Merge pull request #61 from kathyxchen/kipoi-export

Adding an example of how to export a model to Kipoi

Author: kathyxchen

README.md

@@ -7,19 +7,14 @@ Selene is a Python library and command line interface for training deep neural n
 ## Installation
 
 We recommend using Selene with Python 3.6 or above. 
-Package installation should only take a few minutes (less than 10 minutes, typically ~2-3 minutes) with any of these methods (pip, conda, source).  
+Package installation should only take a few minutes (less than 10 minutes, typically ~2-3 minutes) with any of these methods (conda, source). We previously supported installation through pip, but we are refraining from releasing the latest version of Selene with pip until we resolve some issues we are observing with the pip installations of torch and torchvision.   
 
 ### Installing selene with [Anaconda](https://www.anaconda.com/download/) (for Linux):
 
 ```sh
 conda install -c bioconda selene-sdk
 ```
 
-### Installing selene with pip:
-```sh
-pip install selene-sdk
-```
-
 ### Installing selene from source:
 
 First, download the latest commits from the source repository (or download the latest tagged version of Selene for a stable release):
@@ -42,6 +37,8 @@ Otherwise, if you would like to locally install Selene, you can run
 python setup.py install
 ```
 
+### Additional dependency for the CLI
+
 Please install `docopt` before running the command-line script `selene_cli.py` provided in the repository.
 
 ## About Selene

docs/source/overview/faq.md

@@ -10,6 +10,11 @@ The main modules that users may want to extend are
 Please refer to the documentation for these classes.
 If you are interested in contributing new modules to Selene or requesting new functionality, please submit an issue to [Selene](https://github.com/FunctionLab/selene/issues) or e-mail <[email protected]>.
 
+## Exporting a Selene-trained model to Kipoi
+We have provided an example of how to prepare a model for upload to [Kipoi's model zoo](http://kipoi.org/) using a model trained during case study 2. You can use [this example](https://github.com/FunctionLab/selene/tree/master/manuscript/case2/3_kipoi_export) as a starting point for preparing your own model for Kipoi. We have provided a script that can help to automate parts of the process.
+
+We are also working on an export function that will be built into Selene and accessible through the CLI. 
+
 ## Hyperparameter optimization
 Hyperparameter optimization is the process of finding the set of hyperparameters that yields an optimal model against a predefined score (e.g. minimizing a loss function). 
 Hyperparameters are the variables that govern the training process (i.e. these parameters are constant during training, compared to model parameters which are optimized/"tuned" by the training process itself). 

docs/source/overview/installation.md

@@ -1,6 +1,6 @@
 # Installation
 
-We recommend that users either clone and build the repository or install Selene through conda. This is because there is less control over PyTorch dependencies with pip.
+Users can either clone and build the repository locally or install Selene through conda. We previously supported installation through pip, but are refraining from releasing the latest version of Selene through pip due to some issues we are observing when using the pip-installed torch and torchvision dependencies.
 
 ## Installing with Anaconda
 
@@ -9,14 +9,6 @@ To install with conda (recommended for Linux users), run the following command i
 conda install -c bioconda selene-sdk
 ```
 
-## Installing with pip
-
-Selene can be installed with [pip](https://pypi.org/project/pip/).
-To install with pip, simply type the following commands into the terminal:
-```
-pip install selene-sdk
-```
-
 ## Installing from source
 
 Selene can also be installed from source.
@@ -39,4 +31,6 @@ Otherwise, if you would like to locally install Selene, you can run
 python setup.py install
 ```
 
+## Additional dependency for running the CLI 
+
 Please install `docopt` before running the command-line script `selene_cli.py` provided in the repository.