LMI AI Solutions

This repo contains the utils scripts, and several submodules for LMI Technologies Inc. AI modeling development. Currently, the following models are supported in the repo:

• object detection
  • yolov8
  • yolov5
  • efficientnet
  • detectron2
  • tensorflow object detection API

• anomaly detection
  • anomalib
• OCR
  • paddleOCR

Clone this master repo

For users who haven't set up the ssh keys

git clone https://github.com/lmitechnologies/LMI_AI_Solutions.git

For users who have the ssh keys

git clone https://github.com/lmitechnologies/LMI_AI_Solutions.git
# npm using git for https
git config --global url."[email protected]:".insteadOf https://github.com/
git config --global url."git://".insteadOf https://

Clone submodules

Go to the master repo

cd LMI_AI_Solutions

Each submodule is pointing to a specific commit in its ais branch. Clone the submodules to the commit that is specified in this repo

git submodule update --init

(not recommend) if you want to update all submodules to the lastest commit in the ais branch, use the --remote argument

git submodule update --init --remote

Use this repo

1. Activate the environmental file - lmi_ai.env:

source PATH_TO_REPO/lmi_ai.env

where PATH_TO_REPO is the path to the LMI_AI_Solutions repo.

2. Run any scripts in this repo, for example:

python -m label_utils.plot_labels -h

Make contributions to this repo

The ais branch of this repo and that branch of submodules are protected, which means you can't directly commit to that branch. You could create a new branch and open the pull request in order to merge into ais branch.