在 python tuple中使用ops.Concat 进行tensor拼接错误

mindspore.ops.Concat

【操作步骤&问题现象】

1、输入tensor变多之后 concat会报错 

代码：（rois shape变成(6,5)就会报错）

    def construct(self, input_cls, input_reg, rois):

        cls_vector = self.reshape(self.roi_align(input_cls, rois), (rois.shape[0], -1))

        reg_vector = self.reshape(self.roi_align(input_reg, rois), (rois.shape[0], -1))  # (N, 196, 7, 7)

        roi_cls_score = ()

        roi_reg_score = ()

        for per_roi in range(rois.shape[0]):

            cls_score = cls_vector[per_roi, 0::self.reg_stride].sum()/self.classes_num

            for per_class in range(1, self.classes_num):

                cls_score = mnp.append(cls_score, cls_vector[per_roi, per_class * self.bins_num::self.cls_stride].sum()/self.classes_num)

            reg_score = reg_vector[per_roi, 0::self.stride].sum() / 4

            for per_axis in range(1,4):

                reg_score = mnp.append(reg_score, reg_vector[per_roi, per_axis * self.bins_num::self.reg_stride].sum()/4)

            roi_cls_score += (self.softmax(cls_score),)

            roi_reg_score += (reg_score,)

        roi_cls_score = self.concat(roi_cls_score)

        roi_reg_score = self.concat(roi_reg_score)

        return roi_cls_score, roi_reg_score

import mindsporefrom mindspore import context

context.set_context(mode=context.GRAPH_MODE, device_target="Ascend", device_id=4)# roi = PSRoIAlign(7,7,0.0625,2)

input_cls = Tensor(np.random.rand(1, 3969, 80, 38), mindspore.float32)

input_reg = Tensor(np.random.rand(1, 196, 80, 38), mindspore.float32)

rois = Tensor(np.array([[0,1,1,2,2],[1,30,30,50,50],[2,30,30,50,50],[3,30,30,50,50]]), mindspore.float32)  //rois shape变成(6,5)就会报错

cls_score_map, reg_score_map = roi(input_cls,input_reg, rois)  # (N, 81*7*7, 7, 7)print(cls_score_map.shape)print(reg_score_map.shape)

rois shape为(5,5) 时返回 但(6,5)就会报错

具体报错请看附件

(ci3.7) [root@ascend90 fr]# python ps_roi_pool.py

[ERROR] ANALYZER(7820,ffff8983f630,python):2021-10-25-22:09:52.345.962 [mindspore/ccsrc/pipeline/jit/static_analysis/evaluator.cc:77] EnterStackFrame] Exceed function call depth limit 1000, (function call depth: 1001, simulate call depth: 999), please call 'context.set_context(max_call_depth=value)' to adjust this value.

[ERROR] DEBUG(7820,ffff8983f630,python):2021-10-25-22:09:52.346.059 [mindspore/ccsrc/debug/trace.cc:118] TraceGraphEval]

*******************************graph evaluate stack**********************************

[ERROR] DEBUG(7820,ffff8983f630,python):2021-10-25-22:09:52.375.526 [mindspore/ccsrc/debug/trace.cc:122] TraceGraphEval]

解答：

[ERROR] ANALYZER(7820,ffff8983f630,python):2021-10-25-22:09:52.345.962 [mindspore/ccsrc/pipeline/jit/static_analysis/evaluator.cc:77] EnterStackFrame] Exceed function call depth limit 1000, (function call depth: 1001, simulate call depth: 999), please call 'context.set_context(max_call_depth=value)' to adjust this value.

报错提示：超过函数调用深度限制1000，(函数调用深度:1001，模拟调用深度:999)，请调用context.set_context(max_call_depth=value)来调整这个值。