Merge remote-tracking branch 'upstream/develop' into fix/debug_model

# Conflicts:
#	spacy/cli/debug_model.py
This commit is contained in:
svlandeg 2020-09-21 16:58:40 +02:00
parent eb9b447960
commit 447b3e5787
1 changed files with 27 additions and 21 deletions

View File

@ -1,4 +1,4 @@
from typing import Dict, Any, Optional
from typing import Dict, Any, Optional, Iterable
from pathlib import Path
from wasabi import msg
from thinc.api import require_gpu, fix_random_seed, set_dropout_rate, Adam
@ -93,11 +93,10 @@ def debug_model(nlp, model: Model, *, print_settings: Optional[Dict[str, Any]] =
# STEP 1: Initializing the model and printing again
X = _get_docs()
goldY = _get_output(model.ops)
# _set_output_dim(nO=goldY.shape[-1], model=model)
_set_output_dim(nO=7, model=model)
# The output vector might differ from the official type of the output layer
with data_validation(False):
model.initialize(X=X, Y=goldY)
model.initialize(X=X)
if print_settings.get("print_after_init"):
msg.divider(f"STEP 1 - after initialization")
_print_model(model, print_settings)
@ -110,12 +109,15 @@ def debug_model(nlp, model: Model, *, print_settings: Optional[Dict[str, Any]] =
if model.has_ref("tok2vec") and model.get_ref("tok2vec").name == "tok2vec-listener":
tok2vec = nlp.get_pipe("tok2vec")
tok2vec.model.initialize(X=X)
goldY = None
for e in range(3):
if tok2vec:
tok2vec.predict(X)
Y, get_dX = model.begin_update(X)
print("get_dX", get_dX)
dY = get_gradient(goldY, Y)
# simulate a goldY value
if not goldY:
goldY = _simulate_gold(Y)
dY = get_gradient(goldY, Y, model.ops)
get_dX(dY)
model.finish_update(optimizer)
if print_settings.get("print_after_training"):
@ -128,11 +130,20 @@ def debug_model(nlp, model: Model, *, print_settings: Optional[Dict[str, Any]] =
msg.divider(f"STEP 3 - prediction")
msg.info(str(prediction))
msg.good(f"Succesfully ended analysis - model looks good!")
msg.good(f"Succesfully ended analysis - model looks good.")
def get_gradient(goldY, Y):
return Y - goldY
def _simulate_gold(element, counter=1):
if isinstance(element, Iterable):
for i in range(len(element)):
element[i] = _simulate_gold(element[i], counter+i)
return element
else:
return 1/counter
def get_gradient(goldY, Y, ops):
return ops.asarray(Y) - ops.asarray(goldY)
def _sentences():
@ -149,18 +160,13 @@ def _get_docs(lang: str = "en"):
return list(nlp.pipe(_sentences()))
def _get_output(ops):
docs = len(_get_docs())
labels = 6
output = ops.alloc2f(d0=docs, d1=labels)
for i in range(docs):
for j in range(labels):
output[i, j] = 1 / (i+j+0.01)
return ops.xp.asarray(output)
def _get_output_old(xp):
return xp.asarray([i + 10 for i, _ in enumerate(_get_docs())], dtype="float32")
def _set_output_dim(model, nO):
# simulating dim inference by directly setting the nO argument of the model
if model.has_dim("nO") is None:
model.set_dim("nO", nO)
if model.has_ref("output_layer"):
if model.get_ref("output_layer").has_dim("nO") is None:
model.get_ref("output_layer").set_dim("nO", nO)
def _print_model(model, print_settings):