Creating my own version of the dogs_cats_redux notebook in order to make my own entry into the Kaggle competition.
My dir structure is similar, but not exactly the same:
utils
dogscats (not lesson1)
data
(no extra redux subdir)
train
test
In [1]:
#Verify we are in the lesson1 directory
%pwd
Out[1]:
In [2]:
%matplotlib inline
In [3]:
import os, sys
sys.path.insert(1, os.path.join(sys.path[0], '../utils'))
from utils import *
from vgg16 import Vgg16
from PIL import Image
from keras.preprocessing import image
from sklearn.metrics import confusion_matrix
Note: had to comment out vgg16bn in utils.py (whatever that is)
In [4]:
current_dir = os.getcwd()
LESSON_HOME_DIR = current_dir
DATA_HOME_DIR = current_dir+'/data'
In [ ]:
from shutil import copyfile
#Create directories
%cd $DATA_HOME_DIR
# did this once
#%mkdir valid
#%mkdir results
#%mkdir -p sample/train
#%mkdir -p sample/test
#%mkdir -p sample/valid
#%mkdir -p sample/results
#%mkdir -p test/unknown
In [ ]:
%cd $DATA_HOME_DIR/train
In [ ]:
# create validation set by renaming 2000
g = glob('*.jpg')
shuf = np.random.permutation(g)
In [ ]:
NUM_IMAGES=len(g)
NUM_VALID = 2000 # 0.1*NUM_IMAGES
NUM_TRAIN = NUM_IMAGES-NUM_VALID
print("total=%d train=%d valid=%d"%(NUM_IMAGES,NUM_TRAIN,NUM_VALID))
In [ ]:
for i in range(NUM_VALID):
os.rename(shuf[i], DATA_HOME_DIR+'/valid/' + shuf[i])
In [ ]:
# copy a small sample
g = glob('*.jpg')
shuf = np.random.permutation(g)
for i in range(200): copyfile(shuf[i], DATA_HOME_DIR+'/sample/train/' + shuf[i])
In [ ]:
%cd $DATA_HOME_DIR/valid
In [ ]:
g = glob('*.jpg')
shuf = np.random.permutation(g)
for i in range(50): copyfile(shuf[i], DATA_HOME_DIR+'/sample/valid/' + shuf[i])
In [ ]:
!ls {DATA_HOME_DIR}/train/ |wc -l
!ls {DATA_HOME_DIR}/valid/ |wc -l
!ls {DATA_HOME_DIR}/sample/train/ |wc -l
!ls {DATA_HOME_DIR}/sample/valid/ |wc -l
In [ ]:
#Divide cat/dog images into separate directories
%cd $DATA_HOME_DIR/sample/train
%mkdir cats
%mkdir dogs
%mv cat.*.jpg cats/
%mv dog.*.jpg dogs/
%cd $DATA_HOME_DIR/sample/valid
%mkdir cats
%mkdir dogs
%mv cat.*.jpg cats/
%mv dog.*.jpg dogs/
%cd $DATA_HOME_DIR/valid
%mkdir cats
%mkdir dogs
%mv cat.*.jpg cats/
%mv dog.*.jpg dogs/
%cd $DATA_HOME_DIR/train
%mkdir cats
%mkdir dogs
%mv cat.*.jpg cats/
%mv dog.*.jpg dogs/
In [ ]:
# Create single 'unknown' class for test set
%cd $DATA_HOME_DIR/test
%mv *.jpg unknown/
In [ ]:
!ls {DATA_HOME_DIR}/test
In [5]:
%cd $DATA_HOME_DIR
#Set path to sample/ path if desired
path = DATA_HOME_DIR + '/' #'/sample/'
test_path = DATA_HOME_DIR + '/test/' #We use all the test data
results_path=DATA_HOME_DIR + '/results/'
train_path=path + '/train/'
valid_path=path + '/valid/'
In [6]:
vgg = Vgg16()
In [7]:
#Set constants. You can experiment with no_of_epochs to improve the model
batch_size=64
no_of_epochs=2
In [8]:
#Finetune the model
batches = vgg.get_batches(train_path, batch_size=batch_size)
val_batches = vgg.get_batches(valid_path, batch_size=batch_size*2)
vgg.finetune(batches)
#Not sure if we set this for all fits
vgg.model.optimizer.lr = 0.01
In [9]:
#Notice we are passing in the validation dataset to the fit() method
#For each epoch we test our model against the validation set
latest_weights_filename = None
#vgg.model.load_weights('/home/rallen/Documents/PracticalDL4C/courses/deeplearning1/nbs/data/dogscats/models/first.h5')
#vgg.model.load_weights(results_path+'ft1.h5')
latest_weights_filename='ft24.h5'
vgg.model.load_weights(results_path+latest_weights_filename)
if you are training, stay here. if you are loading & creating submission skip down from here.
In [ ]:
# if you have run some epochs already...
epoch_offset=12 # trying again from ft1
for epoch in range(no_of_epochs):
print "Running epoch: %d" % (epoch + epoch_offset)
vgg.fit(batches, val_batches, nb_epoch=1)
latest_weights_filename = 'ft%d.h5' % (epoch + epoch_offset)
vgg.model.save_weights(results_path+latest_weights_filename)
print "Completed %s fit operations" % no_of_epochs
Results of ft1.h5
0 val_loss: 0.2122 val_acc: 0.9830
1 val_loss: 0.1841 val_acc: 0.9855
[[987 7]
[ 20 986]]
--
2 val_loss: 0.2659 val_acc: 0.9830
3 val_loss: 0.2254 val_acc: 0.9850
4 val_loss: 0.2072 val_acc: 0.9845
[[975 19]
[ 11 995]]
Results of first0.h5
0 val_loss: 0.2425 val_acc: 0.9830
[[987 7]
[ 27 979]]
In [ ]:
# only if you have to
latest_weights_filename='ft1.h5'
vgg.model.load_weights(results_path+latest_weights_filename)
In [ ]:
val_batches, probs = vgg.test(valid_path, batch_size = batch_size)
filenames = val_batches.filenames
expected_labels = val_batches.classes #0 or 1
#Round our predictions to 0/1 to generate labels
our_predictions = probs[:,0]
our_labels = np.round(1-our_predictions)
In [ ]:
cm = confusion_matrix(expected_labels, our_labels)
plot_confusion_matrix(cm, val_batches.class_indices)
In [ ]:
#Helper function to plot images by index in the validation set
#Plots is a helper function in utils.py
def plots_idx(idx, titles=None):
plots([image.load_img(valid_path + filenames[i]) for i in idx], titles=titles)
#Number of images to view for each visualization task
n_view = 4
In [ ]:
#1. A few correct labels at random
correct = np.where(our_labels==expected_labels)[0]
print "Found %d correct labels" % len(correct)
idx = permutation(correct)[:n_view]
plots_idx(idx, our_predictions[idx])
In [ ]:
#2. A few incorrect labels at random
incorrect = np.where(our_labels!=expected_labels)[0]
print "Found %d incorrect labels" % len(incorrect)
idx = permutation(incorrect)[:n_view]
plots_idx(idx, our_predictions[idx])
In [ ]:
#3a. The images we most confident were cats, and are actually cats
correct_cats = np.where((our_labels==0) & (our_labels==expected_labels))[0]
print "Found %d confident correct cats labels" % len(correct_cats)
most_correct_cats = np.argsort(our_predictions[correct_cats])[::-1][:n_view]
plots_idx(correct_cats[most_correct_cats], our_predictions[correct_cats][most_correct_cats])
In [ ]:
#3b. The images we most confident were dogs, and are actually dogs
correct_dogs = np.where((our_labels==1) & (our_labels==expected_labels))[0]
print "Found %d confident correct dogs labels" % len(correct_dogs)
most_correct_dogs = np.argsort(our_predictions[correct_dogs])[:n_view]
plots_idx(correct_dogs[most_correct_dogs], our_predictions[correct_dogs][most_correct_dogs])
In [ ]:
#4a. The images we were most confident were cats, but are actually dogs
incorrect_cats = np.where((our_labels==0) & (our_labels!=expected_labels))[0]
print "Found %d incorrect cats" % len(incorrect_cats)
if len(incorrect_cats):
most_incorrect_cats = np.argsort(our_predictions[incorrect_cats])[::-1][:n_view]
plots_idx(incorrect_cats[most_incorrect_cats], our_predictions[incorrect_cats][most_incorrect_cats])
In [ ]:
#4b. The images we were most confident were dogs, but are actually cats
incorrect_dogs = np.where((our_labels==1) & (our_labels!=expected_labels))[0]
print "Found %d incorrect dogs" % len(incorrect_dogs)
if len(incorrect_dogs):
most_incorrect_dogs = np.argsort(our_predictions[incorrect_dogs])[:n_view]
plots_idx(incorrect_dogs[most_incorrect_dogs], our_predictions[incorrect_dogs][most_incorrect_dogs])
In [ ]:
#5. The most uncertain labels (ie those with probability closest to 0.5).
most_uncertain = np.argsort(np.abs(our_predictions-0.5))
plots_idx(most_uncertain[:n_view], our_predictions[most_uncertain])
In [ ]:
In [10]:
batches, preds = vgg.test(test_path, batch_size = batch_size*2)
# Error allocating 3347316736 bytes of device memory (out of memory).
# got this error when batch-size = 128
# I see this pop up to 6GB memory with batch_size = 64 & this takes some time...
In [11]:
#For every image, vgg.test() generates two probabilities
#based on how we've ordered the cats/dogs directories.
#It looks like column one is cats and column two is dogs
print preds[:5]
filenames = batches.filenames
print filenames[:5]
In [12]:
#You can verify the column ordering by viewing some images
Image.open(test_path + filenames[1])
Out[12]:
In [ ]:
#Save our test results arrays so we can use them again later
save_array(results_path + 'test_preds.dat', preds)
save_array(results_path + 'filenames.dat', filenames)
In [ ]:
#Load our test predictions from file
preds = load_array(results_path + 'test_preds.dat')
filenames = load_array(results_path + 'filenames.dat')
In [19]:
#Grab the dog prediction column
isdog = preds[:,1]
print "Raw Predictions: " + str(isdog[:5])
print "Mid Predictions: " + str(isdog[(isdog < .6) & (isdog > .4)])
print "Edge Predictions: " + str(isdog[(isdog == 1) | (isdog == 0)])
In [20]:
#play it safe, round down our edge predictions
#isdog = isdog.clip(min=0.05, max=0.95)
#isdog = isdog.clip(min=0.02, max=0.98)
isdog = isdog.clip(min=0.01, max=0.99)
In [21]:
#Extract imageIds from the filenames in our test/unknown directory
filenames = batches.filenames
ids = np.array([int(f[8:f.find('.')]) for f in filenames])
In [22]:
subm = np.stack([ids,isdog], axis=1)
subm[:5]
Out[22]:
In [23]:
%cd $DATA_HOME_DIR
submission_file_name = 'submission4.csv'
np.savetxt(submission_file_name, subm, fmt='%d,%.5f', header='id,label', comments='')
In [24]:
from IPython.display import FileLink
%cd $LESSON_HOME_DIR
FileLink('data/'+submission_file_name)
Out[24]:
In [ ]: