...
Installed Environment
No Format |
---|
|
Useful commands
No Format |
---|
Useful commands: conda info --envs conda list conda activate |
...
No Format |
---|
#!/bin/bash #PBS -N condatest #PBS -m abe #PBS -M YourEMAIL@griffith.edu.au #PBS -l select=1:ncpus=1:mem=4gb,walltime=10:00:00 module load anaconda/4.00py2.7 source /sw/anaconda/py27-ver400/etc/profile.d/conda.sh ##use /sw/anaconda/py3-ver400/etc/profile.d/conda.sh if using “module load anaconda/4.00py3.5” echo "Starting job: " cd $PBS_O_WORKDIR python <pythonFile> >>>An example of the python file>>>>>>>>> cat pythonFile.py import tensorflow as tf >>>>>>>>>>>>>> |
...
Activate an environment
No Format |
---|
source /sw/anaconda/py27-ver400/etc/profile.d/conda.sh OR source /sw/anaconda/py3-ver400/etc/profile.d/conda.sh if using “module load module load anaconda/4.00py3.5” source activate biopython e.g: source activate biopython discarding /sw/anaconda/2.0.1/bin from PATH prepending /sw/anaconda/2.0.1/envs/biopython/bin to PATH |
...
No Format |
---|
conda remove -n biopython |
...
n060 anaconda installation
...
No Format |
---|
Instructions from: https://medium.com/@danielwzou/installing-spacy-and-other-packages-without-sudo-or-root-privileges-using-anaconda-59571ec0ecad First, navigate to your home directory. module load anaconda/5.3.1py3 wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh sh Miniconda3-latest-Linux-x86_64.sh Now, we have a local installation of the Conda package manager at ~/miniconda3 Let’s add conda to our bash profile so that we can call it from the command line vim .bash_profile i Add this line to the file: export PATH=$HOME/miniconda3/bin:$PATH esc :wq Now, let’s reload our .bash_profile source .bash_profile And check that our installation was completed successfully which conda Which should return ~/miniconda3/bin/conda Now that Conda is installed, we can install Spacy or other Python packages conda install spacy python -m spacy download en Now, let’s make sure that Spacy is installed correctly python import spacy nlp = spacy.load('en') doc = nlp(u'Hello World!') print([w.lemma_ for w in doc]) Now that you have Conda installed, you can use it to install many packages locally. |
...
Anaconda on the new cluster
No Format |
---|
module load anaconda3/2019.07py3
This should have packages like the tensorflow env. You can find with this:
conda info --envs
>>>>>
conda info --envs
# conda environments:
#
/sw/anaconda2/2019.07/envs/bio
/sw/anaconda2/2019.07/envs/bioinfoPy27
base * /sw/anaconda3/2019.07
3point6 /sw/anaconda3/2019.07/envs/3point6
R /sw/anaconda3/2019.07/envs/R
bioinformatics /sw/anaconda3/2019.07/envs/bioinformatics
>>>>>>>
This will give the packages in the base:
conda list
source activate 3point6
Next run: " conda list"
This will list the packages installed in that environment.
It shows among many others:
pytorch 1.3.1 cuda100py36h53c1284_0
tensorboard 2.0.0 pyhb230dea_0
tensorflow 2.0.0 mkl_py36hef7ec59_0
tensorflow-base 2.0.0 mkl_py36h9204916_0
tensorflow-estimator 2.0.0 pyh2649769_0
termcolor 1.1.0 py_2 conda-forge
theano 1.0.4 py36hf484d3e_1000 conda-forge
Remember this cluster has no GPU (currently) |
...
3. https://medium.com/@danielwzou/installing-spacy-and-other-packages-without-sudo-or-root-privileges-using-anaconda-59571ec0ecad
4. http://continuum.io/downloads#all