def testValidationAccuracy():
net = memoizeNet()
trainX, trainY, validX, validY = getData()
[accuracy] = net.evaluate(validX, validY)
printColor("Validation Accuracy: {:.2f}%".format(accuracy * 100), "38;32m")
return accuracy > 0.95
def testLoadData():
trainX, trainY, validX, validY = getData()
return (trainX.shape == (7500, 150, 200, 1) and trainY.shape == (7500, 25)
and validX.shape == (2500, 150, 200, 1)
and validY.shape == (2500, 25))
# -*- coding: utf-8 -*-
"""
Created on Wed Sep 20 07:26:13 2017
@author: lxf96
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
import numpy as np
import load_data
Imgs, labels = load_data.getData()
train_Img = load_data.preprocess(Imgs)
labels = np.zeros((50000))
for i in range(500):
labels[i*100:(i+1)*100]=np.array(range(100))
from keras.layers import Input, Dense, Conv2D, MaxPooling2D, UpSampling2D, Flatten, Reshape, Lambda, concatenate, Activation, BatchNormalization
from keras.layers.advanced_activations import LeakyReLU, PReLU
from keras.optimizers import SGD, Adadelta, Adagrad,Adam, rmsprop
from keras import objectives
from keras.callbacks import TensorBoard
from keras import backend as K
from keras.models import Model
from keras.losses import binary_crossentropy
from keras.callbacks import EarlyStopping
'''
main file for convolutional Autoencoder
'''
from load_data import getData
import numpy as np
import cv2
from ConvAutoencoder_class import ConvAutoencoder
import matplotlib.pyplot as plt
#load dataset
X_train, Y_train, X_test, Y_test = getData()
#creating input tensor
X_train = np.reshape(X_train, (len(X_train), 28, 28, 1))
Y_train = np.reshape(Y_train, (len(Y_train), 28, 28, 1))
X_test = np.reshape(X_test, (len(X_test), 28, 28, 1))
Y_test = np.reshape(Y_test, (len(Y_test), 28, 28, 1))
print("Train data orignal:{},rotated:{}".format(np.shape(X_train),
np.shape(Y_train)))
#train models
autoencoder = ConvAutoencoder()
autoencoder.train(Y_train, X_train, Y_test, X_test, 256, 10)
decoder_image = autoencoder.getDecodedImage(Y_test)
#visualization
plt.figure(figsize=(20, 4))
for i in range(10):
#orignal
from load_data import getData from net import build_net import tflearn netPath = "./net/dni_reader.tfl" net = build_net() net.load(netPath, weights_only=True) trainX, trainY, validX, validY = getData() print "Training Accuracy" print net.evaluate(trainX, trainY)[0] print "Validation Accuracy" print net.evaluate(validX, validY)[0]
