Minor edits.

utils_nlp/azureml/README.md

@@ -5,4 +5,4 @@ The AzureML submodule contains utilities to connect to a
 train, tune and operationalize NLP systems at scale using AzureML.   
 For example, the `DistributedCommunicator` class defined in
 [azureml_bert_util.py](./azureml_bert_util.py) assists in making communication with multiple nodes
-for distributed training possible. [azureml_utils.py](./azureml_utils.py) contains a few helper functions that make it easy to authenticate, create, or retrieve an existing AzureML resource.
+for distributed training possible. [azureml_utils.py](./azureml_utils.py) contains a few helper functions that make it easy to authenticate, create, or retrieve an AzureML resource.

utils_nlp/common/README.md

@@ -1,4 +1,7 @@
 ## [Common](.)
 
-This submodule contains high-level common utilities used across multiple algorithms and frameworks as well as helper functions for managing aspects of different frameworks like pytorch.  
-For example, [pytorch_utils.py](./pytorch_utils.py) contains utilities to interact with PyTorch like getting a device architecture (cpu or gpu), moves a model to a specific device, and handles parallelism when multiple gpus are present.  
+This submodule contains high-level common utilities used across multiple algorithms and 
+frameworks as well as helper functions for managing aspects of different frameworks like pytorch.  
+For example, [pytorch_utils.py](./pytorch_utils.py) contains utilities to interact with PyTorch 
+like getting a device architecture (cpu or gpu), moving a model to a specific device, and handling 
+parallelism when multiple gpus are present.  

utils_nlp/dataset/README.md

@@ -14,11 +14,14 @@ df = load_pandas_df(DATA_FOLDER, file_split ="train", nrows = 1000)
 |Dataset|Dataloader script|
 |-------|-----------------|
 |[Microsoft Research Paraphrase Corpus](https://www.microsoft.com/en-us/download/details.aspx?id=52398)|[msrpc.py](./msrpc.py)|
-|[The Multi-Genre NLI Corpus (MultiNLI)](https://www.nyu.edu/projects/bowman/multinli/)|[multinli.py](./multinli.py)|
+|[The Multi-Genre NLI (MultiNLI) Corpus](https://www.nyu.edu/projects/bowman/multinli/)|[multinli
+.py](
+./multinli.py)|
 |[The Stanford Natural Language Inference (SNLI) Corpus](https://nlp.stanford.edu/projects/snli/)|[snli.py](./snli.py)|
 |[Wikigold NER](https://github.com/juand-r/entity-recognition-datasets/tree/master/data/wikigold/CONLL-format/data)|[wikigold.py](./wikigold.py)|
-|[The Cross-Lingual NLI Corpus (XNLI)](https://www.nyu.edu/projects/bowman/xnli/)|[xnli.py](./xnli.py)|
+|[The Cross-Lingual NLI (XNLI) Corpus](https://www.nyu.edu/projects/bowman/xnli/)|[xnli.py](
+./xnli.py)|
 |[The STSbenchmark dataset](http://ixa2.si.ehu.es/stswiki/index.php/STSbenchmark)|[stsbenchmark.py](./stsbenchmark.py)|
 
 ## Dataset References
-Please see [DatasetReferences.md](DatasetReferences.md) for attributions for datasets used.
+Please see [DatasetReferences.md](DatasetReferences.md) for attributions of datasets used.