conormm

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import begin

l1_nodes = 200
l2_nodes = 100
final_layer_nodes = 10

# define placeholder for data
# also considered as the "visibale layer, the layer that we see"

conormm

init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)

for i in range(1000):
    batch_X, batch_y = mnist.train.next_batch(100)
    # associate data with placeholders
    train_data = {X: batch_X, Y_: batch_y}

    # train model

conormm

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import begin

l1_nodes = 200
l2_nodes = 100
final_layer_nodes = 10

# define placeholder for data
# also considered as the "visibale layer, the layer that we see"

conormm

from keras.datasets import mnist
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation
from keras.utils import np_utils
import numpy as np

l1_nodes = 200
l2_nodes = 100
final_layer_nodes = 10

conormm

library(modelr)
library(dplyr)
library(ggplot2)
library(animation)
library(gganimate)

options(scipen = 999)

n     <- 200 # number of observations
bias  <- 4

conormm

import torch
import torch.nn as nn
from torch.autograd import Variable
import torch.functional as F
from torch.utils.data import Dataset, DataLoader

import numpy as np
from sklearn.datasets import load_boston
from sklearn.preprocessing import StandardScaler

conormm

import torch
import torch.nn as nn
from torch.autograd import Variable
import torch.functional as F
from torch.utils.data import Dataset, DataLoader
from torch.utils.data.sampler import SubsetRandomSampler

import numpy as np
from sklearn.datasets import load_boston
from sklearn.preprocessing import StandardScaler

conormm

logloss <- function(label, p) {
  if (label == 1) {
    x <- -log(p) 
  } else {
    x <- -log(1 - p)
  }
  x
}

conormm

library(tidyverse)

# create range of probabilities
probs <- seq(0, 1, 0.001)

# apply log loss over full range of propabilites
t1 <- logloss(1, probs)
t0 <- logloss(0, probs)

# tidy the data

conormm

df %>% 
  ggplot(aes(probs, logloss, colour = label)) + 
  geom_line(size = .8) + 
  labs(
    title = "Log Loss over estimated probability", 
    x = "Estimated probabilty",
    y = "Log Loss", 
    colour = "Label"
  ) + 
  annotate("text", label = "Log Loss penalises overly confident,