Merge pull request #3625 from co63oc/fix1

fix typos in docs and comments

Author: alanakbik

docs/tutorial/tutorial-basics/part-of-speech-tagging.md

@@ -1,6 +1,6 @@
 # Tagging parts-of-speech
 
-This tutorials shows you how to do part-of-speech tagging in Flair, showcases univeral and language-specific models, and gives a list of all PoS models in Flair.
+This tutorials shows you how to do part-of-speech tagging in Flair, showcases universal and language-specific models, and gives a list of all PoS models in Flair.
 
 ## Language-specific parts-of-speech (PoS)
 
@@ -111,7 +111,7 @@ Universal parts-of-speech are a set of minimal syntactic units that exist across
 will have VERBs or NOUNs. 
 
 
-We ship models trained over 14 langages to tag upos in **multilingual text**. Use like this: 
+We ship models trained over 14 languages to tag upos in **multilingual text**. Use like this: 
 
 ```python
 from flair.nn import Classifier

docs/tutorial/tutorial-embeddings/flair-embeddings.md

@@ -120,7 +120,7 @@ Words are now embedded using a concatenation of three different embeddings. This
 
 We also developed a pooled variant of the [`FlairEmbeddings`](#flair.embeddings.token.FlairEmbeddings). These embeddings differ in that they *constantly evolve over time*, even at prediction time (i.e. after training is complete). This means that the same words in the same sentence at two different points in time may have different embeddings.
 
-[`PooledFlairEmbeddings`](#flair.embeddings.token.PooledFlairEmbeddings) manage a 'global' representation of each distinct word by using a pooling operation of all past occurences. More details on how this works may be found in [Akbik et al. (2019)](https://www.aclweb.org/anthology/N19-1078/).
+[`PooledFlairEmbeddings`](#flair.embeddings.token.PooledFlairEmbeddings) manage a 'global' representation of each distinct word by using a pooling operation of all past occurrences. More details on how this works may be found in [Akbik et al. (2019)](https://www.aclweb.org/anthology/N19-1078/).
 
 You can instantiate and use [`PooledFlairEmbeddings`](#flair.embeddings.token.PooledFlairEmbeddings) like [`FlairEmbeddings`](#flair.embeddings.token.FlairEmbeddings):
 

docs/tutorial/tutorial-training/how-to-train-span-classifier.md

@@ -7,7 +7,7 @@ This tutorial section show you how to train models using the [Span Classifier](#
 
 ## Training an entity linker (NEL) model with transformers
 
-For a state-of-the-art NER sytem you should fine-tune transformer embeddings, and use full document context
+For a state-of-the-art NER system you should fine-tune transformer embeddings, and use full document context
 (see our [FLERT](https://arxiv.org/abs/2011.06993) paper for details).
 
 Use the following script:

examples/ner/README.md

@@ -42,7 +42,7 @@ To use the recently introduced [FLERT](https://arxiv.org/abs/2011.06993) the fol
 # Example
 
 The following example shows how to fine-tune a model for the recently released [Masakhane](https://arxiv.org/abs/2103.11811) dataset for
-the Luo language. We choose XLM-RoBERTa Base for fine-tuning. In this example, the best model (choosen on performance on development set)
+the Luo language. We choose XLM-RoBERTa Base for fine-tuning. In this example, the best model (chosen on performance on development set)
 is used for final evaluation on the test set.
 
 ## Choosing the dataset

flair/models/entity_mention_linking.py

@@ -233,7 +233,7 @@ def process_entity_name(self, entity_name: str) -> str:
 
         entity_name = entity_name.strip()
 
-        # NOTE: Avoid emtpy string if mentions are just punctutations (e.g. `-` or `(`)
+        # NOTE: Avoid empty string if mentions are just punctuations (e.g. `-` or `(`)
         entity_name = original if len(entity_name) == 0 else entity_name
 
         return entity_name

flair/training_utils.py

@@ -155,7 +155,7 @@ def _init_weights_index(self, key, state_dict, weights_to_watch):
 
 
 class AnnealOnPlateau:
-    """A learningrate sheduler for annealing on plateau.
+    """A learningrate scheduler for annealing on plateau.
 
     This class is a modification of
     torch.optim.lr_scheduler.ReduceLROnPlateau that enables