Adam Paszke
30be715900
Add training and evaluation to torch.nn
2016-10-24 22:29:43 +02:00
Sam Gross
98f67e90d5
Fix super call in Container.modules and Container.parameters ( #142 )
2016-10-19 13:21:03 -04:00
Sam Gross
fee67c2e1a
Allow parameters and child modules to be assigned by attribute ( #136 )
...
For example:
self.linear = nn.Linear(10, 20)
self.weight = torch.autograd.Variable(torch.Tensor(10, 20))
2016-10-18 23:34:20 +02:00
Soumith Chintala
e46e05e7c5
fix container doc
2016-10-05 14:53:41 -04:00
Adam Paszke
80e16e44aa
Check container source on load
2016-10-04 17:41:12 -07:00
Adam Paszke
6efefac2df
Add parameter_dict and load_parameter_dict methods for modules
2016-10-04 14:47:56 -07:00
Sam Gross
f4ebc65a12
Add Module.modules() and Module.children() ( #90 )
...
modules(): returns an iterator over all modules in the network
children(): returns an iterator over immediate children
Also fix __getitem__ in Sequential
2016-10-01 21:18:53 -04:00
Sam Gross
ea4f812a12
Fix Container.parameters()
2016-09-30 16:31:36 -07:00
Sam Gross
f5a6a3b0e9
Fix torch.nn.Module._apply with None types ( #66 )
2016-09-28 19:31:07 -04:00
Adam Paszke
7f4ff0e615
Fix type conversions in nn
2016-09-27 15:45:49 -07:00
Adam Paszke
f9d25e8e72
Refactor nn (require specifying parameters explicitly)
2016-09-27 15:22:26 -07:00
Sam Gross
44481354fc
Add back support for child=None in Container constructor ( #55 )
...
It's often useful to have optional child modules, such as the
downsampling operation in ResNets. Add a test for this case:
nn.Container(
child=None,
)
2016-09-26 17:18:02 -04:00
Adam Paszke
4cdeae3283
Return only unique variables from parameters()
2016-09-25 12:23:43 -07:00
Soumith Chintala
b5f7720ab9
docstrings for container and batchnorm
2016-09-16 05:31:36 -04:00
Sam Gross
b738b09606
Clean up Module forward and __call__ ( #14 )
...
* _forward is renamed forward since users should override it
* some __call__ overrides are changed to forward
* function which return a single variable are changed to return that
variable instead of a one-element tuple
2016-09-07 15:41:39 -04:00
Sam Gross
f45213a276
Simplify nn.Container and nn.Sequential
...
- nn.Container.modules is just a python list and used by nn.Sequential
- Every module in nn.Sequential has a name. This fixes Module.type()
- nn.Sequential constructor accepts either a list or an OrderedDict. With a
list, the modules are named "0", "1", "2", ...
2016-08-31 11:15:31 -07:00
Adam Paszke
ea93fb7ac0
Add more nn modules
2016-08-23 19:15:21 -07:00
Adam Paszke
7bcb2a4081
Initial optim version
2016-08-23 19:03:30 -07:00
Adam Paszke
2bf68e72d5
Add hook system to autograd and nn
2016-08-23 13:51:34 -07:00
Adam Paszke
e055ffbdc7
Add nn
2016-08-19 14:56:55 -07:00
