GSoC’22 Mid-term Evaluation Report
| Week | Dates | Main Tasks | Sub tasks completed | Issues Resolved | Blog | Type | Published URL |
|---|---|---|---|---|---|---|---|
| Week 1 | 15th June - 22nd June | Getting to know data and tech stack | [1.] Researching different data sourcs and data collection methods, [2.] Understanding tech stack , [3.] Data parsing for Medline | [1.] Beginning of GSoC Journey, [2.] Community Bonding | Blog | [1.] https://gli-mrunal.github.io/posts/Hola-GSoC-2022/ [2.] https://gli-mrunal.github.io/posts/GSoC-community-bonding/ | |
| Week 2 | 23rd June - 30th June | Medline Data acquisition through e-utility automation for computational neuroscience term | [1.] Researching Pubmed E-utility and data parsing, [2.] XML Pubmed data retrieve through automation | #6 Automate batch retrieval of PubMed MEDLINE xml data for computational+neuroscience term using PubMed E-utility https://github.com/nbdt-journal/automatic-reviewer-assignment/blob/parser_xml_to_csv/scripts/medline_parser/Medline_E-utility.ipynb | PubMed Data | development | https://gli-mrunal.github.io/posts/PubMed-Data/ |
| Week 3 | 1st July - 8th July | Client side protoype and wireframes in figma and fontend development in typescript in Nextjs and tailwindcss | [1.] create wireframes and proptypes for client in Figma, [2.] Program frontend for login and homePage in typesctipt using Nextjs and tailwindcss, [3.] Conenct to firebase for email authentication [4.] Wrap the entire web application by email authorization to only allow access to individuals logged into app to get reviewers’ recommendations. | #8 Client –> login implemented for email authentication using firebase https://github.com/nbdt-journal/automatic-reviewer-assignment/pull/8 | Let’s start building frontend with Firebase Database | development | https://gli-mrunal.github.io/posts/Frontend-UI/ |
| Week 4 | 9th July - 16th July | SciBERT Transformer model from HuggingFace for Neuroscience research abstract vectorization | [1.] Research HuggingFace BERT models [2.] Create embeddings [3.] save Embeddings as .npy for loading them later using aws sagemaker | #14 SciBERT Cosine Similarity - bioRxiv Neuroscience data https://github.com/nbdt-journal/automatic-reviewer-assignment/pull/14 | SciBERT Transformers :hugs: for Neuroscience | Research & Development | https://gli-mrunal.github.io/posts/SciBERT-Transformer-for-Neuroscience/ |
| Week 5 | 17th July - 24th July | Vector Similarity Search | Cosine Similarity Search code implementation | #14 SciBERT Cosine Similarity - bioRxiv Neuroscience data https://github.com/nbdt-journal/automatic-reviewer-assignment/pull/14 | Vector Similarity Search | Development | https://gli-mrunal.github.io/posts/Vector-Similarity-Search/ |
| Week 6 | 25th July - 31st July | Cleaning Code, Publishing jekyll blog documentation, New idea discussion | https://github.com/gli-mrunal/gli-mrunal.github.io/edit/master/_posts/2022-07-28-GSoC-Midterm-Evaluation-Report.md | Midterm Evaluation | Publishing Blog, Documentation | https://gli-mrunal.github.io/posts/GSoC-Midterm-Evaluation-Report/ |