wpm

$ pip install --no-build-isolation --editable ".[transformers,ray]"
Obtaining file:///Users/wmcneill/Documents/src/spaCy
    Preparing wheel metadata ... done
Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.8 in /Users/wmcneill/opt/anaconda3/envs/spaCy/lib/python3.9/site-packages (from spacy==3.1.3) (3.0.8)
Requirement already satisfied: setuptools in /Users/wmcneill/opt/anaconda3/envs/spaCy/lib/python3.9/site-packages (from spacy==3.1.3) (58.0.4)
Requirement already satisfied: typer<0.5.0,>=0.3.0 in /Users/wmcneill/opt/anaconda3/envs/spaCy/lib/python3.9/site-packages (from spacy==3.1.3) (0.4.0)
Requirement already satisfied: requests<3.0.0,>=2.13.0 in /Users/wmcneill/opt/anaconda3/envs/spaCy/lib/python3.9/site-packages (from spacy==3.1.3) (2.26.0)
Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /Users/wmcneill/opt/anaconda3/envs/spaCy/lib/python3.9/site-packages (from spacy==3.1.3) (2.0.5)
Requirement already satisfied: packaging>=20.0 in /Users/wmcneill/opt/anaconda3/envs/spaCy/lib/python3.9/s

wpm

#!/usr/bin/env python
import shutil
from pathlib import Path
from typing import Iterable, Tuple, Optional

import click
from celery import Celery
from celery.utils.log import get_task_logger
from tqdm import tqdm

wpm

def maximal_fully_ordered_sublists(s: List[T]) -> List[List[T]]:
    """
    Find maximum-length sequences of in-order items in a list.

    Let s be a list of items over which there exists a total ordering defined by the < operator.
    Let a fully-ordered sublist s' of s be s with elements removed so that the elements of s' are monotonically
    increasing.
    The maximal fully-ordered sublists of s are the set of fully-ordered sublists such that no sublist is contained in
    another one.

wpm

import json
import re
import time
from random import choice, random
from typing import TextIO, Callable, Sequence, Tuple, Optional

import click

NAME = DATE = str
SPAN_OFFSET = Tuple[int, int]

wpm

import json
from json import JSONDecodeError
from typing import Sequence

import click


class JSONList(click.ParamType):

    def convert(self, value: str, _, __) -> Sequence:

wpm

import json
from json import JSONDecodeError
from typing import Sequence, Iterable, List

import click
import spacy
from spacy.matcher import Matcher


def match_patterns(nlp, patterns: Sequence[dict], corpus: Iterable[str]) -> Iterable[str]:

wpm

def paragraphs(document):
    start = 0
    for token in document:
        if token.is_space and token.text.count("\n") > 1:
            yield document[start:token.i]
            start = token.i
    yield document[start:]

wpm

wpm

"""
Put all the Stanford Sentiment Treebank phrase data into test, training, and dev CSVs.

Socher, R., Perelygin, A., Wu, J. Y., Chuang, J., Manning, C. D., Ng, A. Y., & Potts, C. (2013). Recursive Deep Models
for Semantic Compositionality Over a Sentiment Treebank. Presented at the Conference on Empirical Methods in Natural
Language Processing EMNLP.

https://nlp.stanford.edu/sentiment/
"""

wpm

"""
A minimal implementation of the MNIST handwritten digits classification task in TensorFlow.

This runs MNIST images images through a single hidden layer and softmax loss function.

It demonstrates in a single Python source file the basics of creating a model, training and evaluating data sets, and
writing summaries that can be visualized by TensorBoard.
"""
from __future__ import division