notebooks/spaCy 3.x Tutorial Transformers Spanish.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# spaCy 3.x Tutorial: Transformers Spanish"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(C) 2021-2024 by [Damir Cavar](http:https://damir.cavar.me/) <<dcavar@iu.edu>>**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Version:** 1.3, January 2024"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**License:** [Creative Commons Attribution-ShareAlike 4.0 International License](https://creativecommons.org/licenses/by-sa/4.0/) ([CA BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is a tutorial about developing simple [spaCy](https://spacy.io/) based NLP components for Spanish."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This tutorial was developed as part of my course material for the course L715 Research Seminar on NLU, Knowledge Graphs, and GNNs at [Indiana University at Bloomington](https://www.indiana.edu/)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NLP Pipeline for Spanish"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import spacy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you want to use the GPU in the following, use the following code. It will assign memory allocations to PyTorch, which will be more robust, preventing out-of-memory errors that might occur from competing memory pools (See [spaCy documentation](https://spacy.io/usage/embeddings-transformers))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from thinc.api import set_gpu_allocator, require_gpu\n",
    "\n",
    "set_gpu_allocator(\"pytorch\")\n",
    "require_gpu(0)\n",
    "nlp = spacy.load(\"es_dep_news_trf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for doc in nlp.pipe([\"Este es un texto corto.\", \"Aquí hay algún otro texto.\"]):\n",
    "    tokvecs = doc._.trf_data.tensors[-1]\n",
    "    print(tokvecs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for x in doc:\n",
    "    print(x, x.lemma_, x.pos_, x.tag_, x.dep_, x.shape_, x.is_alpha, x.is_stop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(C) 2021-2024 by [Damir Cavar](http:https://damir.cavar.me/) <<dcavar@iu.edu>>**"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}