feat: upload Maya lora model
Browse files- Maya_config/config_file.toml +18 -6
Maya_config/config_file.toml
CHANGED
@@ -6,17 +6,17 @@ vae = "/content/vae/anime.vae.pt"
|
|
6 |
|
7 |
[additional_network_arguments]
|
8 |
no_metadata = false
|
9 |
-
unet_lr = 0.
|
10 |
-
text_encoder_lr =
|
11 |
network_module = "networks.lora"
|
12 |
-
network_dim =
|
13 |
-
network_alpha =
|
14 |
network_train_unet_only = false
|
15 |
network_train_text_encoder_only = false
|
16 |
|
17 |
[optimizer_arguments]
|
18 |
optimizer_type = "AdamW8bit"
|
19 |
-
learning_rate = 0.
|
20 |
max_grad_norm = 1.0
|
21 |
lr_scheduler = "cosine_with_restarts"
|
22 |
lr_warmup_steps = 0
|
@@ -24,13 +24,25 @@ lr_scheduler_num_cycles = 3
|
|
24 |
|
25 |
[dataset_arguments]
|
26 |
debug_dataset = false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
27 |
|
28 |
[training_arguments]
|
29 |
output_dir = "/content/LoRA/output"
|
30 |
output_name = "Maya"
|
31 |
save_precision = "fp16"
|
32 |
save_every_n_epochs = 3
|
33 |
-
train_batch_size =
|
34 |
max_token_length = 225
|
35 |
mem_eff_attn = false
|
36 |
xformers = true
|
|
|
6 |
|
7 |
[additional_network_arguments]
|
8 |
no_metadata = false
|
9 |
+
unet_lr = 0.0001
|
10 |
+
text_encoder_lr = 5e-5
|
11 |
network_module = "networks.lora"
|
12 |
+
network_dim = 32
|
13 |
+
network_alpha = 16
|
14 |
network_train_unet_only = false
|
15 |
network_train_text_encoder_only = false
|
16 |
|
17 |
[optimizer_arguments]
|
18 |
optimizer_type = "AdamW8bit"
|
19 |
+
learning_rate = 0.0001
|
20 |
max_grad_norm = 1.0
|
21 |
lr_scheduler = "cosine_with_restarts"
|
22 |
lr_warmup_steps = 0
|
|
|
24 |
|
25 |
[dataset_arguments]
|
26 |
debug_dataset = false
|
27 |
+
in_json = "/content/LoRA/meta_lat.json"
|
28 |
+
train_data_dir = "/content/LoRA/train_data"
|
29 |
+
dataset_repeats = 10
|
30 |
+
shuffle_caption = true
|
31 |
+
keep_tokens = 0
|
32 |
+
resolution = "512,512"
|
33 |
+
caption_dropout_rate = 0
|
34 |
+
caption_tag_dropout_rate = 0
|
35 |
+
caption_dropout_every_n_epochs = 0
|
36 |
+
color_aug = false
|
37 |
+
token_warmup_min = 1
|
38 |
+
token_warmup_step = 0
|
39 |
|
40 |
[training_arguments]
|
41 |
output_dir = "/content/LoRA/output"
|
42 |
output_name = "Maya"
|
43 |
save_precision = "fp16"
|
44 |
save_every_n_epochs = 3
|
45 |
+
train_batch_size = 4
|
46 |
max_token_length = 225
|
47 |
mem_eff_attn = false
|
48 |
xformers = true
|