Data Science Stack Exchange
Data Science Stack Exchange

Questions tagged [theano]

Theano is a numerical computation library for Python. Computations are expressed using a Numpy-like syntax and compiled to run efficiently on either CPU or GPU architectures. Theano uses a mathematical computational graph and optimized the graph to speed up the computations using optimized C code. As of release 1.0, Theano will no longer be actively developed.

Theano is a Python library for performing tensor calculations with a numpy-like syntax. Theano can be used to define, optimize, and evaluate mathematical expressions involving tensors (also see tensorflow). Some of the principal features of Theano that set it apart from other libraries have become commonplace in neural network and tensor math libraries:

  • Integrates with Numpy and provides Numpy-like syntax to express models using mathematical syntax
  • Represents calculations as computation graphs and optimizes graphs for better performance and memory utilization
  • Utilizes GPU hardware for data-intensive calculations
  • Performs high-order differentiation automatically (symbolically)
  • Compiles computation graphs into C code for faster execution

Theano enables rapid building of large-scale computational codes with less cognitive overhead and more expressive syntax. However, it is easy enough to learn that it is used in several deep learning courses (including at the University of Montreal, where it was originally developed).

Several projects build on top of Theano, including keras and lasagne.

As of Theano 1.0, Theano will no longer be under active development. See announcement on mailing list.

Resources

Trivia

Theano is named after a high priestess of Athena in the city of Troy, from Homer's Iliad.

45 questions
59
votes
3 answers

How to set batch_size, steps_per epoch, and validation steps?

I am starting to learn CNNs using Keras. I am using the theano backend. I don't understand how to set values to: batch_size steps_per_epoch validation_steps What should be the value set to batch_size, steps_per_epoch, and validation_steps, if I…
Ermene
  • 693
  • 1
  • 6
  • 6
39
votes
7 answers

Using TensorFlow with Intel GPU

Is there any way now to use TensorFlow with Intel GPUs? If yes, please point me in the right direction. If not, please let me know which framework, if any, (Keras, Theano, etc) can I use for my Intel Corporation Xeon E3-1200 v3/4th Gen Core…
James Bond
  • 1,155
  • 2
  • 11
  • 12
24
votes
2 answers

Choosing between TensorFlow or Theano as backend for Keras

Keras supports both TensorFlow and Theano as backend: what are the pros/cons of choosing one versus the other, besides the fact that currently not all operations are implemented with the TensorFlow backend?
Franck Dernoncourt
  • 5,573
  • 9
  • 40
  • 75
14
votes
1 answer

Make Keras run on multi-machine multi-core cpu system

I'm working on Seq2Seq model using LSTM from Keras (using Theano background) and I would like to parallelize the processes, because even few MBs of data need several hours for training. It is clear that GPUs are far much better in parallelization…
chmodsss
  • 1,954
  • 2
  • 17
  • 37
8
votes
3 answers

Convolutional Neural Network not learning EEG data

I have trained a simple CNN (using Python + Lasagne) for a 2-class EEG classification problem, however, the network doesn't seem to learn. loss does not drop over epochs and classification accuracy doesn't drop from random guessing…
Simon
  • 1,071
  • 2
  • 10
  • 28
8
votes
1 answer

Why doesn't training RNNs use 100% of the GPU?

I wonder why training RNNs typically doesn't use 100% of the GPU. For example, if I run this RNN benchmark on a Maxwell Titan X on Ubuntu 14.04.4 LTS x64, the GPU utilization is below 90%: The benchmark was launched using the command: python rnn.py…
Franck Dernoncourt
  • 5,573
  • 9
  • 40
  • 75
6
votes
2 answers

Error on multitask neural nets where all outputs not observed for every example

Let's say I have 2 datasets, each from a set of experiments. Dataset A measures a set of properties X for set S, while dataset B measures properties Y for set T. X and Y are highly correlated, and S and T have (not perfect) overlap. To give an…
jamesmf
  • 3,077
  • 1
  • 16
  • 24
6
votes
1 answer

Keras: How to normalize dataframe with continuous and categorical data?

I have a dataframe with about 50 columns. The columns are either categorical or continuous data. The continuous data can be between 0.000001-1.00000 or they can be between 500,000-5,000,000. The categorical data is usually a name, for example a…
user1367204
  • 201
  • 1
  • 3
  • 6
6
votes
1 answer

Recurrent Neural Network on Panel Data

There are 2 parts to this question. Suppose we are looking at sales $S$ of a product across $> 1000$ stores where a it sells. For each of these $1000$ stores we have 24 months recorded data. We want to be able to predict $S_t \leftarrow…
Sandipan Bhattacharyya
  • 73
  • 1
  • 5
6
votes
3 answers

Stuck on deconvolution in Theano and TensorFlow

I'm captivated by autoencoders and really like the idea of convolution. It seems though that both Theano and TensorFlow only support conv2d to go from an array of 2D-RGB (n 3D arrays) to an array of higher-depth images. That makes sense from the…
Joseph Catrambone
  • 193
  • 1
  • 8
5
votes
1 answer

What does Theano dimension ordering mean?

In this code , line 13 is commented as Theano dimension ordering mode. What does this mean?
pseudomonas
  • 1,032
  • 3
  • 13
  • 30
4
votes
4 answers

Library to count number of objects in image?

I'm reading a paper on counting cells, humans, etc. Is there an off-the-shelf library for Python/Theano or TensorFlow I could use?
FullStack
  • 267
  • 1
  • 4
  • 9
4
votes
3 answers

Training Restricted Boltzmann Machines (RBMs) using gradient descent

Hey I am a little new to the whole RBM entropy/energy training thing, having some trouble understanding and trying to figure whether it is worth the effort needed to understand. Can't RBMs quite easily be trained with a standard gradient descent…
user18886
  • 41
  • 2
4
votes
1 answer

Benchmarking Theano

I sometimes want to benchmark Theano, either to compare different versions of Theano, or to compare two different computing environment with the same version of Theano. Is there any Theano code I could use to benchmark the main functions of Theano?…
Franck Dernoncourt
  • 5,573
  • 9
  • 40
  • 75
3
votes
2 answers

Can PyLearn do everything that Theano can?

Since PyLearn2 is build upon Theano, is it possible to do anything I can do in Theano in Pylearn2? For example, if I have some snippets of Theano code, can I run them as-is in Pylearn2, or would this not work? If it wouldn't, why not?
robert
  • 131
  • 1
1
2 3