Joshuaalbert

# This is an example parset for Factor


[global]
# Full path to working dir where Factor will run (required). All output will be
# placed in this directory
dir_working = /net/para34/data1/albert/products/goods-n-factor

# Full path to directory containing input bands. It will be scanned for all
# .MS and .ms files (required). Note that if the input files are longer than

Joshuaalbert

import dill
from sys import stdout

class Logger(object):
    def __init__(self,logFile=None,standardOut = True):
        if standardOut:
            self.file = [stdout]
        else:
            self.file = []
        if logFile is not None:

Joshuaalbert

02/17/2017 20:01:00.634,com.tactico.backtest.BacktestFactory,SEVERE,start,java.lang.NullPointerException
	at org.deeplearning4j.nn.graph.ComputationGraph.setInput(ComputationGraph.java:205)
	at org.deeplearning4j.nn.graph.ComputationGraph.feedForward(ComputationGraph.java:1021)
	at org.deeplearning4j.nn.graph.ComputationGraph.score(ComputationGraph.java:1443)
	at com.tactico.tm.asyncRL.PredictivePreconditioning_Pairs.evaluate(PredictivePreconditioning_Pairs.java:129)
	at com.tactico.backtest.BacktestTimeSeries.onTimeSeriesChanged(BacktestTimeSeries.java:311)
	at com.tactico.provider.TimeSeriesObservable.notifyTimeSeriesChanged(TimeSeriesObservable.java:18)
	at com.tactico.provider.historical.HistoricalProvider.processEvents(HistoricalProvider.java:244)
	at com.tactico.provider.historical.HistoricalProvider.simulateEvents(HistoricalProvider.java:149)
	at com.tactico.provider.historical.HistoricalProvider.connect(HistoricalProvider.java:98)

Joshuaalbert

package com.tactico.tm.asyncRL;

import org.deeplearning4j.nn.api.OptimizationAlgorithm;
import org.deeplearning4j.nn.conf.ComputationGraphConfiguration;
import org.deeplearning4j.nn.conf.NeuralNetConfiguration;
import org.deeplearning4j.nn.conf.Updater;
import org.deeplearning4j.nn.conf.inputs.InputType;
import org.deeplearning4j.nn.conf.layers.GravesLSTM;
import org.deeplearning4j.nn.graph.ComputationGraph;
import org.deeplearning4j.nn.weights.WeightInit;

Joshuaalbert

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

	
    <groupId>com.tactico</groupId>
    <artifactId>backtestTM</artifactId>
    <version>1.0</version>
    <modelVersion>4.0.0</modelVersion>

Joshuaalbert

package com.tactico.tm.asyncRL;

import org.deeplearning4j.api.storage.StatsStorage;
import org.deeplearning4j.nn.api.Layer;
import org.deeplearning4j.nn.api.OptimizationAlgorithm;
import org.deeplearning4j.nn.conf.ComputationGraphConfiguration;
import org.deeplearning4j.nn.conf.NeuralNetConfiguration;
import org.deeplearning4j.nn.conf.Updater;
import org.deeplearning4j.nn.conf.graph.MergeVertex;
import org.deeplearning4j.nn.conf.graph.rnn.DuplicateToTimeSeriesVertex;

Joshuaalbert

def pickle(dataFile):
    import numpy as np
    import astropy.units as au
    import astropy.time as at
    import astropy.coordinates as ac
    import dill
    
    class RadioArray(object):
        '''Handles the radio array object.'''
        def __init__(self,arrayFile = None,antennaPos=None,name = None,msFile=None,numAntennas=0,earthLocs=None,frequency=120e6):

Joshuaalbert

from scipy.linalg import cho_solve
import numpy as np

def choSolve(L,y,lower=True):
    x = np.copy(y)
    if lower:
        i = 0
        while i < L.shape[0]:
            x[i] /= L[i,i]
            x[i+1:] -= L[i+1:,i]*x[i]

Joshuaalbert

from scipy.linalg import cho_solve
import numpy as np

def choBackSubstitution(L,y,lower=True,modify=False):
    if not modify:
        x = np.copy(y)
    else:
        x = y
    if lower:
        i = 0

Joshuaalbert

#from scipy.linalg import cho_solve
from scipy.linalg.lapack import dpotrs
import numpy as np

def choBackSubstitution(L,y,lower=True,modify=False):
    if not modify:
        x = np.copy(y)
    else:
        x = y
    if lower: