Target
Antgo is a machine learning experiment manage platform, which has been integrated deeply with MLTalker. Antgo provides a one-stop model development, deployment, analyze, auto-optimize and manage environment.
Installation
(RECOMMENDED) use docker
install from pip
pip install antgo
install from source
git clone https://github.com/jianzfb/antgo.git
cd antgo
pip install -r requirements.txt
python setup.py build_ext install
Register
Register in MLTalker.
All user experiment records would be managed by MLTalker in user’s personal page.
Quick Example
1.step create mvp code(cifar10 classification task)
antgo create mvp –name=cifar10
2.step start training process
python3 ./cifar10/main.py –exp=cifar10 –gpu-id=0 –process=train
3.step check training log
in ./output/cifar10/output/checkpoint
4.step export onnx model
python3 ./cifar10/main.py –exp=cifar10 –checkpoint=./output/cifar10/output/checkpoint/epoch_1500.pth –process=export
1 maintainer
Metadata
1 maintainer
Release files for antgo 0.1.24
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| antgo-0.1.24.tar.gz | 3.5 MB | Details |
Release files / antgo-0.1.24.tar.gz
| Download URL | antgo-0.1.24.tar.gz |
|---|---|
| Size | 3.5 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
55bbc9662a500194f67eb3929367854acbbb8afa6abf2c050c3a796ca2421eaa
|
|
BLAKE2b-256 checksum How to use checksums |
d4acc040f6ebdf6dd51e95d208ae1c3cabda60eef6e46813702f801babfef8a4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.17
|
