Created by: xiegegege

为使您的问题得到快速解决，在建立Issue前，请您先通过如下方式搜索是否有相似问题:【搜索issue关键字】【使用labels筛选】【官方文档】

建立issue时，为快速解决问题，请您根据使用情况给出如下信息：

• 标题：简洁、精准描述您的问题，例如“ssd 模型前置lstm报错  ”
• 版本、环境信息：    1）PaddlePaddle版本：paddle1.7.1    2）CPU：请提供CPU型号，MKL/OpenBlas/MKLDNN/等数学库的使用情况    3）GPU：CUDA9.0 CUDNN7.3 V100    4）系统环境：CentOS python3.6
• 模型信息    1）模型名称 2）使用数据集名称 3）使用算法名称 4）模型链接
• 复现信息：如为报错，请给出复现环境、复现步骤
• 问题描述：请详细描述您的问题，同步贴出报错信息、日志/代码关键片段 运行分割库模型unet的评估时，在batch_size=16时有如下报错： File "/opt/_internal/cpython-3.6.0/lib/python3.6/site-packages/paddle/fluid/framework.py", line 2525, in append_op attrs=kwargs.get("attrs", None)) File "/opt/_internal/cpython-3.6.0/lib/python3.6/site-packages/paddle/fluid/layer_helper.py", line 45, in append_op return self.main_program.current_block().append_op(*args, **kwargs) File "/opt/_internal/cpython-3.6.0/lib/python3.6/site-packages/paddle/fluid/layers/nn.py", line 1405, in conv2d "data_format": data_format, File "/ssd1/xiege/test/PaddleSeg/pdseg/models/libs/model_libs.py", line 127, in conv return fluid.layers.conv2d(*args, **kargs) File "/ssd1/xiege/test/PaddleSeg/pdseg/models/modeling/unet.py", line 39, in double_conv conv(data, out_ch, 3, stride=1, padding=1, param_attr=param_attr)) File "/ssd1/xiege/test/PaddleSeg/pdseg/models/modeling/unet.py", line 83, in encode data = double_conv(data, 64) File "/ssd1/xiege/test/PaddleSeg/pdseg/models/modeling/unet.py", line 128, in unet encode_data, short_cuts = encode(input) File "/ssd1/xiege/test/PaddleSeg/pdseg/models/model_builder.py", line 75, in seg_model logits = unet.unet(image, class_num) File "/ssd1/xiege/test/PaddleSeg/pdseg/models/model_builder.py", line 220, in build_model logits = seg_model(image, class_num) File "pdseg/eval.py", line 95, in evaluate test_prog, startup_prog, phase=ModelPhase.EVAL) File "pdseg/eval.py", line 174, in main evaluate(cfg, **args.dict) File "pdseg/eval.py", line 178, in main()

Error Message Summary:

Error: An error occurred here. There is no accurate error hint for this error yet. We are continuously in the process of increasing hint for this kind of error check. It would be helpful if you could inform us of how this conversion went by opening a github issue. And we will resolve it with high priority.

• New issue link: https://github.com/PaddlePaddle/Paddle/issues/new
• Recommended issue content: all error stack information [Hint: CUDNN_STATUS_NOT_SUPPORTED] at (/paddle/paddle/fluid/platform/cudnn_desc.h:155) [operator < conv2d > error] batch_size=1的时候能正常运行，batch_size=8的时候会有显存不足的报错，所以batch_size=16正常来说应该报显存不足的。。不知道为啥报这个错，麻烦帮忙看下吧～ #20087 看到之前有类似报错，但没解决 附配置文件： unet.txt