CS6600 - Intelligent Agents

Fall 2001
Project 1
Neural Networks
Due: 10/03/01


Acknowledgment

We thank Tom Mitchell for his materials.
 

1. Introduction

This assignment gives you an opportunity to apply neural network learning to the problem of face recognition. The face images you will use are faces of students from Tom Mitchell’s Machine Learning classes. You will not need to do significant amounts of coding for this assignment, and you should not let the size of this document scare you, but training your networks will take time. It is advisable to start early. It is also a good idea to read the assignment in its entirety first; this is a very important point since there is alot of technical information at the end of this document which will be very useful to you in completing the assignment.

It is recomended that you work on this project on a computer which uses the Solaris operating system.  You are welcome to attempt this project on another machine, even a non-Solaris machine if you wish.  However, if you use a different operating system, some of the instructions provided here may not work, and some of the files included in the project may require some changes.  You will need to access the following file for this assignment:

http://www.cc.gatech.edu/classes/AY2002/cs6660_fall/neural.tar.gz
There is a link to this file from the class web page. To use the file, enter the following commands:
gunzip neural.tar.gz
tar -xf neural.tar
This will create a new directory called neural which contains all of the code you will need to begin this project and also contains subdirectories which have all of the data you will need for this project.
 

1.1. The face images

The image data can be found in the faces_4 directory. This directory contains 20 subdirectories, one for each person, named by userid. Each of these directories contains several different face images of the same person.  If you look carefully you may see the face of your TA!

You will be interested in the images with the following naming convention:
<userid>_<pose>_<expression>_<eyes>_<scale>.pgm

If you’ve been looking closely in the image directories, you may notice that some images have a .bad suffix rather than the .pgm suffix. As it turns out, 16 of the 640 images taken have glitches due to problems with the camera setup; these are the .bad images. Some people had more glitches than others, but everyone who got "faced" should have at least 28 good face images (out of the 32 variations possible, discounting scale).
 

1.2. Viewing the face images

To view the images, you can use the program xv. This is usually available at /usr/bin/X11/xv.  xv handles a variety of image formats, including the PGM format in which our face images are stored. While we won’t go into detail about xv in this document, we will quickly describe the basics you need to know to use xv.

To start xv, just specify one or more images on the command line, like this:

xv faces_4/glickman/glickman_straight_happy_open_4.pgm
This will bring up an X window displaying the face.  Clicking the right button in the image window will toggle a control panel with a variety of but-tons.  The Dbl Size button doubles the displayed size of the image every time you click on it. This will be useful for viewing the quarter-resolution images, as you might imagine. You can also obtain pixel values by holding down the left button while moving the pointer in the image window. A text bar will be displayed, showing you the image coordinates and brightness value where the pointer is located.

To quit xv, just click on the Quit button or type q in one of the xv windows.
 

1.3. The neural network and image access code

We’re supplying C code for a three-layer fully-connected feedforward neural network which uses the backpropagation algorithm to tune its weights. To make life as easy as possible, we’re also supplying you with an image package for accessing the face images, as well as the top-level pro-gram for training and testing, as a skeleton for you to modify. To help explore what the nets actu-ally learn, you’ll also find a utility program for visualizing hidden-unit weights as images.

Download, gunzip (gunzip <filename.gz>, and untar (tar -xf <filename.tar>) the code in your own directory and then type make. When the compilation is done, you should have one executable pro-gram: facetrain. Briefly, facetrain takes lists of image files as input, and uses these as training and test sets for a neural network. facetrain can be used for training and/or recognition, and it also has the capability to save networks to files.

The code has been compiled and tested successfully on DecStations, Sun SPARC-2s and 5s as well as IBM PowerPC’s. If you wish to use the code on some other platform, feel free, but be aware that the code has only been tested on these platforms.
Details of the routines, explanations of the source files, and related information can be found in Section 3 of this handout.
 

2. The Assignment

You need to turn in a short writeup of your answers (including the source code you wrote/modified as well as the output when applicable) to the questions found in the following sequence of experiments. Please use a color pen to circle, highlight, or underline all code fragments that you wrote/modified. Do not turn in any file you did not modify.
 
  1. Copy all the files in your own directory.
  2. The code you have been given is currently set up to learn to recognize the person with userid glickman. Modify this code to implement a "sunglasses" recognizer; i.e., train a neural net which,when given an image as input, indicates whether the face in the image is wearing sunglasses, or not. Refer to the beginning of Section 3 for an overview of how to make changes to this code.
  3. Train a network using the default learning parameter settings (learning rate 0.3, momentum 0.3) for 75 epochs, with the following command:
    4. nice facetrain -n shades.net -t ./trainset/straightrnd_train.list -1 ./trainset/straightrnd_test1.list -2 ./trainset/straightrnd_test2.list -e 75
    Placing nice before the command name is not required to make it run but it is strongly recomended, especially if you are using a machine which is used by multiple users since it mini-mizes the extent to which this command will have a negative impact on the performance of essentual system tasks and on the processes of other users. In general it is a good idea to use nice on any command which does a lot of computation. You can type man nice to get more information about nice.  The arguments to facetrain are described in Section 3.1.1, but a short description is in order here.  shades.net is the name of the network file which will be saved when training is finished.  straightrnd_train.list, straightrnd_test1.list, and straightrnd_test2.list are text files which specify the training set (70 examples) and two test sets (34 and 52 examples), respectively.  This command creates and trains your net on a randomly chosen sample of 70 of the 156 "straight" images (namely those whose filenames are in the straightrnd_train.list file), and tests it on the remaining 34 and 52 randomly chosen images, respectively. One way to think of this test strategy is that roughly 1/3 of the images (straightrnd_test2.list) have been held over for testing.  The remaining 2/3 have been used for a train and crossvalidate strategy, in which 2/3 of these are being used for as a training set (straightrnd_train.list) and 1/3 are being used for the validation set to decide when to halt training (straightrnd_test1.list).
  4. What code did you modify? What was the maximum classification accuracy achieved on the training set? How many epochs did it take to reach this level? How about for the validation set? The test set? Note that if you run it again on the same system with the same parameters and input, you should get exactly the same results because, by default, the code uses the same seed to the random number generator each time. You will need to read Section 3.1.2 carefully in order to be able to interpret your experiments and answer these questions.
  5. Now, implement a 1-of-20 face recognizer; i.e. implement a neural net that accepts an image as input, and outputs the userid of the person. To do this, you will need to implement a different output encoding (since you must now be able to distinguish among 20 people). (Hint: leave learning rate and momentum at 0.3, and use 20 hidden units).
  6. As before, train the network, this time for 100 epochs:
      7. nice facetrain -n face.net -t ./trainset/straighteven_train.list -1 ./trainset/straighteven_test1.list -2 ./trainset/straighteven_test2.list -e 100
    You might be wondering why you are only training on samples from a limited distribution (the "straight" images). The essential reason is training time. If you have access to a very fast machine, then you are welcome to do these experiments on the entire set (replace straight with all in the command above). Otherwise, stick to the "straight" images.  The difference between the straightrnd_*.list and the straighteven_*.list sets is that while the former divides the images purely randomly among the training and test sets, the latter ensures a relatively even distribution of each individual’s images over the sets. Because we have only 7 or 8 "straight" images per individual, failure to distribute them evenly would result in testing our network the most on those faces on which it was trained the least.
  7. Which parts of the code was it necessary to modify this time? How did you encode the outputs? What was the maximum classification accuracy achieved on the training set? How many epochs did it take to reach this level? How about for the validation and test set?
  8. Now let’s take a closer look at which images the net may have failed to classify:
      9. nice facetrain -n face.net -T -1 ./trainset/straighteven_test1.list -2 ./trainset/straighteven_test2.list
    Do there seem to be any particular commonalities between the misclassified images?
  9. Implement a pose recognizer; i.e. implement a neural net which, when given an image as input, indicates whether the person in the image is looking straight ahead, up, to the left, or to the right.  You will also need to implement a different output encoding for this task. (Hint: leave learning rate and momentum at 0.3, and use 6 hidden units).
  10. Train the network for 100 epochs, this time on samples drawn from all of the images:
      11. nice facetrain -n pose.net -t ./trainset/all_train.list -1 ./trainset/all_test1.list -2 ./trainset/ all_test2.list -e 100
    Since the pose-recognizing network should have substantially fewer weights to update than the face-recognizing network, even those of you with slow machines can get in on the fun of using all of the images. In this case, 260 examples are in the training set, 140 examples are in test1, and 193 are in test2.
  11. How did you encode your outputs this time? What was the maximum classification accuracy achieved on the training set? How many epochs did it take to reach this level? How about for each test set?
  12. Now, try taking a look at how backpropagation tuned the weights of the hidden units with respect to each pixel. First type make hidtopgm to compile the utility on your system. Then, to visualize the weights of hidden unit n, type:
      13. nice hidtopgm pose.net image-filename 32 30 <n>
    This creates a pgm graphic file with image-filename as its name. Invoking xv on the image image-filename should then display the range of weights, with the lowest weights mapped to pixel values of zero, and the highest mapped to 255. If the images just look like noise, try retraining using facetrain_init0 (compile with make facetrain_init0), which initializes the hidden unit weights of a new network to zero, rather than random values.
  13. Do the hidden units seem to weight particular regions of the image greater than others? Do particular hidden units seem to be tuned to different features of some sort?.

3. Documentation

The code for this assignment is broken into several modules: Although you’ll only need to modify code in imagenet.c and facetrain.c, feel free to modify anything you want in any of the files if it makes your life easier or if it allows you to do a nifty exper-iment.

3.1. facetrain

3.1.1. Running facetrain

facetrain has several options which can be specified on the command line. This section briefly describes how each option works. A very short summary of this information can be obtained by running facetrain with no arguments.

3.1.2. Interpreting the output of facetrain

When you run facetrain, it will first read in all the data files and print a bunch of lines regarding these operations. Once all the data is loaded, it will begin training. At this point, the network’s training and test set performance is outlined in one line per epoch. For each epoch, the following performance measures are output:
<epoch> <delta> <trainperf> <trainerr> <t1perf> <t1err> <t2perf> <t2err>
These values have the following meanings:

3.2. Tips

Although you do not have to modify the image or network packages, you will need to know a little bit about the routines and data structures in them, so that you can easily implement new output encodings for your networks. The following sections describe each of the packages in a little more detail. You can look at imagenet.c and facetrain.c to see how the routines are actually used.

In fact, it is probably a good idea to look over facetrain.c first, to see how the training process works. You will notice that load _target() from imagenet.c is called to set up the target vector for training. You will also notice the routines which evaluate performance and compute error statis-tics, performance_on_imagelist() and evaluate_performance(). The first routine iterates through a set of images, computing the average error on these images, and the second routine computes the
error and accuracy on a single image.

You will almost certainly not need to use all of the information in the following sections, so don’t feel like you need to know everything the packages do. You should view these sections as refer-ence guides for the packages, should you need information on data structures and routines.

Another fun thing to do, if you didn’t already try it in the last question of the assignment, is to use the image package to view the weights on connections in graphical form; you will find routines for creating and writing images, if you want to play around with visualizing your network weights.

Finally, the point of this assignment is for you to obtain first-hand experience in working with neural networks; it is not intended as an exercise in C hacking. An effort has been made to keep the image package and neural network package as simple as possible. If you need clarifications about how the routines work, don’t hesitate to ask.
 

3.3. The neural network package

As mentioned earlier, this package implements three-layer fully-connected feedforward neural networks, using a backpropagation weight tuning method. We begin with a brief description of the data structure, a BPNN (BackPropNeuralNet).

All unit values and weight values are stored as doubles in a BPNN. Given a BPNN *net, you can get the number of input, hidden, and output units with net->input_n, net->hidden_n, and net-> output_n, respectively.

Units are all indexed from 1 to n, where n is the number of units in the layer. To get the value of the kth unit in the input, hidden, or output layer, use net->input units[k], net->hidden units[k], or net->output units[k], respectively.

The target vector is assumed to have the same number of values as the number of units in the out-put layer, and it can be accessed via net->target. The kth target value can be accessed by net->tar-get[k].

To get the value of the weight connecting the ith input unit to the jth hidden unit, use net->input weights[i][j]. To get the value of the weight connecting the jth hidden unit to the kth output unit, use net->hidden weights[j][k].

The routines are as follows:

void bpnn initialize(seed)

int seed;
This routine initializes the neural network package. It should be called before any other routines in the package are used. Currently, its sole purpose in life is to initialize the ran-dom number generator with the input seed.
BPNN *bpnn_create(n_in, n_hidden, n_out)
int n_in, n_hidden, n_out;
Creates a new network with n_in input units, n_hidden hidden units, and n_output output units. All weights in the network are randomly initialized to values in the range [-1.0; 1.0].
Returns a pointer to the network structure. Returns NULL if the routine fails.
void bpnn_free(net)
BPNN *net;
Takes a pointer to a network, and frees all memory associated with the network.
void bpnn_train(net, learning_rate, momentum, erro, errh)
BPNN *net;
double learning_rate, momentum;
double *erro, *errh;
Given a pointer to a network, runs one pass of the backpropagation algorithm. Assumes that the input units and target layer have been properly set up. learning rate and momen-tum are assumed to be values between 0.0 and 1.0. erro and errh are pointers to doubles,?which are set to the sum of the delta error values on the output units and hidden units, respectively.
void bpnn feedforward(net)
BPNN *net;
Given a pointer to a network, runs the network on its current input values.
BPNN *bpnn read(filename)
char *filename;
Given a filename, allocates space for a network, initializes it with the weights stored in the network file, and returns a pointer to this new BPNN. Returns NULL on failure.
void bpnn save(net, filename)
BPNN *net;
char *filename;
Given a pointer to a network and a filename, saves the network to that file.

3.4. The image package

The image package provides a set of routines for manipulating PGM images. An image is a rect-angular grid of pixels; each pixel has an integer value ranging from 0 to 255. Images are indexed by rows and columns; row 0 is the top row of the image, column 0 is the left column of the image.

IMAGE *img_open(filename)

char *filename;
Opens the image given by filename, loads it into a new IMAGE data structure, and returns a pointer to this new structure. Returns NULL on failure.
IMAGE *img creat(filename, nrows, ncols)
char *filename; int nrows, ncols;
Creates an image in memory, with the given filename, of dimensions nrows x ncols, and returns a pointer to this image. All pixels are initialized to 0. Returns NULL on failure.
int ROWS(img)
IMAGE *img;
Given a pointer to an image, returns the number of rows the image has.
int COLS(img)
IMAGE *img;
Given a pointer to an image, returns the number of columns the image has.
char *NAME(img)
IMAGE *img;
Given a pointer to an image, returns a pointer to its base filename (i.e., if the full filename is /usr/joe/stuff/foo.pgm, a pointer to the string foo.pgm will be returned).
int img_getpixel(img, row, col)
IMAGE *img;
int row, col;
Given a pointer to an image and row/column coordinates, this routine returns the value of the pixel at those coordinates in the image.
void img_setpixel(img, row, col, value)
IMAGE *img;
int row, col, value;
Given a pointer to an image and row/column coordinates, and an integer value assumed to be in the range [0, 255], this routine sets the pixel at those coordinates in the image to the given value.
int img_write(img, filename)
IMAGE *img;
char *filename;
Given a pointer to an image and a filename, writes the image to disk with the given file-name. Returns 1 on success, 0 on failure.
void img_free(img)
IMAGE *img;
Given a pointer to an image, deallocates all of its associated memory.
IMAGELIST *imgl_alloc()
Returns a pointer to a new IMAGELIST structure, which is really just an array of pointers to images. Given an IMAGELIST *il, il->n is the number of images in the list. il->list[k] is the pointer to the kth image in the list.
void imgl add(il, img)
IMAGELIST *il;
IMAGE *img;
Given a pointer to an imagelist and a pointer to an image, adds the image at the end of the imagelist.
void imgl_free(il)
IMAGELIST *il;
Given a pointer to an imagelist, frees it. Note that this does not free any images to which the list points.
void imgl_load_images_from_textfile(il, filename)
IMAGELIST *il; char *filename;
Takes a pointer to an imagelist and a filename. filename is assumed to specify a file which is a list of pathnames of images, one to a line. Each image file in this list is loaded into memory and added to the imagelist il.

3.5. hidtopgm

hidtopgm takes the following fixed set of arguments:
hidtopgm net-file image-file x y n
net-file is the file containing the network in which the hidden unit weights are to be found.

image-file is the file to which the derived image will be output.

x and y are the dimensions in pixels of the image on which the network was trained.

n is the number of the target hidden unit. n may range from 1 to the total number of hidden units in the network.
 

3.6. outtopgm

outtopgm takes the following fixed set of arguments:

outtopgm net-file image-file x y n

This is the same as hidtopgm, for output units instead of input units. Be sure you specify x to be 1 plus the number of hidden units, so that you get to see the weight w0 as well as weights associated with the hidden units. For example, to see the weights for output number 2 of a network containing 3 hidden units, do this:

outtopgm pose.net pose-out2.pgm 4 1 2

net-file is the file containing the network in which the hidden unit weights are to be found.

image-file is the file to which the derived image will be output.

x and y are the dimensions of the hidden units, where x is always 1 + the number of hidden units specified for the network, and y is always 1.

n is the number of the target output unit. n may range from 1 to the total number of output units for the network.