Skip to content

Commit

Permalink
move data files into project directory
Browse files Browse the repository at this point in the history
  • Loading branch information
jgreener64 committed Feb 5, 2021
1 parent 7208b14 commit 5f1729c
Show file tree
Hide file tree
Showing 4 changed files with 9 additions and 9 deletions.
8 changes: 4 additions & 4 deletions MANIFEST.in
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
include cgdms/cgdms_params_ep45.pt
include protein_data/results/*
include protein_data/train_val/*
include datasets/*
inlude cgdms/cgdms_params_ep45.pt
inlude cgdms/protein_data/results/*
inlude cgdms/protein_data/train_val/*
inlude cgdms/datasets/*
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ The modes are described below but there are other options outlined in the help t
### Generating protein data files

To simulate or calculate the energy of proteins you need to generate files of a certain format.
If you want to use the proteins presented in the paper, the data files are [here](protein_data/results).
If you want to use the proteins presented in the paper, the data files are [here](cgdms/protein_data/results).
Otherwise you will need to generate these files:

```bash
Expand Down Expand Up @@ -158,7 +158,7 @@ See `cgdms.interactions` for the interaction described by each potential, which
Running a simulation takes less than 1 GB of GPU memory for any number of steps.
Training a model takes up to 32 GB of GPU memory once the number of steps is fully scaled up to 2,000.

The lists of training and validation PDB chains are available [here](datasets) and the protein data files [here](protein_data/train_val).
The lists of training and validation PDB chains are available [here](cgdms/datasets) and the protein data files [here](cgdms/protein_data/train_val).

The code in this package is set up to run specific coarse-grained simulations of proteins.
However, the package contains code that could be useful to others wishing to carry out general differentiable simulations with PyTorch.
Expand Down
4 changes: 2 additions & 2 deletions cgdms/cgdms.py
Original file line number Diff line number Diff line change
Expand Up @@ -12,10 +12,10 @@
import os
from random import gauss, random, shuffle

cgdms_dir = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
cgdms_dir = os.path.dirname(os.path.realpath(__file__))
dataset_dir = os.path.join(cgdms_dir, "datasets")
train_val_dir = os.path.join(cgdms_dir, "protein_data", "train_val")
trained_model_file = os.path.join(cgdms_dir, "cgdms", "cgdms_params_ep45.pt")
trained_model_file = os.path.join(cgdms_dir, "cgdms_params_ep45.pt")

n_bins_pot = 140
n_bins_force = n_bins_pot - 2
Expand Down
2 changes: 1 addition & 1 deletion setup.py
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@

setuptools.setup(
name="cgdms",
version="0.1",
version="0.1.1",
author="Joe G Greener",
author_email="j.greener@ucl.ac.uk",
description="Differentiable molecular simulation of proteins with a coarse-grained potential",
Expand Down

0 comments on commit 5f1729c

Please sign in to comment.