update sphinx files and fix some docs issues

Machine-Learning-for-Medical-Language · Jul 6, 2023 · 0c5b7d4 · 0c5b7d4
1 parent 8d24c1c
commit 0c5b7d4
Show file tree

Hide file tree

Showing 12 changed files with 39 additions and 3 deletions.
diff --git a/build_doc_source.sh b/build_doc_source.sh
diff --git a/docs/cnlpt.api.hier_rest.rst b/docs/cnlpt.api.hier_rest.rst
@@ -0,0 +1,6 @@
+cnlpt.api.hier\_rest module
+===========================
+
+.. automodule:: cnlpt.api.hier_rest
+   :members:
+   :show-inheritance:
diff --git a/docs/cnlpt.api.rst b/docs/cnlpt.api.rst
@@ -10,6 +10,7 @@ Submodules
    cnlpt.api.cnlp_rest
    cnlpt.api.dtr_rest
    cnlpt.api.event_rest
+   cnlpt.api.hier_rest
    cnlpt.api.negation_rest
    cnlpt.api.temporal_rest
    cnlpt.api.timex_rest

diff --git a/docs/cnlpt.cnlp_args.rst b/docs/cnlpt.cnlp_args.rst
@@ -0,0 +1,6 @@
+cnlpt.cnlp\_args module
+=======================
+
+.. automodule:: cnlpt.cnlp_args
+   :members:
+   :show-inheritance:
diff --git a/docs/cnlpt.cnlp_metrics.rst b/docs/cnlpt.cnlp_metrics.rst
@@ -0,0 +1,6 @@
+cnlpt.cnlp\_metrics module
+==========================
+
+.. automodule:: cnlpt.cnlp_metrics
+   :members:
+   :show-inheritance:
diff --git a/docs/cnlpt.cnlp_predict.rst b/docs/cnlpt.cnlp_predict.rst
@@ -0,0 +1,6 @@
+cnlpt.cnlp\_predict module
+==========================
+
+.. automodule:: cnlpt.cnlp_predict
+   :members:
+   :show-inheritance:
diff --git a/docs/cnlpt.data.rst b/docs/cnlpt.data.rst
@@ -8,6 +8,7 @@ Submodules
    :maxdepth: 4
 
    cnlpt.data.preprocess_i2b2_2008
+   cnlpt.data.transform_prot
    cnlpt.data.transform_uci_drug
 
 Module contents

diff --git a/docs/cnlpt.data.transform_prot.rst b/docs/cnlpt.data.transform_prot.rst
@@ -0,0 +1,6 @@
+cnlpt.data.transform\_prot module
+=================================
+
+.. automodule:: cnlpt.data.transform_prot
+   :members:
+   :show-inheritance:
diff --git a/docs/cnlpt.rst b/docs/cnlpt.rst
@@ -19,7 +19,10 @@ Submodules
    cnlpt.BaselineModels
    cnlpt.CnlpModelForClassification
    cnlpt.HierarchicalTransformer
+   cnlpt.cnlp_args
    cnlpt.cnlp_data
+   cnlpt.cnlp_metrics
+   cnlpt.cnlp_predict
    cnlpt.cnlp_processors
    cnlpt.thyme_eval
    cnlpt.train_system

diff --git a/docs/transformer_objects.inv b/docs/transformer_objects.inv
diff --git a/docs/transformer_objects.txt b/docs/transformer_objects.txt
@@ -12,3 +12,5 @@ transformers.data.processors.utils.InputExample py:class 1 main_classes/processo
 transformers.tokenization_utils.PreTrainedTokenizer py:class 1 main_classes/tokenizer#transformers.PreTrainedTokenizer -
 transformers.PretrainedConfig py:class 1 main_classes/configuration#$ -
 transformers.AutoConfig py:class 1 model_doc/auto#$ -
+transformers.BatchEncoding py:class 1 main_classes/tokenizer#$ -
+transformers.tokenization_utils_base.BatchEncoding py:class 1 main_classes/tokenizer#transformers.BatchEncoding -
diff --git a/src/cnlpt/cnlp_data.py b/src/cnlpt/cnlp_data.py
@@ -11,7 +11,6 @@
 import torch
 from torch.utils.data.dataset import Dataset
 from transformers import BatchEncoding, InputExample
-# from transformers.data.processors.utils import DataProcessor, InputExample
 from transformers.tokenization_utils import PreTrainedTokenizer
 from datasets import Features
 from dataclasses import dataclass, field, asdict, astuple
@@ -121,7 +120,7 @@ def cnlp_convert_features_to_hierarchical(
     :param pad_id: the tokenizer's ID representing the PAD token
     :param insert_empty_chunk_at_beginning: whether to insert an
         empty chunk at the beginning of the instance
-    :return: an instance of `BatchEncoding` containing the chunked instance
+    :return: an instance of :class:`transformers.BatchEncoding` containing the chunked instance
     """
 
     for ind in range(len(features['input_ids'])):
@@ -740,7 +739,7 @@ def _reconcile_labels_lists(self, processor):
                 new_labels = set(labels)
                 old_labels = set(self.tasks_to_labels[task])
                 if new_labels.isdisjoint(old_labels):
-                    raise Exception("The same task name has disjoint sets of labels in different dataset: %s vs. %s" % (str(str(old_labels), str(new_labels))))
+                    raise Exception("The same task name has disjoint sets of labels in different dataset: %s vs. %s" % (str(old_labels), str(new_labels)))
                 elif new_labels != old_labels:
                     logger.warn("Two different datasets have the same task name but not completely equal label lists: %s vs. %s. We will merge them." (str(old_labels), str(new_labels)))
                     self.tasks_to_labels[task] = old_labels.union(new_labels)