Skip to content

[TorchToLinalg] Adds Quantization Support for ConvTranspose #1881

[TorchToLinalg] Adds Quantization Support for ConvTranspose

[TorchToLinalg] Adds Quantization Support for ConvTranspose #1881

Triggered via pull request April 29, 2024 22:30
Status Success
Total duration 25m 3s
Artifacts

ci.yml

on: pull_request
Matrix: build-test-linux
Fit to window
Zoom out
Zoom in

Annotations

4 warnings
Build and Test (Linux, torch-nightly, assertions)
This self-hosted runner is currently using runner version 2.314.1. This version is out of date. Please update to the latest version 2.316.0
Build and Test (Linux, torch-nightly, assertions)
Node.js 16 actions are deprecated. Please update the following actions to use Node.js 20: actions/checkout@8f4b7f84864484a7bf31766abe9204da3cbe65b3, actions/cache/restore@v3, actions/cache/save@v3. For more information see: https://github.blog/changelog/2023-09-22-github-actions-transitioning-from-node-16-to-node-20/.
Build and Test (Linux, torch-stable, assertions)
This self-hosted runner is currently using runner version 2.314.1. This version is out of date. Please update to the latest version 2.316.0
Build and Test (Linux, torch-stable, assertions)
Node.js 16 actions are deprecated. Please update the following actions to use Node.js 20: actions/checkout@8f4b7f84864484a7bf31766abe9204da3cbe65b3, actions/cache/restore@v3, actions/cache/save@v3. For more information see: https://github.blog/changelog/2023-09-22-github-actions-transitioning-from-node-16-to-node-20/.