Hi everyone,
We deployed a Shiny App (https://modera-climeapp.shinyapps.io/climeapp/)
Unfortunately, It seems to be crashing quite often.
Possibly due to too few ram (8 GB) at the moment.
However, in the next summer school around 20 students will need it.
Is it possible to buy more than 8GB of ram?
At the moment not even 3 people can use it (we have the Basic plan, set it to 8GB ram and tried different setting options.
Could anyone have a help?
I hope one of the experts on this platform responds to your query. My own experience as a student working to receive the Google Data Analytics Certificate by completing some capstone projects using RStudio has been unfavorable so far. I've already had to purchase the Basic Package but even though that was only yesterday, most of the 8GB is already taken up - and I'm still only on my first project and have more experimentation to work out the kinks inherent in any learning experience. Have already resolved to get off R after this case study, if I EVER finish, and just use Tableau and SQL in addition to Excel.
You could just use R/RStudio on the desktop.
I haven't looked at the Google course but if that can be completed efficiently using Excel and Tableau I wouldn't consider it a certificate of much value.
No, it doesn't just train in Tableau and Excel, but in R and SQL as well. It's a 240-hour course, which is several months if one has a full-time occupation too. I've an old computer which would crash with the required Windows 10/11 needed for the RStudio desktop.
I'm in thesame shoe as you. I am also encountering challenges with using R to clean and analyze my "cylicst ride trip" which forms completion requirement for google data analytics capstone project. I had issue(s) using RStudio-Desktop, and moved it to RStudio-Cloud but couldnt complete cleaning because i was on a free sub with max 1gb RAM. I'm back on RStudio Desktop and want to give it a final try with the RDesktop 'version 4.3.1'. If i dont get through! will it be advisable to buy the basic cloud plan of 8gb RAM? considering you already purchased one - I want to be sure of value for money. I will appreciate if we can connect and walk over this together. Thank you!
I've since purchased the 8GB upgrade which is helpful for now, although even that is a small amount of memory for a new person to R (it's necessary to write and rewrite code and try other things to accomplish roughly the same result, all while using up a lot of RAM). My computer is old so I can't replace its Windows 7 with any version more than 8 lest it crashes, and we wouldn't want that.
Now I have a new problem with uploaded Excel spreadsheet - attempting to clean it up a bit using the functions involving read, filter, and write, since to do the same thing in Excel takes waayyy too long and might crash it. But even chataigpt isn't giving me the right code, so far.
Anyway, I sincerely wish there was a class with an instructor and students so that every new problem (which happens every day as a new R person) could be dealt with on the spot and become part of the learning experience. Let me know any questions or comments on the capstone project.
If your PC is very old, you still have some options (depending on how much RAM it has or how much can be added), such as using R without RStudio (you could try VS Code if you really need an IDE) or installing a lightweight Linux distro. There are packages which allow RAM to be replaced by disk space instead (disk.frame, arrow).
I don't know how much RAM is required for the tasks in the projectbut 8GB would surprise me, as many other people have seemed to complete similar tasks without such issues. You should check whether you are using your memory effectively in Posit Cloud. That may not sound particularly helpful, but we haven't got any details to go on.
Thank you for your prompt response. Im stucked at the third step to completing the cleaning process on R.
Im working on the cyclist data trip and got stucked at the "analyze ridership data by type and weekday" I have explored all the hacks shared with me on here - nothing worked (but i do appreciate all the professional input). So, i will try one more time with the annual ride trip (jan-dec) instead of the quarterly (q2,2019 - q1,2020) that has failed to get me where i want to be.
Will update you of my progress and developments.
Thank you
Yomi - how is the R cleaning process going? I never did the quarters, only 12 months - and I picked the most recent 12 months that data was in the course link (July 2022 to June 2023). So, it's 12 spreadsheets, with up to over 800,000 rows. There's a lot of pre-cleaning that can be done in Excel before uploading to R. For example, I'm deleting some columns showing the stations, since my analysis is focusing on days of the week and duration of rides. I'm also deleting the date ended column, since only a small percentage of rides (~5% in July and ~ 0.5% in January) end later than the day they started. Before I delete that column, I reformat the two date columns into just dates, then create a new column and put in a formula to compare the two dates with the return in the new column being "Dates are equal" or "Dates are not equal". Then I upload the spreadsheet to R in order to remove those rows with "Dates are not equal". Anyway, let me know how you're doing. I recommend using the free gpt chat - it is priceless for beginners like me! I've gotten lots of code from it. I still have to ask questions in this community sometimes, because not even chat gpt is always right.
Martin, yes I was surprised that 1 GB didn't cut it, since some form of R is needed for the capstone project(s). I tried numerous times, over several days but continually couldn't get all the way through the 12 spreadsheets (with a combined total of ~5 million rows). I noticed another poster say they have heard my problem of not enough RAM is not uncommon, so I suppose that makes me feel a little better. Since I got the 8 GB plan, I haven't had any problems, knock on wood.
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.
If you have a query related to it or one of the replies, start a new topic and refer back with a link.