David Fan

optimizer_test.py failure on litprotobuf ERROR

13

I installed onnx/optimizer from source as mentioned. Then I run `python onnxoptimizer/test/optimizer_test.py`, I got the failure below: ```python [libprotobuf ERROR google/protobuf/descriptor_database.cc:644] File already exists in database: onnx/onnx-ml.proto [libprotobuf FATAL google/protobuf/descriptor.cc:1371]...

Handle len(cached_x.grad_fn.next_functions) == 1 in cached_cast

8

Using apex for mix precision training, and find one case when `if x.requires_grad and cached_x.requires_grad:`, the tuple `cached_x.grad_fn.next_functions` contains only one element. In this case, we see the error: ```python...

Support masking for LSTM, RNN?

9

In keras or tensorflow, LSTM or RNN support masking which uses a mask value to skip timesteps. This is commonly used in some state-of-the-art models such as keras-bert. So we...

Compile bug: No rule to make target '/usr/lib/librt.a'

1

### Git commit git rev-parse HEAD c5ede3849fc021174862f9c0bf8273808d8f0d39 ### Operating systems Linux ### GGML backends CUDA ### Problem description & steps to reproduce I want to build llama.cpp from source in...