OccupyMars2025
OccupyMars2025
【队名】:OccupyMars2025 【序号】:1 【状态】:报名 【链接】:https://github.com/OccupyMars2025/Paddle
## "Floating-point exception", so change the dtype in all test cases to int64 to check whether the computation logic is right
## After changing the dtype in test cases to int64, the test cases still report "Floating-point exception", it is strange ## Check if the computation of ReshapeCooGradKernel is right ##...
## It seems that the example code in the docstring of python API will also be used in CI test ## It seems that CI test will check your comment...
// /* Caution : 这是原来的计算逻辑,我认为是 错误的, // 这里计算逻辑是: 原tensor的shape是 (10, 20, 30, 40, 50) // 一个非零元素的索引为 (1, 2, 3, 4, 5) // 进行transpose 后, tensor的shape 是 (30, 10, 50, 20,...
## Delete grad kernel test case to check whether the forward kernel is correct
## failed CI,@zhouwei25 之前的CI报错,我进行了一些修改,但是现在的CI报错,我有点看不懂了,能否请飞桨专家给一些方向性的意见。在python\paddle\fluid\tests\unittests\test_sparse_reshape_op.py 里面,我已经取消了backward检测,只保留了一个测试用例。
## 感谢解答,就是这个唯一的单侧都不能跑通,而且报错信息,我有点看不懂,尤其是下面这两个报错信息:
## 我先按照你的意见修改完后,再跑一遍CI,看是否还是报这两个错误
## 根据飞桨专家的意见修改后,还是有一样的错误,如下: ## 在nvidia官网上搜索得到如下解释: https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html ## 猜测可能的报错原因:forward coo cuda kernel 中 用int64_t * 指针给indices的 DenseTensor的元素赋值时,超出范围???