# \x/x\x/x\x/x\x/ -- fully connected layer (relu) W2 [200, 100] B2[100] # · · · · · · · Y2 [batch, 100] # \x/x\x/x\x/ -- fully connected layer (relu) W3 [100, 60] B3[60] # · · · · · Y3 [batch, 60] # \x/x\x/ -- fully connected layer (relu) W4 [60, 30] B4[30] # · · · Y4 [batch, 30] # \x/ -- fully connected layer (softmax) W5 [30, 10] B5[10] # · Y5 [batch, 10] import tensorflow as tf print("Tensorflow version " + tf.__version__) tf.set_random_seed(0) np.random.seed(0) # Read the training / testing dataset and labels xTrain, yTrain, xTest, yTest, yLabels = readDatabase() # Network parameters layer1Size = 200 layer2Size = 100 layer3Size = 60 layer4Size = 30 # Train hyper-parameters learningRate = 0.003 decay = 0.00035 noOfEpochs = 10 batchSize = 100 # Program parameters
required=False, help="show images (0 = False, 1 = True)") args = vars(ap.parse_args()) verbose = args["verbose"] if verbose is None: verbose = False else: if verbose == '1': verbose = True else: verbose = False # Read the training / testing dataset and labels xTrain, yTrain, xTest, yTest, yLabels = readDatabase(reshape=True) # Network parameters firstConvLayerDepth = 4 secondConvLayerDepth = 8 thirdConvLayerDepth = 12 numberOfNeurons = 200 # Training hyperparameters learningRate = 0.001 noOfEpochs = 3 batchSize = 32 numberOfClasses = yTrain.shape[1] featureSize = xTrain.shape[1]
import argparse ap = argparse.ArgumentParser() ap.add_argument("-v", "--verbose", required=False, help="show images") args = vars(ap.parse_args()) verbose = args["verbose"] if verbose is None: verbose = False else: verbose = bool(verbose) sns.set(style='white', context='notebook', palette='deep') xTrain, yTrain, xTest, yTest, yLabels = readDatabase(reshape=True, categoricalValues=False) barValues = yTrain.value_counts() print("\nNumber of training dataset: ") print(xTrain.shape) print("\nNumber of of images per label: ") print(barValues) print("\nNumber of test dataset: ") print(xTest.shape) if verbose: g = sns.countplot(yTrain) plt.show() g = sns.countplot(yTest)