Coder Social home page Coder Social logo

Comments (11)

Kongsea avatar Kongsea commented on July 4, 2024

Sorry, I have not tested it.
Maybe you can have a try.
Besides, I always use Detectron now. Maybe you can have a try to switch to it too.

from fpn_tensorflow.

zhanhuanli avatar zhanhuanli commented on July 4, 2024

@leishi2018 @Kongsea 我想请问如何在VOC2012数据集上运行起代码?比如dataset_name=‘person’,添加标签集合为:elif cfgs.DATASET_NAME == 'person':
NAME_LABEL_MAP = {
'back_ground': 0,
"person": 1,
'aeroplane': 0,
'bicycle': 0,
'bird': 0,
'boat': 0,
'bottle': 0,
'bus': 0,
'car': 0,
'cat': 0,
'chair': 0,
'cow': 0,
'diningtable': 0,
'dog': 0,
'horse': 0,
'motorbike': 0,
'pottedplant': 0,
'sheep': 0,
'sofa': 0,
'train': 0,
'tvmonitor': 0
}
,调整训练参数,可是依旧一直报InvalidArgumentError (see above for traceback): LossTensor is inf or nan : Tensor had NaN values
[[Node: train_op/CheckNumerics = CheckNumericsT=DT_FLOAT, message="LossTensor is inf or nan", _device="/job:localhost/replica:0/task:0/device:GPU:0"]]
[[Node: Momentum/update_resnet_v1_101/block3/unit_17/bottleneck_v1/conv3/BatchNorm/beta/ApplyMomentum/_2540 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge_13388_Momentum/update_resnet_v1_101/block3/unit_17/bottleneck_v1/conv3/BatchNorm/beta/ApplyMomentum", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]
错误,我想我是哪里出来问题,如果你们可以告诉我,万分感谢~

from fpn_tensorflow.

Kongsea avatar Kongsea commented on July 4, 2024

Please refer to this.
Or maybe it's because your learning rate is too large.

from fpn_tensorflow.

zhanhuanli avatar zhanhuanli commented on July 4, 2024

VOC上的四个坐标会不满足关系吗? 我觉得问题可能不出在这个地方, 我的学习率是0.00005,还会太大吗,其余参数如你一样。

from fpn_tensorflow.

zhanhuanli avatar zhanhuanli commented on July 4, 2024

@Kongsea I set lr to a very small value, and also changed the data to tfrecord. I also changed the data set to 20 types - pascal tried, or the above problem occurs. Do you know why this is the case? ? Have you encountered similarities when using VOC datasets? Where am I wrong?

from fpn_tensorflow.

Kongsea avatar Kongsea commented on July 4, 2024

我猜想可能是因为你修改了数据集的label_map,是否会导致class、label还有数据这些地方的数目无法匹配了?

from fpn_tensorflow.

zhanhuanli avatar zhanhuanli commented on July 4, 2024

我是增加了label_map的person,如上贴图所示,这样的形式只有person一个类别,其余都属于背景,这样class和label数据之间应该都能匹配。我也尝试了直接用原本的pascal ,类别设为20类,也就是voc2012的类别数,这应该就不会存在这样的问题,可还是出现这类问题。你是直接拿自己的数据拿去运行的吗?还是你也尝试过coco或者voc

from fpn_tensorflow.

Kongsea avatar Kongsea commented on July 4, 2024

我是直接在自己的数据集上跑的,没有试过COCO或者voc。
而且我也没有试过你这种类别的形式,我是把所有类别都算作有效类别了。
我怀疑你的问题应该就出在这里。你可以试一下看看。

from fpn_tensorflow.

zhanhuanli avatar zhanhuanli commented on July 4, 2024

对的, 我考虑过这个问题 ,我也把所有类别(20类)都当成有效类别测试了,还是如此。 你的数据集私密吗? 方便共享一份吗?

from fpn_tensorflow.

Kongsea avatar Kongsea commented on July 4, 2024

不好意思,因为是公司项目,所以数据不便共享,请多见谅。
另,如果你使用FPN的话,建议尝试Detectron,个人感觉那个更稳定可靠一些。

from fpn_tensorflow.

zhanhuanli avatar zhanhuanli commented on July 4, 2024

好的,没事~ 我也只是做个测验,行,我有空试试,麻烦了~

from fpn_tensorflow.

Related Issues (18)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.