Fei Yao

I think NF90_INT in NetCDF-Fortran corresponds to integers in Fortran90, -999.9 will be converted to -999 during nf90_put_att and it works. I have checked [NF90_SHORT and found that it corresponds...

So is there any a solution to this issue? I encounter the same issues with the following version information on a Linux machine. ``` $ jupyter --version Selected Jupyter core...

> What error did you get? You might need to change `/dev/sda1` to something else due to EC2/EBS updates. Try `/dev/nvme0n1p1` (shown by `df`) Thanks for your reply. I tried...

> Does `/dev/xvda` work? Unfortunately No. Similar error, `An error occurred (InvalidBlockDeviceMapping) when calling the RunInstances operation: Invalid device name C:/Users/s1855106/AppData/Local/Programs/Git/dev/xvda`.

That was the information output from GitBash after running `aws ec2 run-instances`. The device name within my script remains as `--block-Connection reset by 129.215.8.197 port 22VolumeSize=$EBS_SIZE}`. I guess this issue...

I am interested in running a nested simulation at 0.25x0.3125 over China. I found that ```s3://gcgrid/GEOS_0.25x0.3125_CH/GEOS_FP/``` only contains 2013 and 2016, whereas ```ftp://rain.ucis.dal.ca/ctm/GEOS_0.25x0.3125_CH.d/GEOS_FP/``` contains met data from 2012 to now....

Thanks for your attention and discussion! I have solved the Git Bash problem based on your comments. Or more honestly, Git Bash should not be a problem because I previously...

> Effects are errors and so are not part of the prediction. Thanks for your clarification!

I will keep a big eye on this issue. Hausman tests for FE and RE deserves a add! Currently I need to calculate by myself based on the estimates of...

I think this [link](https://stackoverflow.com/questions/30910944/fast-1d-linear-np-nan-interpolation-over-large-3d-array) can provide a good example of @mlubej 's need. I believe that it would be great if scipy can develop interp1d tools ignoring nan values like...