VPP/sap.json at main · qizekun/VPP · GitHub

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
{
    "diffusion_config": {
        "T": 1000,
        "beta_0": 0.0001,
        "beta_T": 0.02
    },
    "pointnet_config": {
        "model_name": "shapenet_dpsr_zero_normal_upsample_10_noise_0_symmetry",
        "in_fea_dim": 4,
        "out_dim": 6,
        "include_t": false,
        "t_dim": 128,
        "model.use_xyz": true,
        "attach_position_to_input_feature": true,
        "include_abs_coordinate": true,
        "include_center_coordinate": true,
        "record_neighbor_stats": false,
        "bn_first": false,
        "bias": true,
        "res_connect": true,

        "point_upsample_factor": 5,
        "first_refine_coarse_points": false,
        "include_displacement_center_to_final_output": false,
        "output_scale_factor": 0.001,

        "include_class_condition": true,
        "num_class": 13,
        "class_condition_dim": 128,

        "bn": true,

        "include_local_feature": false,
        "include_global_feature": false,
        "global_feature_remove_last_activation": false,
        "pnet_global_feature_architecture": "[[4, 128, 256], [512, 1024]]",

        "attention_setting": {
            "use_attention_module": true,
            "attention_bn": true,
            "transform_grouped_feat_out": true,
            "last_activation": true,
            "add_attention_to_FeatureMapper_module": true
        },
        "architecture": {
            "npoint": "[1024, 256, 64, 16]",
            "radius": "[0.1, 0.2, 0.4, 0.8]",
            "neighbor_definition": "nn",
            "nsample": "[32, 32, 32, 32]",
            "feature_dim": "[32, 64, 128, 256, 512]",
            "mlp_depth": 3,
            "decoder_feature_dim": "[128, 128, 256, 256, 512]",
            "include_grouper": false,
            "decoder_mlp_depth": 2,
            "use_knn_FP": true,
            "K": 8
        },
        "condition_net_architecture": null,
        "feature_mapper_architecture": null
    },
    "train_config": {
        "task": "upsample",
        "dataset": "shapenet_psr_dataset",
        "root_directory": "exps/exp_shapenet_psr_generation/refine_and_upsampling_exps/no_normal_exps",
        "output_directory": "checkpoint",
        "tensorboard_directory": "tensorboard",
        "ckpt_iter": "max",
        "epochs_per_ckpt": 10,
        "iters_per_logging": 50,
        "n_epochs": 1000,
        "eval_start_epoch": 0,
        "eval_per_ckpt": 1,
        "learning_rate": 0.0002,
        "loss_type": "mse",
        "conditioned_on_cloud": false,
        "split_dataset_to_multi_gpus": true
    },
    "dpsr_config":{
        "grid_res": 128,
        "psr_sigma": 2,
        "psr_tanh": true,
        "mirror_before_upsampling": true,
        "only_original_points_split": false
    },
    "shapenet_psr_dataset_config": {
        "dataset": "shapenet_psr_dataset",
        "data_dir": "s3://zylyu_datasets/shapenet_psr",
        "categories": null,
        "npoints": 2048,
        "scale": 1,
        "batch_size": 32,
        "eval_batch_size": 32,
        "num_workers": 16,
        "load_psr": true,
        "augmentation": {"noise_magnitude": 0},
        "num_vis_samples": 32,
        "external_vis_dataset": null,
        "include_normals": false
    },
    "dist_config": {
        "dist_backend": "nccl",
        "CUDA_VISIBLE_DEVICES": null
    }
}