iandewancker

import numpy as np
import sklearn.ensemble
from sklearn.neural_network import MLPClassifier

X = np.random.uniform(low=0, high=5000000,size=(10000,2))
y = np.array(X[:,0] > X[:,1],dtype=int)
X_test = np.random.uniform(low=0,high=5000000,size=(10000,2))
y_test = np.array(X_test[:,0] > X_test[:,1],dtype=int)

clf_rf = sklearn.ensemble.RandomForestClassifier(n_estimators=50,criterion='entropy', max_depth=2)

iandewancker

%matplotlib
import matplotlib
matplotlib.rcParams.update({'font.size': 32})
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np; np.random.seed(10)
mean1, cov1 = [0, 2], [(0.5, .25), (.25, 0.5)]
mean2, cov2 = [2.5, -0.5], [(0.01, 0.02), (0.02, 0.01)]
mean3, cov3 = [3.2,  1.0], [(0.01, 0.02), (0.02, 0.01)]

iandewancker

X_test = np.vstack([test_x1, test_x2, test_x3, test_x4, test_x5, test_x6])
y_test = np.array([1.0, 1.0, 1.0, 1.0, -1.0, -1.0])

clf.model.coef_ = clf.model.coef_.reshape(1,clf.model.coef_.shape[0])
# hack to set the classes
try:
	clf.model.fit([],[0,1])
except:
	pass

iandewancker

def replace_dot_header(file_name):
	with open(file_name, 'r+b') as f:
		line = next(f) # grab first line
		old = '.'
		new = '_' 
		f.seek(0) # move file pointer to beginning of file
		f.write(line.replace(old, new))

iandewancker

def test_f(x):
    x1 = x[0]
    y1 = x[1]
    return 2*x1**2 + 2*x1 + 10 + y1**2 - 5*y1

"""
    f : reference to python function that takes 
    x_0 : numpy array of shape (1,N) where N is dependent on function
"""
def finite_difference_grad(f, x_0, h):

iandewancker

X_g = X[np.where(y==1)] 
X_b = X[np.where(y==0)]
M = X.shape[1]
ranges = []
for i in xrange(M):
    ranges.append((np.min(X[:,i]), np.max(X[:,i])))

importances = []
for i in xrange(M):
   g_dist = np.histogram(X_g[:,i],bins=50,density=True,range=ranges[i])[0]     

iandewancker

# load keras model from disk
model_name = "GVC_IncepvtionV3_epoch_6_vanilla_vgg_chute_date_2018_07_27"                                  
json_file = open(model_name+'.json', 'r')                          
model_json = json_file.read()                                      
json_file.close()                                                  
model = tensorflow.keras.models.model_from_json(model_json)        
# load weights into new model                                      
model.load_weights(model_name+".h5")                               
print("Loaded model from disk")

iandewancker

def custom_cat_crossentropy(y_true, y_pred):
    # negate 3x3 sub matrix of income matrix to get cost matrix
    # check https://github.com/tensorflow/tensorflow/blob/r1.10/tensorflow/python/keras/backend.py#L3461
    cost_m = tf.constant([[ 0.00222222,  0.01111111,  0.00222222],
       [ 0.00222222, -0.05888889,  0.00222222],
       [ 0.00222222,  1.51111111,  0.00222222]])
    y_true = tf.matmul(y_true, cost_m)
    return tf.keras.losses.categorical_crossentropy(y_true, y_pred)

iandewancker

import scipy.stats
import numpy as np
import scipy.optimize

obs_data = [
0.08982035928143713
,0.06818181818181818
,0.012987012987012988
,0.05357142857142857
,0.045454545454545456

iandewancker

import glob
images = glob.glob("*.jpg")
import re
import os
import matplotlib.pyplot as plt

def natural_sort(l): 
    convert = lambda text: int(text) if text.isdigit() else text.lower() 
    alphanum_key = lambda key: [ convert(c) for c in re.split('([0-9]+)', key) ] 
    return sorted(l, key = alphanum_key)