Cloud access #60
Cloud access #60
Conversation
|
@noaaroland Sorry to get to this a bit late. First of all, thank you so much for creating these awesome contents! I have couple of comments that might need your help. I am currently viewing the notebook from your branch directly instead of this diff view here. Notebook changes are relatively hard to make line specific suggestions directly. I will try to add comment and suggestion below. |
| - plotly | ||
| - cartopy |
There was a problem hiding this comment.
I like to use cartopy too since most of the scientific figure is probably generated from this. I also like the idea of providing a code example to show how to make your beautiful figures. Sadly, I tried a few time that the binder service seems to struggle a lot when installing the cartopy package with other in the yml file. That is why I use folium and plotly. So for this. I think we can skip the cartopy in the yml file and generate a .mb file that is static in the cookbook so it does not require code execution.
There was a problem hiding this comment.
I understand. cartopy is a pretty heavy package, but not being able to use binder is not really a good reason to avoid packages. Nobody is going to use binder for any serious work, it's only a convenient free placeholder. Having an environment file that works when people download it to their systems seems more important than having it work in binder.
Not sure exactly what you are proposing, but as long as it's clear how to get an environment installed that will run the notebook will probably be fine.
| - r-maps | ||
| - r-reticulate | ||
| - r-dplyr | ||
| - r-ggplot2 |
There was a problem hiding this comment.
for all the r related package I put them in the r-setup branch which is utilized by the binder when creating the computational env. This is to avoid the env build stuck with too many packages and possible incompatibility binder had to solve between r package and python packages. This is only for the package list. The notebook related to R does not have to stay in the r-setup branch it stay in the main branch
There was a problem hiding this comment.
I would rather make a second separate environment.yml which for R users than have them install packages from the notebook, but if you prefer to put these in the notebook that's fine.
|
@noaaroland the notebook is relatively hard to reference in the PR. If you don't mind, I will pull from your fork branch and submit a new PR to incorporate the changes into the current cookbook structure |
However you want to handle it is fine with me. |
This adds two new notebooks that show how to access data in Python and R directly from the two cloud storage locations.
Each example shows data access from both AWS S3 and Google Cloud Storage.
I added all of the R packages needed to run the R notebook to the environment since I felt it was preferable to have the complete environment at the outset rather than have people download those packages in the notebook and then remember to skip that cell the next time.
In the Python example I used cartopy for a nice plot on the map.
I would like a subject matter expert to add some calculations or other R specific code to that notebook, but for now it does show how to access the data and get it organized in an R data.frame.
You'll have to figure out how to link the notebooks and display them in the cookbook. I don't know how that's done.