| { |
| "models": { |
| "denoiser": { |
| "name": "ElasticSLatFlowModel", |
| "args": { |
| "resolution": 32, |
| "in_channels": 32, |
| "out_channels": 32, |
| "model_channels": 1536, |
| "cond_channels": 1024, |
| "num_blocks": 30, |
| "num_heads": 12, |
| "mlp_ratio": 5.3334, |
| "pe_mode": "rope", |
| "share_mod": true, |
| "initialization": "scaled", |
| "qk_rms_norm": true, |
| "qk_rms_norm_cross": true |
| }, |
| "lora": { |
| "r": 8, |
| "alpha": 16 |
| }, |
| "projection": { |
| "image_size": 512 |
| } |
| } |
| }, |
| "dataset": { |
| "name": "SS3DMSLatShape", |
| "args": { |
| "resolution": 512, |
| "image_size": 512, |
| "max_num_cond_views": 16, |
| "min_aesthetic_score": 4.5, |
| "min_tokens": 3000, |
| "normalization": { |
| "mean": [ |
| 0.781296, |
| 0.018091, |
| -0.495192, |
| -0.558457, |
| 1.06053, |
| 0.093252, |
| 1.518149, |
| -0.933218, |
| -0.732996, |
| 2.604095, |
| -0.118341, |
| -2.143904, |
| 0.495076, |
| -2.179512, |
| -2.130751, |
| -0.996944, |
| 0.261421, |
| -2.217463, |
| 1.260067, |
| -0.150213, |
| 3.790713, |
| 1.481266, |
| -1.046058, |
| -1.523667, |
| -0.059621, |
| 2.22078, |
| 1.621212, |
| 0.87723, |
| 0.567247, |
| -3.175944, |
| -3.186688, |
| 1.578665 |
| ], |
| "std": [ |
| 5.972266, |
| 4.706852, |
| 5.44501, |
| 5.209927, |
| 5.32022, |
| 4.547237, |
| 5.020802, |
| 5.444004, |
| 5.226681, |
| 5.683095, |
| 4.831436, |
| 5.286469, |
| 5.652043, |
| 5.367606, |
| 5.525084, |
| 4.730578, |
| 4.805265, |
| 5.124013, |
| 5.530808, |
| 5.619001, |
| 5.10393, |
| 5.41767, |
| 5.269677, |
| 5.547194, |
| 5.634698, |
| 5.235274, |
| 6.110351, |
| 5.511298, |
| 6.237273, |
| 4.879207, |
| 5.347008, |
| 5.405691 |
| ] |
| }, |
| "pretrained_slat_dec": "microsoft/TRELLIS.2-4B/ckpts/shape_dec_next_dc_f16c32_fp16" |
| } |
| }, |
| "trainer": { |
| "name": "CameraConditionedSparseFlowMatchingCFGTrainer", |
| "args": { |
| "max_steps": 200000, |
| "batch_size_per_gpu": 32, |
| "batch_split": 8, |
| "num_workers_per_gpu": 8, |
| "optimizer": { |
| "name": "AdamW", |
| "args": { |
| "lr": { |
| "lora": 3e-05, |
| "new": 0.0001, |
| "old": 3e-05 |
| }, |
| "weight_decay": { |
| "lora": 0.0, |
| "new_weights": 0.01, |
| "old": 0.01 |
| }, |
| "betas": [ |
| 0.9, |
| 0.95 |
| ], |
| "eps": 1e-08 |
| } |
| }, |
| "lr_scheduler": { |
| "name": "LinearWarmupLRScheduler", |
| "args": { |
| "warmup_steps": 2000 |
| } |
| }, |
| "ema_rate": [ |
| 0.9995 |
| ], |
| "snapshot_batch_size": 4, |
| "mix_precision_mode": "amp", |
| "mix_precision_dtype": "bfloat16", |
| "elastic": { |
| "name": "LinearMemoryController", |
| "args": { |
| "target_ratio": 0.75, |
| "max_mem_ratio_start": 0.5 |
| } |
| }, |
| "grad_clip": { |
| "name": "AdaptiveGradClipper", |
| "args": { |
| "max_norm": 1.0, |
| "clip_percentile": 95 |
| } |
| }, |
| "finetune_ckpt": { |
| "denoiser": "checkpoints/shape_gen/ckpts/shape_slat.pt" |
| }, |
| "i_print": 10, |
| "i_val": 500, |
| "i_log": 10, |
| "i_sample": 2000, |
| "i_save": 2000, |
| "p_uncond": 0.1, |
| "t_schedule": { |
| "name": "uniform", |
| "args": {} |
| }, |
| "sigma_min": 1e-05, |
| "image_cond_model": { |
| "name": "DinoV3FeatureExtractor", |
| "args": { |
| "model_name": "camenduru/dinov3-vitl16-pretrain-lvd1689m", |
| "image_size": 512 |
| } |
| }, |
| "max_val_samples": 5, |
| "num_snapshot_samples": 2, |
| "skip_init_snapshot": true |
| } |
| } |
| } |
