PyTorch v0.2.0 released (opens in new tab)

I LOVE PyTorch for experimenting with dynamic deep neural nets (DNNs) -- that is, DNNs that can have different graphs for different input samples. I find it much, MUCH easier to create and tinker with dynamic DNNs using PyTorch than, say, TensorFlow Fold. PyTorch is great for R&D experimentation.

For example, here's how easy it is to construct a fully-connected neural net with a dynamically random number of recurrent hidden layers in PyTorch. Yes, it's a silly example, but it shows how easy it is to construct dynamic DNNs with PyTorch:

  import random
  import torch
  
  class MySillyDNN(torch.nn.Module):
      def __init__(self, input_dim, hidden_dim, output_dim):
          super(MySillyDNN, self).__init__()
          self.input_layer = torch.nn.Linear(input_dim, hidden_dim)
          self.hidden_layer = torch.nn.Linear(hidden_dim, hidden_dim)
          self.output_layer = torch.nn.Linear(hidden_dim, output_dim)

      def forward(self, x, max_recurrences=3):
          hidden_relu = self.input_layer(x).clamp(min=0)
          for r in range(random.randint(0, max_recurrences)):
              hidden_relu = self.hidden_layer(hidden_relu).clamp(min=0)
          y_pred = self.output_layer(hidden_relu)
          return y_pred

It would be a hassle to do something like this with other frameworks like TensorFlow or Theano, which require you to specify the computational graph (including conditionals, if any) before you can run the graph.

PyTorch's define-the-graph-by-running-it approach is sooo nice for quick-n'-dirty experimentation with dynamic graphs.

You can even create and tinker with dynamic graphs interactively on a Python REPL :-)

For those of us who don't know what this is:

PyTorch is a Python package that provides two high-level features:

Tensor computation (like NumPy) with strong GPU acceleration

Deep neural networks built on a tape-based autograd system

You can reuse your favorite Python packages such as NumPy, SciPy and Cython to extend PyTorch when needed.

We are in an early-release beta. Expect some adventures and rough edges.

There was a great podcast with Soumith Chintala on the O'Reilly data show a couple of days back with more info on PyTorch and how it differs from Theano and Tensorflow:

https://www.oreilly.com/ideas/why-ai-and-machine-learning-re...

Also of note is that the lua version of Torch seems to be in maintenance mode now.

https://twitter.com/soumithchintala/status/89417353800895692...

it is my favorite neural network library as well. Tensorflow is very heavy for testing.