Extract Tables from PDF and convert to Excel sheet with Paddle OCR text detection and recognition.

[@JujutsuMan]

Impressive content for Deep Learning OCR! Many thanks!

[@neuralearn]

You're welcome :)

[@vishaldas6346]

Hi, you have done a phenomenol job, by explaining PaddleOCR in detail. Can you please let me know if we can do the training of PaddleOCR on custom datasets for extracting data from tables of different length in pdfs or images.

[@BailinCAI]

impressive, struggling right now for my little side project using ocr, u helped a lot man, appreciate it

[@AmanChauhan-hr1wh]

hii does this notebook working for you actually for me it's not can u please help

[@BailinCAI]

@@AmanChauhan-hr1wh well i just use his method, not totally copy from him, the result i implemented by myself is not really 100% correct so i end it up by using the azure api, it's really 100% correct and the speed of processing is so fast as well

[@aerocyropyros]

Thanks lad, it gave me some ideas how to apply it paddleOCR in my research mate

[@ajithn7336]

Hello neuralearn, thanks for your great tutorial. Could you please proivide notebook access

[@christianrazvan]

Well that is a very simple and readable table, it's easy enough to do it with basic if logic....but try a no border , very near to border content , on a scanned image of a table

[@autozzhao2812]

Great video and solve my problem, in 1:46:49 there is a little doubt about is this the correct way xmax,ymax should be reflect into code is to select the minimum, the opposite goes for the xmin,ymin?

[@malakkhiari1419]

how i can fix this error "ImportError: libcudart.so.10.2: cannot open shared object file: No such file or directory" ? caused by the line of code "import layoutparser as lp"

[@ShivamSingh-sm2oy]

Hey, Thanks for the wonderful tutorial man! can you please provide access to the notebook please.

[@puneetbansal8567]

Hi, Neuralearn, Thanks for creating great tutorial. Its very useful. Can you please provide notebook access ?

[@mohamedmagdy3872]

brilliant work!!, I would like to thank you for giving me access to notebook. keep going broo 💙💙

[@neuralearn]

My Pleasure :) Feel free to check out on our other videos

[@emailvarun]

Hi Thank you for this, can youj please help me with the notebook access please, also can you please help me understand will I be able to cover most of the table formats through this?

[@leonc5510]

Thank you for the tutorial, I have requested the notebook access

[@neuralearn]

Please check your mail :)

[@pokopiko429]

Congrats, one of the best videos I've seen on this topic! Could you please grant me access to the Google Collab?

[@neuralearn]

Please after requesting access, check your mail inbox or spam

[@robertdolovcak9860]

This is first that I hear about PaddleOCR. Seems very good tool. I really appreciate the work you have done and would also want to try this. Can you please allow access to the google collab code for this?

[@neuralearn]

Hello my dear Robert Please check your mail

[@brmaaouia9715]

What if it does detect the table as table but as figure or text ?

[@HueoanThi-nv6ei]

I have problem with layout parser. It seems due to a conflict between paddleocr and protobuf. What should I do to fix it? Thank

[@toto2321]

thank you man the best who explain what it is actually happening thank you so much

[@neuralearn]

You're welcome:)

[@VaishnaviVatsa-c9n]

hey ,it is very helpful video but I am facing issue while reconstruction .can you please help me out?

[@pavitrabiradar-h3p]

Hello , Thanks for sharing this vedio, is this method will work for nested tables?

[@_keto444]

40:40 i got the following error: TypeError: int() argument must be a string, a bytes-like object or a real number, not 'list' how can i solve it?

[@EarningsApps]

pls give access to notebook ...great and informative tutorial !!

[@neuralearn]

Please check your mail :)

[@RohitSharma-to7yy]

Hi. The content is very impressive. Would love to see the notebook and add upon this to create table in google docs instead. Please share the notebook

[@henrydo9731]

I have a question that if I have a table but it's in 2 pages (half of it is in 1st page and the other is in 2nd page), how could I solve this problem

[@siddharthpatel2193]

Can I get code? I followed video and wrote code and everything is working but due to some issue, out_array at end is same value. Update: Solved Thanks, this is best tutorial on this topic (saying this after going through countless tutorials, research papers and blogs in past 3 months).

[@neuralearn]

:)

[@tapendrabaduwal7423]

Hello.... if their is unstructure table which is not in an order of n*m dimension cell.....then this method will work?

[@neuralearn]

It depends on the table in question. Nonetheless, you can always modify this method to suit your specific table

[@tapendrabaduwal7423]

@@neuralearn It is possible to work in all types of table at one short?

[@neuralearn]

No, it's not possible!

[@tapendrabaduwal7423]

@@neuralearn Thank you soo much

[@neuralearn]

You're welcome :)

[@PurushothamReddy-ff6vp]

Hello, I'm facing trouble when there are multiple lines within the same row, it is considering them as new rows.. how do i fix this?. Thank you!

[@moez.mazhar]

Hi, I've followed your procedure as is but I'm getting "ValueError: Can't convert Python sequence with mixed types to Tensor." on the Non-Max Suppression portion. Can you tell me what might be causing that please?

[@Jean-nf1yh]

Broo, this is awesome, thank you very much!!!

[@neuralearn]

You're welcome :)

[@etarhunisuhaib2031]

Thanks for this video, let's say we have a page with free text and tables, once we have our tables, how can we extract the remaining text ? when im using parser it also extract the table text from the page. i want to use your approche for tables and i want to extract only the remaining text.

[@Ankur-be7dz]

for only extracting texts use pdfminer

[@andrewlachance2062]

just match the consecutive text from the table and parse the PDFs skipping over the text

[@quocvuong6752]

Thank you so much, I really appreciate the informative video. Could you please allow access to google collab? It would be super helpful.

[@neuralearn]

Hello my dear Quốc, please check your mail :)

[@niroshiniedayaratne4066]

Thank you very much for this! Very insightful!

[@neuralearn]

Glad it was helpful :)

[@adillaanam4058]

hi! tysm for the video. would you pls allow access to the notebook? ty!!

[@ZaheerH4ck3r]

Bro you're doing good work

[@neuralearn]

Thanks for the kind words :)

[@ZaheerH4ck3r]

@@neuralearn I have a question I've pdf file which is 560 pages long and which has data that other libraries do convert into excel file but its like garbage. If I use this model i'll be able to convert it?

[@neuralearn]

I think you should just go ahead and try. Its free :)

[@statosys]

Request access for colab notebook, thank you so much.

[@francescodimartino8896]

Amazing job! Could you please share with me the google Collab? 🙏

[@kikigaming4595]

how to intall layout parser ? from the github now it doesn't have any file such as layout parser

[@bharattyagi1405]

Hi, could you please provide access to the collab notebook.

[@dinaharan0213]

Hi,i am installed paddlepaddle instead of paddlepaddle-gpu bcoz i dont have gpu in my local system. I getting "AttributeError: module 'numpy' has no attribute 'int'". Is it possible to run this project in local system without gpu.

[@edwinjoe6044]

I facing this error too...☹

[@neuralearn]

Hello my dear Dinaharan, here is a notebook which works for cpu runtime: colab.research.google.com/drive/1vZHrahaaubhWMz83jlPuvA1na_v98fUP

[@dinaharan0213]

Hi, I am very happy to get your rply and wonder of your help.I am glad to have youtuber like you. I really liked your efforts for your subscribers. Thank you very much. 🤗😇👏👏👏

[@neuralearn]

My pleasure :)

[@pratikmore4044]

I am getting the following error and not sure how can I resolve this: Error: Can not import paddle core while this file exists: /usr/local/lib/python3.10/dist-packages/paddle/fluid/libpaddle.so Tried reinstalling paddlepaddle but that didn't work.

[@owaisasghar2033]

Sir issue solved?

[@Conta-Data]

@Neuralearn Can you send me the example file so I can use it?

[@ss_d25]

Hi, great video. Can you please provide access to this notebook? Thanks a lot in advance.

[@neuralearn]

Hi, check your mail box or spam

[@therafee]

@neuralearn hello could you indicate me where is test.pdf file?? I have access to de notebook but it throws error I got: PDFPageCountError: Unable to get page count. I/O Error: Couldn't open file '/content/bahdanau attention.pdf': No such file or directory

[@kenjeroldarellano4617]

Hi, Neuralearn, Thanks for creating a very useful tutorial. Can you please provide notebook access for my study?

[@harshithprakash2433]

Awesome video and interesting approach towards the problem , would you mind giving me access to that notebook..?

[@neuralearn]

Hello my dear Harshith, please check your mail :)

[@nealgilmore337]

Hello @neuralearn - love the demo! Can you provide me access to the Colab?

[@neuralearn]

Done!

[@therafee]

Why do we need to clone paddle repository at 15:57

[@alirezaghasrimanesh2431]

Thanks for yor great toturial!!!!

[@youssefmouknii5033]

Thank you so much for this video , Could you please allow access to google collab?

[@neuralearn]

Hello my dear Youssef glad this video is helpful :) Please check your mail inbox or spam

[@cissemy]

Great. Is it possible to use this model for matrix recognition ? how many rows and columns, elements of matrix ?

[@aishwaryadinesh7641]

Hi, I'm getting this error - (External) CUDA error(100), no CUDA-capable device is detected. [Hint: 'cudaErrorNoDevice'. This indicates that no CUDA-capable devices were detected by the installed CUDA driver. ] (at /paddle/paddle/phi/backends/gpu/cuda/cuda_info.cc:66). Can you help me out w this please?

[@revanthkumar3406]

Hey, Really Great Video ❤, can u provide access to notebook

[@neuralearn]

Hello my dear Kumar, Please check your mail inbox or spam

[@AniketRana-iz5ms]

hi , not able run this code in jupyter notebook , may u help to run this in local system, like what was the procedure for that

[@neuralearn]

Hello my dear Rana, what issues did you face, while running the code locally?

[@AniketRana-iz5ms]

@@neuralearn it asking for gpu as a requirement, but i want to run this code on jupyter notebook with cpu

[@balasubramaniyang6506]

Hi Nice Explanation, Can you provide access.It's very helpfull for us.

[@khushibaghel220]

Hey! I want to try out your tutorial. Could you please give access of your notebook

[@neuralearn]

Hello check your mail :)

[@ajaychinni3148]

Please approve the access request for the Google Collab notebook. I am very interested in the code

[@IsratjahanFateha9106]

Can I have the access of your Colab Notebook please? I have requested for the access yesterday

[@neuralearn]

Hi, check your mail box or spam

[@dinaharan0213]

Hi , I installed all packages step by step as you guided. But i getting error "from . import libpaddle". can you help me ?

[@neuralearn]

Hello my dear Dinaharan, we just shared the notebook with you. Please rerun again and tell us if you have any issues

[@dinaharan0213]

@@neuralearn Hi thankyou for shared your notebook with me😊.l completely installed all packages. Now i am getting this "ValueError: (InvalidArgument) Device id must be less than GPU count, but received id is: 0. GPU count is: 0. [Hint: Expected id < GetGPUDeviceCount(), but received id:0 >= GetGPUDeviceCount():0.] (at /paddle/paddle/phi/backends/gpu/cuda/cuda_info.cc:242)" can you please solve it

[@neuralearn]

Hello Dinaharan, your error indicates you aren't using a GPU as your hardware accelerator. On Colab, go to runtime(it's at the top left corner), then select change runtime, and finally pick the hardware accelerator to be GPU.

[@dinaharan0213]

@@neuralearn Hi thankyou for make your time with me 😊. I really appreciat it when you taught me.

[@neuralearn]

You're welcome :)

[@anirbanghorai3699]

EXCELLENT! CAN YOU PLS POST A VIDEO ON Paddle OCR custom training (both detection +recognition)steps? I have my own data ..want to do a transfer learning

[@neuralearn]

We are glad this was helpful :) We shall work on that and publish as soon as possible!

[@anirbanghorai3699]

@@neuralearn glad you responded..waiting for the custom training video

[@snehitdua153]

Hey, can you please provide the link for the pdf used in the video? Thanks

[@beratoren7627]

This was an amazing tutorial ! I really want to try and further tweak this. Can you please grant me access to the Google Colab Code?

[@neuralearn]

Hello please check your mail inbox or spam

[@snehitdua153]

I'm getting error in loading the model ValueError: (InvalidArgument) Device id must be less than GPU count, but received id is: 0. GPU count is: 0. [Hint: Expected id < GetGPUDeviceCount(), but received id:0 >= GetGPUDeviceCount():0.] (at /paddle/paddle/phi/backends/gpu/cuda/cuda_info.cc:242)

[@rrrsranjith]

Excellent video 🔥

[@neuralearn]

Glad you loved it :)

[@xavier6649]

Hey Great Work , can you give access to your Colab Drive ? Thanks

[@neuralearn]

Please check your mail :)

[@aishwaryachowta6598]

Thank you for the tutorial !!!

[@neuralearn]

The pleasure is ours :)

[@khaoulafattah]

thank you for the explanation @Neuralearn , can u please provide me access to the colab ?

[@neuralearn]

Please check your mail inbox or spam :)

[@PadmajaPhadke-e1l]

I want convert CSV file into Json file, { field 1: {col1:text, col2:text, col3:text},{field2:{col1:text,col2:text, col3:text} in this format. Can you please help me to create this Json file. Thank You

[@kibtiachowdhury6011]

Hi, I want to get only paragraph text without any figure and table from any type pdf. How can I solve this?

[@neuralearn]

You can pick text by changing [if l.type == 'Table':] ----to --> [if l.type == 'Text:]

[@kinetic_kane9033]

Hello can I please get viewing access to the colab notebook?

[@neuralearn]

hello Kane, please demand access and check your mail in 5 minutes

[@ayushbansal999]

Hi, please could you provide me with the access to this colab notebook

[@neuralearn]

Hello my dear Ayush, Please check your mail inbox or spam

[@KartikSharma-hd7rd]

Excellent tutorial, can you please access grant for google colab notebook :)

[@neuralearn]

Sure:) Check your mail!

[@walkwithus6536]

Hi, if we have multiple tables (huge tables) then this method will work?

[@neuralearn]

Yes, it should work. I think it's best to try it for yourself :)

[@NileshKumar-ug1hl]

Hi, Can you please provide the notebook access?

[@manoubilahbib2572]

That was awesome, thanks

[@chafikhermouche5136]

Hello, thank you for the tutorial !! Can I get the code please ??

[@snehalvats382]

Hey there! it is a wonderful video on how to work with ocr and table. i have requested for notebook access could you please provide me with the access? thank you once again for this tutorial

[@neuralearn]

hello my dear Snehal, please check your mail :)

[@snehalvats382]

@@neuralearn dear team. I have not yet received the confirmation. It's the same email as the one I'm replying with.

[@kanakjaiswal136]

It was excellently explained. I wanted to try it out but got many errors. So, Could you please grant me access to the google Colab code?

[@neuralearn]

Done!

[@AdilKhan-c5q]

Very informative tutorial. I really appreciate the work you have done with this code. I also want to try this. Can you please allow access to the google collab code for this?

[@neuralearn]

hello my dear Adil, Please check your mail :)

[@josephebenezer8869]

Hi, could you grant me access to the notebook please?

[@rupakjha539]

Hi Neuralearn team, can u please provide me the google colab code access

[@sameerdeshmukh1527]

Thank you. Please can you grant me access to notebook?

[@neuralearn]

Please check your mail :)

[@edwinjoe6044]

Hi @Neuralearn. I am getting this "ValueError: (InvalidArgument) Device id must be less than GPU count, but received id is: 0. GPU count is: 0. [Hint: Expected id < GetGPUDeviceCount(), but received id:0 >= GetGPUDeviceCount():0.] (at /paddle/paddle/phi/backends/gpu/cuda/cuda_info.cc:242)" I am having intel® hd graphics 2500 graphics card so I can't run the project in my system how to run the program in my system.

[@neuralearn]

Hello my dear Joe, here is a notebook which works for cpu runtime: colab.research.google.com/drive/1vZHrahaaubhWMz83jlPuvA1na_v98fUP

[@edwinjoe6044]

@@neuralearn Thank you bro Thanks for the support 🤗

[@pmshadow]

hi, excellent video!!! can you please give me access to the notebook? I have requested right now :) Thanks in advance!!

[@neuralearn]

Done!

[@pmshadow]

@@neuralearn thank you very much!

[@neuralearn]

You're welcome :)

[@pmshadow]

@@neuralearn I checked now, but it was not available for me, can you please double check? thanks!

[@neuralearn]

Please check your inbox, we sent you a mail

[@dishaparmar2609]

amazing video..! very helpful ..! could you please provide source code?

[@malakkhiari1419]

How to get access to your notebook?

[@neuralearn]

Done!

[@LouisWong423]

Thank you so much for your tutorial! Can you please grant me access to the Google Colab Code?

[@neuralearn]

Yes sure!

[@hussainahmedsiddiqui3742]

Amazing tutorial, is this code available for use? I would appreciate it!

[@neuralearn]

Please check your mail :)

[@AshishGupta-bd6hu]

Device ID must be less than GPU count, but received Id is:0 GPU count is :0, what does it mean when I run model.detect(image)

[@AshishGupta-bd6hu]

I am running this on my local machine

[@neuralearn]

Hello my dear Ashish, try out this notebook: colab.research.google.com/drive/1vZHrahaaubhWMz83jlPuvA1na_v98fUP

[@AshishGupta-bd6hu]

@@neuralearn thanks for your response, I have sent you access request

[@Sara-fp1zw]

can you please give me the access to notebook?

[@stevevu8654]

it's fascinating. would you mind giving me the access to the colab code?

[@neuralearn]

Hello my dear Steve. Please check your mail :)

[@sudeshkumar5600]

Hi, It is very interesting and to me. I really want to try this out. Could you please grant me access to the google colab code?

[@neuralearn]

Done!

[@AnkushPomendkar-s6f]

This tutorial is very helpful and informative . Can you share this code with me ?

[@neuralearn]

Hi, check your mail box or spam

[@SaniyaFarash]

Very informative video. Can you please share the code with me ? It would be very helpful.

[@youseffarouk6189]

how can i use paddle ocr for receipts ?

[@maniekm2808]

Hi @neuralearn love the tutorial! Can you provide me access to the code?

[@neuralearn]

Please after requesting access, check your mail inbox or spam

[@maniekm2808]

Thank You , chcecked but didn’t get nothing

[@Ga_Yoon-z4k]

Thank you! can i get the code? could you please provide access?

[@neuralearn]

Yes, sure

[@italykiduniya]

The script takes a lot of time to execute is there anyway to optimize it ?

[@neuralearn]

Hello my dear Yadav. To better answer your question, will love to know which part of the script takes a lot of time ? is it the installation? detection?recognition?or reconstruction?

[@ganeshrajv130]

@@neuralearn i have requested for the colab code could you please accept it

[@ganeshrajv130]

@@neuralearn want to try this is so interesting

[@neuralearn]

Done!

[@jyo8507]

What an amazing work. This will be a great tutorial for me in this area of work. I’m trying to access google colab notebook. Could you please grant me permission to access google colab notebook?

[@neuralearn]

Glad it was helpful! Please check your mail

[@jyo8507]

@@neuralearn I am not able to access the notebook. I requested for access.

[@neuralearn]

Please check your mail again

[@jyo8507]

Thanks a lot

[@neuralearn]

You're welcome

[@amilaviraj1014]

This is very informative tutorial! Could you please give me access to the Google Colab Code?

[@neuralearn]

Hi my dear Amila Please check your mail inbox or spam :)

[@tommy-dz1yg]

amazing vid!!!!

[@neuralearn]

Glad you enjoyed it :) More on the way!!!