PyTorch's implementation of the paper "End to End Learning for Self-Driving Cars"

Introduction

This is the PyTorch implementation of the paper "End to End Learning for Self-Driving Cars" by NVIDIA. The paper can be found here.

This code is based on the original implementation by Pankaj Karki. The original implementation is in TensorFlow.

To install the conda environment, run the following command: conda install pytorch torchvision torchaudio -c pytorch.

The dataset used in this project is the Dataset from Sully Chen. It has a size of 2.3GB approximately.

To train the model, run the following command: python train.py.

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
capture_data		capture_data
docker		docker
inference		inference
notebooks		notebooks
runs		runs
scripts		scripts
.gitignore		.gitignore
Dockerfile		Dockerfile
README.md		README.md
calculate_dataset_std_mean.py		calculate_dataset_std_mean.py
chart.py		chart.py
config.py		config.py
convert_to_jpg.py		convert_to_jpg.py
cross_validation.py		cross_validation.py
dataset_loader.py		dataset_loader.py
export_model_to_onnx.py		export_model_to_onnx.py
inference_run_dataset.py		inference_run_dataset.py
inference_run_video.py		inference_run_video.py
loss_comparison.png		loss_comparison.png
model.py		model.py
plot_loss_results.py		plot_loss_results.py
requirements.txt		requirements.txt
steering_wheel_image.jpg		steering_wheel_image.jpg
train.py		train.py
utils.py		utils.py
validate.py		validate.py