https://github.com/vietanhdev/traffic-sign-classification-uff-tensorrt

Traffic sign classification training and conversion code to uff - TensorRT
https://github.com/vietanhdev/traffic-sign-classification-uff-tensorrt

Last synced: 8 months ago
JSON representation

Traffic sign classification training and conversion code to uff - TensorRT

• Host: GitHub
• URL: https://github.com/vietanhdev/traffic-sign-classification-uff-tensorrt
• Owner: vietanhdev
• Created: 2020-11-15T13:42:14.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2021-03-14T01:33:02.000Z (over 5 years ago)
• Last Synced: 2025-04-08T16:38:04.028Z (about 1 year ago)
• Language: Python
• Size: 26.4 KB
• Stars: 6
• Watchers: 2
• Forks: 4
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
### Traffic sign classification

The purpose of this project is to train a traffic sign classification model for an advanced driver-assistance system. [More about this project](https://aicurious.io/posts/adas-jetson-nano-intro-and-hardware/). The inference code was integrated in [this repository](https://github.com/vietanhdev/car-smart-cam).

#### Step 1: Initialize environment

Create anaconda enviroment:

```conda create --name  python=3.6```

Activate created environment and install all requirements:

```pip install requirements.txt```

#### Step 2: Train the models

- Prepare dataset like below structure:

![Dataset preparation](dataset-preparation.png)

- Train the model:

```

python train.py

```

#### Step 3: Convert to UFF

- Modify model paths in `convert_h5_to_pb.py` and `convert_pb_to_uff.py`.

- Convert `.h5` model to `.pb`, and finnally `.uff`: 

```

pip install requirements-h5-to-uff.txt

python convert_h5_to_pb.py

python convert_pb_to_uff.py

```