Spaces:
Running
Running
File size: 3,632 Bytes
966ae59 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 |
method: "svgdreamer"
image_size: 600 # canvas size
path_svg: ~ # if you want to load a svg file and train from it
num_stages: 1 # training stages, you can train x strokes, then freeze them and train another x strokes etc
skip_sive: True # optimize from scratch without SIVE init
color_init: 'rand' # if skip_live=True, then use color_init to init target_img
style: "iconography" # "iconography", "pixelart", "low-poly", "painting", "sketch", "ink"
# lr and optim
lr_stage_one: # SIVE stage
point: 1 # control points
width: 0.1 # stroke width
color: 0.01 # fill color and stroke color
bg: 0.01 # bg in render_warp
optim:
name: 'adam'
betas: [ 0.9, 0.9 ]
eps: 1e-6
lr_schedule: True # use lr_scheduler
schedule:
name: 'linear'
keep_ratio: 0.2
decay_ratio: 0.4
lr_stage_two: # VPSD stage
point: 1
width: 0.1
color: 0.01
bg: 0.01
lr_schedule: True # use lr_scheduler
optim:
name: 'adam'
betas: [ 0.9, 0.9 ]
eps: 1e-6
schedule:
name: 'cosine'
warmup_steps: 10
warmup_start_lr: 0.02
warmup_end_lr: 0.8
cosine_end_lr: 0.4
# primitives
num_paths: 256 # number of strokes
trainable_bg: False # set the background to be trainable
width: 3 # stroke width
num_segments: 4
segment_init: 'circle' # 'random'
radius: 20
coord_init: 'random' # 'random', 'naive', place the first control point
grid: 50 # divide the canvas into n grids
path_reinit: # reinitializing paths
use: True
freq: 100 # every 50 iterations
stop_step: 1000 # for VPSD fine-tuning
opacity_threshold: 0.05
area_threshold: 64
# diffusion
model_id: "sd21b" # sd14, sd15, sd21, sd21b, sdxl
ldm_speed_up: False
enable_xformers: True
gradient_checkpoint: False
cpu_offload: True
num_inference_steps: 50
guidance_scale: 7.5 # sdxl default 5.0
K: 4
lora_path: ~
# VPSD loss
guidance:
use: True
type: 'vpsd'
n_particle: 1 # 4, 8, 16
vsd_n_particle: 1 # the batch size of particles
particle_aug: False # do data enhancement for the input particles
num_iter: 2000 # total iterations
guidance_scale: 7.5 # CFG value
grad_scale: 1.0 # increase or decrease the gradient
grad_clip_val: ~ # eg: 10, clip the gradient of VPSD
t_range: [ 0.02, 0.98 ]
# 'randint': random time steps, this may have a more authentic style.
# 'max_0.5_900': annealing from 0.98 to 0.5 after 900 steps, this may have a more colorful results.
t_schedule: 'max_0.5_1000' # or 'randint'
# phi model config
phi_single: False # if False new an unet model to estimate noise
phi_model: 'lora' # 'lora', 'unet_simple'
use_attn_scale: ${x.guidance.phi_single} # use lora_attn_scale or not
lora_attn_scale: 1.0 # the scale of the attn based lora layer
phi_guidance_scale: 1.0
phi_t: False # different t for phi fine-tuning
phi_update_step: 1 # enable multi-update phi model or not
phi_lr: 0.0001 # learning rate of phi model
phi_scheduler: 'ddim' # 'dpm-solver'
phi_n_particle: 1 # the batch size of phi_model
# ReFL config
phi_ReFL: False # enable reward feed back learning
n_phi_sample: 1 # number of samples used in ReFL
phi_sample_step: 200 # the phi log step
phi_infer_step: 50 # the phi num_inference_steps
# phi model optim
phi_optim:
name: 'adamw'
betas: [ 0.9, 0.999 ]
eps: 1e-8
weight_decay: ~ # 1e-5
# phi model lr learning schedule
phi_schedule:
use: False
name: 'cosine'
warmup_steps: 50
warmup_start_lr: 0.00001
warmup_end_lr: 0.0001
total_step: 800
cosine_end_lr: 0.0001
# reward model
reward_path: './checkpoint/ImageReward'
# xing loss for closed-form paths
xing_loss:
use: False
weight: 0.01
|