andrewssobral

'''This script goes along the blog post
"Building powerful image classification models using very little data"
from blog.keras.io.

It uses data that can be downloaded at:
https://www.kaggle.com/c/dogs-vs-cats/data

In our setup, we:
- created a data/ folder
- created train/ and validation/ subfolders inside data/

andrewssobral

'''This script goes along the blog post
"Building powerful image classification models using very little data"
from blog.keras.io.

It uses data that can be downloaded at:
https://www.kaggle.com/c/dogs-vs-cats/data

In our setup, we:
- created a data/ folder
- created train/ and validation/ subfolders inside data/

andrewssobral

'''This script goes along the blog post
"Building powerful image classification models using very little data"
from blog.keras.io.

It uses data that can be downloaded at:
https://www.kaggle.com/c/dogs-vs-cats/data

In our setup, we:
- created a data/ folder
- created train/ and validation/ subfolders inside data/

andrewssobral

'''Functional Keras is a more functional replacement for the Graph API.
'''

###################
# 2 LSTM branches #
###################
a = Input(input_shape=(10, 32))  # output is a TF/TH placeholder, augmented with Keras attributes
b = Input(input_shape=(10, 32))
encoded_a = LSTM(32)(a)  # output is a TF/TH tensor
encoded_b = LSTM(32)(b)

andrewssobral

import theano
from keras.models import Sequential
from keras.layers.core import Dense, Activation

X_train, y_train = ... # load some training data
X_batch = ... # a batch of test data

# this is your initial model
model = Sequential()
model.add(Dense(20, 64))

andrewssobral

from keras.models import Sequential
from keras.layers import Dense

x, y = ...
x_val, y_val = ...

# 1-dimensional MSE linear regression in Keras
model = Sequential()
model.add(Dense(1, input_dim=x.shape[1]))
model.compile(optimizer='rmsprop', loss='mse')

andrewssobral

-- Xception model
-- a Torch7 implementation of: https://arxiv.org/abs/1610.02357
-- E. Culurciello, October 2016

require 'nn'
local nClasses = 1000

function nn.SpatialSeparableConvolution(nInputPlane, nOutputPlane, kW, kH)
   local block = nn.Sequential()
   block:add(nn.SpatialConvolutionMap(nn.tables.oneToOne(nInputPlane), kW,kH, 1,1, 1,1))

andrewssobral

import numpy as np


def low_rank_approx(SVD=None, A=None, r=1):
    """
    Computes an r-rank approximation of a matrix
    given the component u, s, and v of it's SVD

    Requires: numpy
    """

andrewssobral

"""
Low rank approximation for the lena image
"""
import numpy as np
import scipy as sp
from scipy import linalg
import pylab as pl

X = sp.lena().astype(np.float)
pl.gray()

andrewssobral

from __future__ import division
from numpy import *

class AdaBoost:

    def __init__(self, training_set):
        self.training_set = training_set
        self.N = len(self.training_set)
        self.weights = ones(self.N)/self.N
        self.RULES = []