typos

doc/Coordinates-in-PDF.md

@@ -42,7 +42,7 @@ Example with cURL:
 
 ### Batch processing
 
-We recommand to use the above service mode for best performance and range of options. 
+We recommend to use the above service mode for best performance and range of options. 
 
 Generating coordinates can also been obtained with the batch mode by adding the parameter ```-teiCoordinates``` with the command ```processFullText```.
 
@@ -161,4 +161,4 @@ Example 2:
 
 The above ```@coords``` XML attributes introduces 4 bounding boxes to define the area of the bibliographical reference (typically because the reference is on several line).
 
-As side note, in traditionnal TEI encoding an area should be expressed using SVG. However it would have make the TEI document quickly unreadable and extremely heavy and we are using this more compact notation. 
+As side note, in traditional TEI encoding an area should be expressed using SVG. However it would have make the TEI document quickly unreadable and extremely heavy and we are using this more compact notation. 

doc/Deep-Learning-models.md

@@ -42,7 +42,7 @@ Using Deep Learning model in GROBID with a normal installation/build is not stra
 
 The most simple solution is to use the ["full" GROBID docker image](Grobid-docker.md), which allows to use Deep Learning models without further installation and which provides automatic GPU support. 
 
-However if you need a "local" library installation and build, prepare a lot of coffee, here are the step-by-step instructions to get a working local Deep Learning GROBID.
+However, if you need a "local" library installation and build, prepare a lot of coffee, here are the step-by-step instructions to get a working local Deep Learning GROBID.
 
 #### Classic python and Virtualenv
 
@@ -82,7 +82,7 @@ Indicate the GROBID model that should use a Deep Learning implementation in the
 
 The default Deep Learning architecture is `BidLSTM_CRF`, which is the best sequence labelling RNN architecture (basically a slightly revised version of [(Lample et al., 2016)](https://arxiv.org/abs/1603.01360) with Glove embeddings). However for GROBID, an architecture also exploiting features (in particular layout features, which are not captured at all by the pretrained language models) gives usually better results and the prefered choise is `BidLSTM_CRF_FEATURES`. If you wish to use another architecture, you need to specify it in the same config file. 
 
-For instance to use a model integrating a fine-tuned transformer, you can select a `BERT_CRF` fine-tuned model (basically the transformer layers with CRF as final activation layer) and indicate in the field `transformer` the name of the transformer model in the [Hugging Face transformers Hub](https://huggingface.co/models) to be use to instanciate the transformer layer, typically [allenai/scibert_scivocab_cased](https://huggingface.co/allenai/scibert_scivocab_cased) for `SciBERT` in the case of scientific articles:
+For instance to use a model integrating a fine-tuned transformer, you can select a `BERT_CRF` fine-tuned model (basically the transformer layers with CRF as final activation layer) and indicate in the field `transformer` the name of the transformer model in the [Hugging Face transformers Hub](https://huggingface.co/models) to be used to instantiate the transformer layer, typically [allenai/scibert_scivocab_cased](https://huggingface.co/allenai/scibert_scivocab_cased) for `SciBERT` in the case of scientific articles:
 
 ```yaml
   models:
@@ -105,7 +105,7 @@ Normally by setting the Python environment path in the config file (e.g. `python
 
 <span>4.</span> Install [JEP](https://github.com/ninia/jep) manually and preferably globally (outside a virtual env. and not under `~/.local/lib/python3.*/site-packages/`). 
 
-We provide an installation script for Linux under `grobid-home/scripts`. This script should be launched from grobid root directory (`grobid/`), e.g.:
+We provide an installation script for Linux under `grobid-home/scripts`. This script should be launched from GROBID root directory (`grobid/`), e.g.:
 
 ```shell
 ./grobid-home/scripts/install_jep_lib.sh 

doc/End-to-end-evaluation.md

@@ -6,7 +6,7 @@ For an end-to-end evaluation, covering the whole extraction process from the par
 
 * against JATS-encoded (NLM) articles, such as [PubMed Central](http://www.ncbi.nlm.nih.gov/pmc), [bioRxiv](https://www.biorxiv.org), [PLOS](https://plos.org/ ) or [eLife](https://elifesciences.org/ ). For example, PubMed Central provides both PDF and fulltext XML files in the [NLM](http://www.ncbi.nlm.nih.gov/pmc/pmcdoc/tagging-guidelines/article/style.html) format. Keeping in mind some limits described bellow, it is possible to estimate the ability of Grobid to extract and normalize the content of the PDF documents for matching the quality of the NLM file. bioRxiv is used in Grobid to evaluate more precisely performance on preprint articles. 
 
-* against TEI documents produced by [Pub2TEI](https://github.com/kermitt2/Pub2TEI). Pub2TEI is a set of XSLT that permit to tranform various _native_ XML publishers (including Elsevier, Wiley, Springer, etc. XML formats) into a common TEI format. This TEI format can be used as groundtruth structure information for evaluating GROBID output, keeping in mind some limits described bellow. 
+* against TEI documents produced by [Pub2TEI](https://github.com/kermitt2/Pub2TEI). Pub2TEI is a set of XSLT that permit to tranform various _native_ XML publishers (including Elsevier, Wiley, Springer, etc. XML formats) into a common TEI format. This TEI format can be used as ground-truth structure information for evaluating GROBID output, keeping in mind some limits described bellow. 
 
 For actual benchmarks, see the [Benchmarking page](Benchmarking.md). We describe below the datasets and how to run the benchmarks.  
 

doc/Frequently-asked-questions.md

@@ -87,9 +87,9 @@ release the aggregate under a license that prohibits users from exercising right
 each program's individual license would grant them.
 ```
 
-For convenience it is no problem to ship the pdfalto executables with GROBID - same as a docker image which ships typically a mixture of GPL and Apache/MIT stuff calling each others like crazy and much more "deeply" than in our case.
+For convenience, it is no problem to ship the pdfalto executables with GROBID - same as a docker image which ships typically a mixture of GPL and Apache/MIT stuff calling each others like crazy and much more "deeply" than in our case.
 
-Finally as the two source codes are shipped in different repo with clear licensing information, exercising the rights that each program's individual license grants them is fully respected.
+Finally, as the two source codes are shipped in different repo with clear licensing information, exercising the rights that each program's individual license grants them is fully respected.
 
 The only possible restriction would be: