I have published this as a full on online course in my website along with interactive assessments and course completion certificates (of course free). I have explained the process of signing up to this course here : czcams.com/video/nnIBXwSyhQQ/video.html course landing page : courses.vizkumar.com
Glad you liked it. Feel free to also checkout the course I created on the same topic. It includes interactive quizzes etc. You can also generate a course completion certificate as well. Also there is a discord group which you can join and ask any doubts / clarifications. Happy to help!
Thanks for your feedback bro. Glad you liked it. Feel free to also checkout the course I created on the same topic. It includes interactive quizzes etc. You can also generate a course completion certificate as well. Also there is a discord group which you can join and ask any doubts / clarifications. Happy to help!
Bro oru doubt. Ennoda friend who is a data scientist at Amazon. He was telling me that, nowadays we just use prompts to increase the performance of LLM to reach like F1 score of 90. Furthur finetuning LLMs is not even desirable cause it takes weeks to run but still will result only in an addition of may be a score or 1 or 2 in F1. What do you think of that?
Hi 👋. There are various reasons for fine tuning. Recent research do suggest that prompt engg based fine tuning is preferable than training based fine tuning. One example why need fine tuning is domain based LLMs . For eg if you wanna personalize a general LLM for medical domain , then prompt Engg based fine tuning may not yield best results. It needs a traditional approach. Look at Microsoft’s phi small language model. Same use case
Hi Bro, in the video I used open AI’s gpt-3.5-turbo model. It has limited exposure tanglish (which is what I demonstrated) . Right now more advanced models are available. So gpt4 will be more powerful and more fluent in Tamil and Tanglish language.
Hi bro, create an account at openai and access this link to create your openai API key . Link : help.openai.com/en/articles/4936850-where-do-i-find-my-secret-api-key
I have published this as a full on online course in my website along with interactive assessments and course completion certificates (of course free).
I have explained the process of signing up to this course here : czcams.com/video/nnIBXwSyhQQ/video.html
course landing page : courses.vizkumar.com
thank you brother
at the same time wish you for your channel 💐💐💐💐
Thank you for making my day 😊🤩
Muthalla GitHub yeppati login panrathupodonga bro😢
Explained so well, thanks bro
Glad you liked it. Feel free to also checkout the course I created on the same topic. It includes interactive quizzes etc. You can also generate a course completion certificate as well. Also there is a discord group which you can join and ask any doubts / clarifications. Happy to help!
Thank you sir
Glad you liked it 🙏
Anna,send the prompt to openai model kudukum pothu,ratelimit exceeded nu varuthu.What to do
Hi can you join CIT discord server and post your query under “machine-learning” channel? That way you can get necessary help.
Very Good explanation 👍
Glad you like it 😊
Very nice
Glad you liked it 😊
Excellent. learned a lot. Thanks.
Thanks for your feedback bro. Glad you liked it. Feel free to also checkout the course I created on the same topic. It includes interactive quizzes etc. You can also generate a course completion certificate as well. Also there is a discord group which you can join and ask any doubts / clarifications. Happy to help!
Bro oru doubt. Ennoda friend who is a data scientist at Amazon. He was telling me that, nowadays we just use prompts to increase the performance of LLM to reach like F1 score of 90. Furthur finetuning LLMs is not even desirable cause it takes weeks to run but still will result only in an addition of may be a score or 1 or 2 in F1. What do you think of that?
Hi 👋. There are various reasons for fine tuning. Recent research do suggest that prompt engg based fine tuning is preferable than training based fine tuning. One example why need fine tuning is domain based LLMs . For eg if you wanna personalize a general LLM for medical domain , then prompt Engg based fine tuning may not yield best results. It needs a traditional approach. Look at Microsoft’s phi small language model. Same use case
@@conceptsintamil Got it. Thanks
Nice
Thank you 😊
Nice, can you please explain with Node JS
Sure bro. I’ll try to include more JS examples next time
Really nice.
Glad you liked it
Bro chapgpt ya tanglishla respond vainga
Hi Bro, in the video I used open AI’s gpt-3.5-turbo model. It has limited exposure tanglish (which is what I demonstrated) . Right now more advanced models are available. So gpt4 will be more powerful and more fluent in Tamil and Tanglish language.
Bro ungaloda api key display aaguthu 14:27
😂😂😂 . Thanks bro. I deleted the API key when I uploaded this video . Ungalukku kalugu paarva bro 😎
Where can I find the API keys?
Hi bro, create an account at openai and access this link to create your openai API key . Link : help.openai.com/en/articles/4936850-where-do-i-find-my-secret-api-key
@@conceptsintamil I got this error
RateLimitError: You exceeded your current quota, please check your plan and billing details.
It means you should buy some quota bro. Initially they will give 5$ credit. Once you consume it you need to buy additional quota to use openai api
Bro in Senior QA whether prompt engineering QA role opening is there or only development is there
Hi Bro, prompts should be developed with safety in mind. Hence there are roles related to AI safety or responsible AI etc. You can try those roles bro
Ithu AI prompt thana
Ama bro.
Idhuku job vacancy india la yepidi bro apply pandrathu what are the skills we need? And plz share job title?
India layum ithu popular aitrukku bro. Search for prompt engineer or NLP engineer or AI/ML engineer etc in Indian job vacancy sites such as Naukri etc
I’m planning to publish a IT roadmap 2023 video. In that I’ll share details about different job titles and skills needed. Pls stay tuned 😁🙏
Nice
Glad you liked it 😊