Merge branch 'develop' into fix/default-corpus-values

spacy/cli/train.py

@@ -49,35 +49,11 @@ def train_cli(
         config = util.load_config(config_path, overrides=overrides, interpolate=False)
     msg.divider("Initializing pipeline")
     with show_validation_error(config_path, hint_fill=False):
-        nlp = init_pipeline(config, output_path, use_gpu=use_gpu)
+        nlp = init_nlp(config, use_gpu=use_gpu)
     msg.divider("Training pipeline")
     train(nlp, output_path, use_gpu=use_gpu, silent=False)
 
 
-def init_pipeline(
-    config: Config, output_path: Optional[Path], *, use_gpu: int = -1
-) -> Language:
-    init_kwargs = {"use_gpu": use_gpu}
-    if output_path is not None:
-        init_path = output_path / "model-initial"
-        if not init_path.exists():
-            msg.info(f"Initializing the pipeline in {init_path}")
-            nlp = init_nlp(config, **init_kwargs)
-            nlp.to_disk(init_path)
-            msg.good(f"Saved initialized pipeline to {init_path}")
-        else:
-            nlp = util.load_model(init_path)
-            if must_reinitialize(config, nlp.config):
-                msg.warn("Config has changed: need to re-initialize pipeline")
-                nlp = init_nlp(config, **init_kwargs)
-                nlp.to_disk(init_path)
-                msg.good(f"Re-initialized pipeline in {init_path}")
-            else:
-                msg.good(f"Loaded initialized pipeline from {init_path}")
-        return nlp
-    return init_nlp(config, **init_kwargs)
-
-
 def verify_cli_args(config_path: Path, output_path: Optional[Path] = None) -> None:
     # Make sure all files and paths exists if they are needed
     if not config_path or not config_path.exists():

spacy/training/corpus.py

@@ -28,6 +28,7 @@ def create_docbin_reader(
 ) -> Callable[["Language"], Iterable[Example]]:
     if path is None:
         raise ValueError(Errors.E913)
+    util.logger.debug(f"Loading corpus from path: {path}")
     return Corpus(
         path,
         gold_preproc=gold_preproc,

spacy/training/initialize.py

@@ -94,6 +94,7 @@ def init_vocab(
     if vectors is not None:
         load_vectors_into_model(nlp, vectors)
         logger.info(f"Added vectors: {vectors}")
+    logger.info("Finished initializing nlp object")
 
 
 def load_vectors_into_model(