I tried to break down the Generative AI Stack in 6 minutes (with watsonx)
HTML-код
- Опубликовано: 7 июн 2024
- Whaddayasay, fine tuning video next?
Also, you can try it out here: dataplatform.cloud.ibm.com/wx...
👨💻 Sign up for the Full Stack course and use RUclips50 to get 50% off:
www.coursesfromnick.com/bundl...
🐍 Get the free Python course
go.coursesfromnick.com/python
Hopefully you enjoyed this video.
💼 Find AWESOME ML Jobs: www.jobsfromnick.com
🤖 Get the Code: github.com/nicknochnack/Nopenai
Oh, and don't forget to connect with me!
LinkedIn: bit.ly/324Epgo
Facebook: bit.ly/3mB1sZD
GitHub: bit.ly/3mDJllD
Patreon: bit.ly/2OCn3UW
Join the Discussion on Discord: bit.ly/3dQiZsV
Happy coding!
Nick 
Наука
![Dimension 20: Never Stop Blowing Up Trailer [Dropout Exclusive Series]](http://i.ytimg.com/vi/xYeXpbHnGEM/mqdefault.jpg)
Full disclosure team I work for IBM, not a paid vid just wanted to share some of the sick tech I'm working with!
When it comes to deep learning in industries, is the usage mainly based on frameworks like Keras and TensorFlow, or do they often start from scratch?
Adding external sources to your LLM is referred to as "Retrieval Augmented Generation (RAG)" and helps your model getting the facts straight (+ you can citate the source of the information). Nicholas highlighted this concept with the embeddings and vector stores, though you could also let the LLM generate the embedding at inference time just how it does with prompts. The vector store however helps with quickly finding relevant content to what you specified in the prompt as you can quickly calculate similarities through the distances of the embeddings in the vector space (the closer the vector representation of a word to another word, the similar the word meanings). Fun fact: a lot of the bigger models calculate the embeddings for sub-word tokens, so that the model can deal with unknown words it did not see while training (out-of-vocabulary problem). Nice video and greetings from a fellow IBMer!
Best explanation!! I learned so much from your channel! Thanks so much!!!!
Nicole has this video actually solves all the shoes I had while integrating different API and services thank you so much I had to watch it one time more next slow now
Hey Nick, all of your videos are really great for learning. Have you thought about making a video where you try to teach an AI to battle Pokemon? I think there's a lot of room for teaching and novelty in something like that.
It's really impressive how fast these AI are developing. I'm glad to be watching your videos so I can try and keep up to date lol
Trust me you're extremely ahead of the curve.
This is the coolest intro to LLMs at just 4.3 mins! Wow!!
When it comes to deep learning in industries, is the usage mainly based on frameworks like Keras and TensorFlow, or do they often start from scratch?
Thank so much for your guidance.
great video man, thank you for going deep
Really impressive video on WatsonX.
Great explaination!!
Hi could you do a video about estimating the body measurements through images or pose estimation to find keypoint co ordinantes .
do you have any video where we can build a visual try on using Deep Learning or AI? or planning to make one?
great video - thank you
had a query, is there a open source GenAI tool to generate code ? I see gpt based (openAI) is license or restricted now
Very good explanation. Commented from Bulgarian🇧🇬🇧🇬🇧🇬
Hello nick, I see your course desn't cover SQL topic for beginners to advanced. Do you plan to add this in the future?
Can you explain how to use ready-made templates in huggingface, such as gpt2 on our platforms in WordPress? We want an explanation from scratch and be understandable even for beginners. There is no understandable explanation for this on the Internet. I hope the next video will be
Dude you're the spitting image of Prince with that wig on!
Hi nick, i have been working as an AI engineer for an year but now it seems like I don't know about the basics ... and sometimes i feel like i am just making projects and not creating anything...
What should i do... 😢 ???
Hey Nick , can you make a video of using watsonx to make finetuned qa bots
Can we do image caption generation using data2vec learning?
Hey nick please help me with the idea on how to get this things started, I'm a Jr. Data Scientist fresher in KPMG and joined this month.
Please can you point out how get start with these kind of things, it will be grateful to you.
Love what you are doing and an inspiration to all people like me thank you 😊
Can you do a trading bot for machine leaning I tried with finrl
Hey Bro can you give us the new updated cheat sheet(2023) for data science like 2 years back.
Brother!!! can you make a YOLO object detection from scrach for all YOLO v1 to YOLO v4,It will be a great help,
Basically i am confused about the anchor boxes and parsing of annotations.
I am working on a GAN project and i need your guidance on it, How to reach u.
I'm impressed with this video, and I want to thank you for your effort. Have you thought about making a video about DemoGPT?
Amazing video, can you make a siri or google assistant where it regonize your commond.?
Hi Nicholas, I am a 20 year old fascinated by the thing you do. Can you tell how to start in this field?
Specifically for your tensorflow kivy face ID project? Also can you please mention free resources
Code master!!! Give us the lipnet working with our own videos! Pleaseeeeeeee men
if can't that solve this error when I am going to zip my image and labels which contain n classes and (n x 4) bbox code is throwing me error like that
# first this is my code
train = tf.data.Dataset.zip((train_images, train_labels))
train = train.shuffle(5000)
train = train.batch(8)
train = train.prefetch(4)
# error
InvalidArgumentError: {{function_node __wrapped__IteratorGetNext_output_types_3_device_/job:localhost/replica:0/task:0/device:CPU:0}} Cannot batch tensors with different shapes in component 2. First element had shape [4] and element 2 had shape [10,4]. [Op:IteratorGetNext] name:
Dude this is crazy but unrelated question, what is your skincare routine lol
LOL Ayo Neutrogena and La Roche Posay where the sponsorships at?
Nah, tbh my skin was pretty bad for quite a while because I was eating a ton of sugar and trying to make up for it with products. I don’t use anything now , don’t eat sugar or drink alcohol and run everyday and it’s cleared up massively.
hey nick please make multiple face detection custom model
good video
Thanks a mil!!
Hey, I'm in your LinkedIn. I want to talk with you and work with you, take me as your intern. I just like you and your work way too much!
Usually companies hire interns since they can pay people
Individuals don't typically hire interns as it'll be up to the individual to pay your salary unless you are working for free
plz make a VIDEO on ML and DEEP LEARNING setup in mac book air M1 in entire youtube there is no such a perfect video for this setup
Hi, weldone.
Please, I was following one of your tutorials on Deep facial recognition series part 2,
I made a comment there, please, I'll appreciate if you respond promptly.
Thanks
The Bulgarian word for pizza is пица but is pronounced 'pizza'!
Video quality is quite Poor...