Skip to content

Conversation

@Ziminli
Copy link
Collaborator

@Ziminli Ziminli commented Feb 13, 2025

  • Deleted test_utils.py and combined its content to utils.py
  • Optimized matmul.py by rearranging the structure and applying appropriate abstraction
  • Add infiniDeviceEnum_str_map in device.py
  • Add debug functions and options
  • Add compile flags for num_prerun and num_iterations
  • Add print statement for create_workspace() for displaying the obtained workspace size

CPU test passed:
image

Nvidia test passed:
image

@Ziminli Ziminli self-assigned this Feb 13, 2025
@Ziminli Ziminli linked an issue Feb 13, 2025 that may be closed by this pull request
@Ziminli Ziminli changed the title New Optimized Matmul Test issue1: New Optimized Matmul Test Feb 13, 2025
@Ziminli Ziminli closed this Feb 14, 2025
@Ziminli Ziminli deleted the optimize_matmul_test branch February 14, 2025 08:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants