报错：float64类型不能被安全地转换成整型。
报错原因：当前版本的numpy不支持使用浮点数作为步长，需要降低numpy版本。

Traceback (most recent call last):
  File "./pytorch/train.py", line 662, in <module>
    fire.Fire()
  File "/home/autopilot/anaconda3/envs/pointpillars/lib/python3.7/site-packages/fire/core.py", line 141, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/home/autopilot/anaconda3/envs/pointpillars/lib/python3.7/site-packages/fire/core.py", line 471, in _Fire
    target=component.__name__)
  File "/home/autopilot/anaconda3/envs/pointpillars/lib/python3.7/site-packages/fire/core.py", line 681, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "./pytorch/train.py", line 417, in train
    raise e
  File "./pytorch/train.py", line 405, in train
    result = get_coco_eval_result(gt_annos, dt_annos, class_names)
  File "../second/utils/eval.py", line 911, in get_coco_eval_result
    gt_annos, dt_annos, current_classes, overlap_ranges, compute_aos)
  File "../second/utils/eval.py", line 719, in do_coco_style_eval
    min_overlaps[:, i, j] = np.linspace(*overlap_ranges[:, i, j])
  File "<__array_function__ internals>", line 6, in linspace
  File "/home/autopilot/anaconda3/envs/pointpillars/lib/python3.7/site-packages/numpy/core/function_base.py", line 120, in linspace
    num = operator.index(num)
TypeError: 'numpy.float64' object cannot be interpreted as an integer

可以通过pip list或者pip show numpy来查看当前numpy版本
解决办法:

pip install -U numpy==1.17.0 -i https://pypi.tuna.tsinghua.edu.cn/simple