sphinx-quickstart on Wed Aug 14 16:39:23 2019. You can adapt this file completely to your liking, but it should at least contain the root toctree directive.
- github_url
Welcome to ToR[e]cSys documentation!¶
Recommendation System in PyTorch.
This package is an implementation of several famous recommendation system algorithm in PyTorch, including click-through-rate prediction, learning-to-ranking and embedding. ToR[e]csys also comes with data loader, layers-level implementation, self-defined architecture of models. It’s open-source software, released under the MIT license.
Minimal Requirements¶
Numpy >= 1.17.0
Pandas >= 0.24.2
PyTorch >= 1.2
Installation¶
Install with pip:
pip install torecsys
Install with source code:
git clone https://github.com/p768lwy3/torecsys.git
cd ./torecsys
python setup.py build
python setup.py install
Notations in documentation¶
Notation |
Refer to |
---|---|
T |
torch.Tensor |
B |
batch size |
E |
embedding size |
H_i |
output size of i-th hidden layer |
N |
number of fields |
I |
input sizes of any layers required inputs_size |
O |
output size of any layers required outputs_size |
V |
total number of words in a vocabulary set |
S |
total number of samples, e.g. negative samples |