Merge branch 'develop' of https://github.com/explosion/spaCy into develop

spacy/errors.py

@@ -287,6 +287,8 @@ class Errors(object):
     E108 = ("As of spaCy v2.1, the pipe name `sbd` has been deprecated "
             "in favor of the pipe name `sentencizer`, which does the same "
             "thing. For example, use `nlp.create_pipeline('sentencizer')`")
+    E109 = ("Model for component '{name}' not initialized. Did you forget to load "
+            "a model, or forget to call begin_training()?")
 
 
 @add_codes

spacy/pipeline.pyx

@@ -293,10 +293,16 @@ class Pipe(object):
         Both __call__ and pipe should delegate to the `predict()`
         and `set_annotations()` methods.
         """
+        self.require_model()
         scores, tensors = self.predict([doc])
         self.set_annotations([doc], scores, tensors=tensors)
         return doc
 
+    def require_model(self):
+        """Raise an error if the component's model is not initialized."""
+        if getattr(self, 'model', None) in (None, True, False):
+            raise ValueError(Errors.E109.format(name=self.name))
+
     def pipe(self, stream, batch_size=128, n_threads=-1):
         """Apply the pipe to a stream of documents.
 
@@ -313,6 +319,7 @@ class Pipe(object):
         """Apply the pipeline's model to a batch of docs, without
         modifying them.
         """
+        self.require_model()
         raise NotImplementedError
 
     def set_annotations(self, docs, scores, tensors=None):
@@ -325,6 +332,7 @@ class Pipe(object):
 
         Delegates to predict() and get_loss().
         """
+        self.require_model()
         raise NotImplementedError
 
     def rehearse(self, docs, sgd=None, losses=None, **config):
@@ -495,6 +503,7 @@ class Tensorizer(Pipe):
         docs (iterable): A sequence of `Doc` objects.
         RETURNS (object): Vector representations for each token in the docs.
         """
+        self.require_model()
         inputs = self.model.ops.flatten([doc.tensor for doc in docs])
         outputs = self.model(inputs)
         return self.model.ops.unflatten(outputs, [len(d) for d in docs])
@@ -519,6 +528,7 @@ class Tensorizer(Pipe):
         sgd (callable): An optimizer.
         RETURNS (dict): Results from the update.
         """
+        self.require_model()
         if isinstance(docs, Doc):
             docs = [docs]
         inputs = []
@@ -600,6 +610,7 @@ class Tagger(Pipe):
             yield from docs
 
     def predict(self, docs):
+        self.require_model()
         if not any(len(doc) for doc in docs):
             # Handle case where there are no tokens in any docs.
             n_labels = len(self.labels)
@@ -644,6 +655,7 @@ class Tagger(Pipe):
             doc.is_tagged = True
 
     def update(self, docs, golds, drop=0., sgd=None, losses=None):
+        self.require_model()
         if losses is not None and self.name not in losses:
             losses[self.name] = 0.
 
@@ -904,6 +916,7 @@ class MultitaskObjective(Tagger):
         return model
 
     def predict(self, docs):
+        self.require_model()
         tokvecs = self.model.tok2vec(docs)
         scores = self.model.softmax(tokvecs)
         return tokvecs, scores
@@ -1042,6 +1055,7 @@ class ClozeMultitask(Pipe):
         return sgd
 
     def predict(self, docs):
+        self.require_model()
         tokvecs = self.model.tok2vec(docs)
         vectors = self.model.output_layer(tokvecs)
         return tokvecs, vectors
@@ -1061,6 +1075,7 @@ class ClozeMultitask(Pipe):
         pass
 
     def rehearse(self, docs, drop=0., sgd=None, losses=None):
+        self.require_model()
         if losses is not None and self.name not in losses:
             losses[self.name] = 0.
         predictions, bp_predictions = self.model.begin_update(docs, drop=drop)
@@ -1105,9 +1120,11 @@ class SimilarityHook(Pipe):
             yield self(doc)
 
     def predict(self, doc1, doc2):
+        self.require_model()
         return self.model.predict([(doc1, doc2)])
 
     def update(self, doc1_doc2, golds, sgd=None, drop=0.):
+        self.require_model()
         sims, bp_sims = self.model.begin_update(doc1_doc2, drop=drop)
 
     def begin_training(self, _=tuple(), pipeline=None, sgd=None, **kwargs):
@@ -1171,6 +1188,7 @@ class TextCategorizer(Pipe):
             yield from docs
 
     def predict(self, docs):
+        self.require_model()
         scores = self.model(docs)
         scores = self.model.ops.asarray(scores)
         tensors = [doc.tensor for doc in docs]

spacy/syntax/_parser_model.pyx

@@ -205,7 +205,9 @@ class ParserModel(Model):
             return
         smaller = self.upper
         larger = Affine(new_output, smaller.nI)
-        larger.W *= 0
+        # Set nan as value for unseen classes, to prevent prediction.
+        larger.W.fill(self.ops.xp.nan)
+        larger.b.fill(self.ops.xp.nan)
         # It seems very unhappy if I pass these as smaller.W?
         # Seems to segfault. Maybe it's a descriptor protocol thing?
         smaller_W = smaller.W
@@ -254,8 +256,23 @@ class ParserStepModel(Model):
         if mask is not None:
             vector *= mask
         scores, get_d_vector = self.vec2scores.begin_update(vector, drop=drop)
+        # We can have nans from unseen classes.
+        # For backprop purposes, we want to treat unseen classes as having the
+        # lowest score.
+        # numpy's nan_to_num function doesn't take a value, and nan is replaced
+        # by 0...-inf is replaced by minimum, so we go via that. Ugly to the max.
+        scores[self.ops.xp.isnan(scores)] = -self.ops.xp.inf
+        self.ops.xp.nan_to_num(scores, copy=False)
 
         def backprop_parser_step(d_scores, sgd=None):
+            # If we have a non-zero gradient for a previously unseen class,
+            # replace the weight with 0.
+            new_classes = self.ops.xp.logical_and(
+                self.vec2scores.ops.xp.isnan(self.vec2scores.b),
+                d_scores.any(axis=0)
+            )
+            self.vec2scores.b[new_classes] = 0.
+            self.vec2scores.W[new_classes] = 0.
             d_vector = get_d_vector(d_scores, sgd=sgd)
             if mask is not None:
                 d_vector *= mask
@@ -400,6 +417,8 @@ cdef class precompute_hiddens:
             state_vector, mask = self.ops.maxout(state_vector)
 
         def backprop_nonlinearity(d_best, sgd=None):
+            # Fix nans (which can occur from unseen classes.)
+            d_best[self.ops.xp.isnan(d_best)] = 0.
             if self.nP == 1:
                 d_best *= mask
                 d_best = d_best.reshape((d_best.shape + (1,)))

spacy/syntax/nn_parser.pyx

@@ -227,7 +227,13 @@ cdef class Parser:
             for doc in batch_in_order:
                 yield doc
                 
+    def require_model(self):
+        """Raise an error if the component's model is not initialized."""
+        if getattr(self, 'model', None) in (None, True, False):
+            raise ValueError(Errors.E109.format(name=self.name))
+
     def predict(self, docs, beam_width=1, beam_density=0.0, drop=0.):
+        self.require_model()
         if isinstance(docs, Doc):
             docs = [docs]
         if not any(len(doc) for doc in docs):
@@ -375,6 +381,7 @@ cdef class Parser:
         return [b for b in beams if not b.is_done]
 
     def update(self, docs, golds, drop=0., sgd=None, losses=None):
+        self.require_model()
         if isinstance(docs, Doc) and isinstance(golds, GoldParse):
             docs = [docs]
             golds = [golds]