Deep-Learning Driven Adaptive Molecular Simulations
Project description
DeepDriveMD-F (DeepDriveMD-pipeline)
DeepDriveMD-F: Deep-Learning Driven Adaptive Molecular Simulations (file-based continual learning loop)
Details can be found in the ducumentation. For more information, please see our website.
How to run
Setup
Install deepdrivemd
into a virtualenv with:
python3 -m venv env
source env/bin/activate
pip install --upgrade pip setuptools wheel
pip install -e .
Then, install pre-commit hooks: this will auto-format and auto-lint on commit to enforce consistent code style:
pre-commit install
pre-commit autoupdate
In some places, DeepDriveMD relies on external libraries to configure MD simulations and import specific ML models.
For MD, install the mdtools
package found here: https://github.com/braceal/MD-tools
For ML (specifically the AAE model), install the molecules
package found here: https://github.com/braceal/molecules/tree/main
Generating a YAML input spec:
First, run this command to get a sample YAML config file:
python -m deepdrivemd.config
This will write a file named deepdrivemd_template.yaml
which should be adapted for the experiment at hand. You should configure the molecular_dynamics_stage
, aggregation_stage
, machine_learning_stage
, model_selection_stage
and agent_stage
sections to use the appropriate run commands and environment setups.
Running an experiment
Then, launch an experiment with:
python -m deepdrivemd.deepdrivemd -c <experiment_config.yaml>
This experiment should be launched
Note on input data
The input PDB and topology files should have the following structure:
ls data/sys*
data/sys1:
comp.pdb comp.top
data/sys2:
comp.pdb comp.top
Where the topology files are optional and only used when molecular_dynamics_stage.task_config.solvent_type
is "explicit". Only one system directory is needed but an arbitrary number are supported. Also note that the system directory names are arbitrary. The path to the data
directory should be passed into the config via molecular_dynamics_stage.initial_pdb_dir
.
MIT License
Copyright (c) 2021 DeepDriveMD-F
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for deepdrivemd-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3fa9ac26b2262111243bb5c7a506f4f602c4b172cc3583b80f3fb05c6fe4f289 |
|
MD5 | 24e3d7e3c0ae0ff7b09ac4a98a55fd75 |
|
BLAKE2b-256 | 1d782b520a650fcac791b282d2b7a42f896b051ca6a17c2287f196015682b73d |