In [1]:
from __future__ import division, print_function, absolute_import
import tflearn
from tflearn.layers.core import input_data, dropout, fully_connected
from tflearn.layers.conv import conv_2d, max_pool_2d
from tflearn.layers.normalization import local_response_normalization
from tflearn.layers.estimator import regression
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
%matplotlib inline
In [2]:
import LoadDataset
In [3]:
IMG_WIDTH = 64 # Side for each transformed Image
IMG_HEIGHT = 32
IMG_DEPTH = 3 # RGB files
In [21]:
DATSET_SIZE = 100
NUM_LABELS = 2
In [22]:
data = LoadDataset.getFixedNumDigistsDataSet(DATSET_SIZE, NUM_LABELS)
DATSET_SIZE = data[0]
imgs = data[1]
labels = data[2]
In [23]:
#print (imgs.shape)
import copy
plt.rcParams['figure.figsize'] = (1.0,1.0)
print (labels[10])
img = copy.copy(imgs[10])
#print (img)
#img[img[:,:,:] > 125] = 1
#img[img[:,:,:] <= 125] = 0
plt.imshow(img)
#print (imgs[9].shape)
Out[23]:
In [24]:
def dense_to_one_hot(labels_dense, num_classes=10):
"""Convert class labels from scalars to one-hot vectors."""
num_labels = labels_dense.shape[0]
index_offset = np.arange(num_labels) * num_classes
labels_one_hot = np.zeros((num_labels, num_classes))
index_update = [int(x) for x in index_offset + labels_dense.ravel()]
labels_one_hot.flat[index_update] = 1
return labels_one_hot
In [25]:
# Get the dataset
X = imgs.reshape([-1, IMG_HEIGHT, IMG_WIDTH, IMG_DEPTH])
Y = labels
In [26]:
# Generate validation set
ratio = 0.8 # Train/Test set
randIdx = np.random.random(DATSET_SIZE) <= ratio
print (randIdx)
#print (sum(map(lambda x: int(x), randIdx)))
X_train = X[randIdx]
Y_train = Y[randIdx]
X_test = X[randIdx == False]
Y_test = Y[randIdx == False]
Y_train = [dense_to_one_hot(Y_train[:,idx], num_classes= 11) for idx in range(Y_train.shape[1])]
Y_test = [dense_to_one_hot(Y_test[:,idx], num_classes= 11) for idx in range(Y_test.shape[1])]
del X, Y # release some space
In [27]:
#print (np.array(Y_train).shape)
In [28]:
# Building convolutional network
network = input_data(shape=[None, IMG_HEIGHT, IMG_WIDTH, IMG_DEPTH], name='input')
network = conv_2d(network, 32, 3, activation='relu', regularizer="L2")
network = max_pool_2d(network, 2)
network = local_response_normalization(network)
network = conv_2d(network, 64, 3, activation='relu', regularizer="L2")
network = max_pool_2d(network, 2)
network = local_response_normalization(network)
# Training heads
allHeads = []
for idx in range(NUM_LABELS):
fc = fully_connected(network, 128, activation='tanh')
fc = dropout(fc, 0.8)
softmax = fully_connected(fc, 11, activation='softmax')
networkOut = regression(softmax, optimizer='adam', learning_rate=0.01,
loss='categorical_crossentropy', name='target' + str(idx))
allHeads.append(networkOut)
network = tflearn.merge(allHeads, mode='elemwise_sum')
In [29]:
model = tflearn.DNN(network, tensorboard_verbose=1)
feedTrainDict = {'target'+ str(i): Y_train[i] for i in range(NUM_LABELS)}
feedTestList = [Y_test[i] for i in range(NUM_LABELS)]
model.fit({'input': X_train}, feedTrainDict,
validation_set= (X_test, feedTestList), n_epoch=5, snapshot_step=100, show_metric=True, run_id='convnet_mnist')
In [ ]: