AncTe/ginka-generator
1
0
Fork 0
mirror of https://github.com/unanmed/ginka-generator.git synced 2026-05-14 04:41:12 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

chore: 参数调整

Browse Source
This commit is contained in:
unanmed 2026-04-27 15:20:28 +08:00
parent abbad781ab
commit 638c107884
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
ginka/train_vq.py
View File

@ -38,14 +38,14 @@ from shared.image import matrix_to_image_cv
BATCH_SIZE = 64
NUM_CLASSES = 16
MASK_TOKEN = 15
GENERATE_STEP = 12 # 推理时 MaskGIT 迭代步数
GENERATE_STEP = 18 # 推理时 MaskGIT 迭代步数
MAP_SIZE = 13 * 13
MAP_H = MAP_W = 13
LABEL_SMOOTHING = 0.0
# VQ-VAE 超参
VQ_L = 64 # summary token 数量（即 z 的序列长度）
VQ_K = 1 # codebook 大小
VQ_L = 16 # summary token 数量（即 z 的序列长度）
VQ_K = 2 # codebook 大小
VQ_D_Z = 64 # codebook 嵌入维度
VQ_D_MODEL= 128
VQ_NHEAD = 4
@ -55,12 +55,12 @@ VQ_BETA = 0.25 # commit loss 权重
VQ_GAMMA = 0.1 # entropy loss 权重
# MaskGIT 超参
MG_D_MODEL = 192
MG_D_MODEL = 256
MG_NHEAD = 8
MG_LAYERS = 4
MG_DIM_FF = 512
MG_Z_DROPOUT = 0.15 # 训练时以此概率把 z 替换为随机噪声
MG_STRUCT_DROPOUT= 0.15 # 训练时以此概率将结构标签替换为 null（无条件占位）
MG_DIM_FF = 1024
MG_Z_DROPOUT = 0.1 # 训练时以此概率把 z 替换为随机噪声
MG_STRUCT_DROPOUT= 0.1 # 训练时以此概率将结构标签替换为 null（无条件占位）
# 验证时对每条样本额外采样的 z 数量（0 = 只用真实 z）
N_Z_SAMPLES = 3