When cd scene_seg,sh tool/train.sh s3dis pointnet2_paconv_cuda.I encountered this problem:RuntimeError: Ninja is required to load C++ extensions

[BaymaxOolong]

Traceback (most recent call last):
File "tool/train.py", line 19, in
from model.pointnet2.paconv import PAConv
File "/202221623116/PAConv/scene_seg/model/pointnet2/paconv.py", line 9, in
from lib.paconv_lib.functional import assign_score_withk as assign_score_cuda
File "/202221623116/PAConv/scene_seg/lib/paconv_lib/functional.py", line 1, in
from . import functions
File "/202221623116/PAConv/scene_seg/lib/paconv_lib/functions/init.py", line 1, in
from .assignscore import *
File "/202221623116/PAConv/scene_seg/lib/paconv_lib/functions/assignscore.py", line 4, in
from .. import src
File "/202221623116/PAConv/scene_seg/lib/paconv_lib/src/init.py", line 20, in
], build_directory=gpu_path, verbose=False)
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1092, in load
keep_intermediates=keep_intermediates)
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1303, in _jit_compile
is_standalone=is_standalone)
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1374, in _write_ninja_file_and_build_library
verify_ninja_availability()
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1430, in verify_ninja_availability
raise RuntimeError("Ninja is required to load C++ extensions")
RuntimeError: Ninja is required to load C++ extensions
[2023-12-03 03:10:43,879 INFO test_s3dis.py line 25 75059] arch: pointnet2_paconv_seg
base_lr: 0.05
block_size: 1.0
classes: 13
color_augment: 0.0
correlation_loss: True
correlation_loss_scale: 10.0
cuda: True
data_name: s3dis
data_root: dataset/s3dis
epochs: 100
evaluate: True
fea_dim: 6
hidden: [8, 16, 16]
ignore_label: 255
kernel_input: neighbor
lr_multidecay: True
m: 16
manual_seed: None
model_path: exp/s3dis/pointnet2_paconv_cuda/model/best_train.pth
momentum: 0.9
multiplier: 0.1
names_path: data/s3dis/s3dis_names.txt
no_transformation: False
norm_no_trans: True
num_point: 4096
paconv: [True, True, True, True, False, False, False, False]
print_freq: 100
resume: None
sample_rate: 1.0
save_folder: exp/s3dis/pointnet2_paconv_cuda/result/best_epoch/val5_0.5
save_freq: 1
save_path: exp/s3dis/pointnet2_paconv_cuda/model
score_input: ed7
split: val
start_epoch: 0
step_epoch: 30
stride_rate: 0.5
sync_bn: True
test_area: 5
test_batch_size: 8
test_gpu: [0]
test_list: dataset/s3dis/list/val5.txt
test_list_full: dataset/s3dis/list/val5_full.txt
test_workers: 4
train_batch_size: 16
train_batch_size_val: 8
train_full_folder: dataset/s3dis/trainval_fullarea
train_gpu: None
train_list: dataset/s3dis/list/train12346.txt
train_workers: 8
use_xyz: True
val_list: dataset/s3dis/list/val5.txt
weight: None
weight_decay: 0.0001
[2023-12-03 03:10:43,879 INFO test_s3dis.py line 28 75059] => creating model ...
[2023-12-03 03:10:43,879 INFO test_s3dis.py line 29 75059] Classes: 13
/202221623116/PAConv/scene_seg/lib/pointops/functions/pointops.py:13: UserWarning: Unable to load pointops_cuda cpp extension.
warnings.warn("Unable to load pointops_cuda cpp extension.")
Traceback (most recent call last):
File "/202221623116/PAConv/scene_seg/lib/pointops/functions/pointops.py", line 8, in
import pointops_cuda
ModuleNotFoundError: No module named 'pointops_cuda'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "tool/test_s3dis.py", line 198, in
main()
File "tool/test_s3dis.py", line 36, in main
from model.pointnet2.pointnet2_paconv_seg import PointNet2SSGSeg as Model
File "/202221623116/PAConv/scene_seg/model/pointnet2/pointnet2_paconv_seg.py", line 6, in
from model.pointnet2.pointnet2_paconv_modules import PointNet2FPModule
File "/202221623116/PAConv/scene_seg/model/pointnet2/pointnet2_paconv_modules.py", line 7, in
from lib.pointops.functions import pointops
File "/202221623116/PAConv/scene_seg/lib/pointops/functions/pointops.py", line 35, in
], build_directory=pointops_cuda_src, verbose=False)
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1092, in load
keep_intermediates=keep_intermediates)
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1303, in _jit_compile
is_standalone=is_standalone)
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1374, in _write_ninja_file_and_build_library
verify_ninja_availability()
File "/root/anaconda3/envs/paconv/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1430, in verify_ninja_availability
raise RuntimeError("Ninja is required to load C++ extensions")
RuntimeError: Ninja is required to load C++ extensions

How should I solve this problem.
Thanks!