![Jeff Heaton](/img/default-banner.jpg)
- 635
- 7 306 531
Jeff Heaton
United States
Приєднався 26 лис 2006
Videos about my machine learning projects.
** Follow Me on Social Media!
GitHub: github.com/jeffheaton
Twitter: jeffheaton
Instagram: jeffheatondotcom
Discord: discord.gg/3bjthYv
Patreon: www.patreon.com/jeffheaton
** Follow Me on Social Media!
GitHub: github.com/jeffheaton
Twitter: jeffheaton
Instagram: jeffheatondotcom
Discord: discord.gg/3bjthYv
Patreon: www.patreon.com/jeffheaton
Model Drift and Retraining (13.3)
Some thoughts on the future of reinforcement learning.
Code for This Video:
github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_13_3_retrain.ipynb
~~~~~~~~~~~~~~~ COURSE MATERIAL ~~~~~~~~~~~~~~~
📖 Textbook - Coming soon
😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/
▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.html
🏫 WUSTL Course Site - sites.wustl.edu/jeffheaton/t81-558/
~~~~~~~~~~~~~~~ CONNECT ~~~~~~~~~~~~~~~
🖥️ Website: www.heatonresearch.com/
🐦 Twitter - jeffheaton
😸🐙 GitHub - github.com/jeffheaton
📸 Instagram - jeffheatondotcom
🦾 Discord: discord.gg/3bjthYv
▶️ Subscribe: ua-cam.com/users/heatonresearch
~~~~~~~~~~~~~~ SUPPORT ME 🙏~~~~~~~~~~~~~~
🅿 Patreon - www.patreon.com/jeffheaton
🙏 Other Ways to Support (some free) - www.heatonresearch.com/support.html
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#reinforcementlearning #gymnasium #PyTorch #DeepLearning #PyTorchTutorial #MachineLearning
Code for This Video:
github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_13_3_retrain.ipynb
~~~~~~~~~~~~~~~ COURSE MATERIAL ~~~~~~~~~~~~~~~
📖 Textbook - Coming soon
😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/
▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.html
🏫 WUSTL Course Site - sites.wustl.edu/jeffheaton/t81-558/
~~~~~~~~~~~~~~~ CONNECT ~~~~~~~~~~~~~~~
🖥️ Website: www.heatonresearch.com/
🐦 Twitter - jeffheaton
😸🐙 GitHub - github.com/jeffheaton
📸 Instagram - jeffheatondotcom
🦾 Discord: discord.gg/3bjthYv
▶️ Subscribe: ua-cam.com/users/heatonresearch
~~~~~~~~~~~~~~ SUPPORT ME 🙏~~~~~~~~~~~~~~
🅿 Patreon - www.patreon.com/jeffheaton
🙏 Other Ways to Support (some free) - www.heatonresearch.com/support.html
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#reinforcementlearning #gymnasium #PyTorch #DeepLearning #PyTorchTutorial #MachineLearning
Переглядів: 459
Відео
Tensor Processing Units (TPUs) (13.4)
Переглядів 1,3 тис.14 днів тому
Tensor Processing Units (TPUs) are a Google technology that can speed neural network training and inference much like GPUs. This video shows how to use a TPU with PyTorch. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_13_4_tpu.ipynb COURSE MATERIAL Textbook - Coming soon 😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/ ▶️ Play List - ua-cam.com/play/P...
Using Denoising AutoEncoders (13.1)
Переглядів 73821 день тому
This video presents autoencoders in Python! In this video, we dive deep into the world of autoencoders, a type of artificial neural network used for unsupervised learning. We'll start by exploring the basics: what autoencoders are, how they work, and why they're so crucial in the realm of machine learning and data compression. Then, we'll shift our focus to Python, demonstrating how to implemen...
Anomaly Detection (13.2)
Переглядів 960Місяць тому
This video introduces Anomaly Detection using Autoencoders in PyTorch. In this session, we'll delve deep into the fascinating world of deep learning, particularly focusing on the powerful technique of autoencoders and their application in identifying anomalies in data. We'll start by introducing the basic concepts of autoencoders, explaining how these neural networks learn to compress and recon...
Future of Reinforcement Learning (12.5)
Переглядів 971Місяць тому
Some thoughts on the future of reinforcement learning. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_12_5_rl_future.ipynb COURSE MATERIAL Textbook - Coming soon 😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/ ▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.html 🏫 WUSTL Course Site - sites.wustl.edu/jeffheaton/t81-558/ CONNECT 🖥️ Web...
Atari Games with Stable Baselines Neural Networks (12.4)
Переглядів 9613 місяці тому
Learn to use stablebaselines to teach agents to play Atari games using gymnasium in Python. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_12_4_atari.ipynb COURSE MATERIAL Textbook - Coming soon 😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/ ▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.html 🏫 WUSTL Course Site - sites.wustl.edu/j...
NVIDIA GTC 2024 Jensen Huang Keynote Blackwell Reactions
Переглядів 9223 місяці тому
Link to sign up to GTC 2024: nvda.ws/3R2hdIj Giveaway Instructions Video: ua-cam.com/video/23YCQGR3ayY/v-deo.html&ab_channel=JeffHeaton Session Picks: ua-cam.com/video/gRLrARn4TxY/v-deo.html&ab_channel=JeffHeaton CONNECT 🖥️ Website: www.heatonresearch.com/ 🐦 Twitter - jeffheaton 😸🐙 GitHub - github.com/jeffheaton 📸 Instagram - jeffheatondotcom 🦾 Discord: discord.gg/3bj...
NVIDIA GEFORCE GTX 4080 GTX GPU Giveaway for GTC 2024
Переглядів 1,6 тис.3 місяці тому
Link to sign up to GTC 2024: nvda.ws/3R2hdIj Giveaway Email: jheaton.giveaway@gmail.com Session Picks: ua-cam.com/video/gRLrARn4TxY/v-deo.html&ab_channel=JeffHeaton CONNECT 🖥️ Website: www.heatonresearch.com/ 🐦 Twitter - jeffheaton 😸🐙 GitHub - github.com/jeffheaton 📸 Instagram - jeffheatondotcom 🦾 Discord: discord.gg/3bjthYv ▶️ Subscribe: ua-cam.com/users/heatonresear...
Course Overview: Applications of Generative AI (1.1)
Переглядів 1,5 тис.3 місяці тому
This course covers the dynamic world of Generative Artificial Intelligence providing hands-on practical applications of Large Language Models (LLMs) and advanced text-to-image networks. Using Python as the primary tool, students will interact with OpenAI's models for both text and images. The course begins with a solid foundation in generative AI principles, moving swiftly into the utilization ...
GTC 2024 Top Picks, NVIDIA
Переглядів 1,8 тис.3 місяці тому
Link to sign up to GTC 2024: nvda.ws/3R2hdIj Session Picks: GTC 2024 Keynote [S62542] register.nvidia.com/flow/nvidia/gtcs24/attendeeportaldigital/page/sessioncatalog?search=S62542 Can My Model Be Hacked? Understanding and Mitigating Security Vulnerabilities within LLMs [S61512] register.nvidia.com/flow/nvidia/gtcs24/attendeeportaldigital/page/sessioncatalog?search=S61512 Accelerating Pandas wi...
NVIDIA 2024 is Just Around the Corner... Giveaway Kickoff
Переглядів 6694 місяці тому
Link to sign up to GTC 2024: nvda.ws/3R2hdIj CONNECT 🖥️ Website: www.heatonresearch.com/ 🐦 Twitter - jeffheaton 😸🐙 GitHub - github.com/jeffheaton 📸 Instagram - jeffheatondotcom 🦾 Discord: discord.gg/3bjthYv ▶️ Subscribe: ua-cam.com/users/heatonresearch SUPPORT ME 🙏 🅿 Patreon - www.patreon.com/jeffheaton 🙏 Other Ways to Support (some free) - www.heatonresearch.com/supp...
Stable Baselines Q-Learning (12.3)
Переглядів 7304 місяці тому
Stablebaselines allows you to quickly adapt deep reinforcement models to gymnasium environments. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_12_3_pytorch_reinforce.ipynb COURSE MATERIAL Textbook - Coming soon 😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/ ▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.html 🏫 WUSTL Course Site - ...
Introduction to Q-Learning (12.2)
Переглядів 2,3 тис.5 місяців тому
Learn to use qlearning to solve the cart problem in gymnasium with Python. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_12_2_qlearningreinforcement.ipynb COURSE MATERIAL Textbook - Coming soon 😸🐙 GitHub - github.com/jeffheaton/app_deep_learning/ ▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.html 🏫 WUSTL Course Site - sites.wustl.edu/j...
LLM Memory with LangChain (6.3)
Переглядів 2,7 тис.5 місяців тому
LLMs do not have the sort of memory where you often see ChatGPT recall information from earlier in the conversation. This sort of memory must be added by the host program. In this video I will show how LangChain can add this sort of memory to a conversation. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_06_3_llm_memory.ipynb COURSE MATERIAL Textbook - com...
LangChain and Accessing the OpenAI LLM API (6.2)
Переглядів 1,2 тис.5 місяців тому
In this video we will see LangChain and how it can be used to access a variety of LLM's and can make several access patterns easier. Code for This Video: github.com/jeffheaton/app_deep_learning/blob/main/t81_558_class_06_2_chat_gpt.ipynb COURSE MATERIAL Textbook - coming soon 😸🐙 GitHub - github.com/jeffheaton/app_deep_learning ▶️ Play List - ua-cam.com/play/PLjy4p-07OYzuy_lHcRW8lPTLPTTOmUpmi.h...
Applications of Deep Neural Networks PyTorch Course Overview (1.1, Spring 2024)
Переглядів 4,1 тис.5 місяців тому
Applications of Deep Neural Networks PyTorch Course Overview (1.1, Spring 2024)
Introduction to Introduction to Gymnasium (12.1)
Переглядів 2,7 тис.6 місяців тому
Introduction to Introduction to Gymnasium (12.1)
Future Directions in Artificial Intelligence (13.5)
Переглядів 2 тис.6 місяців тому
Future Directions in Artificial Intelligence (13.5)
Training a Model in Hugging Face (11.5)
Переглядів 2,1 тис.6 місяців тому
Training a Model in Hugging Face (11.5)
Hugging Face Introduction in Python (11.2)
Переглядів 1,7 тис.7 місяців тому
Hugging Face Introduction in Python (11.2)
Introduction to Natural Language Processing and Hugging Face (11.1)
Переглядів 1,1 тис.7 місяців тому
Introduction to Natural Language Processing and Hugging Face (11.1)
Predicting with Meta Prophet (10.5)
Переглядів 1 тис.7 місяців тому
Predicting with Meta Prophet (10.5)
Up to 150x GPU PANDAS Speedup with No Code Changes
Переглядів 4 тис.7 місяців тому
Up to 150x GPU PANDAS Speedup with No Code Changes
Transformer-Based Time Series with PyTorch (10.3)
Переглядів 12 тис.8 місяців тому
Transformer-Based Time Series with PyTorch (10.3)
LSTM-Based Time Series with PyTorch (10.2)
Переглядів 4,5 тис.8 місяців тому
LSTM-Based Time Series with PyTorch (10.2)
Time Series Data Encoding for Deep Learning, PyTorch (10.1)
Переглядів 2,3 тис.8 місяців тому
Time Series Data Encoding for Deep Learning, PyTorch (10.1)
Bayesian Hyperparameter Optimization for PyTorch (8.4)
Переглядів 1,3 тис.8 місяців тому
Bayesian Hyperparameter Optimization for PyTorch (8.4)
Hi Jeff interesting topic. I actually discussing the topic, that many acadamic (but also industrial) dataset are never tested for data drifts (especially if they are tabular). Basically one simply assumes there is no data drift and a common full random k-forld cross validation with test dataset is executed and the results are seen as stable. However, they are not. My approach for the dataset of my specific application was, that I compressed the dataset with a PCA (but any other dimensional reduction method would work as well) to plot it in a 2D scatter plot, but as color index I display class and data point index. By this, anyone could see that there is a clear correlation between time and the clustered derived by the PCA and this delivered a perfect explanations for the experimental results, that also have shown amongst others that reported reality in academic paper is often not the true reality that one would face in the real world deploying a model.
The willingness to listen and learn from others is commendable, fostering a culture of mutual growth.💋
❤I like your tutorial
Thank you!!
the information I was looking for, thanks
best budget ai card right now is the 4060ti with 16gb ram at 400 bucks
NVIDIA Quadro RTX5000 This video card has Fermi Architecture. I think it is outdated for using many python libraries like RAPIDS, etc.
i got an 11$ google collab subscription works well for me . if i needed to upgrade ill just pay more monthly
Hi Thanks a lot. This command worked for me: conda install pytorch::pytorch torchvision torchaudio -c pytorch
Is GEFORCE RTX 2050 is cuda supported??
You really should add a photo sensitivity warning for this video, it didn't go well for me
Какие настройки? Какое разрешение выходного изображения? Хватит ли 8 гб ОЗУ для локальной генерации через Stable-diffusion ?
guess what my 'wsl --install' command having errors
this is great. thanks! what would be the implications of having 2 or 3 different GPUs?
OMG, do you hear that things power supplies through his MIC just by touching the case WOW! ??
thanks for the explanation, what if we use LSTM along with transformer(attention mechanism), it would be helpful or just make the model complex?
😢😢😢
Guess what Wsl --install command is not recognized
Thank you very much! Finally a video straight to the point and clearly explains everything. So helpful. Really Appreciate it.
I am going to start at 45 by taking 2y master in data science after 24 years working in corporate finance...crazy idea, what do you think?
literally no information on TPUs - no speed or perfornance comparisons, no tech details... what did I just watch?
It would be interesting to have a video on the techniques for dealing with VRAM limitations for local AI models. I guess some of the Nvidia tools are addressing this issue? What does it take to run some of the open source LLMs on a 12Gb GPU? What are the options if it's 8Gb VRAM? is a flat no go, or does it run, but slower? Forgive me if there's already a video detailing this... 🙏
Are you in New builder? This is the first time I've seen a video and I'm not trying to be mean I just am unsure. The best way to put on thermal paste hands down is to use the spreader and make a thin layer evenly spread that way there's no guessing to it. From what it looks like you're going to have too much. Also the information and recommendation about water cooling is inaccurate. Well I don't fault you for it as it was advice that you were given but it being a data driven setup is even more of a reason to run water cooling. They're extremely reliable. Maybe 10 years ago they weren't. I'm building three deep learning AI systems concurrently right now. All of them will have full custom water cooling systems built by me and each one will have dual pumps as there's lies the only risk and even that is rare. That way if for some reason one pump goes out I still have no issues and worst case scenario it'll throttle and shut itself off anyways without damaging it
This is like demonstrating the work we've done for an assignment.. "We did this then did this.. " Please teach the things you have done in the code so we can learn them and apply to our own applications.
is there a way to run it using wsl but not using jupyter? i hate having notebooks
Wow, this video is what I have been looking for. I am currently trying to build a dual 3090 platform for fine-tuning and inference, and most importantly, I want to see if SLI is necessary!
Hi, Jeff. It's been 5 years but I still find this content useful. May I ask what version of tensorflow, pandas, numpy, and sklearn libraries you used to execute the notebook included in the video?. Thank you very much!
What about today value
Coral cards use TPUs, a low cost, low power method to get processing (mainly video) in a computer
Thank you Jeff.
64FPS? Why does it seem it's half of it?
I don't understand, the balls are entering through a hole in the middle, of course it's going to be more balls in the middle
Appreciate this very much!
Your explanation is very simple wow great job man, all respect
I have a 3080 10gb, and 3060 12gb, I was wanting to try and run together.
it is not too late at 50 years old. I remember working in the library when I was fifteen in 1994 and the internet was just coming out. the computer monitor only display text. a few years later we were able to get picture on the computer. my understanding is that it wasn't long ago. it is just the advancement of computer technology that has move very fast. we have been bombarded with so much technology within this 30 years time period. From smartphone to AI and streaming service to machine learning. The hardest part is keeping my mind updated.. it is easy for younger kids because they have time and college money to study all day long.
Exactly. For example, tensorflow doesn't support AMD GPU!
Can this work on windows 10?
Thank you Jeff !
just in time for my thesis thank you very much ❤️
this may not be the video for it but can you please please help me out I am using windows 11 Nvidia drivers is 555.99 which supports CUDA 12.5 but Cuda 12.5 does not support TensorFlow 2.10 and im stuck in this infinte loop of fixing one thing after another
I've done everything you do in the video exactly and yet I get a "False".
Thanks man this was a good video. Nothing as frustrating as installing Tensorflow on windows.
Can we use that video card to crack the passwords ? Like big library brute force atack 💙Please make a video abut 'it!
Thanks Jeff
Thank you so much I feel so grateful for this course =)
Watching your videos (old) somewhere 5 years ago, now I am having a few papers on it. Thank you for providing an interesting walk- around....🎉 Keep it up 💪
One key step is: reboot before test the tf installation.
The graphics are OK I guess..
there is no option for background execution