dataset: ucf101_img
data_path: train_datasets/UCF-101_tiny
frame_data_txt: train_datasets/imageslist.txt
pretrained_model_path: pretrained_models
results_dir: ./results_img
pretrained: null
model: LatteIMG-XL/2
num_frames: 16
image_size: 256
num_sampling_steps: 250
frame_interval: 3
fixed_spatial: false
attention_bias: true
learn_sigma: true
extras: 2
save_ceph: true
use_image_num: 8
learning_rate: 0.0001
ckpt_every: 10000
clip_max_norm: 0.1
start_clip_iter: 100000
local_batch_size: 4
max_train_steps: 1000000
global_seed: 3407
num_workers: 8
log_every: 50
lr_warmup_steps: 0
resume_from_checkpoint: null
gradient_accumulation_steps: 1
num_classes: 101
use_compile: false
mixed_precision: false
enable_xformers_memory_efficient_attention: false
gradient_checkpointing: false