v3.5.1
This is a patch release containing the following changes to v3.5:
- Fixed potential page fault in matmul on Intel Datacenter Max Series GPUs (a9c525d)
- Fixed potential stack overflow issue in convolution implementation for Intel GPUs (0fb7e6e)
- Added test cases for matmul with compressed weights (015ccb1)
- Extended Graph API
LayerNorm
operation with zero points support (dc2701a) - Fixed primitive creation error for depthwise convolution backpropagation on Intel GPUs (4a045e4, b529d22)