gcr

json = require 'cjson'

function buildNcduLayer(name, module)
   local result = nil
   if torch.isTensor(module) then
      if module:numel() ~= 0 then
         local strt = {name..': [' .. torch.typename(module) .. ' of size '}
         for i=1,module:nDimension() do
            table.insert(strt, module:size(i))
            if i ~= module:nDimension() then

gcr

json = require 'cjson'

function buildNcduLayer(name, module)
   local result = nil
   if torch.isTensor(module) then
      if module:numel() ~= 0 then
         local strt = {name..': [' .. torch.typename(module) .. ' of size '}
         for i=1,module:nDimension() do
            table.insert(strt, module:size(i))
            if i ~= module:nDimension() then

gcr

th> x = torch.randn(3,3)
                                                                      [0.0001s]
th> x
-0.9764  1.3443  0.4054
 1.7598  1.9367 -0.6121
-0.1593 -0.0788 -0.2321
[torch.DoubleTensor of size 3x3]

                                                                      [0.0002s]
th> x:mean(1)

gcr

Little Machines lyrics

Portal – Lights

(sweet guitar solo)
In the throes of the ruby red river that flows through the darkening
concrete devouring ghosts that float by, I'm with them.
Up in the current and taken in whole, I've been struggling,
strangling, others in tow. Have my eyes gone missing?

gcr

#!/usr/bin/env python

from IPython.parallel.client import client
import argparse
import sys
import numpy as np
import datetime
import time

REDBULLET = u"\x1b[1;31m\u25cf\x1b[0m".encode('utf-8')

gcr

------- AlexNet: Using my own weight initialization
model = nn.Sequential()
model:add(cudnn.SpatialConvolution(3,96,11,11,4,4,2,2))
    model.modules[#model.modules].weight:normal(0, 0.01)
    model.modules[#model.modules].bias:fill(0)
model:add(cudnn.ReLU())
model:add(inn.SpatialCrossResponseNormalization(5, 0.0001, 0.75, 1))
model:add(nn.SpatialMaxPooling(3,3,2,2))
model:add(cudnn.SpatialConvolution(96,256,5,5,1,1,2,2))
    model.modules[#model.modules].weight:normal(0, 0.01)

gcr

gcr

import gzip
import re
def load_all_movies(filename):
    """
    Load and parse 'plot.list.gz'. Yields each consecutive movie as a dictionary:
        {"title": "The movie's title",
         "year": The decade of the movie, like 1950 or 1980,
         "identifier": Full key of IMDB's text string,
         "summary": "The movie's plot summary"
        }

gcr

# Based on scitools meshgrid
def meshgrid(*xi, **kwargs):
    """
    Return coordinate matrices from coordinate vectors.

    Make N-D coordinate arrays for vectorized evaluations of
    N-D scalar/vector fields over N-D grids, given
    one-dimensional coordinate arrays x1, x2,..., xn.

    .. versionchanged:: 1.9

gcr

{
 "metadata": {
  "name": "",
  "signature": "sha256:78d15dad8876f550bbf2e9f489381146f98b271e5ec4e609219ad11ddaa39658"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [