5 ms·

Brainstorm – Deep Learning library, successor to PyBrain

78 points by albertzeyer 11y ago

albertzeyer 11y ago
Schmidhuber already kind of announced this here: https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_am_j%C3%BCrgen_schmidhuber_ama/cp433y3 https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_a...
ilyaeck 11y ago
How does it compare to other similar libraries (Caffe, Theano, Groundhog, etc.)?
- aashu_dwivedi 11y ago
  why successor to pybrain ? is it developed by the same devs ?
  - xtacy 11y ago
    Yes, that is correct. Here's the linked reddit post that confirms it: https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_am_j%C3%BCrgen_schmidhuber_ama/cp433y3 https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_a...
- albertzeyer 11y ago
  Theano (Python) is a symbolic mathematical framework which can do symbolic automatic differentiation - not more and not less. This is a very useful base for all kinds of NNs but also for any mathematical models. This will give you a computation graph which is then also optimized by Theano and which can be evaluated on several different backends, e.g. CUDA, i.e. it can be calculated on the GPU. This is also important for common NNs for performance. Torch (Lua) is similar, but is not symbolic, and when you want to calculate the gradient, it will do the backpropagation through the same graph, just backwards. It's like Theano only a mathematical framework, which is very useful for all kinds of NNs but also any other mathematical models. There are many libs based on Theano, e.g. like Groundhog, Keras, Lasagne, etc. There are also Torch based NN utils. You usually code your model directly in Python / Lua. Thus that is very flexible. Caffe is a C++ framework with CUDA support. You describe your models by some declarative config. It's thus much less flexible. PyBrain (Python) is similar to Groundhog etc, but not based on Theano but on pure Python / Numpy code. Brainstorm (Python) is similar to PyBrain, and also not based on Theano but is has several own custom backends, including CUDA.
  - ilurk 11y ago
    In your opinion, would there be any advantage in using Brainstorm over Theano or even Torch?
    - p1esk 11y ago
      What do you want to do?
scoot 11y ago
Semi related question (on the basis that this post will attract folks with the relevant background to answer) - how would someone with an interest in the possibilities afforded by the field of machine learning, but without the theoretical or mathematical background tneeded to understand the typical discussion on these topics, get started using it in a practical way? Is the a "machine learning for dummies" resource you would recommend?
- ivan_ah 11y ago
  Well you'll certainly need to learn about the math behind these eventually, but you can get started and go quite far simply as a user of a ML library. You can use "systems thinking" and think of ML components as black boxes that do something for you. A good place to start would be the scikit learn tutorials on youtube: https://www.youtube.com/watch?v=r4bRUvvlaBw https://www.youtube.com/watch?v=r4bRUvvlaBw I remember going through them and thinking---wow, this covers everything from my ML class in just a few iPython notebooks... I recommend you start with the basics (datasets, features, training/test/validation data splits) and don't worry too much about the actual choice of model---there will always be shiny new models with better performance but sometimes using the "old stuff" is good enough. Once you get past the basics want to learn the theory, you can take an online course or find a good book, e.g. https://www.cs.ubc.ca/~murphyk/MLbook/ https://www.cs.ubc.ca/~murphyk/MLbook/ (advanced, but very comprehensive).
  - chriswarbo 11y ago
    > I recommend you start with the basics (datasets, features, training/test/validation data splits) and don't worry too much about the actual choice of model---there will always be shiny new models with better performance but sometimes using the "old stuff" is good enough. Or, conversely, it may be that all models are just as bad. This seems to be the case in my domain (formal proofs), where the bottleneck seems to be data representation; it doesn't matter which learning algorithm you use, when your feature selection has stripped out all of the learnable information ;)
- lachenmayer 11y ago
  http://www.metacademy.org/ http://www.metacademy.org/ is a fantastic resource.

78 points by albertzeyer 11y ago

Schmidhuber already kind of announced this here: https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_am_j%C3%BCrgen_schmidhuber_ama/cp433y3 https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_a...

ilyaeck 11y ago

How does it compare to other similar libraries (Caffe, Theano, Groundhog, etc.)?

aashu_dwivedi 11y ago
why successor to pybrain ? is it developed by the same devs ?
- xtacy 11y ago
  Yes, that is correct. Here's the linked reddit post that confirms it: https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_am_j%C3%BCrgen_schmidhuber_ama/cp433y3 https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_a...
albertzeyer 11y ago
Theano (Python) is a symbolic mathematical framework which can do symbolic automatic differentiation - not more and not less. This is a very useful base for all kinds of NNs but also for any mathematical models. This will give you a computation graph which is then also optimized by Theano and which can be evaluated on several different backends, e.g. CUDA, i.e. it can be calculated on the GPU. This is also important for common NNs for performance. Torch (Lua) is similar, but is not symbolic, and when you want to calculate the gradient, it will do the backpropagation through the same graph, just backwards. It's like Theano only a mathematical framework, which is very useful for all kinds of NNs but also any other mathematical models. There are many libs based on Theano, e.g. like Groundhog, Keras, Lasagne, etc. There are also Torch based NN utils. You usually code your model directly in Python / Lua. Thus that is very flexible. Caffe is a C++ framework with CUDA support. You describe your models by some declarative config. It's thus much less flexible. PyBrain (Python) is similar to Groundhog etc, but not based on Theano but on pure Python / Numpy code. Brainstorm (Python) is similar to PyBrain, and also not based on Theano but is has several own custom backends, including CUDA.
- ilurk 11y ago
  In your opinion, would there be any advantage in using Brainstorm over Theano or even Torch?
  - p1esk 11y ago
    What do you want to do?

scoot 11y ago

Semi related question (on the basis that this post will attract folks with the relevant background to answer) - how would someone with an interest in the possibilities afforded by the field of machine learning, but without the theoretical or mathematical background tneeded to understand the typical discussion on these topics, get started using it in a practical way? Is the a "machine learning for dummies" resource you would recommend?

ivan_ah 11y ago
Well you'll certainly need to learn about the math behind these eventually, but you can get started and go quite far simply as a user of a ML library. You can use "systems thinking" and think of ML components as black boxes that do something for you. A good place to start would be the scikit learn tutorials on youtube: https://www.youtube.com/watch?v=r4bRUvvlaBw https://www.youtube.com/watch?v=r4bRUvvlaBw I remember going through them and thinking---wow, this covers everything from my ML class in just a few iPython notebooks... I recommend you start with the basics (datasets, features, training/test/validation data splits) and don't worry too much about the actual choice of model---there will always be shiny new models with better performance but sometimes using the "old stuff" is good enough. Once you get past the basics want to learn the theory, you can take an online course or find a good book, e.g. https://www.cs.ubc.ca/~murphyk/MLbook/ https://www.cs.ubc.ca/~murphyk/MLbook/ (advanced, but very comprehensive).
- chriswarbo 11y ago
  > I recommend you start with the basics (datasets, features, training/test/validation data splits) and don't worry too much about the actual choice of model---there will always be shiny new models with better performance but sometimes using the "old stuff" is good enough. Or, conversely, it may be that all models are just as bad. This seems to be the case in my domain (formal proofs), where the bottleneck seems to be data representation; it doesn't matter which learning algorithm you use, when your feature selection has stripped out all of the learnable information ;)
lachenmayer 11y ago
http://www.metacademy.org/ http://www.metacademy.org/ is a fantastic resource.