-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathREADME_Install_Packages.txt
More file actions
39 lines (29 loc) · 1.66 KB
/
README_Install_Packages.txt
File metadata and controls
39 lines (29 loc) · 1.66 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Setup instructions:
1. Create a conda environment with python 3.8 within which to run the notebook.
$ conda create --name myenv_3_8FT python=3.8.0
2. Activate the environment.
$ conda activate myenv_3_8_FT or source activate myenv_3_8_FT if on Palmetto
3. Install jupyter in the new environment.
$ conda install jupyter
4. Creating a new kernel associated with the new environment which will allow you run the notebook in the env.
$ python -m ipykernel install --user --name python_custom --display-name "<name of your env>"
5. Install the required python packages in the environment.
$ python -m pip install numpy==1.24.4
$ python -m pip install pandas==2.0.3
$ python -m pip install matplotlib==3.7.4
$ python -m pip install mdtraj==1.9.9
$ python -m pip install fretraj==0.2.10
6. Directory Configuration
Before running the Jupyter Notebook, make sure to update the YAML file (`config.yaml`) with the correct directory paths. Follow these steps:
a. Open the `config.yaml` file in a text editor.
b. Locate the section containing directory paths (e.g., `work_directory`, `output_directory`, etc.).
c. Update each path according to your local machine or project structure. Ensure that you provide the correct absolute paths.
# when Frank tried the following instructions, they did not work. matplotlib was installed with an outdated version which used np.float.
installed in this order:
conda install numpy
conda install pandas
conda install matplotlib
conda install fretraj -c conda-forge (this installs mdtraj as well; if not, can pre-install it)
if needed:
python -m pip install copy, json,
this should allow running the notebook