Corey Lammie

Hi @YBRua and @matifali, My apologies for my late reply- I'll look into this later this week and implement a fix for this behavior. Will a static method which can...

Hi @frmar440, I have implemented the base code/logic for a RNN module here: https://github.com/coreylammie/MemTorch/blob/master/memtorch/mn/RNN.py. Although currently, only legacy forward passes are currently enabled, it should be fairly straight forward to...

Hi @wwwe1ty, This may be related to #120. The following logic is used to configure and launch CUDA kernels when active crossbars are simulated: https://github.com/coreylammie/MemTorch/blob/master/memtorch/cu/tile_matmul_kernels.cu#L124-L156. While logic is used to...

Hi @wwwe1ty and @dgr-b, Thank you for your patience, and for letting me know about the CPU tuning bug. I'm actively working on this when I have time, however, this...

This has been partially actioned in #134. I have implemented logic for the legacy routine (for both standard and bidirectional RNN modules with an arbitrary number of layers). Logic is...

Hi @Philippe-Drolet, I have prioritized the implementation of `torch.nn.RNN`, `torch.nn.RNNCell`, `torch.nn.LTSM`, `torch.nn.LTSMCell`, `torch.nn.GRU`, and `torch.nn.GRUCell` modules, so I will likely be unable to work on this issue in the near-future....

Hi @Philippe-Drolet, Sure- my preferred method of debugging is to use `cuda-memcheck`. The `cuda-memcheck` tool can be used to pin-point the exact line/kernel and respective error message, as long as...

Hi @McLemoreT and @RagingFlames, Apologies for the late response! I have self-assigned this issue, so we can discuss it here, rather than re-opening #103. Could you both please confirm that...

Hi @McLemoreT, In many files within `MemTorch`, the following statement is used: `self.device = torch.device("cpu" if "cpu" in memtorch.__version__ else "cuda")` to select the device to be used. Looking at...

Hi @Philippe-Drolet, `PyTorch`, and by extension, `MemTorch`, do not currently officially support `CUDA` 11.5. Current progress is being tracked here: https://github.com/pytorch/pytorch/issues/68259. I have personally been able to compile `MemTorch` with...