Data

Tools

Indexes

Applications

Experiments

Open Source Science

futr3d

https://github.com/caojiaxi0505/futr3d

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (3.6%) to scientific vocabulary

Scientific Fields

Artificial Intelligence and Machine Learning Computer Science - 69% confidence
Last synced: 4 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: caojiaxi0505
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Size: 205 MB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created 8 months ago · Last pushed 7 months ago
Metadata Files
Readme License Citation

README.md

FUTR3D

``` FUTR3D causal-conv1d/ # checkpoints/ # configs/ # data/ # mamba/ # mamba_experimental/ # mamba mmdet3d/ # | apis/ | core/ | datasets/ | models/ | ops/ | utils/ plugin/ # | dssmss/ | futr3d/ tools/

```

conda create -n futr3d python=3.8 -y conda activate futr3d conda install pytorch==1.13.0 torchvision==0.14.0 torchaudio==0.13.0 pytorch-cuda=11.6 -c pytorch -c nvidia pip install -U openmim mim install mmengine mim install mmcv-full==1.7.0 pip install mmdet==2.27.0 -i https://pypi.tuna.tsinghua.edu.cn/simple pip install mmsegmentation==0.30.0 -i https://pypi.tuna.tsinghua.edu.cn/simple pip install nuscenes-devkit -i https://pypi.tuna.tsinghua.edu.cn/simple cd FUTR3D/ pip install -v -e . pip install numpy==1.23.5 -i https://pypi.tuna.tsinghua.edu.cn/simple pip install yapf==0.40.1 -i https://pypi.tuna.tsinghua.edu.cn/simple cd causal-conv1d pip install -v -e . cd ../ cd mamba pip install -v -e .

| iter | batch size | num GPUs | mAP | NDS | SyncBN | log | |:---:|:---:|:---:|:---:|:---:|:---:|:---:| | 1 | 2 | 1 | 0.6393 | 0.6925 | x | log | | 14 | 4 | 2 | 0.5347 | 0.5979 | x | log | | 40 | 1 | 4 | 0.3683 | 0.3664 | v | log |

SPLIT 1

SPLIT 14

| Method | iter | mAP | NDS | | | cfg | log | |:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:| | FUTR3D-L-bs4x2 | 14 | 0.5347 | 0.5979 | baselinebackboneSECONDSyncBN | FPN128256 | cfg | log | | FUTR3D-hednetbackbone-bs4x2 | 14 | 0.5614 | 0.6135 | backbonehednetSyncBN | hednet412Conv2d2ConvTranspose2dChannel256FPN256 | cfg | log | | FUTR3D-hednetbackbone-bs2x2 | 14 | 0.5499 | 0.5961 | backbonehednetSyncBN | hednet412Conv2d2ConvTranspose2dChannel256FPN256 | cfg | log | | FUTR3D-hednetbackbone4-secondmamba1-bs2x2 | 14 | 0.5441 | 0.5903 | backbonehednetSyncBN | hednet412Conv2d2ConvTranspose2dChannel256SECONDMambaBlockFPN256 | cfg | log |

SPLIT 40

| Method | iter | mAP | NDS | | | cfg | log | memory | time | |:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:| | FUTR3D-L-bs1x4 | 40 | 0.3683 | 0.3664 | baselinebackboneSECONDSyncBN | FPN128256 | cfg | log | 3567 | 4h42min | | FUTR3D-hetnetbackbone-bs1x4 | 40 | 0.4125 | 0.4050 | backbonehednetSyncBN | hednet412Conv2d2ConvTranspose2dChannel256FPN256 | cfg | log | 5774 | 6h1min | | FUTR3D-hednetmiddleencoder128-hednetbackbone-bs1x4 | 40 | 0.4131 | 0.3893 | backbonemiddle encoderhednetSyncBN | hednet412Conv2d2ConvTranspose2dChannel128FPN128 | cfg | log | 5625 | 6h10min | | FUTR3D-hednetmiddleencoder256-hednetbackbone-bs1x4 | 40 | 0.4722 | 0.4382 | backbonemiddle encoderhednetSyncBN | hednet412Conv2d2ConvTranspose2dChannel256FPN256 | cfg | log | 8071 | 8h6min |

GPUGPUSample6019Samples

SECOND-Backbone

Block0 Sequential( (0): Conv2d(256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (1): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) (3): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (5): ReLU(inplace=True) (6): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (7): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (8): ReLU(inplace=True) (9): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (10): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (11): ReLU(inplace=True) (12): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (13): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (14): ReLU(inplace=True) (15): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (16): BatchNorm2d(128, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (17): ReLU(inplace=True) ) Block1 Sequential( (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (1): BatchNorm2d(256, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (2): ReLU(inplace=True) (3): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (4): BatchNorm2d(256, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (5): ReLU(inplace=True) (6): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (7): BatchNorm2d(256, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (8): ReLU(inplace=True) (9): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (10): BatchNorm2d(256, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (11): ReLU(inplace=True) (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (13): BatchNorm2d(256, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (14): ReLU(inplace=True) (15): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (16): BatchNorm2d(256, eps=0.001, momentum=0.01, affine=True, track_running_stats=True) (17): ReLU(inplace=True) )

FPS: 4.56 mAP: 0.5339
mATE: 0.3768 mASE: 0.2713 mAOE: 0.3987 mAVE: 0.4499 mAAE: 0.2031 NDS: 0.5969 Eval time: 51.1s Per-class results: Object Class AP ATE ASE AOE AVE AAE
car 0.799 0.210 0.162 0.134 0.405 0.217 truck 0.498 0.415 0.215 0.169 0.353 0.222 bus 0.601 0.435 0.211 0.138 0.991 0.274 trailer 0.351 0.615 0.222 0.568 0.294 0.178 constructionvehicle 0.185 0.794 0.463 0.975 0.145 0.357 pedestrian 0.756 0.244 0.287 0.615 0.425 0.134 motorcycle 0.558 0.261 0.260 0.361 0.729 0.232 bicycle 0.399 0.193 0.268 0.535 0.257 0.010 trafficcone 0.597 0.211 0.347 nan nan nan
barrier 0.595 0.390 0.277 0.093 nan nan

SPLIT14

GPU:2 & batch size:2

| Method | numGPUs | batchsize | SyncBN | mAP | NDS | time | |:---:|:---:|:---:|:---:|:---:|:---:|:---:| | lidar0075v900q | 2 | 2 | v | 0.5082 | 0.5534 | 7h21min(val4) | | lidar0075v900q_hednetbackbone4 | 2 | 2 | v | 0.5499 | 0.5961 | 10h29min(val4)(+42.63%) |

GPU:4 & batch size:4

| Method | numGPUs | batchsize | SyncBN | mAP | NDS | time | |:---:|:---:|:---:|:---:|:---:|:---:|:---:| | lidar0075v900q | 2 | 4 | v | 0.5343 | 0.5977 | 6h22min(val4) | | lidar0075v900q | 2 | 4 | x | 0.5347 | 0.5979 | 8h48min(val20) | | lidar0075v900q_hednetbackbone4 | 2 | 4 | v | 0.5614 | 0.6135 | 8h47min(val4)(+37.96%) |

SPLIT40

Full dataset

| Method | mAP | NDS | time | infer time | GPUs x batchsize | LR | extra explanation | log | pth | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | lidar0075v900q | 63.3 | 68.9 | - | - | - | - | baseline | - | - | | lidar0075v900qsplit11x2bn2d | 63.93 | 69.25 | 6d22h2min | 16min25s (6.1FPS) | 1 x 2 | 6.25e-6 (1/16) | baseline replicate | log | pth | |

1/5 Split

1/14 Split

Owner

  • Name: caojiaxi
  • Login: caojiaxi0505
  • Kind: user

GitHub Events

Total
  • Push event: 4
Last Year
  • Push event: 4

Dependencies

causal-conv1d/causal_conv1d.egg-info/requires.txt pypi
  • ninja *
  • packaging *
  • torch *
causal-conv1d/setup.py pypi
  • ninja *
  • packaging *
  • torch *
mamba/setup.py pypi
  • causal_conv1d >=1.4.0
  • einops *
  • ninja *
  • packaging *
  • torch *
  • transformers *
  • triton *
requirements/build.txt pypi
requirements/docs.txt pypi
  • docutils ==0.16.0
  • m2r *
  • mistune ==0.8.4
  • myst-parser *
  • sphinx ==4.0.2
  • sphinx-copybutton *
  • sphinx_markdown_tables *
requirements/mminstall.txt pypi
  • mmcv-full >=1.4.8,<=1.6.0
  • mmdet >=2.24.0,<=3.0.0
  • mmsegmentation >=0.20.0,<=1.0.0
requirements/optional.txt pypi
  • open3d *
  • spconv *
  • waymo-open-dataset-tf-2-1-0 ==1.2.0
requirements/readthedocs.txt pypi
  • mmcv >=1.4.8
  • mmdet >=2.24.0
  • mmsegmentation >=0.20.1
  • torch *
  • torchvision *
requirements/runtime.txt pypi
  • lyft_dataset_sdk *
  • networkx >=2.2,<2.3
  • numba ==0.53.0
  • numpy *
  • nuscenes-devkit *
  • plyfile *
  • scikit-image *
  • tensorboard *
  • trimesh >=2.35.39,<2.35.40
requirements/tests.txt pypi
  • asynctest * test
  • codecov * test
  • flake8 * test
  • interrogate * test
  • isort * test
  • kwarray * test
  • pytest * test
  • pytest-cov * test
  • pytest-runner * test
  • ubelt * test
  • xdoctest >=0.10.0 test
  • yapf * test
requirements.txt pypi
setup.py pypi