rknn-toolkit2 icon indicating copy to clipboard operation
rknn-toolkit2 copied to clipboard

RKNN: [00:09:08.335] Unsupport type bits 64 for depthwise

Open Naughty-Galileo opened this issue 2 years ago • 3 comments

print('--> Building model')
        ret = self.model.build(do_quantization=False)
        if ret != 0:
            print('Build model failed!')
            exit(ret)
        print('done')

加载模型时出现这个错误 E RKNN: [00:09:08.335] Unsupport type bits 64 for depthwise 已放弃 (核心已转储) 请问该怎么处理?

Naughty-Galileo avatar Jul 01 '23 16:07 Naughty-Galileo

请问您解决这个问题了吗,我也遇到了

czy-1234 avatar Dec 03 '23 11:12 czy-1234

同问, 楼上解决了吗?

walletiger avatar Apr 13 '24 13:04 walletiger

可以查看模型中是否有计算index的代码行的数据类型指定为了int64,如果有的话需要为计算过程为float,index过程为int64

Sicily-love avatar Aug 08 '24 13:08 Sicily-love