ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from——报错简记

numpy 版本导致报错yolov5PyTorch to ONNX and TorchScript formats 代码运行：python export.py --weights yolov5s.pt --img 640 --batch 1由于一些库使用 pip 安装之后，numpy 版本发生冲突，报错如下：ValueError: numpy.ndarray size changed, may i

墨理学AI

36269人浏览 · 2021-08-09 17:25:05

墨理学AI · 2021-08-09 17:25:05 发布

文章目录

📔 numpy 版本导致报错

yolov5 PyTorch to ONNX and TorchScript formats 代码运行：

python export.py --weights yolov5s.pt --img 640 --batch 1

由于一些库使用 pip 安装之后，numpy 版本发生冲突，报错如下：

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

核心原因是 numpy 版本与当前一些库不匹配
通常是因为安装的 numpy 版本过低，把 numpy 更新到最新版本即可

解决方法如下：

pip install --upgrade numpy

📕 CoreML: export failure: Unsupported numpy type: float32

ONNX: export failure: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking arugment for argument index in method wrapper_index_select)
scikit-learn version 0.24.2 is not supported. Minimum required version: 0.17. Maximum required version: 0.19.2. Disabling scikit-learn conversion API.

CoreML: starting export with coremltools 4.1...
Tuple detected at graph output. This will be flattened in the converted model.
Converting graph.
Adding op '1' of type const

CoreML: export failure: Unsupported numpy type: float32