explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-12 18:26:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'upstream/develop' into feature/docs-layers

Browse Source 
# Conflicts:
#	website/docs/usage/layers-architectures.md
This commit is contained in:
svlandeg 2020-09-02 13:17:11 +02:00
commit 422df9c2e2
9 changed files with 79 additions and 77 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
spacy/cli/_util.py
View File

@ -297,9 +297,7 @@ def ensure_pathy(path):
return Pathy(path)
def git_sparse_checkout(
repo: str, subpath: str, dest: Path, *, branch: Optional[str] = None
):
def git_sparse_checkout(repo: str, subpath: str, dest: Path, *, branch: str = "master"):
if dest.exists():
msg.fail("Destination of checkout must not exist", exits=1)
if not dest.parent.exists():
@ -323,21 +321,30 @@ def git_sparse_checkout(
# This is the "clone, but don't download anything" part.
cmd = (
f"git clone {repo} {tmp_dir} --no-checkout --depth 1 "
"--filter=blob:none" # <-- The key bit
f"--filter=blob:none " # <-- The key bit
f"-b {branch}"
)
if branch is not None:
cmd = f"{cmd} -b {branch}"
run_command(cmd, capture=True)
# Now we need to find the missing filenames for the subpath we want.
# Looking for this 'rev-list' command in the git --help? Hah.
cmd = f"git -C {tmp_dir} rev-list --objects --all --missing=print -- {subpath}"
ret = run_command(cmd, capture=True)
missings = "\n".join([x[1:] for x in ret.stdout.split() if x.startswith("?")])
repo = _from_http_to_git(repo)
# Now pass those missings into another bit of git internals
run_command(
f"git -C {tmp_dir} fetch-pack --stdin {repo}", capture=True, stdin=missings
)
missings = " ".join([x[1:] for x in ret.stdout.split() if x.startswith("?")])
cmd = f"git -C {tmp_dir} fetch-pack {repo} {missings}"
run_command(cmd, capture=True)
# And finally, we can checkout our subpath
run_command(f"git -C {tmp_dir} checkout {branch} {subpath}")
cmd = f"git -C {tmp_dir} checkout {branch} {subpath}"
run_command(cmd)
# We need Path(name) to make sure we also support subdirectories
shutil.move(str(tmp_dir / Path(subpath)), str(dest))
def _from_http_to_git(repo):
if repo.startswith("http://"):
repo = repo.replace(r"http://", r"https://")
if repo.startswith(r"https://"):
repo = repo.replace("https://", "git@").replace("/", ":", 1)
repo = f"{repo}.git"
return repo

5
spacy/cli/project/clone.py
View File

@ -43,7 +43,7 @@ def project_clone(name: str, dest: Path, *, repo: str = about.__projects__) -> N
git_sparse_checkout(repo, name, dest)
except subprocess.CalledProcessError:
err = f"Could not clone '{name}' from repo '{repo_name}'"
msg.fail(err)
msg.fail(err, exits=1)
msg.good(f"Cloned '{name}' from {repo_name}", project_dir)
if not (project_dir / PROJECT_FILE).exists():
msg.warn(f"No {PROJECT_FILE} found in directory")
@ -78,6 +78,7 @@ def check_clone(name: str, dest: Path, repo: str) -> None:
if not dest.parent.exists():
# We're not creating parents, parent dir should exist
msg.fail(
f"Can't clone project, parent directory doesn't exist: {dest.parent}",
f"Can't clone project, parent directory doesn't exist: {dest.parent}. "
f"Create the necessary folder(s) first before continuing.",
exits=1,
)

3
spacy/lang/hu/punctuation.py
View File

@ -7,6 +7,7 @@ _concat_icons = CONCAT_ICONS.replace("\u00B0", "")
_currency = r"\$¢£€¥฿"
_quotes = CONCAT_QUOTES.replace("'", "")
_units = UNITS.replace("%", "")
_prefixes = (
LIST_PUNCT
@ -26,7 +27,7 @@ _suffixes = (
r"(?<=[0-9])\+",
r"(?<=°[FfCcKk])\.",
r"(?<=[0-9])(?:[{c}])".format(c=_currency),
r"(?<=[0-9])(?:{u})".format(u=UNITS),
r"(?<=[0-9])(?:{u})".format(u=_units),
r"(?<=[{al}{e}{q}(?:{c})])\.".format(
al=ALPHA_LOWER, e=r"%²\-\+", q=CONCAT_QUOTES, c=_currency
),

5
spacy/pipeline/_parser_internals/_state.pxd
View File

@ -42,6 +42,7 @@ cdef cppclass StateC:
RingBufferC _hist
int length
int offset
int n_pushes
int _s_i
int _b_i
int _e_i
@ -49,6 +50,7 @@ cdef cppclass StateC:
__init__(const TokenC* sent, int length) nogil:
cdef int PADDING = 5
this.n_pushes = 0
this._buffer = <int*>calloc(length + (PADDING * 2), sizeof(int))
this._stack = <int*>calloc(length + (PADDING * 2), sizeof(int))
this.shifted = <bint*>calloc(length + (PADDING * 2), sizeof(bint))
@ -335,6 +337,7 @@ cdef cppclass StateC:
this.set_break(this.B_(0).l_edge)
if this._b_i > this._break:
this._break = -1
this.n_pushes += 1
void pop() nogil:
if this._s_i >= 1:
@ -351,6 +354,7 @@ cdef cppclass StateC:
this._buffer[this._b_i] = this.S(0)
this._s_i -= 1
this.shifted[this.B(0)] = True
this.n_pushes -= 1
void add_arc(int head, int child, attr_t label) nogil:
if this.has_head(child):
@ -431,6 +435,7 @@ cdef cppclass StateC:
this._break = src._break
this.offset = src.offset
this._empty_token = src._empty_token
this.n_pushes = src.n_pushes
void fast_forward() nogil:
# space token attachement policy:

4
spacy/pipeline/_parser_internals/stateclass.pyx
View File

@ -36,6 +36,10 @@ cdef class StateClass:
hist[i] = self.c.get_hist(i+1)
return hist
@property
def n_pushes(self):
return self.c.n_pushes
def is_final(self):
return self.c.is_final()

9
spacy/pipeline/tagger.pyx
View File

@ -289,7 +289,14 @@ class Tagger(Pipe):
err = Errors.E1006.format(name="Tagger")
raise ValueError(err)
self.set_output(len(self.labels))
self.model.initialize(X=doc_sample)
if doc_sample:
label_sample = [
self.model.ops.alloc2f(len(doc), len(self.labels))
for doc in doc_sample
]
self.model.initialize(X=doc_sample, Y=label_sample)
else:
self.model.initialize()
if sgd is None:
sgd = self.create_optimizer()
return sgd

11
spacy/pipeline/transition_parser.pyx
View File

@ -279,14 +279,14 @@ cdef class Parser(Pipe):
# Chop sequences into lengths of this many transitions, to make the
# batch uniform length.
# We used to randomize this, but it's not clear that actually helps?
cut_size = self.cfg["update_with_oracle_cut_size"]
states, golds, max_steps = self._init_gold_batch(
max_pushes = self.cfg["update_with_oracle_cut_size"]
states, golds, _ = self._init_gold_batch(
examples,
max_length=cut_size
max_length=max_pushes
)
else:
states, golds, _ = self.moves.init_gold_batch(examples)
max_steps = max([len(eg.x) for eg in examples])
max_pushes = max([len(eg.x) for eg in examples])
if not states:
return losses
all_states = list(states)
@ -302,7 +302,8 @@ cdef class Parser(Pipe):
backprop(d_scores)
# Follow the predicted action
self.transition_states(states, scores)
states_golds = [(s, g) for (s, g) in zip(states, golds) if not s.is_final()]
states_golds = [(s, g) for (s, g) in zip(states, golds)
if s.n_pushes < max_pushes and not s.is_final()]
backprop_tok2vec(golds)
if sgd not in (None, False):

12
spacy/tests/pipeline/test_textcat.py
View File

@ -84,9 +84,8 @@ def test_overfitting_IO():
# Simple test to try and quickly overfit the textcat component - ensuring the ML models work correctly
fix_random_seed(0)
nlp = English()
textcat = nlp.add_pipe("textcat")
# Set exclusive labels
textcat.model.attrs["multi_label"] = False
textcat = nlp.add_pipe("textcat", config={"model": {"exclusive_classes": True}})
train_examples = []
for text, annotations in TRAIN_DATA:
train_examples.append(Example.from_dict(nlp.make_doc(text), annotations))
@ -103,9 +102,8 @@ def test_overfitting_IO():
test_text = "I am happy."
doc = nlp(test_text)
cats = doc.cats
# note that by default, exclusive_classes = false so we need a bigger error margin
assert cats["POSITIVE"] > 0.8
assert cats["POSITIVE"] + cats["NEGATIVE"] == pytest.approx(1.0, 0.1)
assert cats["POSITIVE"] > 0.9
assert cats["POSITIVE"] + cats["NEGATIVE"] == pytest.approx(1.0, 0.001)
# Also test the results are still the same after IO
with make_tempdir() as tmp_dir:
@ -113,8 +111,8 @@ def test_overfitting_IO():
nlp2 = util.load_model_from_path(tmp_dir)
doc2 = nlp2(test_text)
cats2 = doc2.cats
assert cats2["POSITIVE"] > 0.8
assert cats2["POSITIVE"] + cats2["NEGATIVE"] == pytest.approx(1.0, 0.1)
assert cats2["POSITIVE"] > 0.9
assert cats2["POSITIVE"] + cats2["NEGATIVE"] == pytest.approx(1.0, 0.001)
# Test scoring
scores = nlp.evaluate(train_examples, scorer_cfg={"positive_label": "POSITIVE"})

78
website/docs/usage/layers-architectures.md
View File

@ -3,8 +3,9 @@ title: Layers and Model Architectures
teaser: Power spaCy components with custom neural networks
menu:
- ['Type Signatures', 'type-sigs']
- ['Defining Sublayers', 'sublayers']
- ['Swapping Architectures', 'swap-architectures']
- ['PyTorch & TensorFlow', 'frameworks']
- ['Thinc Models', 'thinc']
- ['Trainable Components', 'components']
next: /usage/projects
---
@ -22,8 +23,6 @@ its model architecture. The architecture is like a recipe for the network, and
you can't change the recipe once the dish has already been prepared. You have to
make a new one.
![Diagram of a pipeline component with its model](../images/layers-architectures.svg)
## Type signatures {#type-sigs}
<!-- TODO: update example, maybe simplify definition? -->
@ -92,9 +91,13 @@ code.
</Infobox>
## Defining sublayers {#sublayers}
## Swapping model architectures {#swap-architectures}
Model architecture functions often accept **sublayers as arguments**, so that
<!-- TODO: textcat example, using different architecture in the config -->
### Defining sublayers {#sublayers}
​Model architecture functions often accept **sublayers as arguments**, so that
you can try **substituting a different layer** into the network. Depending on
how the architecture function is structured, you might be able to define your
network structure entirely through the [config system](/usage/training#config),
@ -114,62 +117,37 @@ approaches. And if you want to define your own solution, all you need to do is
register a ~~Model[List[Doc], List[Floats2d]]~~ architecture function, and
you'll be able to try it out in any of the spaCy components. ​
<!-- TODO: example of switching sublayers -->
### Registering new architectures
- Recap concept, link to config docs. ​
<!-- TODO: example of swapping sublayers -->
## Wrapping PyTorch, TensorFlow and other frameworks {#frameworks}
<!-- TODO: this is copied over from the Thinc docs and we probably want to shorten it and make it more spaCy-specific -->
Thinc allows you to [wrap models](https://thinc.ai/docs/usage-frameworks)
written in other machine learning frameworks like PyTorch, TensorFlow and MXNet
using a unified [`Model`](https://thinc.ai/docs/api-model) API. As well as
**wrapping whole models**, Thinc lets you call into an external framework for
just **part of your model**: you can have a model where you use PyTorch just for
the transformer layers, using "native" Thinc layers to do fiddly input and
output transformations and add on task-specific "heads", as efficiency is less
of a consideration for those parts of the network.
Thinc allows you to wrap models written in other machine learning frameworks
like PyTorch, TensorFlow and MXNet using a unified
[`Model`](https://thinc.ai/docs/api-model) API. As well as **wrapping whole
models**, Thinc lets you call into an external framework for just **part of your
model**: you can have a model where you use PyTorch just for the transformer
layers, using "native" Thinc layers to do fiddly input and output
transformations and add on task-specific "heads", as efficiency is less of a
consideration for those parts of the network.
<!-- TODO: custom tagger implemented in PyTorch, wrapped as Thinc model, link off to project (with notebook?) -->
Thinc uses a special class, [`Shim`](https://thinc.ai/docs/api-model#shim), to
hold references to external objects. This allows each wrapper space to define a
custom type, with whatever attributes and methods are helpful, to assist in
managing the communication between Thinc and the external library. The
[`Model`](https://thinc.ai/docs/api-model#model) class holds `shim` instances in
a separate list, and communicates with the shims about updates, serialization,
changes of device, etc.
## Implementing models in Thinc {#thinc}
The wrapper will receive each batch of inputs, convert them into a suitable form
for the underlying model instance, and pass them over to the shim, which will
**manage the actual communication** with the model. The output is then passed
back into the wrapper, and converted for use in the rest of the network. The
equivalent procedure happens during backpropagation. Array conversion is handled
via the [DLPack](https://github.com/dmlc/dlpack) standard wherever possible, so
that data can be passed between the frameworks **without copying the data back**
to the host device unnecessarily.
| Framework | Wrapper layer | Shim | DLPack |
| -------------- | ------------------------------------------------------------------------- | --------------------------------------------------------- | --------------- |
| **PyTorch** | [`PyTorchWrapper`](https://thinc.ai/docs/api-layers#pytorchwrapper) | [`PyTorchShim`](https://thinc.ai/docs/api-model#shims) | ✅ |
| **TensorFlow** | [`TensorFlowWrapper`](https://thinc.ai/docs/api-layers#tensorflowwrapper) | [`TensorFlowShim`](https://thinc.ai/docs/api-model#shims) | ❌ <sup>1</sup> |
| **MXNet** | [`MXNetWrapper`](https://thinc.ai/docs/api-layers#mxnetwrapper) | [`MXNetShim`](https://thinc.ai/docs/api-model#shims) | ✅ |
1. DLPack support in TensorFlow is now
[available](<(https://github.com/tensorflow/tensorflow/issues/24453)>) but
still experimental.
<!-- TODO:
- Explain concept
- Link off to notebook ​
-->
<!-- TODO: use same example as above, custom tagger, but implemented in Thinc, link off to Thinc docs where appropriate -->
## Models for trainable components {#components}
<!-- TODO:
- Interaction with `predict`, `get_loss` and `set_annotations`
- Initialization life-cycle with `begin_training`.
- Link to relation extraction notebook.
Example: relation extraction component (implemented as project template)
-->
![Diagram of a pipeline component with its model](../images/layers-architectures.svg)
```python
def update(self, examples):