jiqiujia

[{"place_id":"97994878","licence":"Data \u00a9 OpenStreetMap contributors, ODbL 1.0. http:\/\/www.openstreetmap.org\/copyright","osm_type":"relation","osm_id":"161950","boundingbox":["30.1375217437744","35.0080299377441","-88.4731369018555","-84.8882446289062"],"lat":"33.2588817","lon":"-86.8295337","display_name":"Alabama, United States of America","place_rank":"8","category":"boundary","type":"administrative","importance":0.83507032450272,"icon":"http:\/\/nominatim.openstreetmap.org\/images\/mapicons\/poi_boundary_administrative.p.20.png"}]
[{"place_id":"97421560","licence":"Data \u00a9 OpenStreetMap contributors, ODbL 1.0. http:\/\/www.openstreetmap.org\/copyright","osm_type":"relation","osm_id":"162018","boundingbox":["31.3321762084961","37.0042610168457","-114.818359375","-109.045196533203"],"lat":"34.395342","lon":"-111.7632755","display_name":"Arizona, United States of America","place_rank":"8","category":"boundary","type":"administrative","importance":0.83922181098242,"icon":"http:\/\/nominatim.openst

jiqiujia

{
    "Mississippi": [30.1477890014648, 34.9960556030273, -91.6550140380859, -88.0980072021484],
    "Oklahoma": [33.6191940307617, 37.0021362304688, -103.002571105957, -94.4312133789062],
    "Delaware": [38.4511260986328, 39.8394355773926, -75.7890472412109, -74.9846343994141],
    "Minnesota": [43.4994277954102, 49.3844909667969, -97.2392654418945, -89.4833831787109],
    "Illinois": [36.9701309204102, 42.5083045959473, -91.513053894043, -87.0199203491211],
    "Arkansas": [33.0041046142578, 36.4996032714844, -94.6178131103516, -89.6422424316406],
    "New Mexico": [31.3323001861572, 37.0001411437988, -109.050178527832, -103.000862121582],
    "Indiana": [37.7717399597168, 41.7613716125488, -88.0997085571289, -84.7845764160156],
    "Louisiana": [28.9210300445557, 33.019458770752, -94.0431518554688, -88.817008972168],

jiqiujia

"""
Minimal character-level Vanilla RNN model. Written by Andrej Karpathy (@karpathy)
BSD License
"""
import numpy as np

# data I/O
data = open('input.txt', 'r').read() # should be simple plain text file
chars = list(set(data))
data_size, vocab_size = len(data), len(chars)

jiqiujia

""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym

# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward

jiqiujia

%%reference: http://www.mathworks.com/help/signal/ref/spectrogram.html

dat = load('/media/dl/data1/kaggle_eeg_2016/train_1/1_1_1.mat');
dat = dat.dataStruct.data;
dat = dat(:, 1);

Nx = length(dat); %sample number
nsc = floor(Nx/600); %number of signal sections
nov = floor(nsc/2); %overlap between adajacent windows
nff = max(256,2^nextpow2(nsc)); %number of samples to compute the FFT

jiqiujia

def architecture_string(layer):
    model_arch = ''

    for i, layer in enumerate(lasagne.layers.get_all_layers(layer)):
        name = string.ljust(layer.__class__.__name__, 28)
        model_arch += "  %2i  %s %s  " % (i, name,
                                          lasagne.layers.get_output_shape(layer))

        if hasattr(layer, 'filter_size'):
            model_arch += str(layer.filter_size[0])

jiqiujia

class BatchNormLayer(lasagne.layers.Layer):
    def __init__(self, incoming, b=lasagne.init.Constant(0.), g=lasagne.init.Constant(1.), 
                 W=lasagne.init.Normal(0.05), nonlinearity=relu, **kwargs):
        super(BatchNormLayer, self).__init__(incoming, **kwargs)
        self.nonlinearity = nonlinearity
        k = self.input_shape[1]
        if b is not None:
            self.b = self.add_param(b, (k,), name="b", regularizable=False)
        if g is not None:
            self.g = self.add_param(g, (k,), name="g")

jiqiujia

def create_conf_matrix(expected, predicted, n_classes):
    m = np.zeros((n_classes, n_classes))
    for pred, exp in zip(predicted, expected):
        m[pred][exp] += 1
    return m

def calc_accuracy(conf_matrix):
    t = sum(sum(l) for l in conf_matrix)
    return sum(conf_matrix[i][i] for i in range(len(conf_matrix))) / t

jiqiujia

-- Two dashes start a one-line comment.

--[[
     Adding two ['s and ]'s makes it a
     multi-line comment.
--]]

----------------------------------------------------
-- 1. Variables and flow control.
----------------------------------------------------

jiqiujia

from math import ceil
import numpy as np
from sklearn.cluster import KMeans

import theano
import theano.tensor as T

def compute_reps(extract_fn, X, chunk_size):
    """Compute representations for input in chunks."""
    chunks = int(ceil(float(X.shape[0]) / chunk_size))