如何修改sbatch执行job.sh生成的test.Rout文件存储位置

Got it, let's sort out how to redirect that test.Rout file to a folder of your choice. Looking at your job.sh script, here's what's happening right now and how to adjust it:

First, the reason test.Rout lands in the same directory as job.sh is because when you run R CMD BATCH test.R (or your current command with a small typo—we'll fix that), R defaults to saving the .Rout output in the job's starting working directory unless you tell it otherwise.

Here are two simple, reliable ways to change this:

1. Explicitly Define the Output Path in R CMD BATCH #

The R CMD BATCH command lets you specify exactly where to save the .Rout file by adding your target path as the second argument.

First, fix the typo in your current script:
Your existing line R CMD BATCH test.R test.R would overwrite your input test.R file—definitely not what you want! Replace that with a path to your desired output location.

Example: Save to a dedicated output folder #

Let's say you want to store test.Rout in a folder called r_output (we'll add a line to create this folder if it doesn't exist):

#!/bin/bash
#SBATCH --job-name="test"
#SBATCH --nodes=1 # number of nodes
#SBATCH --ntasks-per-node=10 # number of cores
#SBATCH --time=01:00:00 # walltime
#SBATCH --output=error/job.out
#SBATCH --error=error/error.err

# Create the output directory if it doesn't exist
mkdir -p r_output

module load R
# Directly specify where to save test.Rout
R CMD BATCH test.R r_output/test.Rout

This will drop test.Rout straight into the r_output folder. You can also use an absolute path (like /home/yourusername/projects/results/test.Rout) if you want it saved to a fixed, specific location.

2. Switch the Job's Working Directory #

If you prefer to keep using R's default .Rout naming but want all job-related files (including this one) to go to a different directory, use the cd command to change the job's working directory early in your script.

Example: #

#!/bin/bash
#SBATCH --job-name="test"
#SBATCH --nodes=1 # number of nodes
#SBATCH --ntasks-per-node=10 # number of cores
#SBATCH --time=01:00:00 # walltime
#SBATCH --output=error/job.out
#SBATCH --error=error/error.err

# Create target directory and switch to it
mkdir -p /path/to/your/target/folder
cd /path/to/your/target/folder

module load R
# Make sure to use the full path to your test.R script here
R CMD BATCH /path/to/your/original/test.R

Now R will save test.Rout in the directory you switched to with cd.

Quick Side Note #

That print(2+2) line in your current job.sh won't run as intended—it's an R command but is being executed as bash. If you want that code to run, move it inside your test.R file instead.

内容的提问来源于stack exchange，提问作者problème0123