-
In addition to LTO/Graphite, I also build with [Auto-Parallelisation](https://gcc.gnu.org/wiki/AutoParInGCC) where possible. I've converted my own custom flag management hacks over to gentooLTO inclu…
-
Migrated from json into utteranc.es
-
CNN or Preprocessor could offer a way to profile the speed of
- data access/loading
- sample preprocessing
- parallelized preprocessing
- forward pass
and log information to wandb
-
First I think this is a great project. I've been thinking about something like this for a while. Two related things have had me thinking about this
https://www-cs.stanford.edu/~matei/papers/2019/usen…
-
Hi,
I have a question that is related internals of `data.table` and potentially related to feature request. (cross posting SO)
I have a large data with more than 1 billion observations, and I need…
-
Dear Wannier-Berri developers,
Recently, I attempted to run Wannier-Berri using multiple cores. I used the following command to generate and submit a slurm script: python3.9 -m wannierberri.utils…
-
Just as the title says... it is simply a `./configure.ac` automatism, since `AX_OPENMP` automatically defines `HAVE_OPENMP`, which activates openmp threads in the code at preprocessing time in src/thr…
-
Hi,
I've used the command snakemake -j 60 --use-conda to run bacLife but noticed prokka only uses 1 thread.
I checked the Snakefile and changed -cpus from 1 to 60:
rule Prokka_annotation:
…
-
In case of dataframe based task code, it is sometimes possible to batch computation based on some grouping columns. This can be used to reduce the memory footprint (not the whole input and output data…
-
Hi,
As I cannot write this:
```
out(x, y, 0) = selection[0];
out(x, y, 1) = selection[1];
out(x, y, 2) = selection[2];
out(x, y, 3) = selection[3];
```
Is there a better way to write this?
…