Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) by way of Textual Inversion (https://arxiv.org/abs/2208.01618) for Stable Diffusion (https://arxiv.org/abs/2112.10752). Tweaks focused on training faces, objects, and styles.
MIT License
3.2k
stars
554
forks
source link
Does this program support training based on muti-GPUS ? #203
i run it on a server with 3GPUs, it is working now, but it seems this program only used one GPU.
Fri May 3 12:09:24 2024
+-----------------------------------------------------------------------------+ | NVIDIA-SMI 520.61.05 Driver Version: 520.61.05 CUDA Version: 11.8 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 Tesla V100S-PCI... Off | 00000000:17:00.0 Off | 0 | | N/A 81C P0 60W / 250W | 22334MiB / 32768MiB | 100% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 1 Tesla V100S-PCI... Off | 00000000:4B:00.0 Off | 0 | | N/A 57C P0 44W / 250W | 32194MiB / 32768MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 2 Tesla V100S-PCI... Off | 00000000:CA:00.0 Off | 0 | | N/A 42C P0 26W / 250W | 8MiB / 32768MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 1482 G /usr/lib/xorg/Xorg 4MiB | | 0 N/A N/A 2075466 C python3 22326MiB | | 1 N/A N/A 1482 G /usr/lib/xorg/Xorg 4MiB | | 1 N/A N/A 2062468 C ...onda3/envs/SGA/bin/python 32186MiB | | 2 N/A N/A 1482 G /usr/lib/xorg/Xorg 4MiB | +-----------------------------------------------------------------------------+