README.md

metadata

title: Agalma
emoji: 📚
colorFrom: green
colorTo: green
sdk: streamlit
sdk_version: 1.32.0
app_file: app.py
pinned: false
short_description: AGALMA | ἄγαλμα
license: cc-by-4.0
tags:
  - '#AncientGreek'
  - '#LanguageModels'
  - '#Linguistics'

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

File Descriptions

app.py

The main application file for the Streamlit app. This file contains the core logic and layout for the Agalma interface.

requirements.txt

A list of Python dependencies required to run the app.

models/

A directory containing pre-trained word2vec models. These models are used for generating vector representations of words in the interface.

3d_models/

A directory containing pre-trained 3D representations of the models in the ./models directory. Dimensions were reduced using t-SNE technique.

images/

A directory containing all necessary images for the layout of the interface.

corpora/

A directory containing compressed corpora files to faster access all the words from the corpora.

lemma_list_raw/

A directory containing .txt files with all lemmas occuring in each time slice.

all_lemmas.pkl.gz

A compressed pickle file containing all lemmas that are in the LSJ dictionary.

autocomplete.py

A Python file that contains functions to optimize the input suggestions given when a user is typing as input.

compress_words.py

A Python file that contains functions on compressing words. The compressed files obtained are used to faster obtain words that are in the corpora.

lsj_dict.json

A JSON file that contains relevant info in .xml files of the LSJ dictionary.

lsj_dict.py

A Python file that contains function to convert .xml file representations of the LSJ dictionary into a JSON file.

plots.py

A Python file containing functions to create 3D plots.

word2vec.py

A Python file containing functions needed for computations with vectors.