multiprocessing for spacy nlp.pipe

spacy_multiprocess_pipe.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import spacy\n",
    "from spacy.tests.util import assert_docs_equal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3000"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ndata=3000\n",
    "texts=[\"If you’ve been modifying the pipeline, vocabulary, vectors and entities, or made updates to the model, you’ll eventually want to save your progress\"]*ndata\n",
    "len(texts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp=spacy.load(\"en_core_web_sm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6.44 s, sys: 957 ms, total: 7.4 s\n",
      "Wall time: 7.4 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "docs0=list(nlp.pipe(texts,n_process=1))\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.19 s, sys: 102 ms, total: 1.29 s\n",
      "Wall time: 5.72 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "docs1=list(nlp.pipe(texts,n_process=2))\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "for doc0, doc1 in zip(docs0, docs1):\n",
    "    assert_docs_equal(doc0, doc1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp=spacy.load(\"en_core_web_md\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6.47 s, sys: 1.1 s, total: 7.57 s\n",
      "Wall time: 7.58 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "docs0=list(nlp.pipe(texts,n_process=1))\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.2 s, sys: 147 ms, total: 1.35 s\n",
      "Wall time: 5.79 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "docs1=list(nlp.pipe(texts,n_process=2))\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "for doc0, doc1 in zip(docs0, docs1):\n",
    "    assert_docs_equal(doc0, doc1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp=spacy.load(\"en_core_web_lg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 6.35 s, sys: 1.13 s, total: 7.48 s\n",
      "Wall time: 7.48 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "docs0=list(nlp.pipe(texts,n_process=1))\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.18 s, sys: 158 ms, total: 1.34 s\n",
      "Wall time: 5.68 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "docs1=list(nlp.pipe(texts,n_process=2))\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "for doc0, doc1 in zip(docs0, docs1):\n",
    "    assert_docs_equal(doc0, doc1)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "spacy",
   "language": "python",
   "name": "spacy"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}