| """ | |
| Sieve of Eratosthenes (https://en.wikipedia.org/wiki/Sieve_of_Eratosthenes) using list comprehensions | |
| """ | |
| upper_bound = 100 | |
| noprimes = {j for i in range(2,8) for j in range(i*2, upper_bound, i)} | |
| primes = [x for x in range(2, upper_bound) if x not in noprimes] |
| """ | |
| Inspired from [1] and [2], but neither fit my need. | |
| The following lets you decorate a function with a time limit, kills it when it doesn't finish in that time. | |
| The needed part for me was that I had to get the result if it didn't get killed and None if it did. | |
| The following accomplished this. | |
| [1] http://stackoverflow.com/a/26664130 | |
| [2] https://gist.github.com/aaronchall/6331661fe0185c30a0b4 |
| import numpy as np | |
| import theano.tensor as T | |
| pred = np.arange(30, dtype=np.float32).reshape(2,3,5) | |
| pred /= pred.sum(axis=-1, keepdims=True) | |
| ### target matrix to match pred | |
| target1 = np.ones_like(pred) | |
| target1[:,:,:-1] = 0 |
| from __future__ import print_function | |
| from keras.layers import Layer, Input, Embedding, RepeatVector, Flatten, Dense | |
| from keras.layers import TimeDistributed as Distribute | |
| from keras.activations import softmax | |
| from keras.engine import merge, InputSpec | |
| import keras.backend as K | |
| class ProbabilityTensor(Layer): | |
| """ function for turning 3d tensor to 2d probability matrix """ |
| from __future__ import print_function | |
| from keras.layers import Input, Embedding, LSTM, Activation | |
| from keras.layers import TimeDistributed as Distribute | |
| import numpy as np | |
| def outmask(x): | |
| return x._keras_history[0].outbound_nodes[0].output_masks[0] | |
| def inmask(x): | |
| return x._keras_history[0].inbound_nodes[0].output_masks[0] |
| B = self.igor.batch_size | |
| R = self.igor.rnn_size | |
| S = self.igor.max_sequence_len | |
| V = self.igor.vocab_size | |
| E = self.igor.embedding_size | |
| ### loaded from glove | |
| emb_W = self.igor.embeddings.astype(theano.config.floatX) | |
| ## dropout parameters | |
| p_emb = self.igor.p_emb_dropout |
| from __future__ import print_function | |
| from keras.layers import Dense, TimeDistributed | |
| import numpy as np | |
| import theano | |
| def shared(x): | |
| x_ = theano.shared(x) | |
| x_._keras_shape = x.shape | |
| return x_ | |
| dense_func = Dense(1) |
##VGG16 model for Keras
This is the Keras model of the 16-layer network used by the VGG team in the ILSVRC-2014 competition.
It has been obtained by directly converting the Caffe model provived by the authors.
Details about the network architecture can be found in the following arXiv paper:
Very Deep Convolutional Networks for Large-Scale Image Recognition
K. Simonyan, A. Zisserman
| from __future__ import print_function, division | |
| import yaml | |
| import time | |
| import sys | |
| import numpy as np | |
| import itertools | |
| from keras.utils.np_utils import to_categorical | |
| from baal.utils import loggers | |
| try: |
| from __future__ import print_function | |
| import theano | |
| import theano.tensor as T | |
| import numpy as np | |
| import random | |
| A = np.arange(1,31).reshape(5,6).astype(np.float32) | |
| sizes = np.zeros(5, dtype='int16') | |
| for i in range(5): | |
| sizes[i] = random.randint(1,5) |