自定义构建无法与新的 RTX 5000 系列一起使用

{"severity": "ERROR", "message": "异常在 /status [GET] 中", "request": {"method": "GET", "path": "/status", "filename": "", "api_key": "", "remote_addr": "172.19.0.5"}, "logger": "embedding-calculator", "module": "app", "traceback": "追踪（最近一次调用为上一行）：\n 文件 "/usr/local/lib/python3.8/dist-packages/flask/app.py\”，行 2447，在 wsgi_app 中\n response = self.full_dispatch_request()\n 文件 "/usr/local/lib/python3.8/dist-packages/flask/app.py\”，行 1945，在 full_dispatch_request 中\n self.try_trigger_before_first_request_functions()\n 文件 "/usr/local/lib/python3.8/dist-packages/flask/app.py\”，行 1993，在 try_trigger_before_first_request_functions 中\n func()\n 文件 "/app/ml/./src/_endpoints.py\”，行 52，在 init_model 中\n detector(\n 文件 "/app/ml/./src/services/facescan/plugins/mixins.py\”，行 46，在 call 中\n faces = self._fetch_faces(img, det_prob_threshold)\n 文件 "/app/ml/./src/services/facescan/plugins/mixins.py\”，行 53，在 _fetch_faces 中\n boxes = self.find_faces(img, det_prob_threshold)\n 文件 "/app/ml/./src/services/facescan/plugins/insightface/insightface.py\”，行 103，在 find_faces 中\n model = self._detection_model\n 文件 "/usr/local/lib/python3.8/dist-packages/cached_property.py\”，行 36，在 get 中\n value = obj.dict[self.func.name] = self.func(obj)\n 文件 "/app/ml/./src/services/facescan/plugins/insightface/insightface.py\”，行 80，在 _detection_model 中\n model.prepare(ctx_id=self._CTX_ID, nms=self._NMS)\n 文件 "/usr/local/lib/python3.8/dist-packages/insightface/app/face_analysis.py\”，行 32，在 prepare 中\n self.det_model.prepare(ctx_id, nms)\n 文件 "/usr/local/lib/python3.8/dist-packages/insightface/model_zoo/face_detection.py\”，行 223，在 prepare 中\n out = model.get_outputs()[0].asnumpy()\n 文件 "/usr/local/lib/python3.8/dist-packages/mxnet/ndarray/ndarray.py\”，行 2568，在 asnumpy 中\n check_call(_LIB.MXNDArraySyncCopyToCPU(\n 文件 "/usr/local/lib/python3.8/dist-packages/mxnet/base.py\”，行 246，在 check_call 中\n raise get_last_ffi_error()\nmxnet.base.MXNetError: 追踪（最近一次调用为上一行）：\n 文件 "/work/mxnet/include/mshadow/././././cuda/tensor_gpu-inl.cuh\”，行 128\n名称: 检查失败: err == cudaSuccess (209 与 0 比较): MapPlanKernel ErrStr:no kernel image is available for execution on the device\n", "build_version": "dev"} Traceback (最近一次调用为上一行): File "/usr/local/lib/python3.8/dist-packages/flask/app.py", line 2447, in wsgi_app response = self.full_dispatch_request() File "/usr/local/lib/python3.8/dist-packages/flask/app.py", line 1945, in full_dispatch_request …