who seith, I
desire to han matere of governaunce over comunalitees, for vertu,
stille, ne sholde nat elden;' that is to seyn, that [him] leste that, 5
or he wex olde, his vertu, that lay now ful stille, ne should nat
perisshe unexercised in governaunce of comune; for which men
mighten speken or wryten of his goode governement.

Philosophye. 'For sothe,' quod she, 'and that is a thing that

may drawen to governaunce swiche hertes as ben worthy and 10
noble of hir nature; but natheles, it may nat drawen or tollen
swiche hertes as ben y-brought to the fulle perfeccioun of vertu,
that is to seyn, coveitise of glorie and renoun to han wel administred
the comune thinges or don gode desertes to profit of the
comune. For see now and considere, how litel and how voide of 15
alle prys is thilke glorie. Certein thing is, as thou hast lerned by
the demonstracioun of astronomye, that al the environinge of the
erthe aboute ne halt nat but the resoun of a prikke at regard of the
greetnesse of hevene; that is to seyn, that yif ther were maked
comparisoun of the erthe to the greetnesse of hevene, men wolden 20
iugen in al, that the erthe ne helde no space. Of the whiche litel
regioun of this worlde, the ferthe partye is enhabited with livinge
bestes that we knowen, as thou thyself hast y-lerned by Tholomee
that proveth it. And yif thou haddest with-drawen and abated in
thy thought fro thilke ferthe partye as moche space as the see and 25
the mareys contenen and over-goon, and as moche space as the
regioun of droughte over-streccheth, that is to seyn, sandes and
desertes, wel unnethe sholde ther dwellen a right streit place to
the habitacioun of men. And ye thanne, that ben environed and
closed with-in the leste prikke of thilke prikke, thinken ye to 30
manifesten your renoun and don youre name to ben born forth?
But your glorie, that is so narwe and so streite y-throngen in-to so
litel boundes, how mochel coveiteth it in largesse and in greet
doinge? And also sette this there-to: that many a nacioun,
dyverse of tonge and of maneres and eek of resoun of hir livinge, 35
ben enhabited in the clos of thilke litel habitacle; to the whiche
naciouns, what for difficultee of weyes and what for dyversitee of
langages, and what for defaute of unusage and entrecomuninge of
marchaundise, nat only the names of singuler men ne may nat
strecchen, but eek the fame of citees ne may nat strecchen. At 40
the laste, certes, in the tyme of Marcus Tullius, as him-self writ in
his book, that the renoun of the comune of Rome ne hadde nat
yit passed ne cloumben over the mountaigne that highte Caucasus;
and yit was, thilke tyme, Rome wel waxen and greetly redouted of
the Parthes and eek of other folk enhabitinge aboute. Seestow 45
nat thanne how streit and how compressed is thilke glorie that ye
travailen aboute to shewe and to multiplye? May thanne the
glorie of a singuler Romaine strecchen thider as the fame of the
name of Rome may nat climben ne passen? And eek, seestow nat
that the maneres of dyverse folk and eek hir lawes ben discordaunt 50
among hem-self; so that thilke thing that som men
iugen worthy of preysinge, other folk iugen that it is worthy of
torment? And ther-of comth it that, though a man delyte him in
preysinge of his renoun, he may nat in no wyse bringen forth ne
spreden his name to many maner poeples. There-for every man 55
oughte to ben apayed of his glorie that is publisshed among his
owne neighbours; and thilke noble renoun shal ben restreyned
within the boundes of o manere folke. But how many a man,
that was ful noble in his tyme, hath the wrecched and nedy
foryetinge of wryteres put out of minde and don awey! Al be 60
it so that, certes, thilke wrytinges profiten litel; the whiche
wrytinges long and derk elde doth awey, bothe hem and eek hir
autours. But ye men semen to geten yow a perdurabletee, whan
ye thenken that, in tyme to-cominge, your fame shal lasten. But
natheles, yif thou wolt maken comparisoun to the endeles spaces 65
of eternitee, what thing hast thou by whiche thou mayst reioysen
thee of long lastinge of thy name? For yif ther were maked comparisoun
of the abydinge of a moment to ten thousand winter,
for as mochel as bothe the spaces ben ended, yit hath the
moment som porcioun of it, al-though it litel be. But natheles, 70
thilke selve noumbre of yeres, and eek as many yeres as
ther-to may be multiplyed, ne may nat, certes, ben comparisoned
to the perdurabletee that is endeles; for of thinges that han ende
may be maked comparisoun, but of thinges that ben with-outen
ende, to thinges that han ende, may be maked no comparisoun. 75
And forthy is it that, al-though renoun, of as long tyme as ever
thee list to thinken, were thought to the regard of eternitee, that
is unstaunchable and infinit, it ne sholde nat only semen litel, but
pleynliche right naught. But ye men, certes, ne conne don
nothing a-right, but-yif it be for the audience of poeple and for 80
ydel rumours; and ye forsaken the grete worthinesse of conscience
and of vertu, and ye seken your guerdouns of the smale wordes of
straunge folk.

Have now heer and understonde, in the lightnesse of swich

pryde and veine glorie, how a man scornede festivaly and merily 85
swich vanitee. Whylom ther was a man that hadde assayed
with stryvinge wordes another man, the whiche, nat for usage of
verray vertu but for proud veine glorie, had taken up-on him
falsly the name of a philosophre. This rather man. that I spak
of thoughte he wolde assaye, wher he, thilke, were a philosophre 90
or no; that is to seyn, yif that he wolde han suffred lightly in
pacience the wronges that weren don un-to him. This feynede
philosophre took pacience a litel whyle, and, whan he hadde
received wordes of outrage, he, as in stryvinge ayein and reioysinge
of him-self, seyde at the laste right thus: "understondest 95
thou nat that I am a philosophre?" That other man answerde
ayein ful bytingly, and seyde: "I hadde wel understonden it, yif
thou haddest holden thy tonge stille." But what is it to thise
noble worthy men (for, certes, of swiche folke speke I) that seken
glorie with vertu? What is it?' quod she; 'what atteyneth fame 100
to swiche folk, whan the body is resolved by the deeth at the
laste? For yif it so be that men dyen in al, that is to seyn, body
and sowle, the whiche thing our resoun defendeth us to bileven,
thanne is ther no glorie in no wyse. For what sholde thilke glorie
ben, whan he, of whom thilke glorie is seyd to be, nis right naught 105
in no wyse? And yif the sowle, whiche that hath in it-self science
of goode werkes, unbounden fro the prison of the erthe, wendeth
frely to the hevene, despyseth it nat thanne alle erthely occupacioun;
and, being in hevene, reioyseth that it is exempt fro alle
erthely thinges? As who seith, thanne rekketh the sowle of no 110
glorie of renoun of this world.

Pr. VII. 4. A. desired. 5. I supply him (to make sense). // Ed. leste; C. A. list.
6. A. wex; C. wax. 7. C. perise; A. perisshe. // Ed. vnexercysed; C. A.
vnexcercised. 17. A. om. 1st the. // C. om. of. 21. A. that erthe helde. 26. A.
and mareys. // C. spaces (for space). 28. C. vel; A. wel. 32. C. narwh; A.
narwe. 36. A. cloos. 37. C. deficulte; A. difficulte. // C. deficulte (repeated); A.
Ed. diuersite. 38. A. om. and after vnusage. 39. Ed. synguler; C. A. syngler. //
A. om. nat (bef. 1st strecchen). 41. C. marchus; A. Marcus. // Ed. Tullius; C.
A. Tulius. // C. writ; A. writeth. 43. C. om. yit. // A. hyȝt. 44. C. thikke; A.
thilk. // A. wexen. 45. C. sestow; A. Sest thou. 48. Ed. synguler; C. singler; A.
singlere. // A. strecchen; C. strechchen. 49. C. seysthow; A. sest thou; Ed.
seest thou. 51. C. thinge; A. thing. 56. A. paied. // Ed. publysshed; C.
publyssed; A. puplissed. 57. A. neyȝbores; Ed. neyghbours; C. nesshebours.
59. A. nedy and wrecched. 63. A. autours; Ed. auctours; C. actorros (!). // A.
Ed. ye men semen; C. yow men semeth. 64. A. thenke; C. thinken. // A.
comyng (om. to-). 65. A. space (Lat. spatia). 69. C. A. Ed. insert for bef. yit
(wrongly). 70. A. it a litel. 73. C. -durablyte; A. -durablete. // A. eenles (for
endeles). 74, 75. A. om. but of ... comparisoun. 77. A. by (for 2nd to). 82. C.
A. gerdouns; Ed. guerdones. 84. A. whiche (for swich). 89. A. speke. 90. C.
weere he; A. where he; Ed. wheder he. 91. A. om. that. 94. C. resseyuyd; A.
receiued. 95. C. vnderstondow. 97. A. om. it. 98. C. glosses it by s. fama. 102.
A. om. it. 103. C. deffendeth; A. defendith. 105. A. for (for whan). 107. C.
glosses erthe by i. corporis. 108. C. glosses it by i. anima. 110, 111. A. om. As
who ... this world.
 Metre VII.

Quicunque solam mente praecipiti petit.

Who-so that, with overthrowinge thought, only seketh glorie of

fame, and weneth that it be sovereyn good: lat him loken up-on
the brode shewinge contrees of hevene, and up-on the streite site
of this erthe; and he shal ben ashamed of the encrees of his
name, that may nat fulfille the litel compas of the erthe. O! 5
what coveiten proude folk to liften up hir nekkes in ydel in the
dedly yok of this worlde? For al-though that renoun y-sprad,
passinge to ferne poeples, goth by dyverse tonges; and al-though
that grete houses or kinredes shynen with clere titles of honours;
yit, natheles, deeth despyseth alle heye glorie of fame: and deeth 10
wrappeth to-gidere the heye hevedes and the lowe, and maketh
egal and evene the heyeste to the loweste. Wher wonen now the
bones of trewe Fabricius? What is now Brutus, or stierne
Catoun? The thinne fame, yit lastinge, of hir ydel names, is
marked with a fewe lettres; but al-though that we han knowen 15
the faire wordes of the fames of hem, it is nat yeven to knowe
hem that ben dede and consumpte. Liggeth thanne stille, al
outrely unknowable; ne fame ne maketh yow nat knowe. And
yif ye wene to liven the longer for winde of your mortal name,
whan o cruel day shal ravisshe yow, thanne is the seconde deeth 20
dwellinge un-to yow.' Glose. The first deeth he clepeth heer the
departinge of the body and the sowle; and the seconde deeth he
clepeth, as heer, the stintinge of the renoun of fame.

3. C. cyte (for site); A. sete (error for site; Lat. situm). 6. A. liften vpon hire
nekkes in ydel and dedely. 7. A. om. that. 9. A. om. that. // C. cler; A. clere.
13. A. stiern; Ed. sterne. 17. A. Ed. consumpt. 18. A. vtterly. 21. Ed. to (for
un-to); A. in. // A. Ed. the; C. om. (after heer).

Prose VIII.

Set ne me inexorabile contra fortunam.

'But for as mochel as thou shalt nat wenen', quod she, 'that I
bere untretable bataile ayeins fortune, yit som-tyme it bifalleth that
she, deceyvable, deserveth to han right good thank of men; and
that is, whan she hir-self opneth, and whan she descovereth hir
frount, and sheweth hir maneres. Peraventure yit understondest 5
thou nat that I shal seye. It is a wonder that I desire to telle,
and forthy unnethe may I unpleyten my sentence with wordes; for
I deme that contrarious Fortune profiteth more to men than
Fortune debonaire. For alwey, whan Fortune semeth debonaire,
than she lyeth falsly in bihetinge the hope of welefulnesse; but 10
forsothe contrarious Fortune is alwey soothfast, whan she sheweth
hir-self unstable thorugh hir chaunginge. The amiable Fortune
deceyveth folk; the contrarie Fortune techeth. The amiable
Fortune bindeth with the beautee of false goodes the hertes of
folk that usen hem; the contrarie Fortune unbindeth hem by the 15
knowinge of freele welefulnesse. The amiable Fortune mayst
thou seen alwey windinge and flowinge, and ever misknowinge of
hir-self; the contrarie Fortune is atempre and restreyned, and wys
thorugh exercise of hir adversitee. At the laste, amiable Fortune
with hir flateringes draweth miswandringe men fro the sovereyne 20
good; the contrarious Fortune ledeth ofte folk ayein to soothfast
goodes, and haleth hem ayein as with an hooke. Wenest thou
thanne that thou oughtest to leten this a litel thing, that this aspre
and horrible Fortune hath discovered to thee the thoughtes of thy
trewe freendes? For-why this ilke Fortune hath departed and uncovered 25
to thee bothe the certein visages and eek the doutous
visages of thy felawes. Whan she departed awey fro thee, she
took awey hir freendes, and lafte thee thyne freendes. Now whan
thou were riche and weleful, as thee semede, with how mochel
woldest thou han bought the fulle knowinge of this, that is to seyn, 30
the knowinge of thy verray freendes? Now pleyne thee nat thanne
of richesse y-lorn, sin thou hast founden the moste precious kinde
of richesses, that is to seyn, thy verray freendes.

Pr. VIII. A. omits to end of bk. iii. pr. 1. 3. C. desseyuable. // C. desserueth. 7.

So C.; Ed. vnplyten. 13. C. desseyueth. 17. C. maysthow. 30. C. woldesthow.

Metre VIII.
 Quod mundus stabili fide.

That the world with stable feith varieth acordable chaunginges;

that the contrarious qualitee of elements holden among hem-self
aliaunce perdurable; that Phebus the sonne with his goldene
chariet bringeth forth the rosene day; that the mone hath commaundement
over the nightes, which nightes Hesperus the eve-sterre 5
hath brought; that the see, greedy to flowen, constreyneth
with a certein ende hise flodes, so that it is nat leveful to strecche
hise brode termes or boundes up-on the erthes, that is to seyn, to
covere al the erthe:—al this acordaunce of thinges is bounden with
Love, that governeth erthe and see, and hath also commaundements 10
to the hevenes. And yif this Love slakede the brydeles,
alle thinges that now loven hem to-gederes wolden maken a bataile
continuely, and stryven to fordoon the fasoun of this worlde, the
whiche they now leden in acordable feith by faire moevinges.
This Love halt to-gideres poeples ioigned with an holy bond, and 15
knitteth sacrement of mariages of chaste loves; and Love endyteth
lawes to trewe felawes. O! weleful were mankinde, yif thilke
Love that governeth hevene governed youre corages!'

Me. VIII. 6. C. hat. 7. C. lueful; Ed. leful. 8. erthes; Lat. terris.

Explicit Liber secundus.

 BOOK III.
Prose I.

Iam cantum illa finierat.

By this she hadde ended hir song, whan the sweetnesse of hir
ditee hadde thorugh-perced me that was desirous of herkninge,
and I astoned hadde yit streighte myn eres, that is to seyn, to
herkne the bet what she wolde seye; so that a litel here-after I
seyde thus: 'O thou that art sovereyn comfort of anguissous 5
corages, so thou hast remounted and norisshed me with the
weighte of thy sentences and with delyt of thy singinge; so that
I trowe nat now that I be unparigal to the strokes of Fortune:
as who seyth, I dar wel now suffren al the assautes of Fortune, and
wel defende me fro hir. And tho remedies whiche that thou 10
seydest her-biforn weren right sharpe, nat only that I am nat
a-grisen of hem now, but I, desirous of heringe, axe gretely to
heren the remedies.'

Than seyde she thus: 'That felede I ful wel,' quod she, 'whan
that thou, ententif and stille, ravisshedest my wordes; and I 15
abood til that thou haddest swich habite of thy thought as thou
hast now; or elles til that I my-self hadde maked to thee the
same habit, which that is a more verray thing. And certes, the
remenaunt of thinges that ben yit to seye ben swiche, that first
whan men tasten hem they ben bytinge, but whan they ben 20
receyved withinne a wight, than ben they swete. But for thou
seyst that thou art so desirous to herkne hem, with how gret
brenninge woldest thou glowen, yif thou wistest whider I wol
leden thee!'

'Whider is that?' quod I. 25

'To thilke verray welefulnesse,' quod she, 'of whiche thyn herte
dremeth; but for as moche as thy sighte is ocupied and distorbed
by imaginacioun of erthely thinges, thou mayst nat yit seen thilke
selve welefulnesse.'
'Do,' quod I, 'and shewe me what is thilke verray welefulnesse, 30
I preye thee, with-oute taryinge.'

'That wole I gladly don,' quod she, 'for the cause of thee;
but I wol first marken thee by wordes and I wol enforcen me to
enformen thee thilke false cause of blisfulnesse that thou more
knowest; so that, whan thou hast fully bi-holden thilke false 35
goodes, and torned thyn eyen to that other syde, thou mowe knowe
the cleernesse of verray blisfulnesse.

Pr. I. 3. C. streyhte; Ed. streyght. 5. C angwissos. 7. C. weyhte; Ed. weight. //

C. sentenses; Ed. sentences. 8. C. vnparygal; Ed. vnperegall. 10. C. deffende;
Ed. defende. 11. C. hir-; Ed. here-. 12. C. desiros; Ed. desyrous. 17. C. Ed.
had. 21. C. resseyued. 22. C. wit; Ed. with. 23. C. woldesthow; Ed. woldest
thou. 26. C. thynge (!); Ed. thyn; Lat. tuus. 28. C. herthely; Ed. erthly. 31. C.
tarynge; Ed. taryeng; Lat. cunctatione. 33. C. the (for thee); Ed. om.

Metre I.

Qui serere ingenuum uolet agrum.

Who-so wole sowe a feeld plentivous, lat him first delivere it fro
thornes, and kerve asunder with his hook the busshes and the
fern, so that the corn may comen hevy of eres and of greynes.
Hony is the more swete, yif mouthes han first tasted savoures that
ben wikkid. The sterres shynen more agreably whan the wind 5
Nothus leteth his ploungy blastes; and after that Lucifer the
day-sterre hath chased awey the derke night, the day the fairere
ledeth the rosene hors of the sonne. And right so thou, bi-holdinge
first the false goodes, bigin to with-drawen thy nekke
fro the yok of erthely affecciouns; and after-ward the verray goodes 10
shollen entren in-to thy corage.'

Me. I. 1. A. of (for fro). 2. A. bushes; Ed. busshes; C. bosses. 3. C. heres; A.

eres. 5. A. wikke. // C. agreablely. 7. C. dirke; A. derke. 8. A. om. And. 10. C.
verre; A. verrey.
 Prose II.

Tunc defixo paullulum uisu.

Tho fastnede she a litel the sighte of hir eyen, and with-drow
hir right as it were in-to the streite sete of hir thought; and bigan
to speke right thus: 'Alle the cures,' quod she, 'of mortal folk,
whiche that travaylen hem in many maner studies, goon certes by
diverse weyes, but natheles they enforcen hem alle to comen only 5
to oon ende of blisfulnesse. And blisfulnesse is swiche a good,
that who-so that hath geten it, he ne may, over that, no-thing
more desyre. And this thing is forsothe the sovereyn good that
conteyneth in him-self alle maner goodes; to the whiche good yif
ther failede any thing, it mighte nat ben cleped sovereyn good: 10
for thanne were ther som good, out of this ilke sovereyn good, that
mighte ben desired. Now is it cleer and certein thanne, that
blisfulnesse is a parfit estat by the congregacioun of alle goodes;
the whiche blisfulnesse, as I have seyd, alle mortal folk enforcen
hem to geten by diverse weyes. For-why the coveitise of verray 15
good is naturelly y-plaunted in the hertes of men; but the miswandringe
errour mis-ledeth hem in-to false goodes. Of the
whiche men, som of hem wenen that sovereyn good be to liven
with-oute nede of any thing, and travaylen hem to be haboundaunt
of richesses. And som other men demen that sovereyn good be, 20
for to ben right digne of reverence; and enforcen hem to ben
reverenced among hir neighbours by the honours that they han
y-geten. And some folk ther ben that holden, that right heigh
power be sovereyn good, and enforcen hem for to regnen, or elles
to ioignen hem to hem that regnen. And it semeth to some other 25
folk, that noblesse of renoun be the sovereyn good; and hasten
hem to geten glorious name by the arts of werre and of pees.
And many folk mesuren and gessen that sovereyn good be Ioye
and gladnesse, and wenen that it be right blisful thing to ploungen
hem in voluptuous delyt. And ther ben folk that entrechaungen 30
the causes and the endes of thise forseyde goodes, as they that
desiren richesses to han power and delytes; or elles they desiren
power for to han moneye, or for cause of renoun. In thise thinges,
and in swiche othre thinges, is torned alle the entencioun of
desiringes and of werkes of men; as thus: noblesse and favour 35
of people, whiche that yeveth to men, as it semeth hem, a maner
cleernesse of renoun; and wyf and children, that men desiren for
cause of delyt and of merinesse. But forsothe, frendes ne sholden
nat be rekned a-mong the godes of fortune, but of vertu; for it is
a ful holy maner thing. Alle thise othre thinges, forsothe, ben 40
taken for cause of power or elles for cause of delyt.

Certes, now am I redy to referren the goodes of the body to

thise forseide thinges aboven; for it semeth that strengthe and
gretnesse of body yeven power and worthinesse, and that beautee
and swiftnesse yeven noblesses and glorie of renoun; and hele of 45
body semeth yeven delyt. In alle thise thinges it semeth only
that blisfulnesse is desired. For-why thilke thing that every man
desireth most over alle thinges, he demeth that it be the sovereyn
good; but I have defyned that blisfulnesse is the sovereyn good;
for which every wight demeth, that thilke estat that he desireth 50
over alle thinges, that it be blisfulnesse.

Now hast thou thanne biforn thyn eyen almest al the purposed
forme of the welefulnesse of man-kinde, that is to seyn, richesses,
honours, power, and glorie, and delyts. The whiche delyt only
considerede Epicurus, and iuged and establisshed that delyt is 55
the sovereyn good; for as moche as alle othre thinges, as him
thoughte, bi-refte awey Ioye and mirthe fram the herte. But I
retorne ayein to the studies of men, of whiche men the corage
alwey reherseth and seketh the sovereyn good, al be it so that
it be with a derked memorie; but he not by whiche path, right 60
as a dronken man not nat by whiche path he may retorne him to
his hous. Semeth it thanne that folk folyen and erren that
enforcen hem to have nede of nothing? Certes, ther nis non other
thing that may so wel performe blisfulnesse, as an estat plentivous
of alle goodes, that ne hath nede of non other thing, but that is 65
suffisaunt of himself unto him-self. And folyen swiche folk thanne,
that wenen that thilke thing that is right good, that it be eek right
worthy of honour and of reverence? Certes, nay. For that thing
nis neither foul ne worthy to ben despised, that wel neigh al the
entencioun of mortal folk travaylen for to geten it. And power, 70
oughte nat that eek to ben rekened amonges goodes? What
elles? For it is nat to wene that thilke thing, that is most worthy
of alle thinges, be feble and with-oute strengthe. And cleernesse
of renoun, oughte that to ben despised? Certes, ther may no
man forsake, that al thing that is right excellent and noble, that it ne 75
semeth to ben right cleer and renomed. For certes, it nedeth nat
to seye, that blisfulnesse be [nat] anguissous ne drery, ne subgit to
grevaunces ne to sorwes, sin that in right litel thinges folk seken
to have and to usen that may delyten hem. Certes, thise ben
the thinges that men wolen and desiren to geten. And for this 80
cause desiren they richesses, dignitees, regnes, glorie, and delices.
For therby wenen they to han suffisaunce, honour, power, renoun,
and gladnesse. Than is it good, that men seken thus by so many
diverse studies. In whiche desyr it may lightly ben shewed how
gret is the strengthe of nature; for how so that men han diverse 85
sentences and discordinge, algates men acorden alle in lovinge the
ende of good.

Pr. II. 2. C. cyte; A. sete; Lat. sedem. 5. C. enforsen; A. enforced; Ed.

enforcen. 6. A. om. And blisfulnesse. 10. A. om. cleped. 14. C. enforsen; A.
enforcen. 18. A. is (for be). 20. C. ben; A. be. 22. C. nesshebors; A.
neyghbours. 23. A. halden. // C. heyh; A. heyȝe; Ed. hye. 24: A. to b (for be).
28. C. by (for be); A. Ed. be. 29. A. om. thing. 32. A. rycchesse. 35. A. om.
1st of. // C. fauor; A. fauour. 36. A. om. to men and hem. 38. A. shollen. 39.
A. Ed. the; C. tho. 45. C. sweft-; A. swifte-. 49. C. deffyned; A. Ed. diffined.
52. A. om. thy eyen; C. thy (for thyn); Ed. thyn. // A. almost. 55. A. om. and
bef. iuged. // C. A. establyssed; Ed. establysshed. 59. A. ins. of after good
(wrongly). 60. C. dirkyd; A. derke; Ed. dyrked. // A. om. but he ... path. // C.
paath (twice). 62. C. foleyen; A. folyen. 65. C. A. ins. it bef. is; Ed. om. 66. C.
A. foleyen; Ed. folyen. 69. C. wel neyh; Ed. wel nygh; A. om. // C. alle; A. Ed.
al. 77. I supply nat. // C. angwyssos. // C. subgyd; A. subgit. 81. A. rycches.
86. C. allegates; A. algates. // A. lyuynge (!).

Metre II.

Quantas rerum flectat habenas.

It lyketh me to shewe, by subtil song, with slakke and delitable
soun of strenges, how that Nature, mighty, enclineth and flitteth
the governements of thinges, and by whiche lawes she, purveyable,
kepeth the grete world; and how she, bindinge, restreyneth alle
thinges by a bonde that may nat ben unbounde. Al be it so that 5
the lyouns of the contre of Pene beren the faire chaynes, and
taken metes of the handes of folk that yeven it hem, and dreden
hir sturdy maystres of whiche they ben wont to suffren betinges:
yif that hir horrible mouthes ben be-bled, that is to seyn, of bestes
devoured, hir corage of time passed, that hath ben ydel and rested, 10
repeyreth ayein; and they roren grevously and remembren on hir
nature, and slaken hir nekkes fram hir chaynes unbounde; and
hir mayster, first to-torn with blody tooth, assayeth the wode
wrathes of hem; this is to seyn, they freten hir mayster. And the
iangelinge brid that singeth on the heye braunches, that is to seyn, 15
in the wode, and after is enclosed in a streyt cage: al-though that
the pleyinge bisinesse of men yeveth hem honiede drinkes and
large metes with swete studie, yit natheles, yif thilke brid, skippinge
out of hir streyte cage, seeth the agreables shadewes of the
wodes, she defouleth with hir feet hir metes y-shad, and seketh 20
mourninge only the wode; and twitereth, desiringe the wode, with
hir swete vois. The yerde of a tree, that is haled a-doun by
mighty strengthe, boweth redily the crop a-doun: but yif that the
hand of him that it bente lat it gon ayein, anon the crop loketh
up-right to hevene. The sonne Phebus, that falleth at even in 25
the westrene wawes, retorneth ayein eftsones his carte, by privee
path, ther-as it is wont aryse. Alle thinges seken ayein to hir
propre cours, and alle thinges reioysen hem of hir retorninge ayein
to hir nature. Ne non ordinaunce nis bitaken to thinges, but that
that hath ioyned the endinge to the beginninge, and hath maked 30
the cours of it-self stable, that it chaungeth nat from his propre
kinde.

Me. II. 3. A. om. the. 8. A. om. betinges. 9. C. horyble. 11. A. that (for 1st
and). 13. A. to-teren. 15. A. Iangland. // A. this (for 2nd that). 16. A. inclosed.
// C. streyht; A. streit. 17. C. pleynynge; A. pleiyng; Lat. ludens. 19. A. Ed.
agreable. 24. C. bent; A. bente. 27. A. in-to (for to). 30. C. hat; A. hath.

Prose III.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com