Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/torch/nngraph.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-02-13include and paths.dofile -> requirerequireSoumith Chintala
2016-01-29Merge pull request #99 from malcolmreynolds/safer_copynestedkoray kavukcuoglu
Safer copynested
2016-01-29avoid string concatenation error when key type is not string or numberMalcolm Reynolds
2016-01-29Change copyNested to error when there are extra keys in output.Malcolm Reynolds
2016-01-25Merge pull request #98 from malcolmreynolds/remove_zeroing_optimisationkoray kavukcuoglu
Don't bother filling a Tensor with zero right before we copy into it
2016-01-25Merge pull request #97 from malcolmreynolds/detect_dangling_connectionskoray kavukcuoglu
Store a reverse mapping when wiring together graph, detect unused nodes.
2016-01-25Don't bother filling a Tensor with zero right before we copy into itMalcolm Reynolds
2016-01-20Store a reverse mapping when wiring together graph, detect unused nodes.Malcolm Reynolds
The connectivity checking code was previously unable to detect the following error case: local input = nn.Identity()() local usedOutput = nn.Linear(20, 10)(input) local unusedOutput = nn.Linear(20, 10)(input) local gmod = nn.gModule({input}, {usedOutput}) With this fix, when gModule is called it will throw an error, because of unusedOutput. This is a backwards incompatible change, but I feel that the current flexibility is error prone, and I can't see any advantage to it. We have flushed out a couple of bugs in internal code with this change.
2015-12-19Merge pull request #96 from fidlej/topic_split_oneSoumith Chintala
Allow to do node:split(1).
2015-12-18Allow to do node:split(1).Ivo Danihelka
2015-11-24Merge pull request #94 from LeeTaewoo/patch-1koray kavukcuoglu
Minor typo
2015-11-24Minor typoTaewoo Lee
2015-11-22Merge pull request #93 from malcolmreynolds/improve_error_messageskoray kavukcuoglu
Make error messages clearer, disallow empty table in inputs.
2015-11-20Fix for Lua 5.2+ which removed table.maxnMalcolm Reynolds
2015-11-19Make error messages clearer, disallow empty table in inputs.Malcolm Reynolds
This is intended to address the common class of errors I see where people make a mistake connecting up their modules, but the error message is either unclear, or doesn't point towards where the mistake actually is. The 'what is this in the input' is now explicit about what the problem is, and if people pass in a nn.Module (meaning they probably forgot a set of parentheses) instead of a nngraph.Node, we say this explicitly. The '1 of split(2) outputs unused' (which previously provided no information about which split was incorrect) now includes file / line number of both the place where the Node was constructed, and the place where :split() was called. Hopefully this should reduce debugging time drastically. Finally, I have disallow passing an empty table as the input connections, ie 'nn.Identity()({})' will error. I cannot see a use case for this (if you have no input connections, just leave the second parens empty). The risk of this is when people do 'nn.Identity()({variableWithTypo})', thinking they have made a connection but actually they haven't. This is likely to cause errors much later on, whereas with this commit it errors straight away. This *could* break existing code, but theres an easy to apply fix that needs to be done at each callsite. Koray has approved this restriction to the API, but I appreciate others may have a view here..
2015-11-12travis fixessoumith
2015-11-12adding graphviz LD path to travissoumith
2015-11-10fix path in travissoumith
2015-11-10contbuild for Lua51, Lua52soumith
2015-11-10install manual version of graphviz in travissoumith
2015-11-10adding graphviz to travisSoumith Chintala
2015-11-04adding Lua52 compatibility fixesSoumith Chintala
2015-10-26kick contbuildSoumith Chintala
2015-10-26travis contbuildSoumith Chintala
2015-10-19Merge pull request #90 from fidlej/topic_nestkoray kavukcuoglu
Add nngraph.nest() utility function.
2015-10-19Add nngraph.nest() utility function.Ivo Danihelka
2015-10-16Merge pull request #88 from dominikgrewe/gmodule_modulesSoumith Chintala
Initialize modules table after read if necessary.
2015-10-16Initialize modules table after read if necessary.Dominik Grewe
2015-10-16Merge pull request #83 from andreaskoepf/container_baseSoumith Chintala
Use nn.Container as base class for gModule
2015-10-16Use nn.Container as base class for gModuleAndreas Köpf
Added modules to container in ctor, removed redundant methods training(), evaluate(), share(), zeroGradParameters(), parameters(), clone() which are now provided by the base classes (nn.gModule -> nn.Container -> nn.Module).
2015-10-14Merge pull request #86 from yozw/masterkoray kavukcuoglu
Use full urls for including images
2015-10-14Merge pull request #87 from yozw/applykoray kavukcuoglu
Adding a :applyToModules() method to gModule, and :training() and :evaluate() should be applied 'self' as well.
2015-10-14Adding a :applyToModules() method to gModule, and :training() and ↵Yori Zwols
:evaluate() should be applied 'self' as well.
2015-10-14Use full urls for including imagesYori Zwols
2015-10-01Merge pull request #56 from adamlerer/type_applySoumith Chintala
Integrate apply() and type() improvements from https://github.com/torch/nn/pull/303
2015-10-01Integrate apply() and type() improvements from ↵Adam Lerer
https://github.com/torch/nn/pull/303
2015-09-11Merge pull request #81 from torch/nnopkoray kavukcuoglu
support for parameter nodes
2015-09-11make sure forward/backward runs can deal with parameter nodes since theynnopKoray Kavukcuoglu
do not have any inputs coming in. add a display function that does not use qt, but browser
2015-09-11support for parameter nodeskoray kavukcuoglu
2015-09-07Merge pull request #79 from LinusU/patch-1Soumith Chintala
tell git to ignore build output
2015-09-07Merge pull request #80 from d11/isTensorkoray kavukcuoglu
Replace utils.istensor with torch.isTensor.
2015-09-07Replace utils.istensor with torch.isTensor.Dan Horgan
torch.isTensor is more precise.
2015-09-07tell git to ignore build outputLinus Unnebäck
2015-09-04Merge pull request #77 from clementfarabet/masterSoumith Chintala
Whitespace cleanup.
2015-09-04Whitespace cleanup.Clement Farabet
2015-07-23Merge pull request #64 from torch/svg_namekoray kavukcuoglu
add randomized naming to svg file names that are auto generated
2015-07-23add randomized naming to svg file names that are auto generatedsvg_namekoray kavukcuoglu
2015-07-23Merge pull request #63 from fidlej/topic_check_split_inputsSoumith Chintala
Added an assert to check the number of inputs to a split.
2015-07-23Added an assert to check the number of inputs to a split.Ivo Danihelka
2015-07-21Merge pull request #62 from hughperkins/fix-gradOutputBuffer-type-changeSoumith Chintala
"nesting.lua:36: bad argument #1 to 'resizeAs'"