add dataset auto-conversion to all main entry scripts

configs/metric/count_entity_labels.yaml

@@ -6,4 +6,5 @@ title: entity label distribution
 field: entities
 labels: ???
 show_histogram: true
-# show_as_markdown: true
+show_as_markdown: true
+document_type: pytorch_ie.documents.TextDocumentWithLabeledEntities

configs/metric/count_text_characters.yaml

@@ -5,4 +5,5 @@ _target_: pytorch_ie.metrics.statistics.FieldLengthCollector
 title: text length (characters)
 field: text
 show_histogram: true
-# show_as_markdown: true
+show_as_markdown: true
+document_type: pytorch_ie.documents.TextBasedDocument

configs/metric/count_text_tokens.yaml

@@ -7,5 +7,7 @@ text_field: text
 tokenizer: bert-base-uncased
 tokenizer_kwargs:
   add_special_tokens: false
+  # strict_span_conversion: false
 show_histogram: true
-# show_as_markdown: true
+show_as_markdown: true
+document_type: pytorch_ie.documents.TextBasedDocument

src/evaluate.py

@@ -76,6 +76,9 @@ def evaluate(cfg: DictConfig) -> Tuple[dict, dict]:
     log.info(f"Instantiating taskmodule <{cfg.taskmodule._target_}>")
     taskmodule: TaskModule = hydra.utils.instantiate(cfg.taskmodule, _convert_="partial")
 
+    # auto-convert the dataset if the taskmodule specifies a document type
+    dataset = taskmodule.convert_dataset(dataset)
+
     # Init pytorch-ie datamodule
     log.info(f"Instantiating datamodule <{cfg.datamodule._target_}>")
     datamodule: PieDataModule = hydra.utils.instantiate(

src/evaluate_documents.py

@@ -70,6 +70,9 @@ def evaluate_documents(cfg: DictConfig) -> Tuple[dict, dict]:
     log.info(f"Instantiating metric <{cfg.metric._target_}>")
     metric: DocumentMetric = hydra.utils.instantiate(cfg.metric, _convert_="partial")
 
+    # auto-convert the dataset if the metric specifies a document type
+    dataset = metric.convert_dataset(dataset)
+
     # Init lightning loggers
     loggers = utils.instantiate_dict_entries(cfg, "logger")
 

src/predict.py

@@ -84,6 +84,9 @@ def predict(cfg: DictConfig) -> Tuple[dict, dict]:
                 pipeline.device
             )
 
+        # auto-convert the dataset if the taskmodule specifies a document type
+        dataset = pipeline.taskmodule.convert_dataset(dataset)
+
     # Init the serializer
     serializer: Optional[DocumentSerializer] = None
     if cfg.get("serializer") and cfg.serializer.get("_target_"):

src/train.py

@@ -107,23 +107,7 @@ def train(cfg: DictConfig) -> Tuple[dict, dict]:
     )
 
     # auto-convert the dataset if the taskmodule specifies a document type
-    if taskmodule.document_type is not None:
-        if issubclass(dataset.document_type, taskmodule.document_type):
-            log.info(
-                f"the dataset is already of the document type that is specified by the taskmodule: "
-                f"{taskmodule.document_type}"
-            )
-        else:
-            log.info(
-                f"convert the dataset to the document type that is specified by the taskmodule: "
-                f"{taskmodule.document_type}"
-            )
-            dataset = dataset.to_document_type(taskmodule.document_type)
-    else:
-        log.warning(
-            "The taskmodule does not specify a document type. The dataset can not be automatically converted "
-            "to a document type."
-        )
+    dataset = taskmodule.convert_dataset(dataset)
 
     # Init pytorch-ie datamodule
     log.info(f"Instantiating datamodule <{cfg.datamodule._target_}>")