Add ClassificationTextfileDatasetReader

combo/data/dataset_readers/classification_textfile_dataset_reader.py

 from typing import Dict, Iterable, Optional
 
-from .dataset_reader import DatasetReader, DatasetReaderInput
 from overrides import overrides
 
+from .dataset_reader import DatasetReader
 from .. import Instance, Tokenizer, TokenIndexer
 from ..fields.label_field import LabelField
 from ..fields.text_field import TextField
@@ -10,12 +10,10 @@ from ..fields.text_field import TextField
 
 class ClassificationTextfileDatasetReader(DatasetReader):
     def __init__(self,
-                 file_path: Optional[DatasetReaderInput] = None,
                  tokenizer: Optional[Tokenizer] = None,
-                 token_indexers: Optional[Dict[str, TokenIndexer]] = None,
-                 separator: str = ',') -> None:
-        super().__init__(file_path, tokenizer, token_indexers)
-        self.__separator = separator
+                 token_indexers: Optional[Dict[str, TokenIndexer]] = None) -> None:
+        super().__init__(tokenizer, token_indexers)
+        self.__separator = None
 
     @property
     def separator(self) -> str:
@@ -27,6 +25,11 @@ class ClassificationTextfileDatasetReader(DatasetReader):
 
     @overrides
     def _read(self) -> Iterable[Instance]:
+        if self.file_path is None:
+            raise ValueError('File path is None')
+        elif self.separator is None:
+            raise ValueError('Separator is None')
+
         with open(self.file_path, 'r') as lines:
             for line in lines:
                 text, label = line.strip().split(self.separator)
@@ -35,3 +38,8 @@ class ClassificationTextfileDatasetReader(DatasetReader):
                 label_field = LabelField(label)
                 fields = {'text': text_field, 'label': label_field}
                 yield Instance(fields)
+
+    def __call__(self, file_path: str, separator: str):
+        self.file_path = file_path
+        self.separator = separator
+        return self

combo/data/dataset_readers/dataset_reader.py

@@ -9,7 +9,8 @@ from typing import Iterable, Iterator, Optional, Union, TypeVar, Dict, List
 from overrides import overrides
 from torch.utils.data import IterableDataset
 
-from combo.data import Instance, Tokenizer
+from combo.data.instance import Instance
+from combo.data.tokenizers import Tokenizer
 from combo.data.token_indexers import TokenIndexer
 
 logger = logging.getLogger(__name__)
@@ -25,11 +26,10 @@ class DatasetReader(IterableDataset):
     of `Instance`s.
     """
     def __init__(self,
-                 file_path: Optional[DatasetReaderInput] = None,
                  tokenizer: Optional[Tokenizer] = None,
                  token_indexers: Optional[Dict[str, TokenIndexer]] = None) -> None:
         super(DatasetReader).__init__()
-        self.__file_path = file_path
+        self.__file_path = None
         self.__tokenizer = tokenizer
         self.__token_indexers = token_indexers
 
@@ -50,7 +50,7 @@ class DatasetReader(IterableDataset):
         return self.__token_indexers
 
     @overrides
-    def __getitem__(self, item) -> Instance:
+    def __getitem__(self, item, **kwargs) -> Instance:
         raise NotImplementedError
 
     @overrides

combo/data/fields/adjacency_field.py

@@ -9,8 +9,8 @@ import textwrap
 
 import torch
 
-from combo.data import Vocabulary
-from combo.data.fields import Field
+from combo.data.vocabulary import Vocabulary
+from combo.data.fields.field import Field
 from combo.data.fields.sequence_field import SequenceField
 from combo.utils import ConfigurationError
 

combo/data/fields/label_field.py

@@ -9,7 +9,7 @@ import logging
 
 import torch
 
-from combo.data import Vocabulary
+from combo.data.vocabulary import Vocabulary
 from combo.data.fields import Field
 from combo.utils import ConfigurationError
 

combo/data/fields/list_field.py

@@ -4,7 +4,7 @@ https://github.com/allenai/allennlp/blob/main/allennlp/data/fields/list_field.py
 """
 from typing import Dict, List, Iterator, Sequence, Any
 
-from combo.data import Vocabulary
+from combo.data.vocabulary import Vocabulary
 from combo.data.fields.field import DataArray, Field
 from combo.data.fields.sequence_field import SequenceField
 from combo.utils import pad_sequence_to_length

combo/data/fields/sequence_label_field.py

@@ -10,7 +10,7 @@ import textwrap
 
 import torch
 
-from combo.data import Vocabulary
+from combo.data.vocabulary import Vocabulary
 from combo.data.fields import Field
 from combo.data.fields.sequence_field import SequenceField
 from combo.utils import ConfigurationError, pad_sequence_to_length

combo/data/fields/sequence_multilabel_field.py

@@ -11,7 +11,7 @@ from typing import Set, List, Callable, Iterator, Union, Dict
 import torch
 from overrides import overrides
 
-from combo.data import Vocabulary
+from combo.data.vocabulary import Vocabulary
 from combo.data.fields import Field
 from combo.data.fields.sequence_field import SequenceField
 from combo.utils import ConfigurationError

combo/data/instance.py

@@ -4,7 +4,7 @@ https://github.com/allenai/allennlp/blob/main/allennlp/data/instance.py
 """
 
 from typing import Dict, MutableMapping, Mapping, Any
-from combo.data import Vocabulary
+from combo.data.vocabulary import Vocabulary
 from combo.data.fields import Field
 from combo.data.fields.field import DataArray
 

requirements.txt

@@ -2,8 +2,10 @@ absl-py~=1.4.0
 base58~=2.1.1
 cached-path~=1.3.3
 conllu~=4.4.1
+conllutils~=1.1.4
 dependency-injector~=4.41.0
 dill~=0.3.6
+importlib-resources~=5.12.0
 overrides~=7.3.1
 torch~=2.0.0
 torchtext~=0.15.1

requirements_no_deps.txt

+lambo @ git+https://gitlab.clarin-pl.eu/syntactic-tools/lambo.git
\ No newline at end of file