-
# Automatic Reference Counting Base Type
We need to introduce an ARC-like style way of flagging types, so whenever those are used, a `Release` call is made, so we can promptly dispose of that data.…
-
Microsoft recently released Phi-3 models in 3 variants (mini, small & medium). Can we add support for this new family of models.
-
sorry can you give your email address or where is it convenient for you to chat? I wanted to ask something about the framework and skills.
dimiy updated
4 years ago
-
Hi, authors. Thanks for providing this repo.
I'm currently using the Orthogonal module and define it as part of my model weights. When I tried to resume training from a checkpoint, an unexpected er…
-
## 🐛 Bug
I've ran into an issue with data transfer rate from GPU to CPU being much slower than expected. After benchmarking, there seems to be a discrepancy in CPU->GPU and GPU->CPU data transfer s…
-
**paddle版本为develop版本**
**报错信息如下**:
[08/31 21:47:28] model_name : STGCN
W0831 21:47:29.443328 4185452 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.6, Driver API Version:…
-
## 🐛 Bug
We have a use-case in Captum in which we need to use a forward hook to extract an intermediate layer activation and then use torch.autograd.grad to compute gradients of this activation wit…
-
A rank zero tensor should be a scalar.
Should N-dim arrays in _sympy.tensor.array_ return the scalar if created with _shape = ()_ ?
There is a bug in the array printer. Rank zero arrays are not corr…
-
It's not an issue but a query. I like to calculate the Christoffel symbols from a given line element and assign the expressions of each of Christoffel symbols to the variables in a user-defined functi…
-
IMO the automatic summation of repeated indices in expressions involving Indexed objects (from tensor/indexed.py), i.e. the Einstein summation convention, should require some sort of encapsulating obj…