def bt_show_image(self):
self.textedit.append("Make prediction ...")
img = load_img(self.image_path).resize((32, 32))
x = img_to_array(img)
x = normalize(x.reshape((1, ) + x.shape))
result = np.argmax(self.cnn.single_predict(x))
self.textedit.append("The food is: {}".format(self.food_list[result]))
# self.show_index = np.random.randint(3, 1000)
# for i in range(3):
self.lb_target2.setPixmap(
QPixmap("./{}/2.jpg".format(result)).scaled(100, 100))
import numpy as np from load import loadLasso, normalize import matplotlib.pyplot as plt from equations_regression import * np.random.seed(40) # load lasso data and normalize: X, y = loadLasso(version="train") Xv, yv = loadLasso(version="test") X = normalize(X) Xv = normalize(Xv) # add bias term (column of ones): x = np.ones((np.shape(X)[0], np.shape(X)[1] + 1)) x[:, 1:np.shape(X)[1] + 1] = X xv = np.ones((np.shape(Xv)[0], np.shape(Xv)[1] + 1)) xv[:, 1:np.shape(Xv)[1] + 1] = Xv # store data dimensions: p = np.shape(x)[0] # samples -> 50 n = np.shape(x)[1] # dimensions -> 101 # set regularization term: alpha = np.logspace(0.3, 0.6, 100) # initialize for cross fold validation: cost_crossfold_ridge_test = np.zeros(len(alpha)) counter = 0 # cross validation:
from imports import *
from load import load_data, mapValues, getgen, normalize
from schedule import getcallbacks, totalepochs
from model import getmodel
Xall, Yall, Xtest, Ytest = load_data()
Xall, Xtest = Xall.astype(np.float32), Xtest.astype(np.float32)
Nall, Ntest = Xall.shape[0], Xtest.shape[0]
print(
"Training data before split: {}\nTest data: {}\nTraining labels before split: {}\nTest labels: {}"
.format(Xall.shape, Xtest.shape, Yall.shape, Ytest.shape))
Xall, Xtest = mapValues(Xall, 0, 255, save=True), mapValues(Xtest, 0, 255)
Xall, Xtest = normalize(Xall), normalize(Xtest)
print("Intensities after scaling: min={}, max={}, mean={}, std={}".format(
np.min(Xall.flatten()), np.max(Xall.flatten()), np.mean(Xall.flatten()),
np.std(Xall.flatten())))
Xtrain, Xval, Ytrain, Yval = train_test_split(Xall,
Yall,
test_size=0.1,
random_state=SEED)
print("Train data: {}, Validation data: {}".format(Xtrain.shape, Xval.shape))
model = getmodel()
if DO_TRAIN:
history = model.fit_generator(generator=getgen().flow(Xtrain,
Ytrain,
batch_size=BS),
steps_per_epoch=int(Xtrain.shape[0] / BS),
epochs=totalepochs(SCHEDULE),
def get_ai_moves(game):
board = get_board_weird(game)
board = numpy.asarray(board)
board = normalize(board)
#board = append_snake(board)
return ann.predict_move(board)
def correlated_inputs():
n = 3 # dimensions
p = 1000 # samples
#w = [2.0, 3.0, 0.0]
w = [-2.0, 3.0, 0.0]
x = np.zeros((p, n))
x[:, 0:2] = np.random.randn(p, 2)
x[:, 2] = (2.0 / 3.0) * x[:, 0] + (2.0 / 3.0) * x[:, 1] + (
1.0 / 3.0) * np.random.randn(p)
y = np.matmul(w, np.transpose(x)) + np.random.randn(p)
return x, y
# load data:
X, y = correlated_inputs()
x = normalize(X)
# store data dimensions:
p = np.shape(x)[0] # samples -> 1000
n = np.shape(x)[1] # dimensions -> 3
# set regularization term:
alpha = np.logspace(-5, 2, 100)
# initialize for cross fold validation:
weights_crossfold = np.zeros((len(alpha), n))
counter = 0
# cross validation:
for decay in alpha:
if counter == 0: