cs231n：assignment1——Q5: Higher Level Representations: Image Features

自己写的cs231n的作业,希望给点意见,支出错误和不足.谢谢

features.ipynb内容:

Image features exercise

Complete and hand in this completed worksheet (including its outputs and any supporting code outside of the worksheet) with your assignment submission. For more details see the assignments page on the course website.

We have seen that we can achieve reasonable performance on an image classification task by training a linear classifier on the pixels of the input image. In this exercise we will show that we can improve our classification performance by training linear classifiers not on raw pixels but on features that are computed from the raw pixels.

All of your work for this exercise will be done in this notebook.

import randomimport numpy as npfrom cs231n.data_utils import load_CIFAR10import matplotlib.pyplot as plt%matplotlib inlineplt.rcParams['figure.figsize'] = (10.0, 8.0) # set default size of plotsplt.rcParams['image.interpolation'] = 'nearest'plt.rcParams['image.cmap'] = 'gray'# for auto-reloading extenrnal modules# see http://stackoverflow.com/questions/1907993/autoreload-of-modules-in-ipython%load_ext autoreload%autoreload 2

Load data

Similar to previous exercises, we will load CIFAR-10 data from disk.

from cs231n.features import color_histogram_hsv, hog_featuredef get_CIFAR10_data(num_training=49000, num_validation=1000, num_test=1000):  # Load the raw CIFAR-10 data  cifar10_dir = 'cs231n/datasets/cifar-10-batches-py'  X_train, y_train, X_test, y_test = load_CIFAR10(cifar10_dir)  # Subsample the data  mask = range(num_training, num_training + num_validation)  X_val = X_train[mask]  y_val = y_train[mask]  mask = range(num_training)  X_train = X_train[mask]  y_train = y_train[mask]  mask = range(num_test)  X_test = X_test[mask]  y_test = y_test[mask]  return X_train, y_train, X_val, y_val, X_test, y_testX_train, y_train, X_val, y_val, X_test, y_test = get_CIFAR10_data()

Extract Features

For each image we will compute a Histogram of Oriented
Gradients (HOG) as well as a color histogram using the hue channel in HSV
color space. We form our final feature vector for each image by concatenating
the HOG and color histogram feature vectors.

Roughly speaking, HOG should capture the texture of the image while ignoring
color information, and the color histogram represents the color of the input
image while ignoring texture. As a result, we expect that using both together
ought to work better than using either alone. Verifying this assumption would
be a good thing to try for the bonus section.

The hog_feature and color_histogram_hsv functions both operate on a single
image and return a feature vector for that image. The extract_features
function takes a set of images and a list of feature functions and evaluates
each feature function on each image, storing the results in a matrix where
each column is the concatenation of all feature vectors for a single image.

from cs231n.features import *#num_color_bins = 40 # Number of bins in the color histogramnum_color_bins = 50feature_fns = [hog_feature, lambda img: color_histogram_hsv(img, nbin=num_color_bins)]X_train_feats = extract_features(X_train, feature_fns, verbose=True)X_val_feats = extract_features(X_val, feature_fns)X_test_feats = extract_features(X_test, feature_fns)# Preprocessing: Subtract the mean featuremean_feat = np.mean(X_train_feats, axis=0, keepdims=True)X_train_feats -= mean_featX_val_feats -= mean_featX_test_feats -= mean_feat# Preprocessing: Divide by standard deviation. This ensures that each feature# has roughly the same scale.std_feat = np.std(X_train_feats, axis=0, keepdims=True)X_train_feats /= std_featX_val_feats /= std_featX_test_feats /= std_feat# Preprocessing: Add a bias dimensionX_train_feats = np.hstack([X_train_feats, np.ones((X_train_feats.shape[0], 1))])X_val_feats = np.hstack([X_val_feats, np.ones((X_val_feats.shape[0], 1))])X_test_feats = np.hstack([X_test_feats, np.ones((X_test_feats.shape[0], 1))])

Done extracting features for 1000 / 49000 imagesDone extracting features for 2000 / 49000 imagesDone extracting features for 3000 / 49000 images...Done extracting features for 46000 / 49000 imagesDone extracting features for 47000 / 49000 imagesDone extracting features for 48000 / 49000 images

Train SVM on features

Using the multiclass SVM code developed earlier in the assignment, train SVMs on top of the features extracted above; this should achieve better results than training SVMs directly on top of raw pixels.

# Use the validation set to tune the learning rate and regularization strengthfrom cs231n.classifiers.linear_classifier import LinearSVM#learning_rates = [1e-9, 1e-8, 1e-7]#regularization_strengths = [1e4,1e5, 1e6, 1e7]regularization_strengths = [5e3,1e4,5e5, 5e6, 1e7]learning_rates = [5e-9, 3e-8, 1e-7, 5e-7]results = {}best_val = -1best_svm = Nonepass################################################################################# TODO:                                                                        ## Use the validation set to set the learning rate and regularization strength. ## This should be identical to the validation that you did for the SVM; save    ## the best trained classifer in best_svm. You might also want to play          ## with different numbers of bins in the color histogram. If you are careful    ## you should be able to get accuracy of near 0.44 on the validation set.       #################################################################################for lr in learning_rates:    for reg_str in regularization_strengths:        svm = LinearSVM()        loss_hist = svm.train(X_train_feats, y_train, learning_rate=lr, reg=reg_str,                              num_iters=1500, verbose=False)        y_train_pred = svm.predict(X_train_feats)        accuracy_train = np.mean(y_train == y_train_pred)        y_val_pred = svm.predict(X_val_feats)        accuracy_val = np.mean(y_val == y_val_pred)        results[(lr, reg_str)] = (accuracy_train, accuracy_val)         if accuracy_val > best_val:            print "lr:",lr            print "reg:", reg_str            best_val = accuracy_val            best_svm = svm#################################################################################                              END OF YOUR CODE                                ################################################################################## Print out results.for lr, reg in sorted(results):    train_accuracy, val_accuracy = results[(lr, reg)]    print 'lr %e reg %e train accuracy: %f val accuracy: %f' % (                lr, reg, train_accuracy, val_accuracy)print 'best validation accuracy achieved during cross-validation: %f' % best_val

lr: 5e-09reg: 5000.0lr: 5e-09reg: 10000.0lr: 5e-09reg: 5000000.0lr: 1e-07reg: 500000.0cs231n/classifiers/linear_svm.py:85: RuntimeWarning: overflow encountered in double_scalars  reg_loss = 0.5 * reg * np.sum(W * W)cs231n/classifiers/linear_svm.py:85: RuntimeWarning: overflow encountered in multiply  reg_loss = 0.5 * reg * np.sum(W * W)cs231n/classifiers/linear_svm.py:130: RuntimeWarning: overflow encountered in multiply  dW += reg * Wcs231n/classifiers/linear_svm.py:83: RuntimeWarning: invalid value encountered in less  margin[margin<0] = 0    # np.where(margin>0, margin, 0)cs231n/classifiers/linear_svm.py:124: RuntimeWarning: invalid value encountered in greater  mask_margin[margin>0] = 1cs231n/classifiers/linear_classifier.py:71: RuntimeWarning: invalid value encountered in subtract  self.W = self.W - learning_rate * gradlr 5.000000e-09 reg 5.000000e+03 train accuracy: 0.114796 val accuracy: 0.117000lr 5.000000e-09 reg 1.000000e+04 train accuracy: 0.126265 val accuracy: 0.127000lr 5.000000e-09 reg 5.000000e+05 train accuracy: 0.126959 val accuracy: 0.120000lr 5.000000e-09 reg 5.000000e+06 train accuracy: 0.419857 val accuracy: 0.427000lr 5.000000e-09 reg 1.000000e+07 train accuracy: 0.416857 val accuracy: 0.426000lr 3.000000e-08 reg 5.000000e+03 train accuracy: 0.081469 val accuracy: 0.076000lr 3.000000e-08 reg 1.000000e+04 train accuracy: 0.092224 val accuracy: 0.093000lr 3.000000e-08 reg 5.000000e+05 train accuracy: 0.421184 val accuracy: 0.425000lr 3.000000e-08 reg 5.000000e+06 train accuracy: 0.412673 val accuracy: 0.399000lr 3.000000e-08 reg 1.000000e+07 train accuracy: 0.389306 val accuracy: 0.385000lr 1.000000e-07 reg 5.000000e+03 train accuracy: 0.103980 val accuracy: 0.089000lr 1.000000e-07 reg 1.000000e+04 train accuracy: 0.128490 val accuracy: 0.107000lr 1.000000e-07 reg 5.000000e+05 train accuracy: 0.425878 val accuracy: 0.436000lr 1.000000e-07 reg 5.000000e+06 train accuracy: 0.365449 val accuracy: 0.378000lr 1.000000e-07 reg 1.000000e+07 train accuracy: 0.361735 val accuracy: 0.376000lr 5.000000e-07 reg 5.000000e+03 train accuracy: 0.409694 val accuracy: 0.426000lr 5.000000e-07 reg 1.000000e+04 train accuracy: 0.421327 val accuracy: 0.421000lr 5.000000e-07 reg 5.000000e+05 train accuracy: 0.403143 val accuracy: 0.407000lr 5.000000e-07 reg 5.000000e+06 train accuracy: 0.105918 val accuracy: 0.096000lr 5.000000e-07 reg 1.000000e+07 train accuracy: 0.100265 val accuracy: 0.087000best validation accuracy achieved during cross-validation: 0.436000

# Evaluate your trained SVM on the test sety_test_pred = best_svm.predict(X_test_feats)test_accuracy = np.mean(y_test == y_test_pred)print test_accuracy

0.421

# An important way to gain intuition about how an algorithm works is to# visualize the mistakes that it makes. In this visualization, we show examples# of images that are misclassified by our current system. The first column# shows images that our system labeled as "plane" but whose true label is# something other than "plane".examples_per_class = 8classes = ['plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']for cls, cls_name in enumerate(classes):    idxs = np.where((y_test != cls) & (y_test_pred == cls))[0]    idxs = np.random.choice(idxs, examples_per_class, replace=False)    for i, idx in enumerate(idxs):        plt.subplot(examples_per_class, len(classes), i * len(classes) + cls + 1)        plt.imshow(X_test[idx].astype('uint8'))        plt.axis('off')        if i == 0:            plt.title(cls_name)plt.show()

Inline question 1:

Describe the misclassification results that you see. Do they make sense?

Neural Network on image features

Earlier in this assigment we saw that training a two-layer neural network on raw pixels achieved better classification performance than linear classifiers on raw pixels. In this notebook we have seen that linear classifiers on image features outperform linear classifiers on raw pixels.

For completeness, we should also try training a neural network on image features. This approach should outperform all previous approaches: you should easily be able to achieve over 55% classification accuracy on the test set; our best model achieves about 60% classification accuracy.

print X_train_feats.shape

(49000, 195)

from cs231n.classifiers.neural_net import TwoLayerNetinput_dim = X_train_feats.shape[1]hidden_dim = 500num_classes = 10#net = TwoLayerNet(input_dim, hidden_dim, num_classes)best_net = Nonebest_stats = None################################################################################# TODO: Train a two-layer neural network on image features. You may want to    ## cross-validate various parameters as in previous sections. Store your best   ## model in the best_net variable.                                              #################################################################################best_acc = -1input_size = 32 * 32 * 3learning_rate_choice = [1.8,1.7, 1.6, 1.5]reg_choice = [0.01, 0.011]num_iters_choice = [800]for learning_rate_curr in learning_rate_choice:    for reg_cur in reg_choice:            for num_iters_curr in num_iters_choice:            print             print "current training learning_rate:",learning_rate_curr            print "current training reg:",reg_cur            net = TwoLayerNet(input_dim, hidden_dim, num_classes)            stats = net.train(X_train_feats, y_train, X_val_feats, y_val,                        num_iters=num_iters_curr, batch_size=1500,                        learning_rate=learning_rate_curr, learning_rate_decay=0.95,                        reg=reg_cur, verbose=True)            val_acc = (net.predict(X_val_feats) == y_val).mean()            print "current val_acc:",val_acc            if val_acc>best_acc:                best_acc = val_acc                best_net = net                best_stats = stats                print                 print "best_acc:",best_acc                print "best learning_rate:",best_net.hyper_params['learning_rate']                print "best reg:",best_net.hyper_params['reg']                print #################################################################################                              END OF YOUR CODE                                #################################################################################

current training learning_rate: 1.8current training reg: 0.01iteration 0 / 800: loss 2.302590iteration 100 / 800: loss 1.455167iteration 200 / 800: loss 1.427213iteration 300 / 800: loss 1.430055iteration 400 / 800: loss 1.407665iteration 500 / 800: loss 1.455174iteration 600 / 800: loss 1.361302iteration 700 / 800: loss 1.352902current val_acc: 0.577best_acc: 0.577best learning_rate: 1.8best reg: 0.01current training learning_rate: 1.8current training reg: 0.011iteration 0 / 800: loss 2.302591iteration 100 / 800: loss 1.544071iteration 200 / 800: loss 1.471075iteration 300 / 800: loss 1.429699iteration 400 / 800: loss 1.425746iteration 500 / 800: loss 1.413695iteration 600 / 800: loss 1.404900iteration 700 / 800: loss 1.394854current val_acc: 0.561current training learning_rate: 1.7current training reg: 0.01iteration 0 / 800: loss 2.302590iteration 100 / 800: loss 1.488707iteration 200 / 800: loss 1.454584iteration 300 / 800: loss 1.368391iteration 400 / 800: loss 1.359419iteration 500 / 800: loss 1.375673iteration 600 / 800: loss 1.400355iteration 700 / 800: loss 1.355049current val_acc: 0.577current training learning_rate: 1.7current training reg: 0.011iteration 0 / 800: loss 2.302591iteration 100 / 800: loss 1.522634iteration 200 / 800: loss 1.476987iteration 300 / 800: loss 1.477775iteration 400 / 800: loss 1.406658iteration 500 / 800: loss 1.417113iteration 600 / 800: loss 1.424151iteration 700 / 800: loss 1.467979current val_acc: 0.569current training learning_rate: 1.6current training reg: 0.01iteration 0 / 800: loss 2.302590iteration 100 / 800: loss 1.506576iteration 200 / 800: loss 1.455776iteration 300 / 800: loss 1.461364iteration 400 / 800: loss 1.366618iteration 500 / 800: loss 1.411004iteration 600 / 800: loss 1.372812iteration 700 / 800: loss 1.387114current val_acc: 0.583best_acc: 0.583best learning_rate: 1.6best reg: 0.01current training learning_rate: 1.6current training reg: 0.011iteration 0 / 800: loss 2.302591iteration 100 / 800: loss 1.466752iteration 200 / 800: loss 1.460388iteration 300 / 800: loss 1.391446iteration 400 / 800: loss 1.370959iteration 500 / 800: loss 1.410860iteration 600 / 800: loss 1.414151iteration 700 / 800: loss 1.416123current val_acc: 0.581current training learning_rate: 1.5current training reg: 0.01iteration 0 / 800: loss 2.302590iteration 100 / 800: loss 1.488719iteration 200 / 800: loss 1.442920iteration 300 / 800: loss 1.421011iteration 400 / 800: loss 1.421810iteration 500 / 800: loss 1.408096iteration 600 / 800: loss 1.351295iteration 700 / 800: loss 1.362655current val_acc: 0.586best_acc: 0.586best learning_rate: 1.5best reg: 0.01current training learning_rate: 1.5current training reg: 0.011iteration 0 / 800: loss 2.302591iteration 100 / 800: loss 1.481648iteration 200 / 800: loss 1.464853iteration 300 / 800: loss 1.436531iteration 400 / 800: loss 1.423766iteration 500 / 800: loss 1.394977iteration 600 / 800: loss 1.397767iteration 700 / 800: loss 1.444636current val_acc: 0.571

#自己加的(insert by myself)# Plot the loss function and train / validation accuracies#根据上面的确定大致范围进行微调test_net = TwoLayerNet(input_dim, hidden_dim, num_classes)test_stats = test_net.train(X_train_feats, y_train, X_val_feats, y_val,                        num_iters=800, batch_size=1500,                        learning_rate=1.6, learning_rate_decay=0.95,                        reg=0.01, verbose=True)print "acc:", (test_net.predict(X_val_feats) == y_val).mean()print "learning_rate:",test_net.hyper_params['learning_rate']print "reg:",test_net.hyper_params['reg']printplt.subplot(2, 1, 1)plt.plot(test_stats['loss_history'])plt.title('Loss history')plt.xlabel('Iteration')plt.ylabel('Loss')plt.subplot(2, 1, 2)plt.plot(test_stats['train_acc_history'], label='train')plt.plot(test_stats['val_acc_history'], label='val')plt.title('Classification accuracy history')plt.xlabel('Epoch')plt.ylabel('Clasification accuracy')plt.show()

iteration 0 / 800: loss 2.302590iteration 100 / 800: loss 1.488089iteration 200 / 800: loss 1.470812iteration 300 / 800: loss 1.441917iteration 400 / 800: loss 1.404342iteration 500 / 800: loss 1.411365iteration 600 / 800: loss 1.352144iteration 700 / 800: loss 1.384254acc: 0.574learning_rate: 1.6reg: 0.01

# Run your neural net classifier on the test set. You should be able to# get more than 55% accuracy.test_acc = (best_net.predict(X_test_feats) == y_test).mean()print test_acctest_acc = (test_net.predict(X_test_feats) == y_test).mean()print test_acc

0.5630.553

Bonus: Design your own features!

You have seen that simple image features can improve classification performance. So far we have tried HOG and color histograms, but other types of features may be able to achieve even better classification performance.

For bonus points, design and implement a new type of feature and use it for image classification on CIFAR-10. Explain how your feature works and why you expect it to be useful for image classification. Implement it in this notebook, cross-validate any hyperparameters, and compare its performance to the HOG + Color histogram baseline.

Bonus: Do something extra!

Use the material and code we have presented in this assignment to do something interesting. Was there another question we should have asked? Did any cool ideas pop into your head as you were working on the assignment? This is your chance to show off!