Made modification for joss review (issue #2)

README.md

@@ -154,11 +154,11 @@ HoverFast train data -o training_model -p /path/to/pytable_files/ -b 16 -n 20 -e
 ```
 
 ```
-docker run -it --gpus all -v /path/to/pytables/:/app petroslk/hoverfast:latest HoverFast train data -l training_metrics -p /app -b 16 -n 20 -e 100
+docker run -it --gpus all -v /path/to/pytables/:/app petroslk/hoverfast:latest HoverFast train data -o training_metrics -p /app -b 16 -n 20 -e 100
 ```
 
 ```
-singularity exec --nv hoverfast_latest.sif HoverFast train data -l training_metrics -p /path/to/pytables/ -b 16 -n 20 -e 100
+singularity exec --nv hoverfast_latest.sif HoverFast train data -o training_metrics -p /path/to/pytables/ -b 16 -n 20 -e 100
 ```
 
 ## Testing

paper/paper.bib

@@ -64,15 +64,17 @@ @article{paszke2017automatic
   author={Paszke, Adam and Gross, Sam and Chintala, Soumith and Chanan, Gregory and Yang, Edward and DeVito, Zachary and Lin, Zeming and Desmaison, Alban and Antiga, Luca and Lerer, Adam},
   year={2017}}
 
-@article{scikit-learn,
-  title={Scikit-learn: Machine Learning in Python},
-  author={Pedregosa, F. and Varoquaux, G. and Gramfort, A. and Michel, V.          and Thirion, B. and Grisel, O. and Blondel, M. and Prettenhofer, P.          and Weiss, R. and Dubourg, V. and Vanderplas, J. and Passos, A. and          Cournapeau, D. and Brucher, M. and Perrot, M. and Duchesnay, E.},
-  journal={Journal of Machine Learning Research}, 
-  volume={12}, 
-  pages={2825--2830}, 
-  year={2011}},
-  url={http://jmlr.org/papers/v12/pedregosa11a.html},
-  }
+@article{van_der_Walt_2014,
+   title={scikit-image: image processing in Python},
+   volume={2},
+   ISSN={2167-8359},
+   url={http://dx.doi.org/10.7717/peerj.453},
+   DOI={10.7717/peerj.453},
+   journal={PeerJ},
+   publisher={PeerJ},
+   author={van der Walt, Stéfan and Schönberger, Johannes L. and Nunez-Iglesias, Juan and Boulogne, François and Warner, Joshua D. and Yager, Neil and Gouillart, Emmanuelle and Yu, Tony},
+   year={2014},
+   month=jun, pages={e453} }
 
 @article{opencv_library,
   author = {Bradski, G.},

paper/paper.md

@@ -88,7 +88,7 @@ To facilitate the knowledge distillation process, HoverFast presents a training
 ## Inference
 HoverFast has a command-line interface (CLI) written in Python 3.11 and utilizes the PyTorch framework [@paszke2017automatic].  We replicated the structure of the HoverNet model as described by Graham et al [@graham2019hover] without the nuclear classification branch. For the backbone, we used a modified 940k parameter Multi-scale UNet [@su2021msu]  in place of HoverNet’s 33.6 million parameter ResNet50, yielding a reduction in model parameter count by a factor of 30 (see **Appendix 1**). 
 
-HoverFast’s post-processing pipeline was heavily optimized using scikit-learn’s [@scikit-learn] regionprops and watershed functions to effectively identify and split merged cells. To improve throughput after batch model inference, regions are processed in parallel using a “multi-worker, single writer” approach. This involves each worker independently (a) post-processing its assigned region, and then (b) generating nuclei polygon coordinates using OpenCV [@opencv_library], before (c) sending to the single writing process for saving as a QuPath [@bankhead2017qupath] compatible gzip-compressed JSON file. A Docker and Singularity container of HoverFast are provided.
+HoverFast’s post-processing pipeline was heavily optimized using scikit-image’s [@van_der_Walt_2014] regionprops and watershed functions to effectively identify and split merged cells. To improve throughput after batch model inference, regions are processed in parallel using a “multi-worker, single writer” approach. This involves each worker independently (a) post-processing its assigned region, and then (b) generating nuclei polygon coordinates using OpenCV [@opencv_library], before (c) sending to the single writing process for saving as a QuPath [@bankhead2017qupath] compatible gzip-compressed JSON file. A Docker and Singularity container of HoverFast are provided.
 
 
 ## Training