{"author":"chenyu","author_email":"chenyu@fastmail.com","author_time":1711684610,"commit_time":1711684610,"committer":"GitHub","committer_email":"noreply@github.com","hash":"793ab0512eff3d3ec458e17497f6018f8eff99d4","message":"use ctypes to truncate float64 and float32 in uops (#3986)\n\nthis fixed the softmax.argmax bug for ops_python as the float is truncated to float32","parents":["101a0c683d58de6e874c2f761b5ee34a64eeae15"],"tree_hash":"63c9221f71fa002632ff050cab049a1a89ec615b"}