3rd Place Solution of Traffic4Cast 2021 Core Challenge

This is the code for our solution to the NeurIPS 2021 Traffic4Cast Core Challenge.

Paper

Our solution is described in the "Solving Traffic4Cast Competition with U-Net and Temporal Domain Adaptation" paper.

If you wish to cite this code, please do it as follows:

@misc{konyakhin2021solving,
      title={Solving Traffic4Cast Competition with U-Net and Temporal Domain Adaptation}, 
      author={Vsevolod Konyakhin and Nina Lukashina and Aleksei Shpilman},
      year={2021},
      eprint={2111.03421},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

Competition and Demonstration Track @ NeurIPS 2021

Learnt parameters

The models' learnt parameters are available by the link: https://drive.google.com/file/d/1zD0CecX4P3v5ugxaHO2CQW9oX7_D4BCa/view?usp=sharing
Please download the archive and unzip it into the weights folder of the repository, so its structure looks like the following:

├── ...
├── traffic4cast
├── weights
│   ├── densenet                 
│   │   ├── BERLIN_1008_1430_densenet_unet_mse_best_val_loss_2019=78.4303.pth                     
│   │   ├── CHICAGO_1010_1730_densenet_unet_mse_best_val_loss_2019=41.1579.pth
│   │   └── MELBOURNE_1009_1619_densenet_unet_mse_best_val_loss_2019=25.7395.pth    
│   ├── effnetb5
│   │   ├── BERLIN_1008_1430_efficientnetb5_unet_mse_best_val_loss_2019=80.3510.pth    
│   │   ├── CHICAGO_1012_1035_efficientnetb5_unet_mse_best_val_loss_2019=41.6425.pth
│   │   ├── ISTANBUL_1012_2315_efficientnetb5_unet_mse_best_val_loss_2019=55.7918.pth    
│   │   └── MELBOURNE_1010_0058_efficientnetb5_unet_mse_best_val_loss_2019=26.0132.pth    
│   └── unet
│       ├── BERLIN_0806_1425_vanilla_unet_mse_best_val_loss_2019=0.0000_v5.pth    
│       ├── CHICAGO_0805_0038_vanilla_unet_mse_best_val_loss_2019=42.6634.pth
│       ├── ISTANBUL_0805_2317_vanilla_unet_mse_best_val_loss_2019=0.0000_v4.pth
│       └── MELBOURNE_0804_1942_vanilla_unet_mse_best_val_loss_2019=26.7588.pth
├── ...

Submission reproduction

To generate the submission file, please run the following script:

# $1 - absolute path to the dataset, $2 device to run inference
sh submission.sh {absolute path to dataset} {cpu, cuda}
# Launch example
sh submission.sh /root/data/traffic4cast cuda

The above sctipt generates the submission file submission/submission_all_unets_da_none_mpcpm1_mean_temporal_{date}.zip, which gave us the best MSE of 49.379068541527 on the final leaderboard.

3rd Place Solution of the Traffic4Cast Core Challenge @ NeurIPS 2021

Related tags

Overview

3rd Place Solution of Traffic4Cast 2021 Core Challenge

Paper

Learnt parameters

Submission reproduction

Owner

PyTorch implementation for MINE: Continuous-Depth MPI with Neural Radiance Fields

NeRD: Neural Reflectance Decomposition from Image Collections

Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics

Code for the SIGGRAPH 2022 paper "DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds."

Code for our paper A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localization,

CN24 is a complete semantic segmentation framework using fully convolutional networks

Gym for multi-agent reinforcement learning

Generalized Data Weighting via Class-level Gradient Manipulation

Deep Unsupervised 3D SfM Face Reconstruction Based on Massive Landmark Bundle Adjustment.

Capstone-Project-2 - A game program written in the Python language

Segmentation-Aware Convolutional Networks Using Local Attention Masks

UpChecker is a simple opensource project to host it fast on your server and check is server up, view statistic, get messages if it is down. UpChecker - just run file and use project easy

This code is 3d-CNN model that can predict environmental value

MultiMix: Sparingly Supervised, Extreme Multitask Learning From Medical Images (ISBI 2021, MELBA 2021)

Deep learning for spiking neural networks

Dimension Reduced Turbulent Flow Data From Deep Vector Quantizers

Jiminy Cricket Environment (NeurIPS 2021)

Reviving Iterative Training with Mask Guidance for Interactive Segmentation

Blender Python - Node-based multi-line text and image flowchart

Practical tutorials and labs for TensorFlow used by Nvidia, FFN, CNN, RNN, Kaggle, AE