twitter-archive torch-autograd issues

twitter-archive / torch-autograd

Autograd automatically differentiates native Torch code

Apache License 2.0

560 stars 115 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Update LICENSE to use full text

#180 juliaferraioli closed 3 years ago
0
docs: fix typo

#179 Jason-Cooke opened 5 years ago
1
Failed to parse generated code when optimize=true

#178 farleylai opened 5 years ago
0
FloatTensor.cdata not currently supported by autograd when implementing SSIM loss

#177 farleylai closed 5 years ago
1
gradfuns.lua dimension problem on line 130

#176 HeylenJonas opened 7 years ago
0
Real optim state hidden because not passed in

#175 reneschaub opened 7 years ago
3
Feature: Autograd for tensor:split

#174 zzhang1987 opened 7 years ago
0
using `optimfn` with wrapped `nn` modules

#173 sebastiangonsal opened 7 years ago
0
mailing list for autograd

#172 biggerlambda closed 7 years ago
0
Autogradient of function cannot handle batched forward-prop

#171 NaimKabir closed 7 years ago
1
Fix Fast-Weights model (A computed every t)

#170 Atcold closed 7 years ago
1
Help with using autograd in training with wrapped NN modules

#169 synchro-- opened 7 years ago
4
ABS still not implemented error but ABS code is actually there?

#168 synchro-- closed 7 years ago
0
Add RecurrentFWNetwork model

#167 Atcold closed 7 years ago
2
Wrong implemention for logsoftmax?

#166 callowbird opened 8 years ago
1
About tying the weights

#165 johnny5550822 opened 8 years ago
0
literally just removing a print statement

#164 currymj closed 8 years ago
0
torch-cl?

#163 ghost opened 8 years ago
0
Add gather and scatter

#162 szagoruyko closed 8 years ago
0
Question about implementing a pairwise L2 distances function

#161 vadimkantorov opened 8 years ago
1
Reusable AutoModule's

#160 vadimkantorov closed 8 years ago
1
"torch.DoubleTensor.t not currently supported". Workarounds?

#159 vadimkantorov closed 8 years ago
5
Adding a torch-signal function

#158 ssakhavi opened 8 years ago
0
"Failed to parse generated code"

#157 abdullahjamal closed 8 years ago
6
Support to torch.FloatTensor.resize

#156 abdullahjamal opened 8 years ago
0
torch.add(tensor1, scalar, tensor2) not supported.

#155 rtqichen opened 8 years ago
1
Fix locals and layer norm gradient edge case.

#154 allanzelener closed 8 years ago
1
discrepency between licenses

#153 korymath closed 8 years ago
1
Make autograd ignore computing gradients for certain functions

#152 MLpatzer opened 8 years ago
2
Autograd UPDATES weight and bias that DOESN'T CONTRIBUTE to the output of the module!

#151 dlmacedo opened 8 years ago
1
I can't pass tensors to AutoModule

#150 dlmacedo opened 8 years ago
0
autograd initialization error

#149 eladrich closed 8 years ago
2
Nil arguments prevent parsing later non-nil arguments

#148 allanzelener opened 8 years ago
5
Remove globals

#147 szagoruyko closed 8 years ago
1
Modules for Autograd

#146 allanzelener closed 8 years ago
3
Fix gradients of sinh and cosh

#145 gbaydin closed 8 years ago
6
Stop gradients

#144 ikostrikov closed 8 years ago
3
Add layers argument to support reloading and fix iteration upper bound.

#143 allanzelener closed 8 years ago
3
Add batch norm option to LSTM.

#142 allanzelener closed 8 years ago
2
Avoid packing and then unpacking again

#141 bartvm closed 8 years ago
0
Accumulate gradient in-place

#140 bartvm closed 8 years ago
2
Fix wrong zero gradient for assignment in direct mode

#139 bartvm closed 8 years ago
0
Source code printing enabled

#138 alexbw closed 8 years ago
0
Add gradient for log1p

#137 bartvm closed 8 years ago
1
add bmm and baddbmm

#136 jakesnell closed 8 years ago
1
Add scalar support to gradcheck

#135 jakesnell closed 8 years ago
0
GPU support

#134 Zealcui closed 8 years ago
2
Corrected some global assignments

#133 ghostcow closed 8 years ago
0
assignment to undeclared variable zeroGrads

#132 ghostcow closed 8 years ago
0
Input is not a package name or nn object

#131 aleSuglia opened 8 years ago
2