configs/pixart_alpha_config/PixArt_xl2_img1024_internalms.py

_base_ = ['../PixArt_xl2_internal.py']
data_root = 'data'
image_list_json = ['data_info.json',]

data = dict(type='InternalDataMS', root='InternData', image_list_json=image_list_json, transform='default_train', load_vae_feat=True)
image_size = 1024

# model setting
model = 'PixArtMS_XL_2'     # model for multi-scale training
fp32_attention = True
load_from = None
vae_pretrained = "output/pretrained_models/sd-vae-ft-ema"
aspect_ratio_type = 'ASPECT_RATIO_1024'         # base aspect ratio [ASPECT_RATIO_512 or ASPECT_RATIO_256]
multi_scale = True     # if use multiscale dataset model training
pe_interpolation = 2.0

# training setting
num_workers=10
train_batch_size = 12   # max 14 for PixArt-xL/2 when grad_checkpoint
num_epochs = 10 # 3
gradient_accumulation_steps = 1
grad_checkpointing = True
gradient_clip = 0.01
optimizer = dict(type='AdamW', lr=2e-5, weight_decay=3e-2, eps=1e-10)
lr_schedule_args = dict(num_warmup_steps=1000)
save_model_epochs=1
save_model_steps=2000

log_interval = 20
eval_sampling_steps = 200
work_dir = 'output/debug'
micro_condition = True