RuntimeError:“topk_cpu“not implemented for ‘Half‘-CFANZ编程社区

RuntimeError:“topk_cpu“not implemented for ‘Half‘

kmoon_b426

阅读 180

2022-04-30

问题：torch.topk (cpu)不支持半精度（FP16）计算

解决：model对象用“float()”进行转换后再操作

相关推荐
SDKB英文
 RuntimeError: log_vml_cpu not implemented for 'Long'
SDKB英文 75 0 0
Mhhao
 RuntimeError: exp_vml_cpu not implemented for 'Long'
Mhhao 70 0 0
九点韶留学
 RuntimeError: expected device cuda:0 and dtype Float but got device cuda:0 and dtype Half
九点韶留学 91 0 0
八怪不姓丑
 【pytorch错误】：Pytorch RuntimeError: “host_softmax” not implemented for 'torch.cuda.LongTensor'
八怪不姓丑 71 0 0
才德的女子
 RuntimeError: “nll_loss_forward_reduce_cuda_kernel_2d_index“ not implemented for ‘Int‘
才德的女子 144 0 0
思考的鸿毛
 Select Half
思考的鸿毛 76 0 0
春意暖洋洋
 异常解决（一）-- RuntimeError: expected device cpu but got device cuda:0
春意暖洋洋 143 0 0
慕容冲_a4b8
 spark rdd topK sort spark获取topK排序
慕容冲_a4b8 74 0 0
思考的鸿毛
 RuntimeError: Expected object of device type cuda but got device type cpu for argument ……
思考的鸿毛 83 0 0
云岭逸人
 setCurrentRole not implemented in FallbackHiveAuthorizer
云岭逸人 45 0 0

精彩评论（0）