Zixianyan's workspace
Runs
2
State
Notes
User
Tags
Created
Runtime
Sweep
adam_beta1
adam_beta2
adam_epsilon
adam_weight_decay
allow_tf32
caption_column
checkpointing_steps
conditioning_image_column
dataloader_num_workers
dataset_name
enable_xformers_memory_efficient_attention
gradient_accumulation_steps
gradient_checkpointing
image_column
learning_rate
logging_dir
lr_num_cycles
lr_power
lr_scheduler
lr_warmup_steps
max_grad_norm
max_train_steps
num_train_epochs
num_validation_images
output_dir
pretrained_model_name_or_path
proportion_empty_prompts
push_to_hub
report_to
resolution
scale_lr
set_grads_to_none
tracker_project_name
train_batch_size
use_8bit_adam
validation_steps
loss
lr
Finished
-
zixianyan
18h 4m 30s
-
0.9
0.999
1.0000e-8
0.01
false
caption
1000
grayscale_image
0
/root/autodl-tmp/data/grayscale_image_512_100k
true
5
false
image
0.00001
logs
1
1
constant
500
1
10002
3
4
/root/autodl-tmp/model/ckpt512_100k
/root/autodl-tmp/model/diffusion-v1-5
0
false
wandb
512
false
false
train_controlnet
6
false
500
0.070684
0.00001
Finished
-
zixianyan
6h 44s
-
0.9
0.999
1.0000e-8
0.01
false
caption
1000
grayscale_image
0
/root/autodl-tmp/data/grayscale_image_512_100k
true
4
false
image
0.00001
logs
1
1
constant
500
1
4167
1
4
/root/autodl-tmp/model/ckpt512_100k
/root/autodl-tmp/model/diffusion-v1-5
0
false
wandb
512
false
false
train_controlnet
6
false
500
0.13848
0.00001
1-2
of 2