- &quot;Why do you want an iPhone&quot;<br><br>- &quot;I don&#39;t know, it looks cool.&quot;
stringlengths
1
962k
Ok i made tutorial 1 and 2 now but it does not give the same responses as on the openAI playground... there it for examples knows &quot;the lord of the rings&quot;... on the playground my chatbots knows the movie. in the programmed tutorial version it does not. i had the impression its the same i am making a connection...
I love your channel... i do not know if i can ask you something if i need help.. i try to build a chatbot too... but i am beginner. i am glad to found your channel., You can jump into the discord channel (link in description) and ask for help from folks there. We have a dedicated chatbot channel. If you want my dedicat...
Hey David! Thanks for the tutorial, I have it up and running. I was just curious about setting up a was of talking to GPT-3 in cmd without the chatbot style here. Just as a total shell? Is this just a case of chopping some of the code here out or is it a whole new set up? Sorry for the noob question :), ah gotcha. yeah...
Dear David, Can you please help me out?ValueError: Invalid header value b&quot;Bearer -H &#39;Authorization: sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx&#39;\n&quot; I tried different things but..., You should post on the OpenAI community for tech support :)
Helpful! 🙏🏻🙏🏻🙏🏻
Great tutorial, David. Quite astonishing that a complex interaction can be facilitated with just a few lines of relatively simple code. Not so long ago, researchers would have tied themselves in knots trying to code for such a non linear interaction that maintained high levels of semantic congruence. I guess by GPT-4, ...
Damn you&#39;re posting quite frequently, it&#39;s getting hard to keep up with all this amazing content lol. <br><br>I wanted to know if you could make a video talking exclusively about working with embedding models?<br><br>I was a little lost watching the last few videos related to fixing the goldfish memory issues o...
morning everybody David Shapiro here everyone has been clamoring for more tutorials and it took me a while to figure out what to give you a tutorial on because my most popular tutorials are like the basics the introductory stuff um obviously everyone's got to start somewhere so I figured let's start there um someone as...
Hey David! Love your videos. I have a question regarding pdfplumber and pdf layouts; can pdfplumber handle scientific paper structures (columns of text instead of horizontal flow of text)? If so, how would you go about this?, Thjat&#39;s a great question, I have no idea. You should ask ChatGPT ;)
One on one consult fee = $150 an hour :)
It would be interesting to see a tutorial about training a model.<br><br>For example I’d like to know if it would be possible to feed a model the entire Italian Penal Code, It is an effective tool for coding and afaik your audince is in the nlp and ai community I highly recommend using copilot in you videos and maybe t...
So with this i can feed my language model with specific pdf ?
3,000 subs! Lets gooooo
Back from your mini-holiday :)
hey everybody david shapiro here with a quick video well i think it's gonna be quick um i've had a few requests for something that's pretty similar um also let me make sure the sound is good okay cool you can hear me um something that's pretty similar so one person asked for something to uh like summarize um documentat...
Please consider supporting me on Patreon! <a href="https://www.patreon.com/daveshap?fan_landing=true">https://www.patreon.com/daveshap?fan_landing=true</a>
Works very well! Thank you - excellent tutorial
45 iterations of api use? total cost ? just curious 👀 as if you proof read the summarized content. rewrite manually. check and then do the final work. check again and again. <br><br>to make it perfect and to stnadard
When you&#39;re breaking into chunks in this fashion, don&#39;t you risk cutting in the middle of a word/sentence which can impact summarization of that chunk?
very cool. subbed
One of my dreams was to be able to click on parts of a summary and then it would expand it into bigger summaries. (the opposite direction). You would have to know which part in the summary corresponds to the original chunk, but that would be a supercool idea!, We talked about that on Discord. Have different layers of c...
Why not just paste whatever scientific article in openai playground and summarize? I am just a beginner. Thank you, too many characters, Same thought, because I really have no idea how to use phyton even after watching the video I still feel lost on how to make it work.
super useful to say the least, thank you so much mate
Have you ever tried improving the summarization by adding the context before and after a chunk? I’ll probably try this myself, but I wonder if the summarization could be improved if you feed the model some “key points” from the text along with every chunk. You can imagine that if you want a high-quality summarization, ...
hello, I am afraid we can lose some key concepts maybe while connecting the chunks, can you please share what you think about this worry ?, @David Shapiro ~ AI hi David, thx for sharing your script. I just tested it to summarize a book, ogirinally 518k size =&gt; 122k (v1) =&gt; 40k (v2) =&gt; 16k (v3) =&gt; 4.4k (v4)....
I don&#39;t know if it was the intention of the video but I wound up reading the whole summary by using . and , to frame by frame though the summary. I guess I know what the book version of Alice in Wonderland is now.
Just found your Channel and your Plan with this series. This is super impressive. Thanks for doing it. 💯
this was really cool but doesnt feel economically viable with the token usage no?, Depends on what you&#39;re trying to achieve and how much it&#39;s worth to you
thank you :)<br><br>I&#39;m super <i>not</i> corporate, and/but this seems essential, chunks depend on surrounding chunks for further summarisation :) clearly yes thank you.
thank you so much
hey everybody David Shapiro here uh back after a Hiatus um I've got a lot going on you'll be excited for some news coming up um first I want to address um a big elephant in the room I just put most of my videos back online um this comes after of course making a recent video explaining why I took them down um I did leav...
Support me on Patreon so that I can do this full time and keep the content ad free! <a href="https://www.patreon.com/daveshap">https://www.patreon.com/daveshap</a>
<a href="https://youtu.be/E_sMa3N44u4?t=1491">https://youtu.be/E_sMa3N44u4?t=1491</a> you might be able to get the JSON-LD entities with IBM Watson which has an API for it.
This video deserves it&#39;s weight in gold.
Chat GPT-3 for MCL concept: <br><br>Important question for any AI/IT/coding specialists:<br><br>I noticed this legal review regarding chat GPT and they trained it on supreme court case law. <br><br>I&#39;m wondering if something similar could be retooled to be used with GPT-3 for the Michigan compilation of laws for Bi...
&quot;frozen&quot; : =&gt; &quot;continue&quot;
I too, have a lot going on. Yep! BIG things are in store for me friends. <i>BIG</i> things.
Lots of great ideas, information, and how-to knowledge here, so thank you! I can&#39;t help but point out though that the title is very misleading, no? There&#39;s no training going on and we can&#39;t query chat GPT directly about all the documents as is implied. Or am I just missing something?, It&#39;s an experim...
insta sub, I know a wise sage when I see one.
I see that6 the other programs that blend images is now a question of copyright, for the pool of images are owned by others. Fed District Court Nth Cal
This method isn&#39;t scalable. You still have to work within the token limit.
If you want to do knowledge graphs I believe generally the mainstream approach is to extract entities (people, places, things) from your documents, and then the graphing tool displays the relationships. There are open source &quot;entity extraction&quot; tools that vary in focus and in quality. (This has nothing to d...
Thanks for the video tutorial. I have been on your github and installed a few! Thank you. So you have a link to the py.exe? I would like to be able to do tat myself! Cheers.
Probably lawyers adviced against including legal texts such as laws. They are smart people and thought deep about the consequences of having a model with knowledge of the law.
You only get about ten to 20 responses at a time though. It glitches and freezes before you get to finish anything
Yed format is also text based and..you have a great visualiser
thanks alot for awesome content. could you do video for fine-tuning customer support chat
is the underpaid kenyan workers thing real?, @David Shapiro ~ AI any chance you looked into this and can shed some light?, ​@David Shapiro ~ AI a TIME article states: &quot;OpenAI Used Kenyan Workers on Less Than $2 Per Hour to Make ChatGPT Less Toxic&quot;<br>its hard to believe anything read on the internet most of t...
Thank you David. Your channel really opened my mind about AI&#39;s. I live in a poor place and I would never be able to pay for the content you teach here for free. Thanks
Ok speaking of developer stories, here are some I asked ChatGPT to provide about itself:<br><br>1. As a user, I want to be able to provide specific context to ChatGPT so that its responses are more accurate and relevant to my needs. <br> <br>2. As a developer, I want to be able to train ChatGPT on my own data sets so t...
why aren&#39;t you just going to Google Scholar and clicking &#39;case law&#39;?
13337 views
Great video. Thanks for sharing your thoughts. Learned a lot.
<a href="https://www.youtube.com/watch?v=E_sMa3N44u4&amp;t=8m41s">8:41</a> actually you can use another AI toy to scrap automatically the whole webpage and sort it as you wish.. it&#39;s called &quot;Browse AI&quot;
I’ve noticed that specifying a word count improves the specificity. Like if I tell it to give me a response that is at least 2500 words
Love it. I am here for it. Thank you for your leadership. Please continue to lead and inspire us. I wish you the perfect new year. Lisa.
<a href="https://www.youtube.com/watch?v=E_sMa3N44u4&amp;t=20m00s">20:00</a> LOL I lived in France for years and I loved to say that the French and Americans are exactly alike: They both think they are the best(at everything)! :)
Glad my idea has made such an impact.
How has this only got 4k views.
in France right now and loving your videos!
One of the benefits that I have read of is memory of the past prompts and responses within an instance. <br><br>How would you test and perhaps front load context and few facts to improve output?
Hello, Im the colombian lawyer that you asked for in <a href="https://www.youtube.com/watch?v=E_sMa3N44u4&amp;t=10m47s">10:47</a> LOL
i’ve learned a lot about spiritually from you which is mad weird for a python channel
Amazing stuff ! Your channel is under rated, looking forward the next videos
Thank you for this, you&#39;ve expanded my understanding greatly.
Hi David, thank you for starting the tutorial series again. I&#39;d never heard of NLP before stumbling on your channel, and I previously found your tutorials and your way of going through the thinking process incredibly invaluable for a non-coder, non-expert in this domain. I&#39;m excited to see your future projects/...
Very Nice project. Is it possible to implement same steps for other languages?, @David Shapiro ~ AI By the way, I wonder how to keep json file as one piece for a court case without losing context and case details. Wonderful project by the way! I was struggling to find use cases for supreme court text, @David Shapiro ~ ...
Thank you. Seriously you have no idea how good the first few minutes of this video is. You get straight into it, no fluff just good value. I&#39;ve spent hours googling and searching for a video on this topic. Most videos are clearly filler and websites are written for SEO with very minimal content. Watched a few of yo...
Hi. I am a lawyer here in Brazil. Thanks a lot.
Thank you so much! very educating!
This video is very helpful for me as I am facing a similar challenge and it opened up some lines of thought for me. So thank you very much!
Great post!
This was a fantastic, informative video and am excited for the next one. Thank you!
ChatGPT can generate nice mermaid code that can easily visualize graphs and flow charts. You may try it.
Wow, cool example.<br>Hey, David, are you going to use these data for fine tuning?<br><br>What are common use cases for lawyers?, No, finetuning is not helpful in this case. Use case for lawyers is researching case law before going before the Supreme Court. According to Devin (over at Legal Eagle) there are trial lawye...
This is incredibly valuable content. Both from a GPT and general problem-solving perspective. Thanks for putting this together and I&#39;m looking forward to part 2.
I&#39;m so stoked you&#39;re making videos again! You have no idea how much I owe you for what I&#39;ve done with the ideas that you&#39;ve given me. Thank you to infinity
Glad someone tackles that topic and for the explanations, but a more focused approach to what you actually want to show, rather than the process would be a great time saver when watching ^^ still, looking forward to your sollition, so far Im not quite sure where you&#39;ll gop with this.
Great that you put it up again…
Both GPT-3 playground and AI dungeon seem recently to have started incredibly creepy mandatory censoring of content. It&#39;s kind of come as a shock and it&#39;s as if nobody has noticed. It&#39;s seems like it&#39;s crippling it&#39;s own quality a lot, and has me consider alternative providers. 🤷🏻‍♂️, @David Shapi...
hello everybody david shapiro here for another video about gpt3 i figured it was time to get out of theory and actually get into the classroom so let's talk about heuristic imperatives heuristic imperatives are what i am using to build safe benevolent agi or ace what an ace is an artificial cognitive entity so let's ju...
dang! Learning to use gpt3 in a different waay each day! Thank you.
This &quot;write a formal definition&quot; pattern looks useful :), Glad it can help! I figured I can hit multiple birds with one stone. I can teach people how to use GPT-3 while also talking about my favorite topics :)
No repetition penalties? I always have both types turned all the way up to 2., On a model this large, i would fix my prompt and not the model. I imagine repetition penalty as hardcoding and try to avoid it. Most of the time rephrasing your prompt is the way to go., @David Shapiro ~ AIinteresting, thanks, I just wanted ...
hey everybody good morning david shapiro here for another video um we are going to pick up on auto muse so last time quick review this is the second video last time we left off with we kind of ran into a problem because because each each of the chunks got successively larger and so what i mean by that is if we look at ...
Hi David, can you post the link for Part 1? I am not able to find it.
<a href="https://www.youtube.com/watch?v=V6LAsdXkWjo&amp;t=30m53s">30:53</a> that joke 🤣
Great effort man! I&#39;m inspired to do my own version!, DO IT
There is a bug when switching between edit and complete mode it doesn’t switch the model. You just need to select the model again.
What was the final price after all of those tokens ?, Not much. Just a few dollars.
Load in all the Games of Thrones books and Write Winds of Winter for George RR Martin. :D
ah, a fine AI novel
hey everybody david shapiro here with another video um today this was uh not what i expected to be working on but i saw someone on the open ai community ask about um doing youtube chapters or something or other and then someone posted it on the discord and i was like you know what it's time for me to do some seo so i c...
I keep having error
You David Shapiro are an awesome human being
Hi David. This comment isn&#39;t directly related to the video, but it&#39;s your latest and therefore the best opportunity to touch base. I watch all of your videos and find them fascinating, and I really appreciate your &quot;state of the industry&quot; videos as they are informative. Perhaps, your channel might be ...
I really love your videos. instead of just showing how cool ai is, you combined and create things and make it even cooler.
So awesome!
hey everybody david shapiro here um someone asked a question on the open ai forum uh after i posted my recent artificial cognition cognitive entity video they asked how will we know when we've achieved agi how will we know when that's complete and the the first answer the short version is um it'll never be complete um ...
<a href="https://www.youtube.com/watch?v=x96Thu0PULg&amp;t=0m37s">0:37</a> me in college 😅