Contributions are welcome! Please submit you issues, pull requests, improvements and comments in British English. You can look at the issues and help to solve them or you can add some missing articles as described below.
Here are the steps to follow for adding one (or multiple) article:
-
Check that the article is not already in the dl4m.bib file.
-
Fork the repo.
-
Add the desired bib entry at the beginning of dl4m.bib. Take care to fill all this field for each bib entry (if there is no information about a field please indicate it as
fieldName = {No}
):- Bib entry type (inproceedings, article, techreport, unpublished,...) (N.B.: Indicate arxiv article as @unpublished and if you know of a possible submission use
note = {Submitted to "NameOfJournalOrConference"}
) - Bib key (in the form AuthorlastnameYear, e.g.
Snow1999
- title (N.B.: Make sure the title is not written in all caps and that each word does not start with a capital letter.)
- author (N.B.: Check that all authors in the pdf files are present in the bib author field and in the same order. Indeed automatic bib tools tend to mess up the order or to forget some authors.)
- year
- booktitle or journal
- dataset (e.g.
dataset = {Inhouse & [Jamendo](http://www.mathieuramona.com/wp/data/jamendo/) & [RWC](https://staff.aist.go.jp/m.goto/RWC-MDB/)},
)- provide the link to the dataset
- if multiple dataset are used, insert a
&
between each dataset
- architecture (if multiple architectures are used, insert a
&
between each of them, e.g.archi = {CNN & VPNN},
) - link (HTML link to the pdf file)
- task (if multiple tasks are performed, insert a
&
between each of them, e.g.task = {SVS & SVD},
). Please refer to the acronyms listed in acronyms.md - dataaugmentation (if used, the type of data augmentation technique, otherwise
No
) - pages (if available)
- code (HTML link to the code if available,
No
instead) - learningrate (if given in the paper or the code, the learning rate, otherwise
No
) - framework (if given in the paper or the code, the framework, otherwise
No
) - reproducible (the reproducibility details of the paper and code)
- activation (if given in the paper or the code, the activation function, otherwise
No
) - epochs (if given in the paper or the code, the number of epoch, otherwise
No
) - batch (if given in the paper or the code, the number of batch, otherwise
No
) - loss (if given in the paper or the code, the loss function, otherwise
No
) - layers (if given in the paper or the code, the number of layers, otherwise
No
) - dropout (if given in the paper or the code, the loss function, otherwise
No
) - momentum (if given in the paper or the code, the momentum, otherwise
No
) - gpu (if given in the paper or the code, the type and number of GPUs, otherwise
No
) - metric (if given in the paper or the code, the metric, otherwise
No
) - computationtime (if given in the paper or the code, the global computation time and per epoch, otherwise
No
) - dimension (if given in the paper or the code, the number of dimension, otherwise
No
) - optimizer (if given in the paper or the code, the optimize function, otherwise
No
) - input (if given in the paper or the code, the input type, otherwise
No
) - month (for conference paper only)
- address (for conference paper only)
- note (optional additional custom notes, if you feel you want to share a great detail you read or give your opinion)
For ease of use you can copy and paste and fill the following lines:
@inproceedings{Snow1999, activation = {}, address = {}, architecture = {}, author = {}, batch = {}, booktitle = {}, code = {}, computationtime = {}, dataaugmentation = {}, dataset = {}, dimension = {}, dropout = {}, epochs = {}, framework = {}, gpu = {}, input = {}, layers = {}, learningrate = {}, link = {}, loss = {}, metric = {}, momentum = {}, month = {}, note = {}, optimizer = {}, pages = {}, reproducible = {}, task = {}, title = {}, year = {}, }
- Bib entry type (inproceedings, article, techreport, unpublished,...) (N.B.: Indicate arxiv article as @unpublished and if you know of a possible submission use
-
Check that you have installed this python package:
- numpy
- matplotlib
- bibtexparser
-
Launch the python script
python dl4m.py
. -
Submit your pull request!
Thanks for spotting it! You can:
- Submit an issue or
- Make a pull request:
- Fork the repo.
- Add or update the corresponding field in dl4m.bib.
- Launch the python script
python dl4m.py
. - Submit your pull request with this title:
[Update][<bib_key>] field added or updated
, e.g.[Update][Snow1999] added task
.
I am looking for a way to display relations between articles automatically like a mindmap. Tell me if you know anything able to handle that.