ebenolson

#!/bin/bash
cd ~/pydata2015
git pull

source ~/miniconda2/bin/activate py27
# register with landing page
curl --form ip="`curl http://169.254.169.254/latest/meta-data/public-ipv4`" http://pydata.ebenolson.com/addip

# start notebook server
ipython notebook --ip="*"

ebenolson

choose-mirror-bin mirror/http/proxy string
d-i base-installer/kernel/override-image string linux-server
d-i clock-setup/utc boolean true
d-i clock-setup/utc-auto boolean true
d-i finish-install/reboot_in_progress note
d-i grub-installer/only_debian boolean true
d-i grub-installer/with_other_os boolean true
d-i partman-auto-lvm/guided_size string max
d-i partman-auto/choose_recipe select atomic
d-i partman-auto/method string lvm

ebenolson

Subtensor{::, ::, int64:int64:} [@A] ''   
 |InplaceDimShuffle{0,1,3} [@B] ''   
 | |ConvOp{('imshp', (1, 1, 8)),('kshp', (1, 3)),('nkern', 1),('bsize', 1),('dx', 1),('dy', 1),('out_mode', 'full'),('unroll_batch', None),('unroll_kern', None),('unroll_patch', True),('imshp_logical', (1, 1, 8)),('kshp_logical', (1, 3)),('kshp_logical_top_aligned', True)} [@C] ''   
 |   |InplaceDimShuffle{0,1,x,2} [@D] ''   
 |   | |Subtensor{::, ::, int64:int64:} [@E] ''   
 |   |   |InplaceDimShuffle{0,1,3} [@F] ''   
 |   |   | |ConvOp{('imshp', (1, 1, 8)),('kshp', (1, 3)),('nkern', 1),('bsize', 1),('dx', 1),('dy', 1),('out_mode', 'full'),('unroll_batch', None),('unroll_kern', None),('unroll_patch', True),('imshp_logical', (1, 1, 8)),('kshp_logical', (1, 3)),('kshp_logical_top_aligned', True)} [@G] ''   
 |   |   |   |InplaceDimShuffle{0,1,x,2} [@H] ''   
 |   |   |   | |Subtensor{::, ::, int64:int64:} [@I] ''   

ebenolson

import numpy as np
import theano
import theano.tensor as T

X = T.tensor3()
W = theano.shared(np.ones((1,1,1,3)).astype('float32'))
W_shape = (1, 1, 1, 3)

input = X

ebenolson

from lasagne.layers import InputLayer
from lasagne.layers import DenseLayer
from lasagne.layers import ConcatLayer
from lasagne.layers import NonlinearityLayer
from lasagne.layers import GlobalPoolLayer
from lasagne.layers.dnn import Conv2DDNNLayer as ConvLayer
from lasagne.layers.dnn import MaxPool2DDNNLayer as PoolLayerDNN
from lasagne.layers import MaxPool2DLayer as PoolLayer
from lasagne.layers import LocalResponseNormalization2DLayer as LRNLayer
from lasagne.nonlinearities import softmax, linear

ebenolson

import numpy as np
import gc
import memory_profiler

import theano
import theano.tensor as T
from theano.sandbox.cuda.dnn import dnn_conv

X = T.tensor4()
W = T.tensor4()

ebenolson

name: "CIFAR10_full"
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 32
  dim: 32
}
layers {
  name: "conv1"

ebenolson

require 'optim'
cj = require('cjson')

a = torch.Tensor({0.1,0.2,0.3})
x0 = torch.Tensor({1, 1, 1})

function f(x) 
    return torch.sum(torch.cmul(a, torch.pow(x, 2)))
end

ebenolson

# Usage: experiment.sh SOURCE_FILE LOG_COMMIT_MESSAGE

if [ $# -eq 0 ]
    then
        echo "Usage: experiment.sh SOURCE_FILE [LOG_COMMIT_MESSAGE]"
        exit -1
fi

if [ $# -eq 2 ]
    then

ebenolson

from lasagne.layers import Layer

class HighwayLayer(Layer):
    def __init__(self, incoming, layer_class, gate_nonlinearity=None,
                 **kwargs):
        super(HighwayLayer, self).__init__(incoming)

        self.H_layer = layer_class(incoming, **kwargs)
        
        if gate_nonlinearity: