Machine Learning with R and TensorFlow
HTML-код
- Опубликовано: 21 авг 2024
- J.J. Allaire's keynote at rstudio::conf 2018 on the R interface to TensorFlow (tensorflow.rst...), a suite of packages that provide high-level interfaces to deep learning models (Keras) and standard regression and classification models (Estimators), as well as tools for cloud training, experiment management, and production deployment. The talk also discusses deep learning more broadly (what it is, how it works, and where it might be relevant to users of R in the years ahead).
Slides: beta.rstudioco...
JJ Allaire: - github.com/jja...
Twitter: @fly_upside_down / fly_upside_down
Related blog post: blog.rstudio.c...
This is very useful. Its amazing how the model was able to associate Tusker to the elephant image at 47:01. This is a popular beer brand in Africa that uses the elephant as its logo
Thanks for this excellent introduction to the subject. I really appreciated the fact that you were NOT trying to sell machine learning as a "magic one size fits all solution available for anybody without statistics kwowledge". Unfortunately too often ML is presented as a magical solution by software sellers and I've seen many dangerous attempts to use it blindly for manufacturing decision making. It's not a way to skip our professional responsibility... we have to do our homework and understand what's happening, why it's happening, identify the sources of variations (and address them). And THEN, we could add on top of a mature, understood, in control mfg process a predictive model (ML or not) to sustain or improve quality and performance. This fact was clear in your excellent presentation on top of the very valuable information that you presented, thank you.
Excellent introduction to the topic! This increases my interest in learning more about the topic. I wanna clap along with the audience at the end too. Thank you!
Thank you. Glad this was featured on R blogger. Learnt a lot
Thank you, J.J Allaire and Rstudio. It would have been nice to add the Q&A session as well.
Thank you for great introduction to tensorflow in R.
Great introduction to ML and why it should be used with great care.
Thank you very much Allaire for a wonderful unpacking of this DL library useful at various applications as expressed with clarity and simplicity. I however could not access the ppt. I am intending to apply it in my further research for image time series analysis.
Excellent video and thanks Allairre.
Jump to 31:20
thanks!A Smooth Start
Nice video
Very informative, thank you (y)
Excellent, thanks a lot !
Thanks, Very Informative Video
Brilliant 👏
The link for the presentation isn't working anymore, does someone have the slides please ?
thanks for letting us know! I'll look into this and see what I can find out.
@@PositPBC Still not working
Great into to the package. I get error history part. Says that there is no loss function as shown on slides. Any idea??
I have a problem in loading dataset from my local drive. If anyone help me to load on CloudML, please let me know.
Slide deck link is now 404 😞
where could we find the presented PPT?
Not a PPT, but you can find slides for another presentation that he gave on the same subject here: beta.rstudioconnect.com/ml-with-tensorflow-and-r/#1
Viedo is very good and precise content for someone who wants to understand the concept. This link is not working. Just a black screen appears and nothing. Could you please check and very the link so that we can download the presentation? 5 Min gone on that link and still nothing is happening :(
The link isn't working anymore, does someone have the presentation please ?
IT IS NOT EASY TO SET UP WORKING ENVIRONMENT IN R and WINDOWS !!!!!!!!!!!!!!!!!!!!!
lot of errors tensorflow, keras etc....(search in google)
Honestly, I have so many problems with windows. Literally have 0 problems on Linux, at this point just do machine learning with R on Linux.
Easier than Python 🐍