liuheng92 / tensorflow_psenet Goto Github PK

This is a tensorflow re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network.My blog:

Home Page: https://blog.csdn.net/liuxiaoheng1992/article/details/87646951

License: MIT License

Python 11.06% Makefile 0.05% C++ 88.87% Objective-C 0.02%

psenet tensorflow ocr python cpp text-detection

tensorflow_psenet's Introduction

PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network

Introduction

This is a tensorflow re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network.

Thanks for the author's (@whai362) awesome work!

Installation

Any version of tensorflow version > 1.0 should be ok.
python 2 or 3 will be ok.

Download

trained on ICDAR 2015 (training set) + ICDAR2017 MLT (training set):

baiduyun extract code: pffd

google drive

This model is not as good as article's, it's just a reference. You can finetune on it or you can do a lot of optimization based on this code.

Database	Precision (%)	Recall (%)	F-measure (%)
ICDAR 2015(val)	74.61	80.93	77.64

Train

If you want to train the model, you should provide the dataset path, in the dataset path, a separate gt text file should be provided for each image, and make sure that gt text and image file have the same names.

Then run train.py like:

python train.py --gpu_list=0 --input_size=512 --batch_size_per_gpu=8 --checkpoint_path=./resnet_v1_50/ \
--training_data_path=./data/ocr/icdar2015/

If you have more than one gpu, you can pass gpu ids to gpu_list(like --gpu_list=0,1,2,3)

Note:

right now , only support icdar2017 data format input, like (116,1179,206,1179,206,1207,116,1207,"###"), but you can modify data_provider.py to support polygon format input
Already support polygon shrink by using pyclipper module
this re-implementation is just for fun, but I'll continue to improve this code.
re-implementation pse algorithm by using c++ (if you use python2, just run it, if python3, please replace python-config with python3-config in makefile)

Test

run eval.py like:

python eval.py --test_data_path=./tmp/images/ --gpu_list=0 --checkpoint_path=./resnet_v1_50/ \
--output_dir=./tmp/

a text file and result image will be then written to the output path.

Examples

About issues

If you encounter any issue check issues first, or you can open a new issue.

Reference

Acknowledge

@rkshuai found a bug about concat features in model.py.

If this repository helps you，please star it. Thanks.

tensorflow_psenet's People

Contributors

Stargazers

Watchers

Forkers

fendaq happog rkshuai peternara cwbjyy afterimagex alannewimage chiukin wuxiaolianggit ieee820 sunkaianna billyzju yisampi yangheng111 llf10811020205 shiyuan0806 runauto ouya-bytes 10183308 jangocheng liuzheng081 yxandam zipinggao templeblock normalappler barongeng jasondoinggreat aaronswei stoneabc leo-xxx xianchen22 wansuiye09 janzd xiaowenhe machinelp zhaoqj2016 xiaowen-ttkx xxfly tkwitty aascode wrccrwx zouyuhan lucaslu1987 maikuraky lijian10086 lfs119 lilianglaoding henrylol loovelj mercurial24 lixingwei1106 jireh-father guoyanhw justbiubiu yudie433 fsluckymao 2016xjtuzyt brianzhu01 fireae owwowoow jadentan jo-hero zzutk bachelorwangwei float123 gds101054108 mathpopo yemika aptlin baijingml zhangzongrui joeytang3377 congjianting wjinhai ioir123ju wyc2015fq yangchao0053 anubhavrohatgi vqur wds8237 cuulee ttyhu vinceyzw akshaypatil15 yanggui19891007 xuweitj xiaoyuw-ust yuxiaomu akarazniewicz jyongsmile sunxingxingtf alexjuncpp liuzhuang1024 12345fengce cqray1990 wanglc2008 ceciliaai imagednn qidailiming kobe6258

tensorflow_psenet's Issues

where can i find pse_cpp ??

can anyone tell me how to compile pse ??
where can i find pse_cpp ?
i am getting this error when i make the file from cmd

FIND: Parameter format not correct
process_begin: CreateProcess(NULL, python3-config --cflags, ...) failed.
process_begin: CreateProcess(NULL, python3-config --ldflags, ...) failed.
g++ -o pse.so -I include -std=gnu++0x -O3 pse.cpp --shared -fPIC
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind116detail13get_internalsEvENKUlNSt15__exception_ptr13exception_ptrEE_clES2[_ZZN8pybind116detail13get_internalsEvENKUlNSt15__exception_ptr13exception_ptrEE_clES2]+0x16c): undefined reference to _imp__PyErr_SetString' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1118gil_scoped_acquire7dec_refEv[__ZN8pybind1118gil_scoped_acquire7dec_refEv]+0x4e): undefined reference to _imp__PyThread_set_key_value'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0xd7a): undefined reference to _imp__PyTuple_SetItem' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0xeb1): undefined reference to _imp___Py_NoneStruct'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0xf5e): undefined reference to _imp__PyDict_New' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x1010): undefined reference to _imp___Py_NotImplementedStruct'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x147d): undefined reference to _imp__PyTuple_Size' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x14ae): undefined reference to _imp__PyTuple_GetItem'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x14c4): undefined reference to _imp__PyObject_Repr' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x1579): undefined reference to _imp__PyDict_Size'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x15dc): undefined reference to _imp__PyDict_Next' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x168d): undefined reference to _imp__PyObject_CallObject'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x17d2): undefined reference to _imp__PyExc_TypeError' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x17dd): undefined reference to _imp__PyErr_SetString'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x1947): undefined reference to _imp__PyErr_Restore' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x19da): undefined reference to _imp__PyExc_TypeError'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x19ef): undefined reference to _imp__PyErr_SetString' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x1d6f): undefined reference to _imp__PyExc_SystemError'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x1d82): undefined reference to _imp__PyErr_SetString' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_[__ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2_]+0x1e11): undefined reference to _imp__PyExc_TypeError'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2[_ZN8pybind1112cpp_function10dispatcherEP7_objectS2_S2]+0x1e1c): undefined reference to _imp__PyErr_SetString' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail8instance15allocate_layoutEv[__ZN8pybind116detail8instance15allocate_layoutEv]+0xed): undefined reference to _imp__PyWeakref_NewRef'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail8instance15allocate_layoutEv[__ZN8pybind116detail8instance15allocate_layoutEv]+0x187): undefined reference to _imp__PyMem_Calloc' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail14clear_instanceEP7_object[__ZN8pybind116detail14clear_instanceEP7_object]+0xee): undefined reference to _imp__PyWeakref_NewRef'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail14clear_instanceEP7_object[__ZN8pybind116detail14clear_instanceEP7_object]+0x68b): undefined reference to _imp__PyObject_ClearWeakRefs' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail14clear_instanceEP7_object[__ZN8pybind116detail14clear_instanceEP7_object]+0x697): undefined reference to _imp___PyObject_GetDictPtr'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind1112cpp_function10initializeIZNS_6detail23all_type_info_get_cacheEP11_typeobjectEUlNS_6handleEE_vJS5_EJEEEvOT_PFT0_DpT1_EDpRKT2_ENUlRNS2_13function_callEE1_4_FUNESJ[_ZZN8pybind1112cpp_function10initializeIZNS_6detail23all_type_info_get_cacheEP11_typeobjectEUlNS_6handleEE_vJS5_EJEEEvOT_PFT0_DpT1_EDpRKT2_ENUlRNS2_13function_callEE1_4_FUNESJ]+0x8a): undefined reference to _imp___Py_NoneStruct' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZZN8pybind1112cpp_function10initializeIZNS_6detail23all_type_info_get_cacheEP11_typeobjectEUlNS_6handleEE_vJS5_EJEEEvOT_PFT0_DpT1_EDpRKT2_ENUlRNS2_13function_callEE1_4_FUNESJ_[__ZZN8pybind1112cpp_function10initializeIZNS_6detail23all_type_info_get_cacheEP11_typeobjectEUlNS_6handleEE_vJS5_EJEEEvOT_PFT0_DpT1_EDpRKT2_ENUlRNS2_13function_callEE1_4_FUNESJ_]+0xfb): undefined reference to _imp___Py_NoneStruct'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x18): undefined reference to _imp__PyFloat_Type' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x2b): undefined reference to _imp__PyType_IsSubtype'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x45): undefined reference to _imp__PyLong_AsLong' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x67): undefined reference to _imp__PyErr_Occurred'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x70): undefined reference to _imp__PyExc_TypeError' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x7b): undefined reference to _imp__PyErr_ExceptionMatches'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x85): undefined reference to _imp__PyErr_Clear' C:\Users\Zebi\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0x91): undefined reference to _imp__PyErr_Clear'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0xa5): undefined reference to _imp__PyNumber_Check' C:\Users\Zebi\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZN8pybind116detail11type_casterIivE4loadENS_6handleEb[__ZN8pybind116detail11type_casterIivE4loadENS_6handleEb]+0xb2): undefined reference to _imp__PyNumber_Long'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW[_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW]+0x33c): undefined reference to _imp__PyList_New' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_[__ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_]+0x3c0): undefined reference to _imp__PyLong_FromSsize_t'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW[_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW]+0x4d4): undefined reference to _imp__PyType_IsSubtype' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_[__ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_]+0x551): undefined reference to _imp__PyType_IsSubtype'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW[_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW]+0x633): undefined reference to _imp__PyExc_ValueError' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_[__ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_]+0x646): undefined reference to _imp__PyErr_SetString'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW[_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW]+0x64c): undefined reference to _imp__PyErr_Clear' C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_[__ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_]+0x658): undefined reference to _imp__PyExc_ValueError'
C:\Users\xxx\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW[_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW]+0x66b): undefined reference to _imp__PyErr_SetString' C:\Users\Zebi\AppData\Local\Temp\ccum3Drs.o:pse.cpp:(.text$_ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_[__ZZN8pybind1112cpp_function10initializeIRPFSt6vectorIS2_IiSaIiEESaIS4_EENS_7array_tIiLi1EEENS7_IhLi1EEEiES6_JS8_S9_iEJNS_4nameENS_5scopeENS_7siblingEA38_cNS_3argESH_NS_5arg_vEEEEvOT_PFT0_DpT1_EDpRKT2_ENKUlRNS_6detail13function_callEE1_clESW_]+0x671): undefined reference to _imp__PyErr_Clear'
Makefile:10: recipe for target 'pse.so' failed
make: *** [pse.so] Error 1

Pre-trained checkpoint

Thank you for sharing this project. I wanted to ask would it be possible for the pretrained model checkpoint to also be available on another hosting website like Drive or Dropbox?

from .pse import pse_cpp ImportError: dynamic module does not define module export function (PyInit_pse)

I have modified Makefile for python3 but still got this error, plz help, thanks !

复现咨询

您好，现在代码完全可以训练和测试吗？复现效果怎么样,与原作者的结果相差大吗？

tensorflow_PSENet-master/pse/pse.so: undefined symbol: _Py_ZeroStruct

Traceback (most recent call last):
File "eval.py", line 243, in
tf.app.run()
File "/home/sas/anaconda3/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 125, in run
_sys.exit(main(argv))
File "eval.py", line 202, in main
boxes, kernels, timer = detect(seg_maps=seg_maps, timer=timer, image_w=w, image_h=h)
File "eval.py", line 100, in detect
mask_res, label_values = pse(kernals, min_area_thresh)
File "/home/sas/桌面/tensorflow_PSENet-master/pse/init.py", line 18, in pse
from .pse import pse_cpp
ImportError: /home/sas/桌面/tensorflow_PSENet-master/pse/pse.so: undefined symbol: _Py_ZeroStruct

About "checkpoint" file

Hi, I want to test the model on my dataset. After I downloaded your model from baiduyun, I found that there are only three files in the folder, there is no "checkpoint" file. I want to know if this is caused by my download error? thanks.

Single-Character detection failure.

While evaluating the results after considerable training, the model skips the single character detection part due to shrinked_poly function in data_provider.py line 234 causing the area of bounding boxes around single characters to go 0. Can anyone suggest the changes to be made in the code to work around this issue?

ImportError: No module named 'Queue'

你好：
我下载了模型和工程代码，在Ubuntu16.04、tensorflow-gpu==1.12.0环境下运行：python3 eval.py --test_data_path=./test_input_0.5/ --gpu_list=0 --checkpoint_path=./model/ --output_dir=./output/
报了如下错误：
Traceback (most recent call last):
File "eval.py", line 8, in
from utils.utils_tool import logger, cfg
File "/media/jack/code/clion_kdevelop/OCR/tensorflow_PSENet/utils/utils_tool.py", line 3, in
import Queue
ImportError: No module named 'Queue'

我尝试通过sudo pip3 install Queue的方式进行安装，但是失败了。查看了整个工程，也没有Queue.py。请问我该如何解决这个问题？谢谢！ @atulak @liuheng92

undefined symbol: _Py_ZeroStruct问题

我修改了makefile里的python3-config，run eval.py，还是报错这个问题。请求帮助。 @liuheng92

中文训练

作者，你好，请问你的训练模型，没有在中文训练集上训练过吗？我再中文图片上感觉效果不好，打算在中文数据集比如天池数据上进行训练，是否可行？需要哪些注意的地方

Not able to download pretrained model

Can we get Drive share ... on the given link you have to download app and login first .
Is there any work around for this ?. it's getting difficult for non-china user to explore PSENet .

训练时长

请问您训练完消耗了多长时间吗？您的GPU是多大内存的？
@liuheng92

使用自己的数据集训练时，model loss 一直为1.0000？

相关信息及问题描述：
(1) 标签文件的格式为：116,1179,206,1179,206,1207,116,1207,"###"，保存为.txt文件；
(2) 标签文件和图片存放在同一个文件夹中；
(3) 模型可以正常训练，model loss从训练开始到结束一直为1.0000；total loss 的值最开始大于
1.0000，随着训练的进行，逐渐减小，直至total loss值变为1.0000，当total loss变为1后就不
再发生变化。即训练结束时，model loss和total loss 的值均为1。(上图是训练初期的训练日志
信息)

    使用训练好的模型测试时，无法检测文本，具体情况是输入图片和输出图片完全相同，类似于模型未对输入图片作任何处理，只是原样输出。请问，出现这个问题的原因是什么，我应该从哪些方面着手来解决该问题？

能否把icdar2017数据集发给我？

你好，作者。请问你能不能吧icdar2017数据集发给我呢，网上下载不到。
这是我的邮箱：[email protected]
万分感谢！

Error loading model using downloaded files

Thanks for sharing your work ! Looks very neat .

I am trying to load the model provided and test it on my images. I get this error when running the eval.py script.
"InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Cannot assign a device for operation model_0/split: ...."

Could you please let me know what the issue is?

it is seem too slowly of training speed?

[meta]: Adding license

@liuheng92, thank you for your work! To enable others use your code for other projects, could you please add a license to your repo, explaining the conditions for fair use? Since the PyTorch implementation is licensed with Apache 2.0, this would look great from the legal side as well. I appreciate your time!

How to get text out of boudry box

I have received text file with coordinates. Please help me to get actual text out of boundry boxes.

a error raise on python3.6+TF1.11Gpu

from .pse import pse_cpp
ImportError: dynamic module does not define module export function (PyInit_pse)

model_path = os.path.join(FLAGS.checkpoint_path, os.path.basename(ckpt_state.model_checkpoint_path)) AttributeError: 'NoneType' object has no attribute 'model_checkpoint_path'

model_path = os.path.join(FLAGS.checkpoint_path, os.path.basename(ckpt_state.model_checkpoint_path))
AttributeError: 'NoneType' object has no attribute 'model_checkpoint_path'

does it because no checkpoint file in model?
pls help

The training speed is very slow. Is this normal

INFO:root:Step 000000, model loss 0.9921, total loss 1.0328, 5.59 seconds/step, 1.43 examples/second
INFO:root:Step 000010, model loss 0.9581, total loss 0.9988, 92.09 seconds/step, 0.09 examples/second
INFO:root:Step 000020, model loss 0.8688, total loss 0.9095, 91.93 seconds/step, 0.09 examples/second

Error while running script please help!

tf.app.run()

File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 126, in run
_sys.exit(main(argv))
File "eval.py", line 171, in main
model_path = os.path.join(FLAGS.checkpoint_path, os.path.basename(ckpt_state.model_checkpoint_path))
AttributeError: 'NoneType' object has no attribute 'model_checkpoint_path'

running below command-
python3.6 eval.py --test_data_path=./tmp/images/ --gpu_list=0 --checkpoint_path=./resnet_v1_50/--output_dir=./tmp/

Note- I have resnet_v1_50.ckpt file under resnet_v1_50 directory.

Please help.;

如何修改data_provider.py来支持多边形格式输入？

请问怎么用自己的数据集进行训练？

我想用自己的数据进行训练，数据集应该按照什么格式？

'NoneType' object has no attribute 'model_checkpoint_path'

sas@sas-virtual-machine:~/桌面/tensorflow_PSENet-master$ python3 eval.py
DEBUG:matplotlib.pyplot:Loaded backend qt5agg version unknown.
DEBUG:matplotlib.pyplot:Loaded backend qt5agg version unknown.
DEBUG:matplotlib.pyplot:Loaded backend Qt5Agg version unknown.
/home/sas/桌面/tensorflow_PSENet-master/pse ll
make: 进入目录“/home/sas/桌面/tensorflow_PSENet-master/pse”
make: “pse.so”已是最新。
make: 离开目录“/home/sas/桌面/tensorflow_PSENet-master/pse”
resnet_v1_50/block1 (?, ?, ?, 256)
resnet_v1_50/block2 (?, ?, ?, 512)
resnet_v1_50/block3 (?, ?, ?, 1024)
resnet_v1_50/block4 (?, ?, ?, 2048)
2019-09-18 20:55:14.255332: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
Traceback (most recent call last):
File "eval.py", line 237, in
tf.app.run()
File "/home/sas/anaconda3/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 125, in run
_sys.exit(main(argv))
File "eval.py", line 173, in main
model_path = os.path.join(FLAGS.checkpoint_path, os.path.basename(ckpt_state.model_checkpoint_path))
AttributeError: 'NoneType' object has no attribute 'model_checkpoint_path'
这里出现了'NoneType' object has no attribute 'model_checkpoint_path'的问题，不知道是何问题？

accuary

请问大佬，训练的时候怎样显示accuary呢？ @liuheng92

pretrained_model

我添加了pretrained_model_path，为什么一直报错找不到文件呢？路径核对了没有问题的。

File "/home/ws/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/errors_impl.py", line 528, in exit
c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.NotFoundError: Unsuccessful TensorSliceReader constructor: Failed to find any matching files for ./resnet_v1_50/
@liuheng92

想问一下这个数据集具体是什么的形式？

No module named 'pse.pse'

from .pse import pse_cpp

ModuleNotFoundError: No module named 'pse.pse'

i couldn't use the pse_cpp module ?
is compiling pse is necessesary!! i am getting file not found error in subprocess?
can anyone please help me

Single character detection missing

Hi,

Thanks for the great work!
I encounter a problem with the code where single characters could not be detected. I found a similar issue and it seems not to be resolved #36

Could anyone suggest a solution for detecting single characters? Thanks.

论文中的Lc区域用OHEM进行选择的，貌似代码里计算Lc部分没找到OHEM部分诶

Training model has one file missing？

when I download the trianing model from baiduyun or googledrive,it contains three files :model.ckpt.data-00000-of-00001,model.ckpt.index ,model.ckpt.meta,Isn't there a missing one called “checkpoint“ file?
@liuheng92

请问data_provider中的crop_area作用是什么？是否可以省去这一步？thx

这段代码似乎有问题

tensorflow_PSENet/utils/data_provider/data_provider.py

Lines 297 to 301 in 1afad50

 new_h, new_w, _ = im.shape 

 #max_h_w_i = np.max([new_h, new_w, input_size]) 

 im_padded = np.zeros((new_h, new_w, 3), dtype=np.uint8) 

 im_padded[:new_h, :new_w, :] = im.copy() 

 im = cv2.resize(im_padded, dsize=(input_size, input_size))

参见

https://github.com/novioleo/FOTS/blob/5d1a4fe822f498b9e6252dc19bf309d0e6a317f0/data_loader/datautils.py#L631-L643

How to test the model?

I want to know the accuracy of the model I trained. How do I test the model? @liuheng92

请问model loss和total loss分别降到多少可以停止训练，另外，训练步数一定要100000步吗，感觉得好久才能跑完

FileNotFoundError: [WinError 2] 系统找不到指定的文件。

WARNING: Logging before flag parsing goes to stderr.
I0918 17:56:17.906025 18412 pyplot.py:219] Loaded backend qt5agg version unknown.
I0918 17:56:17.948019 18412 pyplot.py:219] Loaded backend qt5agg version unknown.
I0918 17:56:17.948019 18412 pyplot.py:219] Loaded backend Qt5Agg version unknown.
Traceback (most recent call last):
C:\Users\sas\Desktop\tensorflow_PSENet-master\pse
File "C:/Users/sas/Desktop/tensorflow_PSENet-master/eval.py", line 20, in
from pse import pse
File "C:\Users\sas\Desktop\tensorflow_PSENet-master\pse_init_.py", line 9, in
if subprocess.call(['make', '-C', BASE_DIR]) != 0: # return value
File "C:\software\Anaconda\lib\subprocess.py", line 323, in call
with Popen(*popenargs, **kwargs) as p:
File "C:\software\Anaconda\lib\subprocess.py", line 775, in init
restore_signals, start_new_session)
File "C:\software\Anaconda\lib\subprocess.py", line 1178, in _execute_child
startupinfo)
FileNotFoundError: [WinError 2] 系统找不到指定的文件。
请问作者这块问题如何解决呢？是不是编译c++文件出错？

Training icdar2017, it seems not converging, what's wrong ?

Here is the beginning of the training log.

/root/anaconda3/envs/psenet/bin/python3.5 /data3/xxx/tensorflow_PSENet/train.py --gpu_list=0,1 --input_size=512 --batch_size_per_gpu=4 --checkpoint_path=./resnet_v1_50/ --training_data_path=./icdar2017rctw_train_v1.2/train/
DEBUG:matplotlib:$HOME=/root
DEBUG:matplotlib:CONFIGDIR=/root/.config/matplotlib
DEBUG:matplotlib:matplotlib data path: /root/anaconda3/envs/psenet/lib/python3.5/site-packages/matplotlib/mpl-data
DEBUG:matplotlib:loaded rc file /root/anaconda3/envs/psenet/lib/python3.5/site-packages/matplotlib/mpl-data/matplotlibrc
DEBUG:matplotlib:matplotlib version 3.0.3
DEBUG:matplotlib:interactive is False
DEBUG:matplotlib:platform is linux
DEBUG:matplotlib:loaded modules: ['email.parser', 'tensorflow.contrib.integrate.python.ops', '_collections_abc', 'tensorflow.python.ops.random_ops', 'pkg_resources.extern.packaging.specifiers', 'numpy.polynomial.hermite', 'tensorflow.contrib.distributions.python.ops.bijectors.conditional_bijector_impl', 'tensorflow.contrib.learn.python.learn.datasets.synthetic', 'cv2.face', 'google.protobuf.internal._api_implementation', 'unittest.main', 'tensorflow.contrib.labeled_tensor.python.ops.sugar', 'tensorflow.python.saved_model.utils', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.reader_source', 'tensorflow.contrib.opt.python.training.moving_average_optimizer', 'tensorflow.contrib.solvers.python.ops.least_squares', 'cv2.ogl', 'cv2.instr', 'google.protobuf.text_format', 'tensorflow.contrib.keras.api.keras.datasets.mnist', 'tensorflow.contrib.linalg.python.ops.linear_operator_full_matrix', '_sitebuiltins', 'tensorflow.python.ops.confusion_matrix', 'tensorflow.contrib.metrics.python.metrics.classification', 'tensorflow.core.protobuf.saved_model_pb2', 'posix', 'tensorflow.python.layers.layers', 'cv2.motempl', 'encodings.unicode_escape', 'numpy.core._multiarray_umath', 'tensorflow.contrib.distributions.python.ops.onehot_categorical', 'tensorflow.python.estimator.export.export_output', 'tensorflow.contrib.learn.python.learn.preprocessing.text', 'tensorflow.python.training.session_manager', 'tensorflow.contrib.linear_optimizer.python.ops.sparse_feature_column', 'tensorflow.python.saved_model.loader_impl', 'tensorflow.contrib.seq2seq.python.ops.dynamic_attention_wrapper', 'tensorflow.contrib.keras.api.keras.datasets', 'google.protobuf.internal.enum_type_wrapper', '_hashlib', 'tensorflow.contrib.slim', 'tensorflow.contrib.input_pipeline.python.ops.input_pipeline_ops', 'tensorflow.python.framework.cpp_shape_inference_pb2', 'tensorflow.contrib.tensorboard.plugins.projector', 'tensorflow.python.saved_model.builder', 'tensorflow.contrib.session_bundle.gc', 'tensorflow.contrib.cloud.python', 'cv2.img_hash', 'tensorflow.contrib.metrics.python.ops.metric_ops', 'numpy.core.shape_base', 'tensorflow.contrib.rnn.python.ops', 'tensorflow.contrib.training.python.training.device_setter', 'tensorflow.contrib.distributions.python.ops.conditional_distribution', 'tensorflow.contrib.solvers.python.ops.linear_equations', 'tensorflow.contrib.linalg.python.ops.linear_operator_udvh_update', 'tensorflow.python.saved_model.main_op_impl', 'matplotlib.colors', 'tensorflow.python.platform.googletest', 'tensorflow.contrib.training.python.training.training', 'unittest.result', 'tensorflow.contrib.distributions.python.ops.bijectors.affine', 'tensorflow.contrib.stat_summarizer', 'string', 'tensorflow.contrib.keras.python.keras.regularizers', 'tensorflow.contrib.keras.api.keras.initializers', 'tensorflow.python.ops.gen_script_ops', 'tensorflow.python.training.adagrad', 'numpy.lib.histograms', 'tensorflow.contrib.util.loader', 'tensorflow.contrib.learn.python.learn.learn_io.pandas_io', 'tensorflow.contrib.distributions.python.ops.bijectors.invert_impl', 'tensorflow.core.framework.variable_pb2', 'b7e078e27f11ed49d55125fdb2fb1324', '_string', 'unittest.loader', 'cv2.structured_light', 'tensorflow.contrib.cudnn_rnn.ops.gen_cudnn_rnn_ops', 'distutils.version', 'tensorflow.contrib.framework.python.framework.experimental', 'tensorflow.python.training.slot_creator', 'tensorflow.contrib.keras.api.keras.optimizers', 'multiprocessing.context', 'tensorflow.contrib.keras.python.keras.metrics', 'numpy.polynomial', 'numpy.testing', 'tensorflow.python.training.coordinator', 'google.protobuf.internal.containers', 'tensorflow.python.ops.functional_ops', 'tensorflow.contrib.opt.python.training.lazy_adam_optimizer', 'tensorflow.contrib.keras.python.keras.utils.data_utils', 'tensorflow.contrib.distributions.python.ops.mvn_linear_operator', 'atexit', '_sysconfigdata', 'tensorflow.contrib.learn.python.learn.estimators.kmeans', 'tensorflow.contrib.distributions.python.ops.dirichlet', 'tensorflow.contrib.tensor_forest.python.constants', 'tensorflow.contrib.specs', 'tensorflow.python.framework.subscribe', '_lzma', 'numpy.lib._iotools', 'tensorflow.contrib.distributions.python.ops.laplace', 'numpy.config', 'linecache', 'tensorflow.contrib.layers.python.layers.feature_column', 'tensorflow.contrib.distributions.python.ops.shape', 'tensorflow.contrib.tfprof', 'tensorflow.contrib.seq2seq.python.ops.decoder', 'tensorflow.contrib.graph_editor', 'tensorflow.python.ops.math_ops', 'functools', 'inspect', 'tensorflow.python.training.training_util', 'tensorflow.core.framework.op_def_pb2', 'numpy.lib.shape_base', 'textwrap', 'decimal', 'tensorflow.contrib.distributions.python', 'tensorflow.python.framework.framework_lib', 'tensorflow.core.framework.types_pb2', 'tensorflow.contrib.keras.python.keras.utils.vis_utils', 'tensorflow.python.ops.nn_ops', 'nets.resnet.resnet_utils', 'tensorflow.contrib.keras.python.keras.callbacks', 'pkg_resources.extern.six.moves.urllib', 'tensorflow.contrib.opt.python', 'tensorflow.python.layers.normalization', 'tensorflow.contrib.keras.api.keras.preprocessing.image', 'tensorflow.contrib.tensorboard.plugins.projector.projector_config_pb2', 'tensorflow.contrib.linalg.python.ops.linear_operator_identity', 'threading', 'tensorflow.python.summary.writer.writer', 'matplotlib.cbook.deprecation', 'tensorflow.contrib.distributions.python.ops.binomial', 'tensorflow.contrib.keras.api.keras.callbacks', '_json', 'tensorflow.contrib.factorization.python.ops', 'tensorflow.contrib.keras.python.keras.utils.generic_utils', '_functools', 'pprint', 'tensorflow.contrib.distributions.python.ops.bijectors.softmax_centered_impl', 'tensorflow.contrib.factorization.python.ops.gen_factorization_ops', 'tensorflow.contrib.learn.python.learn.estimators', 'site', 'numpy.lib.arrayterator', 'tensorflow.python.platform.flags', 'tensorflow', 'numpy.ma', 'tensorflow.contrib.keras.python.keras.layers.advanced_activations', 'tensorflow.contrib.testing.python', 'tensorflow.contrib.distributions.python.ops.normal', 'tensorflow.contrib.keras.python.keras.applications.xception', 'tensorflow.contrib.linalg.python.ops', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.boolean_mask', 'tensorflow.python.ops.math_grad', 'imp', 'tensorflow.contrib.cudnn_rnn.python', 'tensorflow.contrib.learn.python.learn.ops.losses_ops', 'errno', 'tensorflow.contrib.slim.python', 'email.iterators', '_signal', 'tensorflow.contrib.lookup.lookup_ops', 'mp_main', 'numpy.core.numerictypes', 'tensorflow.contrib.distributions.python.ops.student_t', 'tensorflow.python.framework.meta_graph', 'tensorflow.contrib.learn.python.learn.utils.export', '78f3f6833449015a20606668f4590bf9', 'tensorflow.contrib.quantization.python.math_ops', 'google.protobuf.pyext.cpp_message', '_bisect', 'tensorflow.contrib.bayesflow.python.ops.monte_carlo', 'tensorflow.contrib.training.python.training.sequence_queueing_state_saver', 'ctypes._endian', 'opcode', 'tensorflow.python.ops.gen_set_ops', 'tensorflow.core.framework', 'numpy.core.info', 'numpy.version', 'tensorflow.contrib.learn.python.learn.estimators.logistic_regressor', 'pkg_resources.extern.packaging._structures', 'numpy.fft.fftpack_lite', 'tensorflow.core.framework.cost_graph_pb2', '_pickle', '_weakrefset', 'tensorflow.contrib.keras.python.keras.layers.convolutional_recurrent', 'genericpath', 'tensorflow.python.training.adagrad_da', 'email.encoders', '_bootlocale', 'email.utils', 'tensorflow.contrib.keras.python.keras.datasets.imdb', 'tensorflow.core.framework.versions_pb2', 'tensorflow.python.training.proximal_gradient_descent', '_io', 'tensorflow.python.framework.versions', 'tensorflow.python.ops.gen_image_ops', 'sre_compile', 'tensorflow.contrib.rnn.python.ops.core_rnn_cell', 'email._parseaddr', 'builtins', 'tensorflow.contrib.sparsemax', 'tensorflow.contrib.session_bundle', 'numpy.polynomial.polyutils', 'tensorflow.python.ops.variable_scope', 'tensorflow.contrib.lookup', 'tensorflow.contrib.keras.python.keras.preprocessing', 'tensorflow.python.ops.metrics', 'tensorflow.contrib.learn.python.learn.preprocessing.categorical', 'nets.model', 'tensorflow.contrib.bayesflow.python.ops.stochastic_gradient_estimators', 'tensorflow.python.ops.gen_spectral_ops', 'tensorflow.core', 'numpy.lib.stride_tricks', 'matplotlib.cbook', 'tensorflow.contrib.solvers.python', 'tensorflow.contrib.legacy_seq2seq', 'tensorflow.contrib.keras.api.keras.regularizers', 'tensorflow.contrib.specs.python.summaries', 'tensorflow.contrib.session_bundle.constants', 'operator', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.batch', 'tensorflow.contrib.opt.python.training.variable_clipping_optimizer', 'pkg_resources.extern.appdirs', 'tensorflow.python.framework.random_seed', 'numpy.core.defchararray', 'tensorflow.python.saved_model.main_op', 'tensorflow.contrib.keras.python.keras.layers.pooling', 'tensorflow.python.training.basic_loops', 'tensorflow.contrib.learn.python.learn.estimators.dnn', 'tensorflow.contrib.keras.python.keras.layers.merge', 'numpy.core.numeric', 'tensorflow.core.protobuf.debug_pb2', 'numpy.core.memmap', 'tensorflow.contrib.keras.api.keras.preprocessing', 'ssl', 'tensorflow.python.ops.ctc_ops', 'sre_constants', 'cv2.rgbd', 'warnings', 'google.protobuf.internal', 'numpy.linalg.lapack_lite', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.sparsify', 'tensorflow.contrib.tensorboard', 'google.protobuf.internal.type_checkers', 'tensorflow.python.saved_model.tag_constants', 'tensorflow.contrib.distributions.python.ops.operator_pd_full', 'tensorflow.contrib.tensor_forest.python.ops', 'email', 'tensorflow.python.training.monitored_session', 'tensorflow.contrib.learn.python.learn.estimators.dynamic_rnn_estimator', 'reprlib', 'tensorflow.contrib.factorization.python.ops.clustering_ops', 'easydict', '_random', 'tensorflow.contrib.keras.api.keras.applications', 'tensorflow.contrib.keras.python.keras.engine.training', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.unary_transforms', 'tensorflow.contrib.seq2seq.python.ops.helper', 'numpy.core._type_aliases', 'numpy.core.records', 'tensorflow.contrib.keras.api.keras.datasets.reuters', 'tensorflow.contrib.testing.python.framework.util_test', 'google.protobuf.reflection', '58660d504b637a90ef22b65c6c82c9ed', 'sre_parse', 'tensorflow.contrib.keras.python.keras.layers.serialization', 'tensorflow.core.protobuf', 'email.header', 'tensorflow.python.ops.gen_ctc_ops', 'tensorflow.core.example.example_pb2', 'tensorflow.contrib.tensor_forest.client.eval_metrics', 'tensorflow.contrib.memory_stats.ops', 'tensorflow.core.protobuf.control_flow_pb2', 'tensorflow.contrib.rnn', '4f6a1f13bcf3a21f78e647afbc3808f7', 'tensorflow.python.framework.op_def_registry', 'tensorflow.contrib.saved_model.python.saved_model', 'multiprocessing', 'numpy.lib.mixins', 'cv2.ximgproc', 'main', 'tensorflow.core.framework.tensor_shape_pb2', 'tensorflow.core.util', '_datetime', 'tensorflow.contrib.distributions', 'tensorflow.contrib.seq2seq.python', 'tensorflow.contrib.keras.python.keras.backend', 'tensorflow.contrib.keras.python.keras.applications', 'tensorflow.python.client.device_lib', 'unittest.signals', 'six.moves.urllib.request', 'tensorflow.python.platform', 'tensorflow.contrib.cudnn_rnn.python.ops', 'unittest.mock', 'tensorflow.core.lib.core.error_codes_pb2', 'tensorflow.contrib.input_pipeline.python.ops', 'pkgutil', 'tensorflow.contrib.layers.python.ops.sparse_feature_cross_op', 'tensorflow.contrib.rnn.ops', 'tensorflow.python.ops.gen_parsing_ops', 'tensorflow.python.ops.gen_data_flow_ops', 'cv2.xphoto', 'locale', 'tensorflow.contrib.keras.python.keras', 'tensorflow.python.training.summary_io', 'tensorflow.contrib.keras.python.keras.wrappers.scikit_learn', 'tensorflow.contrib.training.python.training.hparam', 'tensorflow.contrib.layers', 'tensorflow.contrib.sparsemax.python.ops', 'fnmatch', 'tensorflow.python.framework.graph_util_impl', 'tensorflow.contrib.learn.python.learn.learn_io', 'tensorflow.python.ops.clip_ops', 'tensorflow.python.ops', 'tensorflow.contrib.keras.api.keras', 'tensorflow.contrib.solvers.python.ops.util', 'd99dc76f3a6d0e4eae2cb9be2387d0b2', 'tensorflow.contrib.memory_stats.ops.gen_memory_stats_ops', 'numpy.ma.core', 'tensorflow.python.framework.errors_impl', 'tensorflow.contrib.specs.python', 'tensorflow.contrib.tensorboard.plugins.trace', 'pkg_resources.extern.packaging', 'tensorflow.contrib.learn.python.learn.graph_actions', 'cv2.detail', 'tensorflow.python.ops.sdca_ops', 'tensorflow.python.client.session', 'tensorflow.contrib.rnn.python.ops.lstm_ops', 'base64', 'cv2.xfeatures2d', 'numpy.matrixlib.defmatrix', 'tensorflow.contrib.layers.python.ops.bucketization_op', 'tensorflow.contrib.keras.api.keras.wrappers', 'tensorflow.contrib.layers.python.layers.summaries', 'tensorflow.python.util.deprecation', 'tensorflow.contrib.slim.python.slim.evaluation', 'google.protobuf.internal.api_implementation', 'matplotlib.fontconfig_pattern', 'cv2.datasets', 'tensorflow.python.ops.nn_impl', 'importlib.machinery', 'tensorflow.python.ops.sets_impl', 'tensorflow.contrib.distributions.python.ops.poisson', 'socket', 'encodings', 'google.protobuf.internal.well_known_types', 'numpy.fft.fftpack', 'cv2.ipp', 'tokenize', 'cv2.ft', 'json.decoder', 'tensorflow.python.estimator.run_config', 'token', 'tensorflow.contrib.linalg.python', 'tensorflow.python.training.ftrl', 'tensorflow.python.ops.data_flow_grad', 'tensorflow.contrib.integrate.python', 'tensorflow.contrib.nn.python.ops', 'tensorflow.contrib.framework.python.framework.checkpoint_utils', 'tensorflow.contrib.keras.python.keras.layers.recurrent', 'tensorflow.contrib.distributions.python.ops.bijectors.sigmoid', 'tensorflow.contrib.quantization.python.array_ops', 'tensorflow.contrib.distributions.python.ops.bijectors.softplus_impl', 'abc', 'zipimport', 'tensorflow.contrib.distributions.python.ops.operator_pd_identity', 'tensorflow.contrib.rnn.python.ops.core_rnn_cell_impl', 'tensorflow.contrib.keras.python.keras.engine', 'tensorflow.contrib.tfprof.python.tools.tfprof', 'tensorflow.python.summary.writer', 'google.protobuf.descriptor_pool', 'tensorflow.contrib.linear_optimizer.python.ops', 'tensorflow.contrib.distributions.python.ops.bijectors.power_transform_impl', 'tensorflow.contrib.quantization.python.nn_ops', 'tensorflow.python.training.training', 'tensorflow.contrib.training.python.training.failure_tolerator', 'tensorflow.python.training.adam', 'tensorflow.core.example.feature_pb2', 'tensorflow.contrib.linalg.python.ops.linear_operator_diag', 'tensorflow.contrib.distributions.python.ops.deterministic', 'tensorflow.core.protobuf.saver_pb2', 'tensorflow.contrib.learn.python.learn.utils.input_fn_utils', 'tensorflow.contrib.keras.python.keras.utils.layer_utils', 'tensorflow.contrib.learn.python.learn.dataframe.dataframe', 'tensorflow.python.ops.embedding_ops', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.difference', 'tensorflow.contrib.training.python', 'cv2.bioinspired', 'logging', 'subprocess', 'tensorflow.python.util.nest', 'cv2.ximgproc.segmentation', 'tensorflow.python.ops.partitioned_variables', 'tensorflow.contrib.learn.python.learn.preprocessing.categorical_vocabulary', 'numpy.lib.nanfunctions', 'tensorflow.contrib.keras.python.keras.datasets.cifar10', 'datetime', 'tensorflow.contrib.distributions.python.ops.multinomial', 'utils.utils_tool', 'urllib', '_bz2', '_compression', 'encodings.utf_8', 'tensorflow.contrib.distributions.python.ops.bijectors.inline_impl', 'tensorflow.contrib.sparsemax.python', 'numpy.polynomial.polynomial', 'uu', 'tensorflow.contrib.labeled_tensor', 'tensorflow.python.client.timeline', 'tensorflow.core.framework.graph_pb2', 'tensorflow.contrib.linalg.python.ops.linear_operator', 'google.protobuf.symbol_database', 'tensorflow.contrib.losses', 'tensorflow.contrib.tensor_forest.python', 'six.moves', 'tensorflow.core.framework.tensor_pb2', 'numpy.core._add_newdocs', 'tensorflow.contrib.learn.python.learn.experiment', 'cv2.plot', 'tensorflow.contrib.slim.python.slim.data.dataset', 'tensorflow.python.estimator.estimator_lib', 'tensorflow.contrib.distributions.python.ops.operator_pd_vdvt_update', 'tensorflow.python.lib', 'tensorflow.contrib.graph_editor.edit', 'tensorflow.python.layers', 'tensorflow.contrib.learn.python.learn.learn_io.numpy_io', 'tensorflow.python.ops.nn', 'tensorflow.contrib.learn.python.learn.ops.embeddings_ops', 'tensorflow.contrib.keras.python.keras.losses', 'tensorflow.python.ops.gradients', 'tensorflow.contrib.learn.python.learn.export_strategy', 'tensorflow.python.ops.control_flow_grad', 'tensorflow.python.training.server_lib', 'tensorflow.contrib.solvers', 'tensorflow.contrib.integrate.python.ops.odes', 'tensorflow.contrib.layers.python.layers.embedding_ops', 'numpy.polynomial.hermite_e', 'tensorflow.contrib.tensor_forest.python.ops.data_ops', 'tensorflow.python.framework.common_shapes', 'tensorflow.contrib.layers.python.layers.utils', 'numpy.lib.index_tricks', 'tensorflow.contrib.learn.python.learn.dataframe.series', 'numpy.linalg.info', 'tensorflow.python.ops.io_ops', 'tensorflow.contrib.keras.python.keras.utils.conv_utils', 'tensorflow.python.framework.op_def_library', 'tensorflow.contrib.cloud.python.ops.gen_bigquery_reader_ops', 'tensorflow.python.ops.gen_resource_variable_ops', 'tensorflow.contrib.keras.python.keras.constraints', 'a0c893eca6b687fe52f72782f634c697', 'tensorflow.contrib.linalg.python.ops.linear_operator_addition', 'tensorflow.python.ops.gen_functional_ops', 'tensorflow.contrib.labeled_tensor.python.ops.ops', 'tensorflow.python.summary.plugin_asset', 'pkg_resources.extern.pyparsing', 'tensorflow.contrib.slim.python.slim.queues', 'hashlib', 'ast', 'numpy.linalg', 'tensorflow.contrib.distributions.python.ops.operator_pd', 'tensorflow.contrib.specs.python.specs_lib', 'tensorflow.contrib.keras.python.keras.models', 'urllib.error', 'cv2.Error', 'tensorflow.contrib.metrics.python.ops.confusion_matrix_ops', 'tensorflow.contrib.integrate', 'tensorflow.contrib.image.python.ops', 'tensorflow.python.platform.benchmark', 'tensorflow.contrib.crf.python', 'tensorflow.python.layers.utils', 'tensorflow.contrib.tensorboard.plugins.trace.trace_info_pb2', 'tensorflow.python.summary', 'cv2.dnn', 'tensorflow.contrib.learn.python.learn.estimators._sklearn', 'tensorflow.contrib.distributions.python.ops.bijectors.bijector', 'numpy', '21de8d59637e674d4b2e45e5064942a0', 'tensorflow.python.ops.script_ops', 'tensorflow.contrib.keras.python.keras.layers.normalization', 'tensorflow.contrib.factorization.python.ops.gen_clustering_ops', 'tensorflow.contrib.memory_stats', 'email.message', 'numpy.linalg.linalg', 'matplotlib', 'tensorflow.contrib.linear_optimizer', 'pkg_resources.extern.packaging.markers', 'tensorflow.contrib.layers.python', 'tensorflow.contrib.rnn.python.ops.fused_rnn_cell', 'tensorflow.contrib.learn.python.learn.dataframe', 'tensorflow.contrib.training.python.training', 'numpy.lib.info', 'tensorflow.contrib.image.python.ops.image_ops', 'tensorflow.core.protobuf.meta_graph_pb2', 'heapq', 'tensorflow.contrib.framework.python.ops.variables', 'csv', 'tensorflow.python.estimator', 'tensorflow.core.lib.core', 'tensorflow.contrib.learn.python.learn.utils.saved_model_export_utils', 'numpy.lib.scimath', 'tensorflow.contrib.metrics.python.ops', 'utils', 'tensorflow.python.ops.gen_linalg_ops', '_ssl', 'struct', 'numpy.core._methods', 'tensorflow.contrib.bayesflow.python.ops.stochastic_tensor', 'tensorflow.contrib.image.python', 'tensorflow.python.estimator.inputs', 'io', 'tensorflow.python.estimator.export', 'tensorflow.contrib.opt.python.training', 'tensorflow.contrib.distributions.python.ops.mixture', 'tensorflow.python.ops.losses.util', 'cv2.aruco', 'pyparsing', 'numpy.fft.info', 'tensorflow.contrib.distributions.python.ops', 'collections.abc', 'posixpath', 'json.encoder', 'pkg_resources.extern.packaging.version', 'tensorflow.contrib.keras.python', 'tensorflow.contrib.layers.python.layers.initializers', 'tensorflow.contrib.layers.python.layers.encoders', 'tensorflow.contrib.bayesflow', 'tensorflow.contrib.keras.api.keras.constraints', 'tensorflow.python.layers.base', 'tensorflow.contrib.legacy_seq2seq.python.ops.seq2seq', 'google.protobuf.pyext', 'tensorflow.core.framework.tensor_description_pb2', 'tensorflow.python.estimator.model_fn', 'tensorflow.python.ops.gen_state_ops', 'tensorflow.contrib.learn.python.learn.models', '_imp', 'tensorflow.contrib.rnn.python.ops.rnn_cell', 'tensorflow.contrib.distributions.python.ops.geometric', 'tensorflow.tools.tfprof.tfprof_log_pb2', 'google.protobuf.json_format', 'tensorflow.python.ops.gen_control_flow_ops', 'tensorflow.python.framework.device', 'tensorflow.contrib.distributions.python.ops.operator_pd_diag', 'tensorflow.contrib.distributions.python.ops.bijectors', 'marshal', 'tensorflow.contrib.learn.python.learn.estimators.model_fn', 'importlib._bootstrap', 'tensorflow.python.user_ops', 'ntpath', 'cv2.optflow', 'uuid', 'xml', 'ipaddress', 'tensorflow.contrib.distributions.python.ops.kullback_leibler', 'tensorflow.python.estimator.inputs.queues', 'tensorflow.contrib.learn.python.learn.preprocessing', 'tensorflow.python.platform.test', 'tensorflow.python.ops.spectral_grad', 'numpy.lib.function_base', 'tensorflow.contrib.distributions.python.ops.logistic', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.csv_parser', 'tensorflow.contrib.keras.python.keras.layers.convolutional', 'email._policybase', 'tensorflow.contrib.keras.python.keras.preprocessing.text', 'numpy.lib.polynomial', '_weakref', 'importlib', '_frozen_importlib_external', 'tensorflow.python.ops.gen_candidate_sampling_ops', 'cv2.reg', 'tensorflow.contrib.rnn.python.ops.rnn', 'tensorflow.python.platform.app', 'tensorflow.contrib.labeled_tensor.python.ops.io_ops', 'tensorflow.tools.tfprof.tfprof_options_pb2', 'tensorflow.python.training.sync_replicas_optimizer', 're', 'dateutil', 'platform', 'tensorflow.python.ops.gradients_impl', 'tensorflow.python.ops.numerics', '_ctypes', 'tensorflow.python.user_ops.user_ops', 'tensorflow.contrib.learn.python.learn.trainable', 'tensorflow.contrib.keras.python.keras.wrappers', 'tensorflow.contrib.keras.python.keras.applications.vgg19', 'email.feedparser', 'numpy.random.mtrand', 'tensorflow.contrib.keras', 'tensorflow.python.saved_model.utils_impl', 'cv2.linemod', 'cv2.text', 'tensorflow.contrib.input_pipeline', 'tensorflow.contrib.keras.api.keras.metrics', 'tensorflow.contrib.ndlstm.python.lstm2d', 'tensorflow.python.training.training_ops', 'tensorflow.python.ops.parsing_ops', 'json.scanner', 'zlib', 'tensorflow.contrib.keras.python.keras.datasets.reuters', 'tensorflow.contrib.distributions.python.ops.bijectors.exp', 'tensorflow.contrib.graph_editor.subgraph', 'tensorflow.python.ops.image_ops_impl', 'lzma', 'tensorflow.contrib.slim.python.slim.data', 'tensorflow.contrib.slim.python.slim.data.dataset_data_provider', 'tensorflow.contrib.distributions.python.ops.mvn_tril', 'tempfile', 'tensorflow.contrib.nn.python.ops.cross_entropy', '_heapq', 'tensorflow.contrib.legacy_seq2seq.python', 'tensorflow.contrib.grid_rnn', 'tensorflow.python.saved_model.loader', 'tensorflow.python.training.rmsprop', 'xml.parsers', 'tensorflow.contrib.keras.python.keras.preprocessing.sequence', 'tensorflow.python.ops.standard_ops', 'numpy.lib.arraypad', 'gc', 'tensorflow.python.training.gen_training_ops', 'ctypes', 'tensorflow.contrib.opt.python.training.external_optimizer', 'google.protobuf.message_factory', 'tensorflow.python.util.protobuf', 'encodings.aliases', 'mtrand', 'tensorflow.contrib.keras.python.keras.initializers', 'tensorflow.core.framework.attr_value_pb2', 'tensorflow.contrib.keras.api.keras.datasets.cifar10', 'tensorflow.contrib.learn.python.learn.datasets.base', 'tensorflow.contrib.crf', 'tensorflow.python.framework.constant_op', 'tensorflow.python.framework.load_library', 'tensorflow.contrib.testing', 'google.protobuf.descriptor_database', 'numpy.testing._private.utils', 'tensorflow.contrib.cudnn_rnn.python.ops.cudnn_rnn_ops', 'tensorflow.python.ops.session_ops', 'calendar', 'google.protobuf.any_pb2', 'tensorflow.python.summary.writer.writer_cache', 'tensorflow.contrib.learn.python.learn.dataframe.queues.feeding_functions', 'tensorflow.contrib.keras.python.keras.applications.inception_v3', 'numpy.ma.extras', 'xml.parsers.expat', 'tensorflow.python.training.supervisor', '_pywrap_tensorflow_print_model_analysis_lib', '_socket', 'numpy.core', 'tensorflow.contrib.session_bundle.exporter', 'zipfile', 'tensorflow.contrib.slim.python.slim.data.data_provider', 'numpy.core._multiarray_tests', 'numpy.core.arrayprint', 'tensorflow.python.training.optimizer', 'tensorflow.python.framework.ops', 'tensorflow.contrib.learn.python.learn.datasets', 'numpy.testing._private.nosetester', 'os.path', 'urllib.parse', 'bisect', 'tensorflow.contrib.losses.python.losses.loss_ops', '_collections', 'tensorflow.contrib.linear_optimizer.python.sdca_optimizer', 'tensorflow.contrib.slim.python.slim', 'tensorflow.contrib.learn.python.learn.utils', 'tensorflow.contrib.rnn.python.ops.core_rnn', 'future', 'numpy.linalg._umath_linalg', 'pwd', 'tarfile', 'tensorflow.contrib.distributions.python.ops.distribution_util', 'tensorflow.contrib.solvers.python.ops.lanczos', 'urllib.request', 'tensorflow.python.ops.gen_nn_ops', 'tensorflow.python.saved_model.signature_def_utils', 'numpy.testing._private', 'tensorflow.contrib.cloud.python.ops', 'tensorflow.python.training.basic_session_run_hooks', 'numpy.testing._private.decorators', 'tensorflow.contrib.keras.api.keras.datasets.boston_housing', 'tensorflow.contrib.training', 'tensorflow.contrib.learn.python.learn.estimators.metric_key', 'tensorflow.contrib.keras.python.keras.applications.imagenet_utils', 'tensorflow.contrib.grid_rnn.python', 'pkg_resources.extern.packaging.requirements', 'tensorflow.core.framework.function_pb2', 'tensorflow.contrib.training.python.training.resample', 'tensorflow.contrib.slim.python.slim.model_analyzer', 'signal', 'tensorflow.python.layers.core', 'nets.resnet', 'xml.parsers.expat.model', 'pyexpat.model', 'tensorflow.python.pywrap_tensorflow', 'tensorflow.contrib.distributions.python.ops.mvn_diag_plus_low_rank', 'tensorflow.contrib.keras.python.keras.utils.np_utils', 'math', '_locale', 'weakref', 'tensorflow.python.framework.tensor_shape', 'tensorflow.contrib.saved_model.python', 'tensorflow.contrib.distributions.python.ops.relaxed_bernoulli', 'difflib', 'tensorflow.contrib.distributions.python.ops.chi2', 'tensorflow.python.framework', 'codecs', 'tensorflow.contrib.session_bundle.manifest_pb2', 'tensorflow.contrib.ndlstm', 'pkg_resources._vendor.six', 'tensorflow.contrib.distributions.python.ops.bijectors.cholesky_outer_product', 'pkg_resources.extern.six.moves', 'numpy.fft.helper', 'tensorflow.python.ops.metrics_impl', 'tensorflow.python.ops.array_ops', 'copy', 'tensorflow.contrib.learn.python.learn.session_run_hook', 'tensorflow.contrib.distributions.python.ops.negative_binomial', 'tensorflow.python.training.device_setter', 'tensorflow.python.estimator.inputs.pandas_io', 'keyword', 'google', 'tensorflow.contrib.tensor_forest', 'pathlib', 'email._encoded_words', 'tensorflow.python.util.compat', 'shutil', 'tensorflow.contrib.tensor_forest.python.ops.tensor_forest_ops', 'tensorflow.python.lib.io.file_io', 'tensorflow.python.training.gradient_descent', '6f42c22f2523ba04a92beabc6382ea6a', 'tensorflow.contrib.learn.python.learn', 'tensorflow.contrib.keras.api.keras.datasets.cifar100', 'tensorflow.python.ops.gen_random_ops', 'swig_runtime_data4', 'tensorflow.python.estimator.inputs.queues.feeding_functions', 'tensorflow.core.framework.step_stats_pb2', 'tensorflow.contrib.distributions.python.ops.bijectors.sigmoid_centered', 'tensorflow.contrib.distributions.python.ops.relaxed_onehot_categorical', 'tensorflow.python.ops.gen_math_ops', 'tensorflow.python.ops.gen_string_ops', 'tensorflow.contrib.learn.python.learn.utils.gc', 'unittest.suite', 'tensorflow.core.example', 'tensorflow.contrib.compiler.jit', 'encodings.latin_1', 'tensorflow.contrib.testing.python.framework', 'argparse', 'cv2.multicalib', 'itertools', 'tensorflow.core.protobuf.rewriter_config_pb2', 'tensorflow.python.ops.resource_variable_ops', 'pkg_resources.extern.packaging._compat', 'tensorflow.contrib.keras.api.keras.applications.resnet50', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.hashes', 'tensorflow.contrib.keras.api.keras.backend', 'de01d2588a57b3f7d752d74e2796a249', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.binary_transforms', 'numpy.polynomial._polybase', 'tensorflow.contrib.framework.python.ops.ops', 'cv2', 'tensorflow.contrib.learn.python.learn.estimators.tensor_signature', 'numpy.lib.twodim_base', 'importlib.util', 'tensorflow.python.saved_model.signature_def_utils_impl', 'tensorflow.contrib.ndlstm.python', 'tensorflow.python.platform.tf_logging', 'tensorflow.contrib.keras.api.keras.utils', 'tensorflow.contrib.copy_graph', 'tensorflow.contrib.slim.python.slim.learning', 'tensorflow.contrib.distributions.python.ops.dirichlet_multinomial', 'tensorflow.python.ops.tensor_array_grad', 'tensorflow.contrib.layers.python.layers.feature_column_ops', 'tensorflow.contrib.distributions.python.ops.bijectors.chain_impl', 'tensorflow.contrib.copy_graph.python', 'os', 'tensorflow.python.ops.spectral_ops', 'tensorflow.contrib.distributions.python.ops.wishart', 'tensorflow.contrib.layers.ops.gen_sparse_feature_cross_op', 'tensorflow.contrib.layers.ops.gen_bucketization_op', 'tensorflow.contrib.ndlstm.python.misc', 'tensorflow.contrib.keras.api', 'tensorflow.contrib.tfprof.python.tools.tfprof.tfprof_logger', 'cv2.data', 'matplotlib.rcsetup', 'tensorflow.contrib.cudnn_rnn', 'tensorflow.contrib.opt', 'tensorflow.python.framework.sparse_tensor', 'tensorflow.contrib.grid_rnn.python.ops', 'tensorflow.contrib.bayesflow.python.ops', 'tensorflow.contrib.learn.python.learn.summary_writer_cache', 'tensorflow.contrib.layers.python.layers.target_column', 'tensorflow.contrib.distributions.python.ops.uniform', 'tensorflow.python.estimator.export.export', 'tensorflow.python.lib.io.python_io', 'tensorflow.contrib.sparsemax.python.ops.sparsemax_loss', 'tensorflow.contrib.distributions.python.ops.bijectors.cholesky_outer_product_impl', '_warnings', 'tensorflow.python.client', 'tensorflow.python.framework.function', 'tensorflow.python.ops.gradient_checker', 'google.protobuf.descriptor_pb2', 'tensorflow.contrib.rnn.python.ops.gru_ops', 'tensorflow.contrib.legacy_seq2seq.python.ops', 'tensorflow.contrib.distributions.python.ops.transformed_distribution', 'tensorflow.contrib.keras.python.keras.applications.resnet50', 'tensorflow.contrib.tensor_forest.python.ops.gen_tensor_forest_ops', 'pyexpat.errors', 'tensorflow.contrib.keras.python.keras.activations', 'pkg_resources._vendor', 'tensorflow.contrib.keras.api.keras.losses', 'unittest.util', 'tensorflow.contrib.slim.python.slim.data.data_decoder', 'pkg_resources._vendor.six.moves', '_compat_pickle', 'cv2.flann', 'tensorflow.contrib.copy_graph.python.util.copy_elements', 'grp', 'urllib.response', 'tensorflow.python.ops.data_flow_ops', 'stat', 'numpy.lib.npyio', 'tensorflow.contrib.training.python.training.sampling_ops', 'utils.data_provider.data_provider', 'tensorflow.contrib.crf.python.ops.crf', 'tensorflow.contrib.keras.api.keras.preprocessing.sequence', 'tensorflow.contrib.framework.python.ops.gen_variable_ops', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.example_parser', 'tensorflow.core.protobuf.queue_runner_pb2', 'google.protobuf.internal.encoder', 'bz2', 'google.protobuf.descriptor', '_operator', 'tensorflow.contrib.learn.python.learn.ops', 'tensorflow.contrib.slim.python.slim.data.prefetch_queue', 'tensorflow.contrib.rnn.ops.gen_lstm_ops', 'tensorflow.core.framework.resource_handle_pb2', 'sys', 'tensorflow.contrib.seq2seq.python.ops.loss', 'email.errors', 'tensorflow.contrib.labeled_tensor.python.ops', 'tensorflow.python.training.input', 'tensorflow.contrib.graph_editor.select', 'numpy.core.function_base', 'numpy.core.einsumfunc', '_csv', 'tensorflow.contrib.framework', 'tensorflow.contrib.framework.python.framework.tensor_util', 'selectors', 'six.moves.urllib', 'pyexpat', 'numpy.core.getlimits', 'tensorflow.core.framework.allocation_description_pb2', 'cv2.ocl', 'cv2.omnidir', 'numpy.random', 'tensorflow.contrib.distributions.python.ops.bijectors.identity_impl', 'tensorflow.contrib.distributions.python.ops.bijectors.invert', 'tensorflow.contrib.framework.python.ops', 'tensorflow.contrib.image.ops', 'tensorflow.contrib.linalg.python.ops.linear_operator_tril', 'tensorflow.contrib.graph_editor.util', 'time', 'tensorflow.contrib.input_pipeline.ops', 'tensorflow.contrib.labeled_tensor.python.ops._typecheck', 'tensorflow.python.saved_model', 'tensorflow.python.ops.weights_broadcast_ops', 'tensorflow.contrib.factorization.python.ops.factorization_ops', 'tensorflow.contrib.layers.python.layers.optimizers', 'tensorflow.contrib.distributions.python.ops.mvn_diag', 'tensorflow.contrib.tensorboard.plugins.trace.trace', '_codecs', 'glob', 'tensorflow.contrib.learn.python.learn.datasets.text_datasets', 'tensorflow.contrib.learn.python.learn.learn_io.data_feeder', 'tensorflow.python.ops.sparse_grad', 'tensorflow.contrib.distributions.python.ops.bijectors.inline', 'tensorflow.contrib.cudnn_rnn.ops', 'pkg_resources.extern', 'tensorflow.python.ops.resources', 'tensorflow.contrib.distributions.python.ops.bijectors.chain', 'tensorflow.contrib.distributions.python.ops.bijectors.softmax_centered', 'tensorflow.contrib.learn.python.learn.estimators.estimator', 'tensorflow.python.training.saver', 'numpy._distributor_init', 'tensorflow.contrib.keras.python.keras.layers.embeddings', 'tensorflow.contrib.distributions.python.ops.special_math', 'tensorflow.contrib.losses.python.losses', 'multiprocessing.process', 'tensorflow.contrib.factorization.python', 'tensorflow.contrib.metrics.python.ops.set_ops', 'tensorflow.contrib.tfprof.python.tools', 'types', 'tensorflow.python.client.client_lib', 'tensorflow.contrib.quantization.python', 'tensorflow.contrib.bayesflow.python.ops.variational_inference', 'tensorflow.contrib.tensor_forest.python.tensor_forest', 'tensorflow.python.ops.string_ops', 'cv2.cuda', 'nets.resnet.resnet_v1', '_ast', 'cython_runtime', 'tensorflow.contrib.compiler', 'tensorflow.contrib.saved_model.python.saved_model.signature_def_utils', 'tensorflow.contrib.framework.python.ops.arg_scope', 'tensorflow.python.ops.init_ops', 'tensorflow.python.ops.linalg_ops', 'dis', 'google.protobuf', 'numpy.lib._datasource', 'google.protobuf.text_encoding', 'tensorflow.python.ops.image_grad', 'tensorflow.contrib.distributions.python.ops.bijectors.bijector_impl', 'matplotlib._color_data', 'tensorflow.contrib.distributions.python.ops.operator_pd_cholesky', 'tensorflow.core.protobuf.config_pb2', 'tensorflow.contrib.learn.python.learn.learn_io.graph_io', 'tensorflow.contrib.distributions.python.ops.inverse_gamma', 'tensorflow.contrib.distributions.python.ops.conditional_transformed_distribution', 'tensorflow.contrib.bayesflow.python.ops.stochastic_variables', 'tensorflow.contrib.learn.python.learn.estimators.dnn_linear_combined', 'tensorflow.contrib.distributions.python.ops.bijectors.affine_linear_operator_impl', 'tensorflow.contrib.losses.python', 'binascii', 'tensorflow.contrib.bayesflow.python.ops.stochastic_graph', 'tensorflow.contrib.keras.python.keras.utils.io_utils', 'tensorflow.python.training.session_run_hook', 'tensorflow.python.summary.summary', 'tensorflow.contrib.specs.python.specs', 'collections', 'tensorflow.python.ops.losses.losses_impl', 'tensorflow.contrib.keras.python.keras.datasets.mnist', 'cv2.cv2', 'tensorflow.contrib.distributions.python.ops.normal_conjugate_posteriors', 'tensorflow.python.ops.check_ops', 'tensorflow.python.platform.sysconfig', 'tensorflow.python.saved_model.builder_impl', 'tensorflow.python.ops.gen_io_ops', 'dateutil._version', 'numpy.core.fromnumeric', 'tensorflow.contrib.keras.api.keras.applications.vgg16', 'tensorflow.python.ops.special_math_ops', 'cv2.videostab', 'numpy.core._string_helpers', 'tensorflow.contrib.learn.python.learn.estimators.prediction_key', 'numpy.polynomial.chebyshev', 'importlib.abc', 'tensorflow.contrib.learn.python.learn.dataframe.transforms', 'tensorflow.python.lib.io', 'tensorflow.python.training.evaluation', 'tensorflow.contrib.bayesflow.python.ops.variational_inference_impl', 'tensorflow.python.util', 'tensorflow.core.util.test_log_pb2', 'tensorflow.contrib.learn.python', 'tensorflow.contrib.distributions.python.ops.gamma', 'tensorflow.contrib.sparsemax.python.ops.sparsemax', 'tensorflow.contrib.solvers.python.ops', 'numbers', 'tensorflow.tools.tfprof.tfprof_output_pb2', 'tensorflow.python.training.momentum', 'numpy.lib._version', 'cv2.fisheye', 'tensorflow.contrib.linalg', 'tensorflow.python.ops.histogram_ops', 'tensorflow.python.estimator.inputs.inputs', 'tensorflow.contrib.keras.python.keras.utils', 'tensorflow.contrib.linear_optimizer.python.ops.sharded_mutable_dense_hashtable', 'tensorflow.contrib.image.ops.gen_image_ops', 'tensorflow.contrib.slim.python.slim.data.parallel_reader', 'tensorflow.python.saved_model.saved_model', 'nets', 'google.protobuf.internal.wire_format', 'tensorflow.python.util.protobuf.compare', 'numpy.core.overrides', 'tensorflow.contrib.keras.python.keras.datasets.cifar', 'tensorflow.python.framework.errors', 'numpy.compat.py3k', 'cv2.ml', 'tensorflow.contrib.framework.python', 'numpy.polynomial.legendre', 'tensorflow.contrib.distributions.python.ops.beta', 'email.quoprimime', 'importlib._bootstrap_external', 'tensorflow.python.framework.tensor_util', 'numpy.core.multiarray', 'tensorflow.contrib.specs.python.params_ops', 'tensorflow.tools', 'tensorflow.python.platform.resource_loader', '_stat', 'numpy.lib.type_check', 'six.moves.urllib.error', 'unittest', 'tensorflow.contrib.training.python.training.tuner', 'tensorflow.contrib.deprecated', 'tensorflow.python.ops.candidate_sampling_ops', '_struct', 'tensorflow.python.ops.losses', 'tensorflow.contrib.keras.api.keras.wrappers.scikit_learn', 'sysconfig', 'tensorflow.python.saved_model.constants', 'tensorflow.python.ops.gen_user_ops', 'tensorflow.contrib.distributions.python.ops.quantized_distribution', 'tensorflow.contrib.keras.python.keras.engine.topology', 'numpy.core._internal', 'tensorflow.contrib.learn.python.learn.estimators.constants', 'numpy.core.machar', 'tensorflow.python.ops.tensor_array_ops', 'numpy._pytesttester', 'tensorflow.contrib.input_pipeline.python', 'tensorflow.contrib.keras.python.keras.datasets', 'tensorflow.contrib.distributions.python.ops.bijectors.softplus', 'tensorflow.contrib.keras.api.keras.models', 'cv2.bgsegm', 'tensorflow.contrib.keras.api.keras.applications.inception_v3', 'tensorflow.python.pywrap_tensorflow_internal', 'tensorflow.contrib.saved_model', 'tensorflow.contrib.keras.api.keras.layers', 'tensorflow.contrib.memory_stats.python', 'tensorflow.contrib.linear_optimizer.python.ops.sdca_ops', 'tensorflow.contrib.learn.python.learn.evaluable', 'tensorflow.contrib.factorization', 'tensorflow.python.framework.graph_io', 'tensorflow.contrib.distributions.python.ops.bijectors.affine_linear_operator', 'tensorflow.contrib.learn.python.learn.estimators.svm', 'numpy.lib.ufunclike', '_frozen_importlib', 'tensorflow.contrib.learn.python.learn.learn_io.dask_io', 'tensorflow.contrib.learn.python.learn.monitors', 'tensorflow.contrib.learn.python.learn.dataframe.queues', 'tensorflow.contrib.layers.python.ops', 'tensorflow.contrib.keras.api.keras.datasets.imdb', 'quopri', 'tensorflow.contrib.crf.python.ops', 'tensorflow.python.ops.nn_grad', 'tensorflow.python.estimator.export.export_lib', 'tensorflow.contrib.distributions.python.ops.bernoulli', 'tensorflow.contrib.metrics.python.metrics', 'encodings.raw_unicode_escape', 'tensorflow.contrib.tfprof.python', 'tensorflow.python.summary.summary_iterator', 'tensorflow.contrib.linalg.python.ops.linear_operator_util', 'numpy.fft', 'tensorflow.contrib.bayesflow.python', 'tensorflow.python.training.queue_runner', 'tensorflow.contrib.tfprof.python.tools.tfprof.pywrap_tensorflow_print_model_analysis_lib', 'tensorflow.contrib.learn.python.learn.basic_session_run_hooks', 'tensorflow.contrib.learn.python.learn.estimators.run_config', 'unittest.runner', 'contextlib', 'pkg_resources', 'tensorflow.python.framework.graph_util', 'gzip', 'tensorflow.contrib.keras.python.keras.optimizers', 'numpy.lib.arraysetops', 'tensorflow.contrib.distributions.python.ops.categorical', 'enum', 'tensorflow.python.ops.gen_logging_ops', 'tensorflow.contrib.distributions.python.ops.bijectors.exp_impl', 'pickle', 'tensorflow.python.ops.summary_ops', 'tensorflow.contrib.layers.python.ops.sparse_ops', 'tensorflow.contrib', 'tensorflow.contrib.util', 'tensorflow.contrib.bayesflow.python.ops.entropy_impl', 'cv2.saliency', 'mpl_toolkits', 'tensorflow.contrib.distributions.python.ops.bijectors.sigmoid_centered_impl', 'tensorflow.contrib.rnn.ops.gen_gru_ops', 'gettext', 'queue', 'tensorflow.python.ops.state_grad', 'tensorflow.contrib.tensor_forest.client.random_forest', 'tensorflow.python.estimator.estimator', 'tensorflow.python.ops.linalg_grad', 'tensorflow.contrib.layers.python.layers', 'tensorflow.python.training.checkpoint_state_pb2', 'tensorflow.contrib.tfprof.python.tools.tfprof.model_analyzer', '_pywrap_tensorflow_internal', 'tensorflow.contrib.learn.python.learn.datasets.mnist', 'numpy.lib.format', 'tensorflow.contrib.labeled_tensor.python', 'tensorflow.contrib.cloud', 'http.client', 'tensorflow.contrib.keras.python.keras.datasets.cifar100', 'tensorflow.python.platform.gfile', 'tensorflow.python.training.moving_averages', 'tensorflow.contrib.bayesflow.python.ops.stochastic_graph_impl', 'tensorflow.core.protobuf.tensorflow_server_pb2', 'tensorflow.contrib.keras.api.keras.applications.vgg19', '_opcode', 'tensorflow.contrib.distributions.python.ops.bijectors.identity', 'tensorflow.contrib.ndlstm.python.lstm1d', 'tensorflow.contrib.learn.python.learn.ops.seq2seq_ops', 'tensorflow.contrib.training.python.training.evaluation', 'email.base64mime', 'tensorflow.contrib.keras.python.keras.layers.noise', 'tensorflow.contrib.training.python.training.hparam_pb2', 'tensorflow.python.ops.state_ops', 'random', 'tensorflow.python.lib.io.tf_record', 'numpy.lib', 'tensorflow.python.estimator.inputs.numpy_io', 'google.protobuf.pyext._message', '_thread', 'cycler', 'tensorflow.contrib.input_pipeline.ops.gen_input_pipeline_ops', 'tensorflow.core.lib', 'tensorflow.python.ops.gen_sparse_ops', 'tensorflow.contrib.keras.python.keras.datasets.boston_housing', 'c25bf247d6a4a71428ea400372969416', 'tensorflow.python.ops.image_ops', 'tensorflow.python.estimator.inputs.queues.feeding_queue_runner', 'tensorflow.contrib.cloud.python.ops.bigquery_reader_ops', '_posixsubprocess', 'tensorflow.contrib.distributions.python.ops.distribution', 'tensorflow.contrib.keras.api.keras.activations', 'tensorflow.python.framework.test_util', 'tensorflow.core.framework.node_def_pb2', 'tensorflow.python.framework.importer', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.in_memory_source', 'matplotlib._version', 'tensorflow.python', 'tensorflow.contrib.keras.python.keras.layers', 'tensorflow.contrib.seq2seq', '_sre', 'plistlib', 'tensorflow.python.ops.logging_ops', 'tensorflow.contrib.factorization.python.ops.gmm', 'pkg_resources._vendor.packaging.about', 'tensorflow.contrib.keras.python.keras.layers.core', 'tensorflow.python.ops.sparse_ops', 'tensorflow.contrib.learn.python.learn.dataframe.transform', 'numpy.core.umath', 'tensorflow.contrib.framework.python.ops.prettyprint_ops', 'tensorflow.python.ops.rnn_cell_impl', 'tensorflow.contrib.keras.api.keras.preprocessing.text', 'distutils', 'tensorflow.contrib.keras.python.keras.layers.local', 'google.protobuf.message', 'tensorflow.contrib.learn.python.learn.estimators.head', 'numpy.ctypeslib', 'email.charset', 'tensorflow.python.ops.variables', 'tensorflow.contrib.tensorboard.plugins', 'google.protobuf.internal.decoder', '78da087da6dda04802f331e8197a652c', 'numpy.core._dtype_ctypes', 'numpy.matrixlib', 'six', 'tensorflow.contrib.nn.python', 'tensorflow.contrib.metrics.python.ops.histogram_ops', 'tensorflow.contrib.learn', 'tensorflow.contrib.learn.python.learn.estimators.linear', 'tensorflow.contrib.graph_editor.reroute', 'json', 'tensorflow.core.framework.device_attributes_pb2', 'tensorflow.python.ops.sets', 'tensorflow.python.training.proximal_adagrad', 'tensorflow.contrib.nn', 'tensorflow.python.ops.control_flow_ops', 'pkg_resources.extern.six', 'traceback', 'tensorflow.python.util.all_util', 'tensorflow.contrib.distributions.python.ops.bijectors.conditional_bijector', 'tensorflow.contrib.keras.python.keras.applications.vgg16', 'tensorflow.contrib.distributions.python.ops.exponential', 'cv2.line_descriptor', 'tensorflow.contrib.training.python.training.feeder', 'utils.data_provider', 'tensorflow.contrib.labeled_tensor.python.ops.core', 'copyreg', 'http', 'numpy.polynomial.laguerre', 'parser', 'unittest.case', 'tensorflow.python.ops.gen_sdca_ops', '_decimal', 'tensorflow.python.framework.registry', 'numpy.globals', 'pkg_resources.py31compat', 'tensorflow.contrib.layers.python.layers.layers', 'tensorflow.python.framework.dtypes', 'tensorflow.contrib.specs.python.specs_ops', 'tensorflow.contrib.tensor_forest.client', 'tensorflow.contrib.graph_editor.transform', 'tensorflow.contrib.rnn.python', 'tensorflow.contrib.bayesflow.python.ops.monte_carlo_impl', 'tensorflow.contrib.seq2seq.python.ops', 'tensorflow.python.ops.rnn', 'tensorflow.contrib.distributions.python.ops.bijectors.sigmoid_impl', 'tensorflow.contrib.bayesflow.python.ops.stochastic_tensor_impl', 'tensorflow.contrib.metrics', 'tensorflow.contrib.linalg.python.ops.linear_operator_composition', 'tensorflow.python.training.learning_rate_decay', 'tensorflow.contrib.image', 'tensorflow.contrib.memory_stats.python.ops', 'tensorflow.tools.tfprof', 'tensorflow.contrib.layers.ops', 'tensorflow.python.ops.template', 'tensorflow.contrib.bayesflow.python.ops.entropy', 'tensorflow.python.ops.losses.losses', 'tensorflow.python.saved_model.signature_constants', 'tensorflow.contrib.metrics.python', 'tensorflow.core.framework.summary_pb2', 'tensorflow.python.summary.writer.event_file_writer', 'tensorflow.contrib.slim.python.slim.data.tfexample_decoder', 'numpy.lib.utils', 'tensorflow.python.util.decorator_utils', 'numpy.core.dtype', 'numpy.compat.inspect', 'tensorflow.contrib.layers.python.layers.regularizers', 'tensorflow.contrib.keras.api.keras.applications.xception', 'tensorflow.contrib.framework.python.framework', 'tensorflow.contrib.copy_graph.python.util', 'tensorflow.python.ops.gen_array_ops', 'tensorflow.contrib.linear_optimizer.python', 'tensorflow.contrib.distributions.python.ops.bijectors.affine_impl', 'tensorflow.python.layers.convolutional', 'tensorflow.contrib.distributions.python.ops.bijectors.power_transform', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.split_mask', 'ctypes.util', 'tensorflow.contrib.labeled_tensor.python.ops.nn', 'tensorflow.contrib.learn.python.learn.estimators.rnn_common', 'tensorflow.contrib.training.python.training.bucket_ops', 'select', 'xml.parsers.expat.errors', 'numpy.compat', 'tensorflow.contrib.learn.python.learn.metric_spec', 'tensorflow.python.layers.pooling', 'tensorflow.contrib.testing.python.framework.fake_summary_writer', 'tensorflow.contrib.memory_stats.python.ops.memory_stats_ops', 'cv2.ppf_match_3d', 'tensorflow.python.ops.array_grad', 'tensorflow.contrib.keras.python.keras.preprocessing.image', 'tensorflow.core.util.event_pb2', 'tensorflow.python.training', 'tensorflow.contrib.factorization.python.ops.gmm_ops', 'tensorflow.contrib.keras.python.keras.layers.wrappers', 'tensorflow.contrib.learn.python.learn.dataframe.tensorflow_dataframe', 'tensorflow.contrib.learn.python.learn.dataframe.transforms.sum', 'tensorflow.contrib.seq2seq.python.ops.basic_decoder', 'tensorflow.python.training.adadelta', 'tensorflow.contrib.quantization', 'numpy.lib.financial', 'tensorflow.contrib.grid_rnn.python.ops.grid_rnn_cell', 'tensorflow.python.training.queue_runner_impl']
DEBUG:matplotlib:CACHEDIR=/root/.cache/matplotlib
DEBUG:matplotlib.font_manager:Using fontManager instance from /root/.cache/matplotlib/fontlist-v300.json
DEBUG:matplotlib.pyplot:Loaded backend TkAgg version unknown.
resnet_v1_50/block1 (?, ?, ?, 256)
resnet_v1_50/block2 (?, ?, ?, 512)
resnet_v1_50/block3 (?, ?, ?, 1024)
resnet_v1_50/block4 (?, ?, ?, 2048)
resnet_v1_50/block1 (?, ?, ?, 256)
resnet_v1_50/block2 (?, ?, ?, 512)
resnet_v1_50/block3 (?, ?, ?, 1024)
resnet_v1_50/block4 (?, ?, ?, 2048)
2019-06-16 02:22:46.345406: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
2019-06-16 02:22:46.345462: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2019-06-16 02:22:46.345472: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2019-06-16 02:22:46.345480: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2019-06-16 02:22:46.345488: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
2019-06-16 02:22:46.948758: I tensorflow/core/common_runtime/gpu/gpu_device.cc:887] Found device 0 with properties:
name: Tesla P4
major: 6 minor: 1 memoryClockRate (GHz) 1.1135
pciBusID 0000:05:00.0
Total memory: 7.43GiB
Free memory: 7.31GiB
2019-06-16 02:22:46.948930: W tensorflow/stream_executor/cuda/cuda_driver.cc:485] creating context when one is currently active; existing: 0xcf23ab0
2019-06-16 02:22:47.529396: I tensorflow/core/common_runtime/gpu/gpu_device.cc:887] Found device 1 with properties:
name: Tesla P4
major: 6 minor: 1 memoryClockRate (GHz) 1.1135
pciBusID 0000:81:00.0
Total memory: 7.43GiB
Free memory: 7.31GiB
2019-06-16 02:22:47.529478: I tensorflow/core/common_runtime/gpu/gpu_device.cc:779] Peer access not supported between device ordinals 0 and 1
2019-06-16 02:22:47.529496: I tensorflow/core/common_runtime/gpu/gpu_device.cc:779] Peer access not supported between device ordinals 1 and 0
2019-06-16 02:22:47.529522: I tensorflow/core/common_runtime/gpu/gpu_device.cc:908] DMA: 0 1
2019-06-16 02:22:47.529530: I tensorflow/core/common_runtime/gpu/gpu_device.cc:918] 0: Y N
2019-06-16 02:22:47.529536: I tensorflow/core/common_runtime/gpu/gpu_device.cc:918] 1: N Y
2019-06-16 02:22:47.529550: I tensorflow/core/common_runtime/gpu/gpu_device.cc:977] Creating TensorFlow device (/gpu:0) -> (device: 0, name: Tesla P4, pci bus id: 0000:05:00.0)
2019-06-16 02:22:47.529558: I tensorflow/core/common_runtime/gpu/gpu_device.cc:977] Creating TensorFlow device (/gpu:1) -> (device: 1, name: Tesla P4, pci bus id: 0000:81:00.0)
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:before shrink poly area:24.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part3/image_2117.jpg
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:8034 training images in ./icdar2017rctw_train_v1.2/train/
INFO:root:before shrink poly area:32.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_6929.jpg
INFO:root:before shrink poly area:30.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part2/image_1501.jpg
INFO:root:before shrink poly area:150.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_7908.jpg
INFO:root:before shrink poly area:26.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_7643.jpg
2019-06-16 02:23:09.599360: I tensorflow/core/common_runtime/gpu/pool_allocator.cc:247] PoolAllocator: After 1640 get requests, put_count=1574 evicted_count=1000 eviction_rate=0.635324 and unsatisfied allocation rate=0.710976
2019-06-16 02:23:09.599395: I tensorflow/core/common_runtime/gpu/pool_allocator.cc:259] Raising pool_size_limit from 100 to 110
INFO:root:Step 000000, model loss 0.7630, total loss 1.0419, 1.77 seconds/step, 4.52 examples/second
INFO:root:Step 000010, model loss 0.6606, total loss 0.9395, 1.93 seconds/step, 4.14 examples/second
2019-06-16 02:23:31.091535: I tensorflow/core/common_runtime/gpu/pool_allocator.cc:247] PoolAllocator: After 2568 get requests, put_count=2121 evicted_count=1000 eviction_rate=0.471476 and unsatisfied allocation rate=0.57243
2019-06-16 02:23:31.091587: I tensorflow/core/common_runtime/gpu/pool_allocator.cc:259] Raising pool_size_limit from 256 to 281
INFO:root:Step 000020, model loss 0.8125, total loss 1.0914, 1.00 seconds/step, 8.00 examples/second
INFO:root:Step 000030, model loss 0.9919, total loss 1.2708, 0.99 seconds/step, 8.06 examples/second
2019-06-16 02:23:50.217105: I tensorflow/core/common_runtime/gpu/pool_allocator.cc:247] PoolAllocator: After 3852 get requests, put_count=3903 evicted_count=1000 eviction_rate=0.256213 and unsatisfied allocation rate=0.261682
2019-06-16 02:23:50.217166: I tensorflow/core/common_runtime/gpu/pool_allocator.cc:259] Raising pool_size_limit from 655 to 720
INFO:root:Step 000040, model loss 0.8503, total loss 1.1292, 1.00 seconds/step, 8.00 examples/second
INFO:root:before shrink poly area:14.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_7948.jpg
INFO:root:Step 000050, model loss 0.7931, total loss 1.0720, 0.98 seconds/step, 8.14 examples/second
INFO:root:before shrink poly area:16.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_677.jpg
INFO:root:before shrink poly area:14.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_677.jpg
INFO:root:before shrink poly area:14.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_686.jpg
INFO:root:Step 000060, model loss 0.8387, total loss 1.1175, 0.99 seconds/step, 8.07 examples/second
INFO:root:Step 000070, model loss 0.5866, total loss 0.8655, 0.99 seconds/step, 8.12 examples/second
INFO:root:Step 000080, model loss 0.4927, total loss 0.7715, 0.99 seconds/step, 8.05 examples/second
INFO:root:before shrink poly area:18.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part2/image_1620.jpg
INFO:root:Step 000090, model loss 0.9037, total loss 1.1825, 1.47 seconds/step, 5.45 examples/second
INFO:root:before shrink poly area:56.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part5/image_5449.jpg
INFO:root:Step 000100, model loss 0.4607, total loss 0.7395, 1.69 seconds/step, 4.74 examples/second
INFO:root:Step 000110, model loss 0.5811, total loss 0.8599, 1.88 seconds/step, 4.25 examples/second
INFO:root:Step 000120, model loss 0.7551, total loss 1.0339, 1.72 seconds/step, 4.64 examples/second
INFO:root:Step 000130, model loss 0.8812, total loss 1.1600, 1.72 seconds/step, 4.64 examples/second
INFO:root:Step 000140, model loss 0.4375, total loss 0.7163, 2.16 seconds/step, 3.71 examples/second
INFO:root:Step 000150, model loss 0.7385, total loss 1.0172, 2.43 seconds/step, 3.30 examples/second
INFO:root:before shrink poly area:58.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part2/image_1006.jpg
INFO:root:Step 000160, model loss 0.7277, total loss 1.0064, 2.43 seconds/step, 3.29 examples/second
INFO:root:Step 000170, model loss 0.5224, total loss 0.8011, 2.43 seconds/step, 3.29 examples/second
INFO:root:Step 000180, model loss 0.9714, total loss 1.2501, 2.43 seconds/step, 3.29 examples/second
INFO:root:before shrink poly area:16.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part3/image_2166.jpg
INFO:root:Step 000190, model loss 0.5887, total loss 0.8674, 2.43 seconds/step, 3.29 examples/second
INFO:root:before shrink poly area:30.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part5/image_4927.jpg
INFO:root:Step 000200, model loss 0.6891, total loss 0.9678, 2.45 seconds/step, 3.27 examples/second
INFO:root:Step 000210, model loss 0.5563, total loss 0.8350, 2.77 seconds/step, 2.89 examples/second
INFO:root:before shrink poly area:26.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part5/image_4695.jpg
INFO:root:before shrink poly area:46.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part5/image_4695.jpg
INFO:root:Step 000220, model loss 0.5175, total loss 0.7961, 2.61 seconds/step, 3.06 examples/second
INFO:root:Step 000230, model loss 0.5861, total loss 0.8648, 2.70 seconds/step, 2.97 examples/second
INFO:root:before shrink poly area:24.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:24.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:24.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:12.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:18.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:10.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:10.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:12.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:10.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:10.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:8.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part1/image_883.jpg
INFO:root:before shrink poly area:10.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part2/image_1937.jpg
INFO:root:Step 000240, model loss 0.9837, total loss 1.2623, 2.71 seconds/step, 2.95 examples/second
INFO:root:Step 000250, model loss 0.7516, total loss 1.0302, 2.75 seconds/step, 2.91 examples/second
INFO:root:Step 000260, model loss 0.4626, total loss 0.7412, 2.68 seconds/step, 2.98 examples/second
INFO:root:Step 000270, model loss 0.8013, total loss 1.0799, 2.74 seconds/step, 2.92 examples/second
INFO:root:Step 000280, model loss 0.4171, total loss 0.6957, 2.86 seconds/step, 2.79 examples/second
INFO:root:Step 000290, model loss 0.5999, total loss 0.8785, 2.80 seconds/step, 2.85 examples/second
INFO:root:before shrink poly area:18.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part5/image_4795.jpg
INFO:root:Step 000300, model loss 0.8744, total loss 1.1530, 2.85 seconds/step, 2.80 examples/second
INFO:root:before shrink poly area:30.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_6528.jpg
INFO:root:before shrink poly area:20.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_6528.jpg
INFO:root:before shrink poly area:60.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_6528.jpg
INFO:root:before shrink poly area:32.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part6/image_6528.jpg
INFO:root:Step 000310, model loss 0.4642, total loss 0.7427, 3.16 seconds/step, 2.53 examples/second
INFO:root:Step 000320, model loss 0.7711, total loss 1.0497, 2.82 seconds/step, 2.84 examples/second
INFO:root:Step 000330, model loss 0.5749, total loss 0.8535, 2.85 seconds/step, 2.81 examples/second
INFO:root:before shrink poly area:14.0 len(shrinked_poly) is 0,image ./icdar2017rctw_train_v1.2/train/part3/image_2001.jpg
INFO:root:Step 000340, model loss 0.8024, total loss 1.0809, 2.79 seconds/step, 2.86 examples/second
INFO:root:Step 000350, model loss 0.8053, total loss 1.0838, 2.85 seconds/step, 2.81 examples/second
INFO:root:Step 000360, model loss 0.5403, total loss 0.8189, 2.88 seconds/step, 2.78 examples/second
INFO:root:Step 000370, model loss 0.5011, total loss 0.7796, 2.83 seconds/step, 2.83 examples/second
INFO:root:Step 000380, model loss 0.4545, total loss 0.7330, 2.90 seconds/step, 2.76 examples/second

关于训练集

博主您好十分佩服您能复现出这个代码，小弟有一个问题想请教一下您，您放的百度云链接你训练好的模型是只在1000张的icdar2015上训练出来的，还是在icdar2015+icdar2017-mlt 8000多张上训练出来的

Windows下运行出错

RuntimeError: Cannot compile pse: H:\tensorflow_PSENet-master\pse

[eval.py] Should the evaluation script clip to the original width and height?

I was going over the evaluation script and noticed that in the end boxes are clipped to the resized image size, which seems strange. Shouldn't text boxes fit the original image size instead?

数据集咨询

作者，你好，我再用天池数据进行训练的时候发现天池数据的txt中坐标值是逆时针排序的，而icdar2015中是按顺时针的，请问在训练的时候有必要将坐标转换为顺时针吗，现在训练并没有报错

请问一下模型最后输出节点名称是什么？

作者你好，我需要将模型转为pb格式，但是转化过程中需要设置output node names，然而我的电脑没办法对模型进行可视化，阅读代码也没找到哪里有设置输出节点，所以想请问一下模型最后输出节点的名称是什么？谢谢了

弯曲文本数据集处理的问题

请问，如何处理弯曲文本数据集CTW1500的数据呢？需要把它转成TFRecords格式送入Tensorflow吗？

test with batch

@liuheng92 Hi, thanks for your work.
I want to test hundreds of images, but with batch, not a single image.
is it possible? what should I do?

请问eval生成的文本文件是什么，为什么会有浮点数，应该如何从文本提取出检测到的图片块呢~

Stuck during training, neither results nor error

environment：Ubuntu 16.04
python 2.7.12
tensorflow 1.9.0
opencv 4.0
GPU：Tesla P100-PCIE-16GB *4

quesiton:
when i run the training,there is no result and no error,Always like this，why? ask for help,thanks.

this is my info:
t93@t93:~/liukui/tensorflow_PSENet$ python train.py --gpu_list=0,1,2,3 --input_size=8 --batch_size_per_gpu=1 --training_data_path=/home/t93/liukui/tensorflow_PSENet/data/train_data/ --learning_rate=0.01
DEBUG:matplotlib:CACHEDIR=/home/t93/.cache/matplotlib
DEBUG:matplotlib.font_manager:Using fontManager instance from /home/t93/.cache/matplotlib/fontList.json
DEBUG:matplotlib.backends:backend TkAgg version unknown
('resnet_v1_50/block1', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(256)]))
('resnet_v1_50/block2', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(512)]))
('resnet_v1_50/block3', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(1024)]))
('resnet_v1_50/block4', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(2048)]))
('resnet_v1_50/block1', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(256)]))
('resnet_v1_50/block2', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(512)]))
('resnet_v1_50/block3', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(1024)]))
('resnet_v1_50/block4', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(2048)]))
('resnet_v1_50/block1', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(256)]))
('resnet_v1_50/block2', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(512)]))
('resnet_v1_50/block3', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(1024)]))
('resnet_v1_50/block4', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(2048)]))
('resnet_v1_50/block1', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(256)]))
('resnet_v1_50/block2', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(512)]))
('resnet_v1_50/block3', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(1024)]))
('resnet_v1_50/block4', TensorShape([Dimension(None), Dimension(None), Dimension(None), Dimension(2048)]))
2019-04-18 08:37:04.457951: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-04-18 08:37:07.624170: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1392] Found device 0 with properties:
name: Tesla P100-PCIE-16GB major: 6 minor: 0 memoryClockRate(GHz): 1.3285
pciBusID: 0000:01:00.0
totalMemory: 15.89GiB freeMemory: 15.60GiB
2019-04-18 08:37:07.853845: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1392] Found device 1 with properties:
name: Tesla P100-PCIE-16GB major: 6 minor: 0 memoryClockRate(GHz): 1.3285
pciBusID: 0000:02:00.0
totalMemory: 15.89GiB freeMemory: 15.60GiB
2019-04-18 08:37:08.093972: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1392] Found device 2 with properties:
name: Tesla P100-PCIE-16GB major: 6 minor: 0 memoryClockRate(GHz): 1.3285
pciBusID: 0000:82:00.0
totalMemory: 15.89GiB freeMemory: 15.60GiB
2019-04-18 08:37:08.354169: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1392] Found device 3 with properties:
name: Tesla P100-PCIE-16GB major: 6 minor: 0 memoryClockRate(GHz): 1.3285
pciBusID: 0000:83:00.0
totalMemory: 15.89GiB freeMemory: 15.60GiB
2019-04-18 08:37:08.357313: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1471] Adding visible gpu devices: 0, 1, 2, 3
2019-04-18 08:37:09.640009: I tensorflow/core/common_runtime/gpu/gpu_device.cc:952] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-04-18 08:37:09.640055: I tensorflow/core/common_runtime/gpu/gpu_device.cc:958] 0 1 2 3
2019-04-18 08:37:09.640063: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] 0: N Y N N
2019-04-18 08:37:09.640067: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] 1: Y N N N
2019-04-18 08:37:09.640071: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] 2: N N N Y
2019-04-18 08:37:09.640075: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] 3: N N Y N
2019-04-18 08:37:09.641406: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1084] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 15123 MB memory) -> physical GPU (device: 0, name: Tesla P100-PCIE-16GB, pci bus id: 0000:01:00.0, compute capability: 6.0)
2019-04-18 08:37:09.642872: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1084] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 15123 MB memory) -> physical GPU (device: 1, name: Tesla P100-PCIE-16GB, pci bus id: 0000:02:00.0, compute capability: 6.0)
2019-04-18 08:37:09.643928: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1084] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:2 with 15123 MB memory) -> physical GPU (device: 2, name: Tesla P100-PCIE-16GB, pci bus id: 0000:82:00.0, compute capability: 6.0)
2019-04-18 08:37:09.645089: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1084] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:3 with 15123 MB memory) -> physical GPU (device: 3, name: Tesla P100-PCIE-16GB, pci bus id: 0000:83:00.0, compute capability: 6.0)
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/
INFO:root:10048 training images in /home/t93/liukui/tensorflow_PSENet/data/train_data/

How to read .pb file

Hi,
I have converted ckpt files to one single pb file with "model_0/Sigmoid:0" as my output layer. Could anyone guide me how to use this pb file in the code. What changes do I need to make to load the .pb model.

Any help would be usefull.

one question

when i test,it appears the following error,please help me:
Traceback (most recent call last):
File "eval.py", line 235, in
tf.app.run()
File "/root/.local/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 48, in run
_sys.exit(main(_sys.argv[:1] + flags_passthrough))
File "eval.py", line 171, in main
model_path = os.path.join(FLAGS.checkpoint_path, os.path.basename(ckpt_state.model_checkpoint_path))
AttributeError: 'NoneType' object has no attribute 'model_checkpoint_path'

How to fix the "Ran out of input error" in data_provider.py?

my enviroment is win10x64+py3.6，the "Ran out of input error" and "Can't pickle local object 'GeneratorEnqueuer.start..data_generator_task'" occur when I run the code,could you please figure out how to fix the error?thx~

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.

	new_h, new_w, _ = im.shape
	#max_h_w_i = np.max([new_h, new_w, input_size])
	im_padded = np.zeros((new_h, new_w, 3), dtype=np.uint8)
	im_padded[:new_h, :new_w, :] = im.copy()
	im = cv2.resize(im_padded, dsize=(input_size, input_size))