Instructions to use torch-pointcloud/act.scanobjectnn-objonly.runpei-dong with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- torch-pointcloud
How to use torch-pointcloud/act.scanobjectnn-objonly.runpei-dong with torch-pointcloud:
import torch_pointcloud as tp model = tp.create_model("act.scanobjectnn-objonly.runpei-dong", pretrained=True) - Notebooks
- Google Colab
- Kaggle
File size: 927 Bytes
f7a3234 | 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 | {
"library_name": "torch-pointcloud",
"name": "act.scanobjectnn-objonly.runpei-dong",
"task": "classification",
"hparams": {
"in_channels": 0,
"num_classes": 15,
"embed_dim": 384,
"depth": 12,
"num_heads": 6,
"num_groups": 128,
"group_size": 32,
"encoder_local_channels": [
128,
256
],
"encoder_global_channels": [
512
],
"pos_embed_channels": [
128
],
"drop_path": 0.1,
"dropout": 0.5,
"act": "gelu",
"spatial_dim": 3
},
"input_keys": [
"x",
"pos",
"batch"
],
"dataset": "scanobjectnn-objonly",
"metrics": {
"OA": 91.22,
"mAcc": 90.26
},
"classes": [
"bag",
"bin",
"box",
"cabinet",
"chair",
"desk",
"display",
"door",
"shelf",
"table",
"bed",
"pillow",
"sink",
"sofa",
"toilet"
],
"author": "runpei-dong",
"license": "MIT"
}
|