Skip to content

Rewrite the softmax op#1911

Merged
snnn merged 8 commits intomasterfrom
snnn/softmax
Sep 25, 2019
Merged

Rewrite the softmax op#1911
snnn merged 8 commits intomasterfrom
snnn/softmax

Conversation

@snnn
Copy link
Copy Markdown
Contributor

@snnn snnn commented Sep 24, 2019

Description:

Copy the tensorflow's softmax implementation to there.

Motivation and Context

  • Why is this change required? What problem does it solve?
    Improve batch perf.
data size old new
1x1000 0.006ms 0.008ms
1000x1000 2.5ms 1.5ms

The old one is not parallelized well.
The new one has much less code.

  • If it fixes an open issue, please link to the issue here.

@snnn snnn requested a review from a team as a code owner September 24, 2019 23:57
Copy link
Copy Markdown
Contributor

@HectorSVC HectorSVC left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

:shipit:

@snnn snnn merged commit 76f7625 into master Sep 25, 2019
@snnn snnn deleted the snnn/softmax branch September 25, 2019 21:29
yuslepukhin pushed a commit that referenced this pull request Mar 17, 2026
The `olive run-pass` command was implemented and available in the CLI
but was not documented in the CLI reference documentation at
`docs/source/reference/cli.rst`.

## Changes Made

- Added "Run Pass" section to `docs/source/reference/cli.rst` between
"Run" and "Finetune" sections
- Used the command's help text description: "Run a single pass on the
input model (supports HuggingFace, ONNX, PyTorch, and Azure ML models)"
- Added proper `argparse` directive pointing to the correct path
`run-pass`
- Maintained consistent formatting with all other CLI command
documentation

## Verification

The documentation now includes all available CLI commands and follows
the same pattern as the existing 11 documented commands. The `run-pass`
command allows users to execute individual optimization passes on
models, complementing the full workflow execution provided by the `run`
command.

Fixes #1910.

<!-- START COPILOT CODING AGENT TIPS -->
---

💡 You can make Copilot smarter by setting up custom instructions,
customizing its development environment and configuring Model Context
Protocol (MCP) servers. Learn more [Copilot coding agent
tips](https://gh.io/copilot-coding-agent-tips) in the docs.

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: devang-ml <47577486+devang-ml@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants