joycex99

(def params
  {:optimizer         (adam/adam)
   :batch-size        100
   :epoch-count       50
   :epoch-size        200000})

(defn train
  []
  (let [network (network/linear-network network-description)
        [train-orig test-ds] (get-train-test-dataset)

joycex99

(def num-features (count (:data (first (create-dataset))))

(def network-description
  [(layers/input num-features 1 1 :id :data) ;width, height, channels
  (layers/linear->relu 20) ; num-output 
  (layers/dropout 0.9)
  (layers/linear->relu 10)
  (layers/linear 2)
  (layers/softmax :id :label)])

joycex99

(defn f-beta
  "F-beta score, default uses F1"
  ([precision recall] (f-beta precision recall 1))
  ([precision recall beta]
    (let [beta-squared (* beta beta)]
      (* (+ 1 beta-squared)
         (try                         ;; catch divide by 0 errors
           (/ (* precision recall)
              (+ (* beta-squared precision) recall))
         (catch ArithmeticException e

joycex99

(defn train
  "Trains network for :epoch-count number of epochs"
  []
  (let [network (network/linear-network network-description)
        [train-orig test-ds] (get-train-test-dataset)
        train-ds (experiment-util/infinite-class-balanced-dataset train-orig
                                            :class-key :label
                                            :epoch-size (:epoch-size params))]
    (experiment-train/train-n network train-ds test-ds
                                      :batch-size (:batch-size params)

joycex99

(defonce get-scaled-variances
  (memoize
    (fn []
      (let [{positives true negatives false} (group-by #(= (:label %) [0.0 1.0]) (create-dataset))
            pos-data (mat/matrix (map #(:data %) positives))
            variances (mat/matrix (map #(matstats/variance %) (mat/columns pos-data)))
            scaled-vars (mat/mul (/ 5000 (mat/length variances)) variances)]
        scaled-vars))))

joycex99

(defn f-beta
  "F-beta score, default uses F1"
  ([precision recall] (f-beta precision recall 1))
  ([precision recall beta]
    (let [beta-squared (* beta beta)]
      (* (+ 1 beta-squared)
         (try                         ;; catch divide by 0 errors
           (/ (* precision recall)
              (+ (* beta-squared precision) recall))
         (catch ArithmeticException e

joycex99

(def params
  {:test-ds-size      50000 ;; total = 284807, test-ds ~= 17.5%
   :optimizer         (adam/adam)   ;; alternately, (adadelta/adadelta)
   :batch-size        100
   :epoch-count       50
   :epoch-size        200000})

(def network-description
  [(layers/input (count (:data (first (create-dataset)))) 1 1 :id :data) ;width, height, channels, args
  (layers/linear->relu 20) ; num-output & args

joycex99

(ns fraud-detection.core
  (:require [clojure.java.io :as io]
           [clojure.string :as string]
           [clojure.data.csv :as csv]
           [clojure.core.matrix :as mat]
           [clojure.core.matrix.stats :as matstats]
           [cortex.nn.layers :as layers]
           [cortex.nn.network :as network]
           [cortex.nn.execute :as execute]
           [cortex.optimize.adadelta :as adadelta]

joycex99

$(document).on('ready page:change', function() {
  // $(document).on('flip', '#flashcard', function() {
  //   axis: 'x',
  //   trigger: 'click',
  //   speed: 400
  // })

  $('#flashcard').flip({
    axis: 'x',
    trigger: 'click',