Merge pull request #1 from deepchem/master

[![Build Status](https://travis-ci.org/deepchem/deepchem.svg?branch=master)](https://travis-ci.org/deepchem/deepchem)

[![Coverage Status](https://coveralls.io/repos/github/deepchem/deepchem/badge.svg?branch=master)](https://coveralls.io/github/deepchem/deepchem?branch=master)

[![Anaconda-Server Badge](https://anaconda.org/deepchem/deepchem/badges/version.svg)](https://anaconda.org/deepchem/deepchem)

[![Anaconda-Server Badge](https://anaconda.org/conda-forge/deepchem/badges/version.svg)](https://anaconda.org/conda-forge/deepchem)

[![PyPI version](https://badge.fury.io/py/deepchem.svg)](https://badge.fury.io/py/deepchem)

[Website](https://deepchem.io/) | [Documentation (master)](https://deepchem.readthedocs.io/en/latest/)) | [Colab Tutorial](https://github.com/deepchem/deepchem/tree/master/examples/tutorials) | [Discussion Forum](https://forum.deepchem.io/) | [Gitter](https://gitter.im/deepchem/Lobby)

DeepChem aims to provide a high quality open-source toolchain

that democratizes the use of deep-learning in drug discovery,

materials science, quantum chemistry, and biology.

  - [Install from source](#install-from-source)

    - [General installation](#general-installation)

    - [Use powershell (Windows)](#use-powershell-windows)

  - [Install using a Docker (WIP)](#install-using-a-docker-with-gpu-wip)

  - [Install using a Docker (WIP)](#install-using-a-docker-wip)

- [FAQ and Troubleshooting](#faq-and-troubleshooting)

- [Getting Started](#getting-started)

- [Contributing to DeepChem](/CONTRIBUTING.md)

  - [Code Style Guidelines](/CONTRIBUTING.md#code-style-guidelines)

  - [Documentation Style Guidelines](/CONTRIBUTING.md#documentation-style-guidelines)

  - [Gitter](#gitter)

- [DeepChem Publications](#deepchem-publications)

- [Examples](/examples)

- [About Us](#about-us)

- [Citing DeepChem](#citing-deepchem)

DeepChem requires these packages on any condition.

- [joblib](https://pypi.python.org/pypi/joblib)

- [NumPy](https://numpy.org/)

- [pandas](http://pandas.pydata.org/)

- [scikit-learn](https://github.com/scikit-learn/scikit-learn.git)

- [tensorflow](https://www.tensorflow.org/)

- [scikit-learn](https://scikit-learn.org/stable/)

- [SciPy](https://www.scipy.org/)

- [TensorFlow](https://www.tensorflow.org/)

  - `deepchem>=2.4.0` requires tensorflow v2

  - `deepchem<2.4.0` requires tensorflow v1

DeepChem has a number of "soft" requirements. These are packages which are needed for various submodules of DeepChem but not for the package as a whole.

- [BioPython](https://biopython.org/wiki/Documentation)

- [OpenAI Gym](https://gym.openai.com/)

- [MDTraj](http://mdtraj.org/)

- [NetworkX](https://networkx.github.io/documentation/stable/index.html)

- [OpenMM](http://openmm.org/)

- [PDBFixer](https://github.com/pandegroup/pdbfixer)

- [Pillow](https://pypi.org/project/Pillow/)

- [pyGPGO](https://pygpgo.readthedocs.io/en/latest/)

- [PyTorch](https://pytorch.org/)

- [RDKit](http://www.rdkit.org/docs/Install.html)

- [simdna](https://github.com/kundajelab/simdna)

- [XGBoost](https://xgboost.readthedocs.io/en/latest/)

### Install via conda (Recommendation)

RDKit is a soft requirement package, but many useful methods like molnet depend on it.

If you use conda, we recommend installing RDKit with deepchem.

`deepchem>=2.4.0`

Coming soon...

`deepchem<2.4.0`

```bash

pip install tensorflow==1.15

conda install -c deepchem -c rdkit -c conda-forge -c omnia deepchem==2.3.0

pip install tensorflow==1.14

conda install -c rdkit -c conda-forge rdkit deepchem==2.3.0

```

If you want GPU support:

```bash

conda install -y -q scikit-learn=0.22

conda install -c deepchem -c rdkit -c conda-forge -c omnia deepchem-gpu==2.3.0

pip install tensorflow-gpu==1.14

conda install -c rdkit -c conda-forge rdkit deepchem==2.3.0

```

### Install via pip (WIP)

`deepchem<2.4.0`

```bash

pip install joblib pandas pillow scikit-learn==0.22 tensorflow==1.15 deepchem==2.2.1.dev54

pip install pandas pillow scikit-learn==0.22 tensorflow==1.14 deepchem==2.2.1.dev54

```

If you want GPU support:

```bash

pip install joblib pandas pillow scikit-learn==0.22 tensorflow-gpu==1.15 deepchem==2.2.1.dev54

pip install pandas pillow scikit-learn==0.22 tensorflow-gpu==1.14 deepchem==2.2.1.dev54

```

### Install from source

```

Before activating deepchem environment, make sure conda has been initialized.  

Check if there is a `(base)` in your command line.  

If not, use `conda init bash` to activate it, then:

Check if there is a `(base)` in your command line. If not, use `conda init bash` to activate it, then:

```

conda activate deepchem

python setup.py install                                # Manual install

nosetests -a '!slow' -v deepchem --nologcapture        # Run tests

python setup.py install

pytest -m "not slow"

```

Check [this link](https://conda.io/projects/conda/en/latest/user-guide/install/index.html) for more information about the installation of conda environments.

```

Before activating deepchem environment, make sure conda-powershell has been initialized.  

Check if there is a `(base)` before `PS` in powershell.  

If not, use `conda init powershell` to activate it, then:

Check if there is a `(base)` before `PS` in powershell. If not, use `conda init powershell` to activate it, then:

```bash

conda activate deepchem

python setup.py install

nosetests -a '!slow' -v deepchem --nologcapture

pytest -m "not slow"

```

### Install using a Docker (WIP)

### Build the image from Dockerfile

We prepare for [sample Dockerfiles](https://github.com/deepchem/deepchem/tree/master/docker) to install deepchem from source codes and conda package manager. Please check them!

We created [sample Dockerfiles](https://github.com/deepchem/deepchem/tree/master/docker) based on the `nvidia/cuda:10.1-cudnn7-devel` image.  

If you want to build your own deepchem environment, these files may be helpful.  

- `docker/x.x.x` : build an image by using conda package manager (x.x.x is a version of deepchem)  

- `docker/master` : build an image from master branch of deepchem source codes

### Use the official deepchem image (WIP)

## Version

2.1.0

2.3.0

    featurizer = dc.feat.one_hot.OneHotFeaturizer(dc.feat.one_hot.zinc_charset)

    one_hots = featurizer.featurize(mols)

    untransformed = featurizer.untransform(one_hots)

    len(smiles) == len(untransformed)

    assert len(smiles) == len(untransformed)

    for i in range(len(smiles)):

      assert smiles[i] == untransformed[i][0]

    conda create -y --name deepchem python=3.6 && \

    . /miniconda/etc/profile.d/conda.sh && \

    conda activate deepchem && \

    conda install -y -q scikit-learn=0.22 && \

    conda install -c deepchem -c rdkit -c conda-forge -c omnia deepchem-gpu==2.3.0 && \

    conda clean -afy

    pip install tensorflow-gpu==1.14 && \

    conda install -c rdkit -c conda-forge rdkit deepchem==2.3.0 && \

    conda clean -afy && \

    rm -rf ~/.cache/pip

RUN echo "conda activate deepchem" >> ~/.bashrc

WORKDIR /root/mydir

# install deepchem with master branch

RUN conda update -n base conda && \

    git clone https://github.com/deepchem/deepchem.git && \

    git clone --depth 1 https://github.com/deepchem/deepchem.git && \

    cd deepchem && \

    . /miniconda/etc/profile.d/conda.sh && \

    bash scripts/install_deepchem_conda.sh deepchem && \

`forum post`_ for Colab quick start guides.

If you'd like to install DeepChem locally, we recommend using

:code:`conda`.  If you have :code:`conda` installed, you can install

DeepChem with GPU support with the one-liner

:code:`conda` and installing RDKit with deepchem. 

RDKit is a soft requirement package, but many useful methods like

molnet depend on it.

.. code-block:: bash

    conda install -y -c deepchem -c rdkit -c conda-forge -c omnia deepchem-gpu

    pip install tensorflow-gpu==1.14

    conda install -y -c rdkit -c conda-forge rdkit deepchem

For CPU only support instead run

.. code-block:: bash

    conda install -y -c deepchem -c rdkit -c conda-forge -c omnia deepchem

    pip install tensorflow==1.14

    conda install -y -c rdkit -c conda-forge rdkit deepchem

Then open your python and try running.