jit


Topic Replies Activity
Making Jit work 9 May 4, 2019
Can't save the LSTM model with torch.jit.script_method 4 April 30, 2019
It is possible to iterate torch._C.graph? 2 April 30, 2019
How to convert the model with no fixed size input to Torch Script? 3 April 29, 2019
Maintaining dropout layer for deployment 4 April 25, 2019
Cannot access nn.Linear.in_features in ScriptModule 3 April 17, 2019
TracerWarning: Converting a tensor to a Python index might cause the trace to be incorrect...This means that the trace might not generalize to other inputs! 5 April 17, 2019
How to create a variable of optional type(e.g., Optional[Tuple[int,int]]), with C++ libtorch jit api? 5 April 17, 2019
Transforming a model based on graph information 2 April 16, 2019
BERT torch graph tensor shape issues 3 April 12, 2019
Namedtuple in C++ interface 2 April 11, 2019
Cannot Obtain Similar DL Prediction Result in Pytorch C++ API Compared to Python 2 April 11, 2019
How to make jit script works with a custom RNN module (renamed) 3 April 8, 2019
Saving validation time inferences 7 April 2, 2019
Pytorch model production 1 April 2, 2019
Difference between writing in a single line 4 March 31, 2019
Should I use @weak_module for my modules? 4 March 28, 2019
Disabling gradient calculation in jit.script_method 1 March 28, 2019
JIT with torch.nn.functional.interpolate 5 March 25, 2019
Error in @torch.jit.script 3 March 23, 2019
TorchScript: Using tensor populated with a different module 3 March 22, 2019
Using Dict[str, Tensor] type as a argument for a jit script_method 2 March 21, 2019
Do items in lists have to consist of only a single type? 4 March 21, 2019
Getting input and output tensors for each node in trace graph 2 March 21, 2019
Where is source code of Tensor.bmm? 2 March 21, 2019
Dropout in RNN for each layer 1 March 15, 2019
Tracing custom RNN model 3 March 7, 2019
JIT doesn't support 'to' keyword 3 March 6, 2019
Get intermediate layer outputs from traced graph 3 March 6, 2019
Cannot move ScriptModule to GPU with to() 9 March 4, 2019