spaCy/spacy/tests/parser/test_parse.py

151 lines
4.7 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals
import pytest
💫 Refactor test suite (#2568) ## Description Related issues: #2379 (should be fixed by separating model tests) * **total execution time down from > 300 seconds to under 60 seconds** 🎉 * removed all model-specific tests that could only really be run manually anyway – those will now live in a separate test suite in the [`spacy-models`](https://github.com/explosion/spacy-models) repository and are already integrated into our new model training infrastructure * changed all relative imports to absolute imports to prepare for moving the test suite from `/spacy/tests` to `/tests` (it'll now always test against the installed version) * merged old regression tests into collections, e.g. `test_issue1001-1500.py` (about 90% of the regression tests are very short anyways) * tidied up and rewrote existing tests wherever possible ### Todo - [ ] move tests to `/tests` and adjust CI commands accordingly - [x] move model test suite from internal repo to `spacy-models` - [x] ~~investigate why `pipeline/test_textcat.py` is flakey~~ - [x] review old regression tests (leftover files) and see if they can be merged, simplified or deleted - [ ] update documentation on how to run tests ### Types of change enhancement, tests ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [ ] My changes don't require a change to the documentation, or if they do, I've added all required information.
2018-07-25 00:38:44 +03:00
from ..util import get_doc, apply_transition_sequence
def test_parser_root(en_tokenizer):
text = "i don't have other assistance"
heads = [3, 2, 1, 0, 1, -2]
deps = ["nsubj", "aux", "neg", "ROOT", "amod", "dobj"]
tokens = en_tokenizer(text)
💫 Refactor test suite (#2568) ## Description Related issues: #2379 (should be fixed by separating model tests) * **total execution time down from > 300 seconds to under 60 seconds** 🎉 * removed all model-specific tests that could only really be run manually anyway – those will now live in a separate test suite in the [`spacy-models`](https://github.com/explosion/spacy-models) repository and are already integrated into our new model training infrastructure * changed all relative imports to absolute imports to prepare for moving the test suite from `/spacy/tests` to `/tests` (it'll now always test against the installed version) * merged old regression tests into collections, e.g. `test_issue1001-1500.py` (about 90% of the regression tests are very short anyways) * tidied up and rewrote existing tests wherever possible ### Todo - [ ] move tests to `/tests` and adjust CI commands accordingly - [x] move model test suite from internal repo to `spacy-models` - [x] ~~investigate why `pipeline/test_textcat.py` is flakey~~ - [x] review old regression tests (leftover files) and see if they can be merged, simplified or deleted - [ ] update documentation on how to run tests ### Types of change enhancement, tests ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [ ] My changes don't require a change to the documentation, or if they do, I've added all required information.
2018-07-25 00:38:44 +03:00
doc = get_doc(tokens.vocab, words=[t.text for t in tokens], heads=heads, deps=deps)
for t in doc:
assert t.dep != 0, t.text
@pytest.mark.xfail
@pytest.mark.parametrize("text", ["Hello"])
def test_parser_parse_one_word_sentence(en_tokenizer, en_parser, text):
tokens = en_tokenizer(text)
doc = get_doc(
tokens.vocab, words=[t.text for t in tokens], heads=[0], deps=["ROOT"]
)
2016-05-02 17:08:11 +03:00
assert len(doc) == 1
with en_parser.step_through(doc) as _:
2016-05-02 17:08:11 +03:00
pass
assert doc[0].dep != 0
@pytest.mark.xfail
def test_parser_initial(en_tokenizer, en_parser):
text = "I ate the pizza with anchovies."
heads = [1, 0, 1, -2, -3, -1, -5]
transition = ["L-nsubj", "S", "L-det"]
tokens = en_tokenizer(text)
apply_transition_sequence(en_parser, tokens, transition)
assert tokens[0].head.i == 1
assert tokens[1].head.i == 1
assert tokens[2].head.i == 3
assert tokens[3].head.i == 3
def test_parser_parse_subtrees(en_tokenizer, en_parser):
text = "The four wheels on the bus turned quickly"
heads = [2, 1, 4, -1, 1, -2, 0, -1]
tokens = en_tokenizer(text)
💫 Refactor test suite (#2568) ## Description Related issues: #2379 (should be fixed by separating model tests) * **total execution time down from > 300 seconds to under 60 seconds** 🎉 * removed all model-specific tests that could only really be run manually anyway – those will now live in a separate test suite in the [`spacy-models`](https://github.com/explosion/spacy-models) repository and are already integrated into our new model training infrastructure * changed all relative imports to absolute imports to prepare for moving the test suite from `/spacy/tests` to `/tests` (it'll now always test against the installed version) * merged old regression tests into collections, e.g. `test_issue1001-1500.py` (about 90% of the regression tests are very short anyways) * tidied up and rewrote existing tests wherever possible ### Todo - [ ] move tests to `/tests` and adjust CI commands accordingly - [x] move model test suite from internal repo to `spacy-models` - [x] ~~investigate why `pipeline/test_textcat.py` is flakey~~ - [x] review old regression tests (leftover files) and see if they can be merged, simplified or deleted - [ ] update documentation on how to run tests ### Types of change enhancement, tests ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [ ] My changes don't require a change to the documentation, or if they do, I've added all required information.
2018-07-25 00:38:44 +03:00
doc = get_doc(tokens.vocab, words=[t.text for t in tokens], heads=heads)
assert len(list(doc[2].lefts)) == 2
assert len(list(doc[2].rights)) == 1
assert len(list(doc[2].children)) == 3
assert len(list(doc[5].lefts)) == 1
assert len(list(doc[5].rights)) == 0
assert len(list(doc[5].children)) == 1
assert len(list(doc[2].subtree)) == 6
def test_parser_merge_pp(en_tokenizer):
text = "A phrase with another phrase occurs"
heads = [1, 4, -1, 1, -2, 0]
deps = ["det", "nsubj", "prep", "det", "pobj", "ROOT"]
tags = ["DT", "NN", "IN", "DT", "NN", "VBZ"]
tokens = en_tokenizer(text)
doc = get_doc(
tokens.vocab, words=[t.text for t in tokens], deps=deps, heads=heads, tags=tags
)
nps = [(np[0].idx, np[-1].idx + len(np[-1]), np.lemma_) for np in doc.noun_chunks]
for start, end, lemma in nps:
doc.merge(start, end, label="NP", lemma=lemma)
assert doc[0].text == "A phrase"
assert doc[1].text == "with"
assert doc[2].text == "another phrase"
assert doc[3].text == "occurs"
@pytest.mark.xfail
def test_parser_arc_eager_finalize_state(en_tokenizer, en_parser):
text = "a b c d e"
# right branching
transition = ["R-nsubj", "D", "R-nsubj", "R-nsubj", "D", "R-ROOT"]
tokens = en_tokenizer(text)
apply_transition_sequence(en_parser, tokens, transition)
assert tokens[0].n_lefts == 0
assert tokens[0].n_rights == 2
assert tokens[0].left_edge.i == 0
assert tokens[0].right_edge.i == 4
assert tokens[0].head.i == 0
assert tokens[1].n_lefts == 0
assert tokens[1].n_rights == 0
assert tokens[1].left_edge.i == 1
assert tokens[1].right_edge.i == 1
assert tokens[1].head.i == 0
assert tokens[2].n_lefts == 0
assert tokens[2].n_rights == 2
assert tokens[2].left_edge.i == 2
assert tokens[2].right_edge.i == 4
assert tokens[2].head.i == 0
assert tokens[3].n_lefts == 0
assert tokens[3].n_rights == 0
assert tokens[3].left_edge.i == 3
assert tokens[3].right_edge.i == 3
assert tokens[3].head.i == 2
assert tokens[4].n_lefts == 0
assert tokens[4].n_rights == 0
assert tokens[4].left_edge.i == 4
assert tokens[4].right_edge.i == 4
assert tokens[4].head.i == 2
# left branching
transition = ["S", "S", "S", "L-nsubj", "L-nsubj", "L-nsubj", "L-nsubj"]
tokens = en_tokenizer(text)
apply_transition_sequence(en_parser, tokens, transition)
assert tokens[0].n_lefts == 0
assert tokens[0].n_rights == 0
assert tokens[0].left_edge.i == 0
assert tokens[0].right_edge.i == 0
assert tokens[0].head.i == 4
assert tokens[1].n_lefts == 0
assert tokens[1].n_rights == 0
assert tokens[1].left_edge.i == 1
assert tokens[1].right_edge.i == 1
assert tokens[1].head.i == 4
assert tokens[2].n_lefts == 0
assert tokens[2].n_rights == 0
assert tokens[2].left_edge.i == 2
assert tokens[2].right_edge.i == 2
assert tokens[2].head.i == 4
assert tokens[3].n_lefts == 0
assert tokens[3].n_rights == 0
assert tokens[3].left_edge.i == 3
assert tokens[3].right_edge.i == 3
assert tokens[3].head.i == 4
assert tokens[4].n_lefts == 4
assert tokens[4].n_rights == 0
assert tokens[4].left_edge.i == 0
assert tokens[4].right_edge.i == 4
assert tokens[4].head.i == 4