steven-mi

def softmax(class_scores):
    """
    Calculate class probability distribution for each digit from given class scores.
    
    :param class_scores: class scores of your function
    :return: probability distribution
    """
    class_scores -= np.max(class_scores)
    return np.exp(class_scores) / np.sum(np.exp(class_scores),axis=1, keepdims=True)
 

steven-mi

# this implementation was given as assignment 3 of the course 
# B55.2 WT Ausgewählte Kapitel sozialer Webtechnologien at HTW Berlin


# third party
import numpy as np
import matplotlib.pyplot as plt

# internal
from deep_teaching_commons.data.fundamentals.mnist import Mnist

steven-mi

import cv2

# open video
video = cv2.VideoCapture('video.mp4')

# get fps and duration
fps = video.get(cv2.CAP_PROP_FPS)      # OpenCV2 version 2 used "CV_CAP_PROP_FPS"
frameCount = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
duration = frameCount/fps

steven-mi

def minibatcher(inputs, targets, batchsize, shuffle=False):
   assert len(inputs) == len(targets)
   if shuffle:
      indices = np.arange(len(inputs))
      np.random.shuffle(indices)
   for start_idx in range(0, len(inputs) - batchsize + 1, batchsize):
      if shuffle:
         excerpt = indices[start_idx:start_idx + batchsize]
      else:
         excerpt = slice(start_idx, start_idx + batchsize)

steven-mi

import tensorflow as tf

def get_weight(shape, name, trainable=True):
    #initial = tf.random_uniform(shape, minval=-0.1, maxval = 0.1)
    initial = tf.contrib.layers.xavier_initializer()(shape)
    return tf.Variable(initial, trainable=trainable, name=name+'_W', dtype=tf.float32)

def get_bias(shape, name, trainable=True):
    """
    filter_height, filter_width, in_channels, out_channels]

steven-mi

import matplotlib.pyplot as plt
#   plot 0     plot 1    plot 2   plot 3
x=[[1,2,3,4],[1,4,3,4],[1,2,3,4],[9,8,7,4]]
y=[[3,2,3,4],[3,6,3,4],[6,7,8,9],[3,2,2,4]]
plots = zip(x,y)

figs={}
axs={}
for idx,plot in enumerate(plots):
    figs[idx]=plt.figure()

steven-mi

export JAVA_7_HOME=$(/usr/libexec/java_home -v1.7)
export JAVA_8_HOME=$(/usr/libexec/java_home -v1.8)
export JAVA_9_HOME=$(/usr/libexec/java_home -v9)

alias java7='export JAVA_HOME=$JAVA_7_HOME'
alias java8='export JAVA_HOME=$JAVA_8_HOME'
alias java9='export JAVA_HOME=$JAVA_9_HOME'

#default java8
export JAVA_HOME=$JAVA_8_HOME

steven-mi

remove Docker container

docker rm tf

stop the container

docker stop tf

steven-mi

How to install virtualenv:

Install pip first

sudo apt-get install python3-pip

Then install virtualenv using pip3

sudo pip3 install virtualenv 

steven-mi

    plt.figure(figsize=(20, 20))
    num_classes = 10

    for c in range(num_classes):
        # Select samples_per_class random keys of the labels == current class
        keys = np.random.choice(np.where(label == c)[0], examples_each_row)
        images = data[keys]
        labels = label[keys]
        for i in range(examples_each_row):
            f = plt.subplot(examples_each_row, num_classes, i * num_classes + c + 1)