coreyauger

import numpy as np
import os
import dtdata as dt
from sklearn.model_selection import train_test_split
from sklearn.decomposition import PCA
from keras.models import Sequential
from keras.layers import Dense, Activation, Dropout
from keras.optimizers import RMSprop, Adam
from keras.callbacks import ModelCheckpoint
from keras import regularizers

coreyauger

def plotTrainingExample(te):
    plt.plot(range(len(te)),te)
    plt.show()

plotTrainingExample(data[INDEX_OF_EXAMPLE,:])

coreyauger

import numpy as np
import os
import dtdata as dt
from sklearn.model_selection import train_test_split
from sklearn.decomposition import PCA
from keras.models import Sequential
from keras.layers import Dense, Activation, Dropout
from keras.optimizers import RMSprop, Adam
from keras.callbacks import ModelCheckpoint

coreyauger

def toClasses(labels, num_classes):
    sorted = np.sort(np.array(labels, copy=True))
    bsize = math.floor( len(sorted) / num_classes )
    buckets = []
    for i in range(num_classes):        
        buckets.append(sorted[i*bsize])
    print("buckets: " + str(buckets))
    targets = np.digitize(labels, buckets) - 1
    one_hot_targets = np.eye(num_classes)[targets]
    print(one_hot_targets)

coreyauger

from sklearn import preprocessing

def scale(data):
    return preprocessing.scale(data)

coreyauger

import numpy as np
import os
import dtdata as dt
from sklearn.neighbors import NearestNeighbors
from sklearn.decomposition import PCA
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(color_codes=True)

coreyauger

def loadData(path, subset = -1):        
    allFiles = glob.glob(os.path.join(path, "data_*.csv"))
    if(subset > 0):
        allFiles = allFiles[0:subset]
    data = []
    for file in allFiles:
        print(file)
        with open(file, 'r') as f:
            data.append( [float(x[1]) for x in list(csv.reader(f))] )   
    return np.array(data)

coreyauger

def centerAroundEntry(data):
    # extract the price at 20 min after entry
    labels = data[:,-1]
    # remove the last 20 min of history from our data..
    data = data[:,0:-20]
    # normalise data to the ENTRY point
    for i in range(data.shape[0]):
        labels[i] = (labels[i]/data[i,-1]) - 1.0
        data[i,] = (data[i,]/data[i,-1]) - 1.0
    return (data, labels)

coreyauger

worker.webrtc().onAddRemoteStream = (uuid, video, dataChannel) ->
    id = $scope.peers.length+1;
    dataChannelList.push(dataChannel)
    $scope.peers.push({
      uuid:uuid
      local: false
      username: ''
      id: id
    })
    jidToPeerId[uuid] = id