TF1.15.5训练ssd_mobilenet_v1_coco报错tensorflow.python.framework.errors_impl.InvalidArgumentError

于 2021-07-13 14:10:47 发布
阅读量1.4k 收藏
点赞数 2
分类专栏： Bug List 文章标签： assertionFailed tensorflow报错
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/BAR_WORKSHOP/article/details/118667735
版权

问题记录

  • Ubuntu 20.04 LTS
  • nvidia-smi 465.27 cuda11.3
  • tensorflow-gpu 1.15.5
     
tensorflow.python.framework.errors_impl.InvalidArgumentError:  assertion failed: [[0.919921875][0.68359375][0.517578125]...] [[0.814453125][0.587890625][0.42578125]...]
	 [[{{node Assert_1/AssertGuard/Assert}}]]
	 [[IteratorGetNext]]

完整报错如下：

Traceback (most recent call last):
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call
    return fn(*args)
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 1349, in _run_fn
    return self._call_tf_sessionrun(options, feed_dict, fetch_list,
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 1441, in _call_tf_sessionrun
    return tf_session.TF_SessionRun_wrapper(self._session, options, feed_dict,
tensorflow.python.framework.errors_impl.InvalidArgumentError: {{function_node Dataset_map_transform_and_pad_input_data_fn_433}} assertion failed: [[0.919921875][0.68359375][0.517578125]...] [[0.814453125][0.587890625][0.42578125]...]
	 [[{{node Assert_1/AssertGuard/Assert}}]]
	 [[IteratorGetNext]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "model_main.py", line 108, in <module>
    tf.app.run()
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/platform/app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "/home/haotian/.local/lib/python3.8/site-packages/absl/app.py", line 312, in run
    _run_main(main, args)
  File "/home/haotian/.local/lib/python3.8/site-packages/absl/app.py", line 258, in _run_main
    sys.exit(main(argv))
  File "model_main.py", line 104, in main
    tf.estimator.train_and_evaluate(estimator, train_spec, eval_specs[0])
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/training.py", line 473, in train_and_evaluate
    return executor.run()
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/training.py", line 613, in run
    return self.run_local()
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/training.py", line 710, in run_local
    self._estimator.train(
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 370, in train
    loss = self._train_model(input_fn, hooks, saving_listeners)
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1161, in _train_model
    return self._train_model_default(input_fn, hooks, saving_listeners)
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1193, in _train_model_default
    return self._train_with_estimator_spec(estimator_spec, worker_hooks,
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1494, in _train_with_estimator_spec
    _, loss = mon_sess.run([estimator_spec.train_op, estimator_spec.loss])
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/training/monitored_session.py", line 750, in run
    return self._sess.run(
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/training/monitored_session.py", line 1255, in run
    return self._sess.run(
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/training/monitored_session.py", line 1360, in run
    raise six.reraise(*original_exc_info)
  File "/home/haotian/.local/lib/python3.8/site-packages/six.py", line 703, in reraise
    raise value
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/training/monitored_session.py", line 1345, in run
    return self._sess.run(*args, **kwargs)
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/training/monitored_session.py", line 1413, in run
    outputs = _WrappedSession.run(
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/training/monitored_session.py", line 1176, in run
    return self._sess.run(*args, **kwargs)
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 955, in run
    result = self._run(None, fetches, feed_dict, options_ptr,
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 1179, in _run
    results = self._do_run(handle, final_targets, final_fetches,
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 1358, in _do_run
    return self._do_call(_run_fn, feeds, fetches, targets, options,
  File "/home/haotian/.local/lib/python3.8/site-packages/tensorflow_core/python/client/session.py", line 1384, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError:  assertion failed: [[0.919921875][0.68359375][0.517578125]...] [[0.814453125][0.587890625][0.42578125]...]
	 [[{{node Assert_1/AssertGuard/Assert}}]]
	 [[IteratorGetNext]]

 

解决方法

检查数据集标签，尤其是自己做了数据增强的，很可能是xml损坏了

  • 检查 xmin、xmax、ymin、ymax 有没有负值；
     
  • 检查 xmin、xmax、ymin、ymax 是否超出图像范围，可以在生成tfrecords的csv_to_tfrecords.py里面找到如下位置，修改成这样：
for index, row in group.object.iterrows():
        # xmins.append(row['xmin'] / width)
        # xmaxs.append(row['xmax'] / width)
        # ymins.append(row['ymin'] / height)
        # ymaxs.append(row['ymax'] / height)

        xmn = row['xmin'] / width
        if xmn < 0.0:
            xmn = 0.0
        elif xmn > 1.0:
            xmn = 1.0
        xmins.append(xmn)

        xmx = row['xmax'] / width
        if xmx < 0.0:
            xmx = 0.0
        elif xmx > 1.0:
            xmx = 1.0
        xmaxs.append(xmx)

        ymn = row['ymin'] / height
        if ymn < 0.0:
            ymn = 0.0
        elif ymn > 1.0:
            ymn = 1.0
        ymins.append(ymn)

        ymx = row['ymax'] / height
        if ymx < 0.0:
            ymx = 0.0
        elif ymx > 1.0:
            ymx = 1.0
        ymaxs.append(ymx)

        classes_text.append(row['class'].encode('utf8'))
        classes.append(class_text_to_int(row['class']))

 

  • 检查 xmin<xmax、ymin<ymax 是否满足，这里有一段修复xmin<xmax、ymin<ymax的程序：
import os
import shutil
import xml.etree.ElementTree as ET

xml_src_dir = "..."    #xml原位置
xml_dst_dir = "..."    #修改后xml的保存位置

xml_files = os.listdir(xml_src_dir)
for xml_src_file in xml_files:
    if xml_src_file.endswith('.xml'):
        shutil.copy(xml_src_dir + xml_src_file, xml_dst_dir)
        xml_dst = ET.parse(xml_dst_dir + xml_src_file)
        for obj in xml_dst.iter('object'):
            xmin=float(obj.find('bndbox').find('xmin').text)
            ymin=float(obj.find('bndbox').find('ymin').text)
            xmax=float(obj.find('bndbox').find('xmax').text)
            ymax=float(obj.find('bndbox').find('ymax').text)
            xmn = min(xmin,xmax)
            ymn = min(ymin,ymax)
            xmx = max(xmin,xmax)
            ymx = max(ymin,ymax)
            xmn = int(xmn)
            ymn = int(ymn)
            xmx = int(xmx)
            ymx = int(ymx)
            obj.find('bndbox').find('xmin').text=str(xmn)
            obj.find('bndbox').find('ymin').text=str(ymn)
            obj.find('bndbox').find('xmax').text=str(xmx)
            obj.find('bndbox').find('ymax').text=str(ymx)
        xml_dst.write(xml_dst_dir + xml_src_file)
tiiaan
关注
  • 2
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
kbd-legacy-1.15.5-16.el7_9.noarch.rpm
01-08
官方离线安装包，测试可用。使用rpm -ivh [rpm完整包名] 进行安装
Python库 | setuptools_scm-1.15.5.tar.gz
04-14
资源分类：Python库 所属语言：Python 资源全名：setuptools_scm-1.15.5.tar.gz 资源来源：官方 安装方法：https://lanzao.blog.csdn.net/article/details/101784059
kbd-1.15.5-16.el7_9.x86_64.rpm
01-08
官方离线安装包，测试可用。使用rpm -ivh [rpm完整包名] 进行安装
coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip
07-28
tensorflow lite example 中的object_detection资源文件，下载后放到app下的build下的zips中，同时修改download_model.gradle文件，屏蔽task downloadZipFile,并把task downloadAndUnzipFile 中的 from zipTree(downloadZipFile.dest) 修改为 from zipTree(new File(buildDir, 'zips/coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip')) 当然如果你没有积分，找到download_model.gradle中task downloadZipFile中的链接地址，使用迅雷下载后做上述操作也是可以的
kbd-misc-1.15.5-16.el7_9.noarch.rpm
01-08
官方离线安装包，测试可用。使用rpm -ivh [rpm完整包名] 进行安装
Tensorflow Object API 提供的ssd-Mobilenet-v1和SNPE ssd-mobilenetv1关于tensorflow版本不兼容问题：
qq_31119155的博客
01-03 1661
  这是一篇求助帖：   snpe Object API convert Tf1.6   train   Tf&gt;1.9 Snpe: file:///home/sy/snpe-1.22.0.212/d...
tensorflow设置layer的执行位置（device）
要努力成为很厉害很厉害的人哦
06-03 1412
1.tensoflow默认的执行位置 https://tensorflow.google.cn/guide/using_gpu 1.1 设备选择的优先级 If a TensorFlow operation has both CPU and GPU implementations, the GPU devices will be given priority when the operatio...
tensorflow InvalidArgumentError 错误处理
热门推荐
门中雀的博客
05-25 1万+
tensorflow.python.framework.errors_impl.InvalidArgumentError: Expected image (JPEG, PNG, or GIF), got unknown format starting with '!(5#*8)-&gt;+-?*,&gt;*' [[Node: DecodeJpeg = DecodeJpeg[acceptable_...
tensorflow精进之路(二十四)——Object Detection API目标检测(中)（COCO数据集训练的模型—ssd_mobilenet_v1_coco模型）
YBAdiam的博客
07-29 6695
1、概述 上一讲简单的讲了目标检测的原理以及TensorflowObjectDetectionAPI的安装，这一节继续讲TensorflowObjectDetectionAPI怎么用。 2、COCO数据集介绍 COCO数据集是微软发布的一个可以用来进行图像识别训练的数据集，图像中的目标都经过精确的segmentation进行位置定位，COCO数据集包括90类目标。Object...
tensorflow-1.15.5+nv21.6-cp36-cp36m-linux_aarch64.whl
06-30
tensorflow-1.15.5+nv21.6-cp36-cp36m-linux_aarch64.whl，适用于nvidia xavier nxt等，jetpack4.5(ubuntu18.04)。
tensorflow-1.15.5-cp37-cp37m-win_amd64.whl
最新发布
03-20
TensorFlow是一个开放源代码的软件库，用于进行高性能数值计算。通过其灵活的架构，它允许用户轻松地部署计算工作在各种平台（CPUs、GPUs、TPUs）上，无论是在桌面、服务器还是移动设备上。TensorFlow最初由Google ...
tensorflow-1.15.5-cp36-cp36m-win_amd64.whl
03-20
TensorFlow是一个开放源代码的软件库，用于进行高性能数值计算。通过其灵活的架构，它允许用户轻松地部署计算工作在各种平台（CPUs、GPUs、TPUs）上，无论是在桌面、服务器还是移动设备上。TensorFlow最初由Google ...
tensorflow-1.15.5-cp37-cp37m-manylinux2010_x86_64.whl
03-20
TensorFlow是一个开放源代码的软件库，用于进行高性能数值计算。通过其灵活的架构，它允许用户轻松地部署计算工作在各种平台（CPUs、GPUs、TPUs）上，无论是在桌面、服务器还是移动设备上。TensorFlow最初由Google ...
Ubuntu 20.04 坚果云无法正常启动
Hello
10-08 5247
问题描述 在Ubuntu 20.04 OS中使用dpkg安装坚果云deb包 sudo dpkg -i nautilus_nutstore_amd64.deb 安装依赖 sudo apt-get install -f 安装完成后无法启动程序   原因 这篇博文中说可能是python版本依赖问题。我查了下python版本，python2已经安装过了，并且python指向的默认版本已经是python2。 ~$ ls -l /usr/bin/python* lrwxrwxrwx 1 root root
sudo apt-get update 出现 “无法安全地用该源进行更新，所以默认禁用该源”
Hello
09-28 4930
问题描述 sudo apt-get update 今天在运行上述命令时报错： 404 Not Found [IP: 91.189.95.83 80] 正在读取软件包列表... 完成 E: 仓库 “http://ppa.launchpad.net/tualatrix/ppa/ubuntu focal Release” 没有 Release 文件。 N: 无法安全地用该源进行更新，所以默认禁用该源。 N: 参见 apt-secure(8) 手册以了解仓库创建和用户配置方面的细节。   原因
git push 提交出现 failed to push some refs to
Hello
09-14 4388
问题描述 $ git push -u origin master error: failed to push some refs to 'codechina.csdn.net:BAR_WORKSHOP/topopt-in-matlab.git' hint: Updates were rejected because the remote contains work that you do hint: not have locally. This is usually caused by another r
git push 提交出现 fatal: Authentication failed for
Hello
09-14 1974
问题描述 $ git push -u origin master remote: HTTP Basic: Access denied fatal: Authentication failed for ... 原因 解决方法
安装tensorflow1.15.5的方法
05-01
以下是安装 TensorFlow 1.15.5 的步骤： ... print(tf.__version__) ``` 如果输出 `1.15.5`，则说明 TensorFlow 安装成功。 注意：Anaconda Prompt 是用来运行命令的终端，你需要在其中输入以上命令。

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值