Package Health Score

61 / 100

security
No known security issues
popularity
Small
maintenance
Inactive
community
Active

Popularity

Small

GitHub Stars: 10.95K
Forks: 2.14K
Contributors: 80

Direct Usage Popularity

Based on project statistics from the GitHub repository for the PyPI package trt-engine-explorer, we found that it has been starred 10,954 times.

Security

No known security issues

Security and license risk for latest version

Release Date: Nov 9, 2023
Direct Vulnerabilities: 0
C
0
H
0
M
0
L
Indirect Vulnerabilities: 0
C
0
H
0
M
0
L
License Risk: 0
H
0
M
0
L

All security vulnerabilities belong to production dependencies of direct and indirect packages.

License

Apache-2.0

Security Policy

We found a way for you to contribute to the project! Looks like trt-engine-explorer is missing a security policy.

You can connect your project's repository to Snyk to stay up to date on security alerts and receive automatic fix pull requests.

Keep your project free of vulnerabilities with Snyk

Maintenance

Inactive

Commit Frequency

Open Issues: 406
Open PR: 36
Last Release: 1 year ago
Last Commit: 9 days ago

Further analysis of the maintenance status of trt-engine-explorer based on released PyPI versions cadence, the repository activity, and other data points determined that its maintenance is Inactive.

An important project maintenance signal to consider for trt-engine-explorer is that it hasn't seen any new versions released to PyPI in the past 12 months, and could be considered as a discontinued project, or that which receives low attention from its maintainers.

As a healthy sign for on-going project maintenance, we found that the GitHub repository had at least 1 pull request or issue interacted with by the community.

Community

Active

Readme: Yes
Contributing.md: Yes
Code of Conduct: No
Contributors: 80
Funding: No

With more than 10 contributors for the trt-engine-explorer repository, this is possibly a sign for a growing and inviting community.

We found a way for you to contribute to the project! Looks like trt-engine-explorer is missing a Code of Conduct.

Embed Package Health Score Badge

Keep your project healthy

Check your requirements.txt

Ensure all the packages you're using are healthy and well-maintained

Snyk Vulnerability Scanner

Get health score & security insights directly in your IDE

Package

Python Versions Compatibility: >=3.8,<4.0

Age: 1 year
Latest Release: 1 year ago
Dependencies: 23 Direct
Versions: 4
Maintainers: 0
Wheels: OS Independent

Readme

trt-engine-explorer

This repository contains Python code (trex package) to explore various aspects of a TensorRT engine plan and its associated inference profiling data.

An engine plan file is a serialized TensorRT engine format. It contains information about the final inference graph and can be deserialized for inference runtime execution. An engine plan is specific to the hardware and software versions of the system used to build the engine.

trex is useful for initial model performance debugging, visualization of plan graphs, and for understanding the characteristics of an engine plan. For in-depth performance analysis, Nvidia ® Nsight Systems ™ is the recommended performance analysis tool.

Features

The trex package contains an API and Jupyter notebooks for viewing and inspecting TensorRT engine-plan files and profiling data.

An engine plan graph (JSON) is loaded to a Pandas dataframe which allows slicing, querying, filtering, viewing and diagraming.
An engine plan graph can be visualized as SVG/PNG files.
Layer linters are an API for flagging potential performance hazards (preview feature).
Four Jupyter notebooks provide:
- An introduction to trex tutorial.
- trex API examples.
- Detailed engine plan performance, characteristics and structure analysis.
- Comparison of two or more engine plans.
Because trex operates on JSON input files, it does not require a GPU.

Caveats

When trtexec times individual layers, the total engine latency (computed by summing the average latency of each layer) is higher than the latency reported for the entire engine. This is due to per-layer measurement overheads.

To measure per-layer execution times, when trtexec enqueues kernel layers for execution in a stream, it places CUDA event objects between the layers to monitor the start and completion of each layer. These CUDA events add a small overhead which is more noticeable with smaller networks (shallow and narrow networks or networks with small activation data).

Supported TensorRT Versions

Starting with TensorRT 8.2, engine-plan graph and profiling data can be exported to JSON files. trex supports TensortRT 8.2 and 8.4.

trex has only been tested on Ubuntu 18.04, 20.04, 22.04, with Python 3.8.
trex does not require a GPU, but generating the input JSON file(s) does require a GPU.

Installation

The instructions below detail how to use a Python3 virtualenv for installing and using trex (Python 3.8+ is required).

1. Clone the trex code repository from TensorRT OSS repository

$ git clone https://github.com/NVIDIA/TensorRT.git

2. Create and activate a Python virtual environment

The commands listed below create and activate a Python virtual environment named env_trex which is stored in a directory by the same name, and configures the current shell to use it as the default python environment.

$ cd TensorRT/tools/experimental/trt-engine-explorer
$ python3 -m virtualenv env_trex
$ source env_trex/bin/activate

3. Install trex in development mode and the Jupyter extensions required for the notebooks

$ python3 -m pip install -e .
$ jupyter nbextension enable widgetsnbextension --user --py

4. Install Graphviz

Generating dot and SVG graphs requires Graphviz, an open source graph visualization software:

$ sudo apt-get --yes install graphviz

Workflow

The typical trex workflow is depicted below:

Convert an external model to a TensorRT INetworkDefinition.
Build a TensorRT engine.
Profile the engine while creating the necessary JSON files.
Explore the engine by loading the JSON files in a trex notebook.

The Python script utils/process_engine.py implements this workflow for ONNX models:

Use trtexec to import an ONNX model and create an engine.
Load the engine and create an engine-graph JSON file.
Use trtexec to profile the engine's inference execution and store the results in an engine profiling JSON file.
Create an engine graph diagram in SVG format.

For more information see TensorRT Engine Inspector and the Tutorial notebook.

Jupyter Server

Launch the Jupyter notebook server as detailed below and open your browser at http://localhost:8888 or http://:8888

$ jupyter-notebook --ip=0.0.0.0 --no-browser

License

The TensorRT Engine Explorer license can be found in the LICENSE file.

trt-engine-explorer dependencies

dtale flask graphviz ipyfilechooser ipywidgets jupyter jupyter-dash jupyterlab kaleido netron numpy onnx openpyxl pandas plotly protobuf pytest qgrid setuptools traitlets werkzeug wheel xlsxwriter

FAQs

What is trt-engine-explorer?

NVIDIA's experimental tool to explore TRT engines. Visit Snyk Advisor to see a full health score report for trt-engine-explorer, including popularity, security, maintenance & community analysis.

Is trt-engine-explorer popular?

The python package trt-engine-explorer receives a total of 98 weekly downloads. As such, trt-engine-explorer popularity was classified as small. Visit the popularity section on Snyk Advisor to see the full health analysis.

Is trt-engine-explorer well maintained?

We found indications that trt-engine-explorer is an Inactive project. See the full package health analysis to learn more about the package maintenance status.

Is trt-engine-explorer safe to use?

The python package trt-engine-explorer was scanned for known vulnerabilities and missing license, and no issues were found. Thus the package was deemed as safe to use. See the full health analysis review.

Last updated on 21 December-2024, at 08:38 (UTC).

Build a secure application checklist

Select a recommended open source package

Minimize your risk by selecting secure & well maintained open source packages

DONE

Scan your app for vulnerabilities

Scan your application to find vulnerabilities in your: source code, open source dependencies, containers and configuration files

SCAN NOW

Fix identified vulnerabilities

Easily fix your code by leveraging automatically generated PRs

AUTO FIX

Monitor for new issues

New vulnerabilities are discovered every day. Get notified if your application is affected

MONITOR APP

trt-engine-explorer

NVIDIA's experimental tool to explore TRT engines For more information about how to use this package see README

Ensure you're using the healthiest python packages

Package Health Score

Popularity

Direct Usage Popularity

Security

Security and license risk for latest version

Maintenance

Commit Frequency

Community

Embed Package Health Score Badge

Keep your project healthy

Check your requirements.txt

Snyk Vulnerability Scanner

Package

Readme

trt-engine-explorer

Features

Caveats

Supported TensorRT Versions

Installation

1. Clone the trex code repository from TensorRT OSS repository

2. Create and activate a Python virtual environment

3. Install trex in development mode and the Jupyter extensions required for the notebooks

4. Install Graphviz

Workflow

Jupyter Server

License

trt-engine-explorer dependencies

FAQs

What is trt-engine-explorer?

Is trt-engine-explorer popular?

Is trt-engine-explorer well maintained?

Is trt-engine-explorer safe to use?

Build a secure application checklist

Select a recommended open source package

Scan your app for vulnerabilities

Fix identified vulnerabilities

Monitor for new issues

Package Health Score

Direct Usage Popularity

Security and license risk for latest version

Commit Frequency

Embed Package Health Score Badge

Keep your project healthy

Check your requirements.txt

Snyk Vulnerability Scanner

Readme

trt-engine-explorer

Features

Caveats

Supported TensorRT Versions

Installation

1. Clone the trex code repository from TensorRT OSS repository

2. Create and activate a Python virtual environment

3. Install trex in development mode and the Jupyter extensions required for the notebooks

4. Install Graphviz

Workflow

Jupyter Server

License

trt-engine-explorer dependencies

FAQs

What is trt-engine-explorer?

Is trt-engine-explorer popular?

Is trt-engine-explorer well maintained?

Is trt-engine-explorer safe to use?

Build a secure application checklist

Select a recommended open source package

Scan your app for vulnerabilities

Source Code

Open Source

Container

Config

Fix identified vulnerabilities

Monitor for new issues