Pasukan Mod
1.0.0
Ini adalah implementasi PyTorch/GPU dari makalah Mod-Squad: Merancang Campuran Pakar Sebagai Pembelajar Multi-Tugas Modular:
@article{chen2022modsquad,
title={Mod-Squad: Designing Mixtures of Experts As Modular Multi-Task Learners},
author={Zitian Chen and Yikang Shen and Mingyu Ding and Zhenfang Chen and Hengshuang Zhao and Erik Learned-Miller and Chuang Gan},
journal={CVPR},
year={2023}
}
Kumpulan Data: Tugas
Contoh pengunduhan dari subset kecil
omnitools.download class_object class_scene depth_euclidean depth_zbuffer edge_occlusion edge_texture keypoints2d keypoints3d nonfixated_matches normal points principal_curvature reshading rgb segment_semantic segment_unsup2d segment_unsup25d --components taskonomy --subset tiny --dest ./taskonomy_tiny/ --connections_total 40 --agree --name [your name] --email [your email]
Silakan masukkan datanya ke ./data
Model default akan disimpan ke ./work_dir, log akan disimpan ke ./log_dir
Lingkungan: timm==0.3.2 pytorch==1.10.2
Instal modul MoE:
cd parallel_linear
pip3 install .
python -m torch.distributed.launch --nnodes=1 --nproc_per_node=2 --master_port 44875 main_mt.py
--batch_size 6
--epochs 100
--input_size 224
--blr 4e-4 --weight_decay 0.05
--warmup_epochs 10
--model mtvit_taskgate_att_mlp_base_MI_twice
--drop_path 0.1
--scaleup
--exp-name scaleup_mtvit_taskgate_att_mlp_base_MI_twice