大量动态shape报错，参考之前别人提的issue，已经把reshape(-1)这种直接动态推理维度的部分修改，也参考哈工大同学提的Top_k的问题和解决方法，减少了很多报错。目前还有大量找不到原因的unknown shape。请问，能不能帮我看看具体原因是什么？

silu 运行超慢 整个网络完结到达 1s多 layer { name: "conv1" type: "Convolution" #bottom: "yolov5_blob" bottom: "blob1" top: "conv_blob1" convolution_param

teacher_forcing_ratio # get the highest predicted token from our predictions top1 = output.argmax(1) # if teacher forcing, use actual next token

parameter_map["auto_tune_mode"].s = tf.compat.as_bytes("RL,GA") custom_op.parameter_map["precision_mode"].s = tf.compat.as_bytes("allow_mix_precision")

## Customer's Demands on Product/Solution ## Gaps to Fill on Product/Solution - Preliminary Discussion - Acceptance Standards #

ile_mode"].b = True custom_op.parameter_map["precision_mode"].s = tf.compat.as_bytes("allow_mix_precision") sess_config.graph_options

from net_s3fd import s3fd import torch path_to_detector = 's3fd.pth' model_weights = torch.load(path_to_detector) model = s3fd() model

一、问题现象（附报错日志上下文）： npu-smi info在所有卡上面的显存占用和gpu占用率在启动模型前后没有显著变化，使用top命令可以看到该模型的python命令占用一个cpu的核心100%，训练速度非常慢 目前不知道怎么把代码设置放到GPU上面跑 二、软件版本:

shown below; layer { bottom: "layer16-conv" top: "yolo1_coords" top: "yolo1_obj" top: "yolo1_classes" name: "yolo1" type: "Yolo"

zas9TCCcDa-TPn3R85sR8Tm3jlhF52bITnZ27THc25cqw87s1xi6Gv82F9rjMCLeHn_gtA-ueu8top52GlkwXguecu5gyqS79PPDy10dRiW9L0Hgabm0ZgEQ5nzQ2w

pooling layer top_model = base_model.output top_model = GlobalAveragePooling2D()(top_model) # or just flatten the layers # top_model =