@@supreetdhanjal334 yes it does but they haven't put the exact amount that will be consumed by the activity yet. I think the licensing model will be out with 23.10 version where this will no longer be in preview. Let me know if this is the answer you're looking for?
Amazing work and great video!! Very well explained as well. Thank you for this content. Do you know if there is a way that we can make the prompts dynamic? Meaning we take the Name of the Field and the Generative Prompt to be extracted as an Argument for example?
Hello, Jose. I've shared a very similar video on how to write an extraction to excel, it just needs to be inside of a datatable for simplicity, you can find that part here ua-cam.com/video/vbElqDB73vA/v-deo.html if you still can't find this to be helpful, just join the discord channel I can show you how to do that there, the discord link is: discord.gg/jUe948xsv4
Awsome Video.. Very Insightful. Thank you for such great Content. I have a question though, in the Enterprise Version, if I don't using the classify activity and only go through with the Extract Document Data activity that you used, is this going to consume any AI Units for this? Thanks again Reda, Really Appreciate your content.
You're welcome, This is a second part video, I created the sequence from scratch in the first video, you can find it here ua-cam.com/video/LmM21lgfCEE/v-deo.htmlsi=LCRDOOSU3G58BnAe Let me know if you have any questions.
Could you please provide the link to download this specific version of UiPath? I'm unable to find the UiPath.DocumentUnderstanding.Activities in the latest version
You should have access to the package on any new version of UiPath, I'm working with the latest version of UiPath and I can see UiPath.DocumentUnderstanding.Activities on all packages inside of manage packages (latest version is v2.7.0 btw)
Thanks for the video. Great job! I would like to ask if it's possible to use your solution to extract information from non-English invoices (like a German one). Does extraction still work well, or should we make some adaptations?
You're welcome. it worked on french documents without adaptations so I suppose it would work on German ones. Try it and let me know if it worked or not.
well the good news is that gen ai is getting better at reading text. but if you find always giving 100% you can tell it to assign the 100% only in the cases where it's absolutely sure. Let me know if this approach helps?
Hi Reda, really helpful video. :) I installed the package version 2.4 but "Extract Document" activity is showing error that my studio is not connected to Orchestrator tenant." Do I need to have Automation Cloud account or any API ? Thank you
Hello Rudravas, Yes you need a UiPath cloud account on cloud.uipath.com you can create an account there for free and connect your uipath studio to that account and once you'll use the activity Extract document the API key will already be inside of the activity without your intervention. Join the discord channel if you have further questions: discord.com/invite/jUe948xsv4
Hi Reda. I did some tests but this error always appears "Extract Document Data: Request headers must contain only ASCII characters." Do you know what it could be? Great Video
Hi Thiago, Glad you enjoyed the video :D I've never got this error, i got a similar error when I had a special character in the values of the column in the activity Extract Document Data. Check if this is the case and use only normal letters inside of it, if this is not the case we should investigate further.
Hello Suraj, I made a video where I showed how to read scanned documents, you can watch it here: ua-cam.com/video/vbElqDB73vA/v-deo.htmlsi=eOtkAuAiM_yHNl1L As per Table extraction I haven't made it yet because Document Understanding is getting major updates and it's going to be so much easier in the next few months. I'm waiting for the update and I will make the video ASAP.
@@suraj_singh321 oh yeah extraction tables using Regex is possible to make now without the need to wait. I will add the video to the pipeline for the next videos. Stay tuned 😁
If it's an electronic signature with a written name then you can add the following: Key: signature Name Value: the name of the electronic signature in the document. If it's a normal signature I don't really know what to extract because most of the time it's not text. Let me know if you mean something else.
@@Anjaan_Ajab I don't have an exact number of pages, but there are limits to both computer vision and document understanding models, I would say if you need it just use it until you can't anymore, it's free after all.
Great video, Subscribed
Glad you liked the video.
This is amazing! Great video, well explained! Keep going! Subscribed!
Thank you, I really appreciate it!
Hey Reda! Great explanation. I have a question: what is your approach to extract the items on the invoice?
Amazing work 👏
Glad you enjoyed the video!
This video was very helpful for me!
My pleasure, I’m glad you found it useful 😁
I had a small query.
Does this consume any AI units?
@@supreetdhanjal334 yes it does but they haven't put the exact amount that will be consumed by the activity yet.
I think the licensing model will be out with 23.10 version where this will no longer be in preview.
Let me know if this is the answer you're looking for?
Subscribing right away!❤
Amazing work and great video!! Very well explained as well. Thank you for this content. Do you know if there is a way that we can make the prompts dynamic? Meaning we take the Name of the Field and the Generative Prompt to be extracted as an Argument for example?
You can now define variables or arguments inside the generative prompt it has been added in the latest version.
Hello Reda, great video.
But are you ready to transfer the data to an Excel sheet? You said you would months ago.
Thank you
Hello, Jose.
I've shared a very similar video on how to write an extraction to excel, it just needs to be inside of a datatable for simplicity, you can find that part here
ua-cam.com/video/vbElqDB73vA/v-deo.html
if you still can't find this to be helpful, just join the discord channel I can show you how to do that there, the discord link is: discord.gg/jUe948xsv4
Awsome Video.. Very Insightful. Thank you for such great Content. I have a question though, in the Enterprise Version, if I don't using the classify activity and only go through with the Extract Document Data activity that you used, is this going to consume any AI Units for this? Thanks again Reda, Really Appreciate your content.
Thank you, yeah both of those activities consume AI Units.
Hi Reda
Great Video :) Could you by any chance create a new video onto how we could add this data to an Excel file
Thanks
Hello,
Thank you,
Yeah I should make a video about it, I got this request multiple times.
Thank you for your informative video. Am I missing or you did not give the step by step sequence elements such as from stratch?
You're welcome,
This is a second part video, I created the sequence from scratch in the first video, you can find it here ua-cam.com/video/LmM21lgfCEE/v-deo.htmlsi=LCRDOOSU3G58BnAe
Let me know if you have any questions.
Could you please provide the link to download this specific version of UiPath? I'm unable to find the UiPath.DocumentUnderstanding.Activities in the latest version
You should have access to the package on any new version of UiPath, I'm working with the latest version of UiPath and I can see UiPath.DocumentUnderstanding.Activities on all packages inside of manage packages (latest version is v2.7.0 btw)
Join the Discord community: discord.gg/jUe948xsv4
l can't join the discord by this link, would you share again? thanks
@@waiwong8843 Thank you for letting me know, Here is the new link: discord.gg/jUe948xsv4
Thanks for the video. Great job! I would like to ask if it's possible to use your solution to extract information from non-English invoices (like a German one). Does extraction still work well, or should we make some adaptations?
You're welcome.
it worked on french documents without adaptations so I suppose it would work on German ones.
Try it and let me know if it worked or not.
Hi Do you have any video related to DU framework using attended and unattended explaining all activities and config as well please
Hello,
Not yet, I'm waiting for some new functionalities to be added so I can make a video about the DU Framework.
so we cannot put condition "send to action center based on confidence" bcos genai is always 100%
well the good news is that gen ai is getting better at reading text. but if you find always giving 100% you can tell it to assign the 100% only in the cases where it's absolutely sure.
Let me know if this approach helps?
Hi Reda, really helpful video. :) I installed the package version 2.4 but "Extract Document" activity is showing error that my studio is not connected to Orchestrator tenant." Do I need to have Automation Cloud account or any API ? Thank you
Hello Rudravas,
Yes you need a UiPath cloud account on cloud.uipath.com
you can create an account there for free and connect your uipath studio to that account and once you'll use the activity Extract document the API key will already be inside of the activity without your intervention.
Join the discord channel if you have further questions: discord.com/invite/jUe948xsv4
@@redamarzouk Thanks so much for your prompt response.
Hi Reda. I did some tests but this error always appears "Extract Document Data: Request headers must contain only ASCII characters." Do you know what it could be? Great Video
Hi Thiago,
Glad you enjoyed the video :D
I've never got this error, i got a similar error when I had a special character in the values of the column in the activity Extract Document Data.
Check if this is the case and use only normal letters inside of it, if this is not the case we should investigate further.
Hi sir make a video on scanned pdf data including table extraction as it is write to excel using regex and data manipulation uipath.
Hello Suraj,
I made a video where I showed how to read scanned documents, you can watch it here: ua-cam.com/video/vbElqDB73vA/v-deo.htmlsi=eOtkAuAiM_yHNl1L
As per Table extraction I haven't made it yet because Document Understanding is getting major updates and it's going to be so much easier in the next few months.
I'm waiting for the update and I will make the video ASAP.
@@redamarzouk if you feel free so follow this regex and data manipulation method if possible
@@suraj_singh321 oh yeah extraction tables using Regex is possible to make now without the need to wait.
I will add the video to the pipeline for the next videos. Stay tuned 😁
@@redamarzouk thanks a lot
Can I get the documents you used for the demonstration?
Hello,
You can find all the documents in the discord server: discord.gg/jUe948xsv4
Channel: Information > #Ressources
How you extract a signature for example ?
If it's an electronic signature with a written name then you can add the following:
Key: signature Name
Value: the name of the electronic signature in the document.
If it's a normal signature I don't really know what to extract because most of the time it's not text.
Let me know if you mean something else.
How to write properly in Excel.
Hello,
I will make a video about this where I will complete the process by writing in Excel.
if you can't show the output then how can able to guess did you get that correct output or not ??????????
Can you point out to which output you're referring to ?
What's the package name?
UiPath.DocumentUnderstanding.Activities
@@redamarzouk Thank you! Let me check it out. how many documents can I process with community edition? any idea?
@@Anjaan_Ajab I don't have an exact number of pages, but there are limits to both computer vision and document understanding models, I would say if you need it just use it until you can't anymore, it's free after all.