Hi, Can anyone help me to resolve above issues? Regards, Samruddhi
On Thursday, September 16, 2021 at 2:58:50 PM UTC+5:30 Samruddhi Dhake wrote: > > Hi, > One more question to add here is, after running 2nd command mentioned > above, I am getting assert in file lstmtrainer.h, but I didn't find this > file in src/training in my folder Tesseract-OCR. > Can you please me with this too? > > Regards, > Samruddhi > On Thursday, September 16, 2021 at 2:54:25 PM UTC+5:30 Samruddhi Dhake > wrote: > >> Hello, >> >> *lstmtraining --model_output="D:\Test\output" >> --continue_from="D:\Test\Dim_test.lstmf" >> --train_listfile="D:\Test\eng.training_files.txt" >> --traineddata="D:\Test\eng\eng.traineddata" --debug_interval -1 >> -max_iterations 10* >> >> After running above command, I got, >> >> Warning: given outputs 111 not equal to unicharset of 110. >> Num outputs,weights in Series: >> 1,36,0,1:1, 0 >> Num outputs,weights in Series: >> C3,3:9, 0 >> Ft16:16, 160 >> Total weights = 160 >> [C3,3Ft16]:16, 160 >> Mp3,3:16, 0 >> Lfys48:48, 12480 >> Lfx96:96, 55680 >> Lrx96:96, 74112 >> Lfx256:256, 361472 >> Fc110:110, 28270 >> Total weights = 532174 >> Built network:[1,36,0,1[C3,3Ft16]Mp3,3Lfys48Lfx96Lrx96Lfx256Fc110] from >> request [1,36,0,1 Ct3,3,16 Mp3,3 Lfys48 Lfx96 Lrx96 Lfx256 O1c111] >> Training parameters: >> Debug interval = -1, weights = 0.1, learning rate = 0.001, momentum=0.5 >> null char=109 >> Loaded 2/2 lines (1-2) of document D:\Test\Dim_test.lstmf >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 0: GROUND TRUTH : +1.5 >> Iteration 0: ALIGNED TRUTH : ++11..55 >> Iteration 0: BEST OCR TEXT : _B_f_t_t_t_t_f >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.855%, delta=27.586%, train=450%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 1: GROUND TRUTH : +1.5 >> Iteration 1: ALIGNED TRUTH : ++11..55 >> Iteration 1: BEST OCR TEXT : _______ >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.839%, delta=27.586%, train=362.5%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 2: GROUND TRUTH : +1.5 >> Iteration 2: ALIGNED TRUTH : ++11..55 >> Iteration 2: BEST OCR TEXT : _+++++++ >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.821%, delta=27.586%, train=325%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 3: GROUND TRUTH : +1.5 >> Iteration 3: ALIGNED TRUTH : ++11..55 >> Iteration 3: BEST OCR TEXT : +++++++ >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.798%, delta=25.862%, train=300%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 4: GROUND TRUTH : +1.5 >> Iteration 4: ALIGNED TRUTH : ++11..55 >> Iteration 4: BEST OCR TEXT : +++++++ >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.765%, delta=24.138%, train=285%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 5: GROUND TRUTH : +1.5 >> Iteration 5: ALIGNED TRUTH : +11..55 >> Iteration 5: BEST OCR TEXT : ++++... >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.704%, delta=22.414%, train=266.667%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 6: GROUND TRUTH : +1.5 >> Iteration 6: ALIGNED TRUTH : +1..555 >> Iteration 6: BEST OCR TEXT : ++. >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.519%, delta=19.704%, train=239.286%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 7: GROUND TRUTH : +1.5 >> Iteration 7: ALIGNED TRUTH : +1..5555 >> Iteration 7: BEST OCR TEXT : +. >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.308%, delta=17.672%, train=215.625%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 8: GROUND TRUTH : +1.5 >> Iteration 8: ALIGNED TRUTH : +1...555 >> Iteration 8: BEST OCR TEXT : . >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=5.101%, delta=16.092%, train=200%(100%), skip ratio=100% >> Encoding of string failed! Failure bytes: ffffffc3 ffffff98 34 32 33 2e >> 31 20 30 2e 30 >> Can't encode transcription: '├ÿ423.1 0.0' in language '' >> Iteration 9: GROUND TRUTH : +1.5 >> Iteration 9: ALIGNED TRUTH : +1...55 >> Iteration 9: BEST OCR TEXT : .......5 >> File D:\Test\Dim_test.lstmf line 1 : >> Mean rms=4.888%, delta=14.828%, train=200%(100%), skip ratio=100% >> At iteration 10/10/20, Mean rms=4.888%, delta=14.828%, char train=200%, >> word train=100%, skip ratio=100%, New worst char error = 200 wrote >> checkpoint. >> >> Finished! Error rate = 100 >> >> Later, running command, >> lstmtraining --stop_training --continue_from="D:\Test\output_checkpoint" >> --traineddata="D:\Test\eng\eng.traineddata >> --model_output="D:\Test\abc.traineddata" >> >> It am getting. >> mgr_.Init(traineddata_path.c_str()):Error:Assert failed:in file >> ../../../../../src/training/lstmtrainer.h, line 96 >> >> >> My question is, does lstmtraining first generates output_checkpoints and >> then giving those to lstmtraining --stop_training, does it generates >> .trainneddata??? >> Thankyou. >> >> Regards, >> Samruddhi >> > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/0f2c640a-c5bc-4a79-917c-30a7ca342d8dn%40googlegroups.com.
