Joint Deep Matcher for Points and Lines 🖼️💥🖼️ (ICCV 2023).
Key parameters and specifications
Key parameters and specifications
Parameter
Value
Language
Python
Category
Model Development
Subcategory
Domain-specific training
Stars
686
Forks
188
Contributors
4
Created date
2015-11-27
Updated date
2026-10-10
Stars in 1 day
0
Stars in 7 days
1
Rank
15,009
Previous rank
14,838
Rank change
-171
7-day stars rank
10,937
Previous 7-day stars rank
10,991
7-day stars rank change
54
Total stars rank
15,118
Previous total stars rank
15,090
Total stars rank change
-28
Country
Switzerland
README summary
Magpie is a deep learning tool for multi-label text classification. It has been developed at CERN to assign subject categories to High Energy Physics abstracts and extract keywords from them. Very short introduction Short introduction To train the model you need to have a large corpus of labeled data in a text format encoded as UTF-8. An example corpus can be found under data/hep-categories directory. Magpie looks for files containing the text to predict on and
corresponding files with assigned labels in separate lines. A pair of files containing the labels and the text should have the same name and differ only in extensions e. Before you train the model, you need to build appropriate word vector representations for your corpus. In theory, you can train them on a different corpus or reuse already trained ones (tutorial), however Magpie enables you to do that as well. Then you need to fit a scaling matrix to normalize
input data, it is specific to the trained word2vec representation. Here's the one liner: You would usually want to combine those two steps, by simply running: If you plan to reuse the trained word representations, you might want to save them and pass in the constructor to next time. For the training, just type: By providing the argument, the model splits data into train & test datasets (in this example into 80/20 ratio) and evaluates itself after every epoch displaying
it's current loss and accuracy. The default value of is 0 meaning that all the data will be used for training. If your data doesn't fit into memory, you can also run which has a similar API, but is more memory efficient. Trained models can be used for prediction with methods: Saving & loading the model A object consists of three components - the word2vec mappings, a scaler and a model. In order to train Magpie you can either provide the word2vec
mappings and a scaler in advance or let the program compute them for you on the training data. Usually you would want to train them yourself on a full dataset and reuse them afterwards. You can use the provided functions for that purpose: When you want to reinitialize your trained model, you can run: or just pass the objects directly! Installation The package is not on PyPi, but you can get it directly from GitHub: If you encounter any problems
with the installation, make sure to install the correct versions of dependencies listed in file. Disclaimer & citation The neural network models used within Magpie are based on work done by Yoon Kim and subsequently Mark Berger. Contact If you have any problems, feel free to open an issue.
Frequently asked questions
How many stars and forks are recorded?
inspirehep/magpie has 686 stars and 188 forks.
What is the repository rank?
inspirehep/magpie is ranked 15,009.
What language and category are recorded?
inspirehep/magpie has 686 stars, uses Python and belongs to Model Development.
Which repository is closest in rank?
inspirehep/magpie is compared with mokemokechicken/reversi-alpha-zero, ranked 15,011 with 686 stars.
README
Magpie is a deep learning tool for multi-label text classification. It learns on the training corpus to assign labels to arbitrary text and can be used to predict those labels on unknown data. It has been developed at CERN to assign subject categories to High Energy Physics abstracts and extract keywords from them.
To train the model you need to have a large corpus of labeled data in a text format encoded as UTF-8. An example corpus can be found under data/hep-categories directory. Magpie looks for .txt files containing the text to predict on and corresponding .lab files with assigned labels in separate lines. A pair of files containing the labels and the text should have the same name and differ only in extensions e.g.
$ ls data/hep-categories
1000222.lab 1000222.txt 1000362.lab 1000362.txt 1001810.lab 1001810.txt ...
Before you train the model, you need to build appropriate word vector representations for your corpus. In theory, you can train them on a different corpus or reuse already trained ones (tutorial), however Magpie enables you to do that as well.
from magpie import Magpie
magpie = Magpie()
magpie.train_word2vec('data/hep-categories', vec_dim=100)
Then you need to fit a scaling matrix to normalize input data, it is specific to the trained word2vec representation. Here's the one liner:
magpie.fit_scaler('data/hep-categories')
You would usually want to combine those two steps, by simply running:
If you plan to reuse the trained word representations, you might want to save them and pass in the constructor to Magpie next time. For the training, just type:
By providing the test_ratio argument, the model splits data into train & test datasets (in this example into 80/20 ratio) and evaluates itself after every epoch displaying it's current loss and accuracy. The default value of test_ratio is 0 meaning that all the data will be used for training.
If your data doesn't fit into memory, you can also run magpie.batch_train() which has a similar API, but is more memory efficient.
Trained models can be used for prediction with methods:
>>> magpie.predict_from_file('data/hep-categories/1002413.txt')
[('Experiment-HEP', 0.47593361),
('Gravitation and Cosmology', 0.055745006),
('Theory-HEP', 0.02692855)]
>>> magpie.predict_from_text('Stephen Hawking studies black holes')
[('Gravitation and Cosmology', 0.96627593),
('Experiment-HEP', 0.64958507),
('Theory-HEP', 0.20917746)]
Saving & loading the model
A Magpie object consists of three components - the word2vec mappings, a scaler and a keras model. In order to train Magpie you can either provide the word2vec mappings and a scaler in advance or let the program compute them for you on the training data. Usually you would want to train them yourself on a full dataset and reuse them afterwards. You can use the provided functions for that purpose:
