Comments (7)
FLming
commented on May 24, 2024
什么版本的TF？请提供具体点的信息
from crnn.tf2.
LoeWHJ
commented on May 24, 2024
今天下载的您的代码和提供的模型，我的tf时2.2.0的
错误提示是：
(py36) D:\whj\projects\OCR\CRNN.tf2-master>python crnn/demo.py -i ./example/images/ -t D:/whj/projects/OCR/CRNN.tf2-master/example/table.txt -m ./model/exported_model.h5
2020-11-03 11:17:38.872236: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2020-11-03 11:17:38.875483: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Traceback (most recent call last):
File "crnn/demo.py", line 41, in <module>
model = keras.models.load_model(args.model, compile=False)
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\saving\save.py", line 184, in load_model
return hdf5_format.load_model_from_hdf5(filepath, custom_objects, compile)
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\saving\hdf5_format.py", line 178, in load_model_from_hdf5
custom_objects=custom_objects)
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\saving\model_config.py", line 55, in model_from_config
return deserialize(config, custom_objects=custom_objects)
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\layers\serialization.py", line 109, in deserialize
printable_module_name='layer')
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 373, in deserialize_keras_object
list(custom_objects.items())))
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\engine\sequential.py", line 398, in from_config
custom_objects=custom_objects)
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\layers\serialization.py", line 109, in deserialize
printable_module_name='layer')
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 362, in deserialize_keras_object
config, module_objects, custom_objects, printable_module_name)
File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 321, in class_and_config_for_serialized_keras_object
raise ValueError('Unknown ' + printable_module_name + ': ' + class_name)
ValueError: Unknown layer: Functional
…------------------ 原始邮件 ------------------
发件人: "FLming/CRNN.tf2" <[email protected]>;
发送时间: 2020年11月3日(星期二) 中午11:33
收件人: "FLming/CRNN.tf2"<[email protected]>;
抄送: "🔥"<[email protected]>;"Author"<[email protected]>;
主题: Re: [FLming/CRNN.tf2] 我在运行demo.py，用您提供的模型时，出现错误 (#10)
什么版本的TF？请提供具体点的信息
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
from crnn.tf2.
FLming
commented on May 24, 2024
具体原因有眉目了，tf2.3是能工作的，我给的模型是在导出时加了一层softmax的。2.2居然不支持functional层的加载吗？我稍后验证一下。
…------------------ 原始邮件 ------------------
发件人: LoeWHJ <[email protected]>
发送时间: 2020年11月3日 11:36
收件人: FLming/CRNN.tf2 <[email protected]>
抄送: Yiming Huang <[email protected]>, Assign <[email protected]>
主题: 回复：[FLming/CRNN.tf2] 我在运行demo.py，用您提供的模型时，出现错误 (#10)
今天下载的您的代码和提供的模型，我的tf时2.2.0的
错误提示是：
(py36) D:\whj\projects\OCR\CRNN.tf2-master&gt;python crnn/demo.py -i ./example/images/ -t D:/whj/projects/OCR/CRNN.tf2-master/example/table.txt -m ./model/exported_model.h5
2020-11-03 11:17:38.872236: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2020-11-03 11:17:38.875483: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Traceback (most recent call last):
&nbsp; File "crnn/demo.py", line 41, in <module&gt;
&nbsp; &nbsp; model = keras.models.load_model(args.model, compile=False)
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\saving\save.py", line 184, in load_model
&nbsp; &nbsp; return hdf5_format.load_model_from_hdf5(filepath, custom_objects, compile)
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\saving\hdf5_format.py", line 178, in load_model_from_hdf5
&nbsp; &nbsp; custom_objects=custom_objects)
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\saving\model_config.py", line 55, in model_from_config
&nbsp; &nbsp; return deserialize(config, custom_objects=custom_objects)
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\layers\serialization.py", line 109, in deserialize
&nbsp; &nbsp; printable_module_name='layer')
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 373, in deserialize_keras_object
&nbsp; &nbsp; list(custom_objects.items())))
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\engine\sequential.py", line 398, in from_config
&nbsp; &nbsp; custom_objects=custom_objects)
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\layers\serialization.py", line 109, in deserialize
&nbsp; &nbsp; printable_module_name='layer')
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 362, in deserialize_keras_object
&nbsp; &nbsp; config, module_objects, custom_objects, printable_module_name)
&nbsp; File "D:\Program Files (x86)\anaconda3\envs\py36\lib\site-packages\tensorflow\python\keras\utils\generic_utils.py", line 321, in class_and_config_for_serialized_keras_object
&nbsp; &nbsp; raise ValueError('Unknown ' + printable_module_name + ': ' + class_name)
ValueError: Unknown layer: Functional
------------------&nbsp;原始邮件&nbsp;------------------
发件人: "FLming/CRNN.tf2" <[email protected]&gt;;
发送时间:&nbsp;2020年11月3日(星期二) 中午11:33
收件人:&nbsp;"FLming/CRNN.tf2"<[email protected]&gt;;
抄送:&nbsp;"🔥"<[email protected]&gt;;"Author"<[email protected]&gt;;
主题:&nbsp;Re: [FLming/CRNN.tf2] 我在运行demo.py，用您提供的模型时，出现错误 (#10)
什么版本的TF？请提供具体点的信息
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
—
You are receiving this because you were assigned.
Reply to this email directly, view it on GitHub, or unsubscribe.
from crnn.tf2.
FLming
commented on May 24, 2024
问题已复现。
from crnn.tf2.
FLming
commented on May 24, 2024
相关问题，确认使用functional建立的h5模型无法在2.2被加载，SavedModel格式是可行的。但迁移学习时，SavedModel还是不够方便。所以可能的话，请更新TensorFlow到2.3以上。2.2中部分功能依旧可以使用，只是导出模型时需要使用SavedModel使加载功能能够工作，或者通过代码构建模型以使h5权重能够正常加载。
Tensorflow 2.3 Changelog:
Functional models now get constructed if any tensor in a layer call's arguments/keyword arguments comes from a keras input.
Previously the functional api would only work if all of the elements in the first argument to the layer came from a keras input.
from crnn.tf2.
FLming
commented on May 24, 2024
确认2.3下导出的SavedModel格式模型，依旧无法在2.2版本下工作。:sweat_smile:
from crnn.tf2.
LoeWHJ
commented on May 24, 2024
把tf 升级到了2.3可以用了
…------------------ 原始邮件 ------------------
发件人: "FLming/CRNN.tf2" <[email protected]>;
发送时间: 2020年11月3日(星期二) 中午1:09
收件人: "FLming/CRNN.tf2"<[email protected]>;
抄送: "🔥"<[email protected]>;"Author"<[email protected]>;
主题: Re: [FLming/CRNN.tf2] 我在运行demo.py，用您提供的模型时，出现错误 (#10)
确认2.3下导出的SavedModel格式模型，依旧无法在2.2版本下工作。😅
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
from crnn.tf2.
Related Issues (20)
- 请问作者训练了多少个epoch之后，sequence_accuracy开始有值？ HOT 4
- UnicodeDecodeError: 'gbk' codec can't decode byte 0xa6 in position 363: illegal multibyte sequence HOT 5
- Load model in OpenCV HOT 3
- Prediction accuracy in ONNX Runtime HOT 1
- About Mjsynth HOT 1
- What is validation and test accuracy on MJSynth dataset? HOT 1
- after special characters are predicting <UNK> HOT 1
- tf2.3 not find "keras.optimizers.schedules.CosineDecay" HOT 1
- Ragged Tensors in Keras HOT 3
- Is 'Sequence Accuracy' referring only to exact matches? HOT 4
- Does CRNN support text line images? HOT 1
- predict not True HOT 1
- can i help me HOT 3
- 作者你好，请问你的SavedModel是用什么数据训练的？ HOT 1
- 你好，首先很感谢你的工作，获益很多，请教一下您有试过添加白名单功能吗 HOT 1
- chinese predicting problems
- How to get the Last layer (ctc_greedy_decoder)? HOT 1
- Invalid argument error HOT 2
- Expected concatenating dimensions in the range [-1, 1), but got 1 [Op:ConcatV2] name: concat HOT 1
- Handling invalid image path or corrupted image files.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
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.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from crnn.tf2.