svecon

import requests

def DownloadImage(pic_url_prefix, pic_name):
    with open(pic_name, 'wb') as handle:
        response = requests.get(pic_url_prefix+pic_name, stream=True)

        if not response.ok:
            print(pic_url_prefix+pic_name, response)

        for block in response.iter_content(1024):

svecon

def remove_layers(model, n)
    for _ in range(n):
        model.pop()
    model.outputs = [model.layers[-1].output]
    model.layers[-1].outbound_nodes = []
    model.compile(loss='binary_crossentropy', optimizer='adam')
    return model
    
# Example
new_model = remove_layers(model.layers_by_depth[1][0])

svecon

# https://medium.com/@nikasa1889/a-guide-to-receptive-field-arithmetic-for-convolutional-neural-networks-e0f514068807#1780
import math
def receptive_field(n, layers):
    """
    Parameters
    ----------
    n: int, input size
    layers: array of triplets: (k,s,p)
        k: kernel size
        s: stride

svecon

from IPython.core.display import display, HTML
display(HTML("<style>.container { width:100% !important; }</style>"))
display(HTML("<style> div.cell {display: -webkit-box; -webkit-box-orient: horizontal;-webkit-box-align: stretch;\
            display: -moz-box;-moz-box-orient: horizontal;-moz-box-align: stretch;\
            display: box;box-orient: horizontal;box-align: stretch;}</style>"))
display(HTML("<style> div.input{min-width:50%;}</style>"))
display(HTML("<style> div.code_cell{width: 70%;float: left;}"
   +"div.text_cell{width: 30%;float: right;}"
   +"div.text_cell div.prompt {display: none;}</style>"))

svecon

A mathematician and a lady strike conversation during a flight. The lady asks the mathematician to guess the age of her three daughters. Mathematician, being unable to answer the questions, asks for some hint(s).
The lady replies: The product of their ages is 36.
Still unable to answer, mathematician asks for another hint.
Hint 2: The sum of their ages is equal to your (i.e. mathematician's) seat number.
Mathematician still cann't answer the question.
Final hint: My (i.e. the lady's) youngest daughter has blue eyes.
Listening to the final hint, mathematician is able to answer the question.

x*y*z = 36 = 2*2*3*3

svecon

Mahalanobis distance dM(x,x′)^2 = (x − x′)^T M (x − x′)
	where M is cone of symmetric PSD d*d matrix

Must-link / cannot-link constraints (sometimes called positive / negative pairs):
	S = {(xi, xj) : xi and xj should be similar},
	D = {(xi, xj) : xi and xj should be dissimilar}.
Relative constraints (sometimes called training triplets):
	R = {(xi, xj , xk) : xi should be more similar to xj than to xk}.

Optimization problem that has the following general form: min M { ℓ(M, S,D,R) + λR(M) }

svecon

import tensorflow as tf

class EvoNetwork:
    def __init__(self, input_size, weights, use_biases=False):
        with tf.Graph().as_default():
            self.inputs = tf.placeholder(tf.float32, [None, input_size])

            hidden_layer = self.inputs
            for i, weight in enumerate(weights):
                with tf.variable_scope('hidden_layer_{}'.format(i)):

svecon

# Converting date to number of days
# Useful when you need to add few days and get a date
# https://en.wikipedia.org/wiki/Julian_day
def toJND(day,month,year):
	a = (14-month)//12
	y = year + 4800 - a
	m = month + 12*a - 3

	return day + (153*m+2)//5 + 365*y + y//4 - 32083

svecon

[HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Services\TimeBroker]
"Start"=dword:00000003

[HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Services\SysMain]
"DisplayName"="Superfetch"
"Start"=dword:00000003

svecon

(function (doc) {
  var xhr = new XMLHttpRequest();
  xhr.onload = function () {
    var div = doc.createElement('div');
    div.innerHTML = this.responseText;
    // div.style.display = 'none'; -- causes gradients not to render!
    div.style.visibility = 'hidden';
    div.style.position = 'absolute';
    div.style.height = 0;
    div.style.width = 0;