Coding Week 9 Meeting
Attendees
- Gaurav Mishra
- Vasudev
- Ayush Bharadwaj
- Shreya Singh
- Kaushlendra Pratap Singh
- Omar AbdelSamea
Discussions
-
Clutter Removal Preview was given and discussed.
-
Implementation of the trained NER model was discussed using a CUSTOM training data set.
-
Next target for Final evaluations were discussed:
- Training the model using a custom dataset.
- Integrating the ALgorithm with FOSSOLOGY.
- Documenting everything and revamping it according to the integration.
-
CUSTOM [ORG] list extraction process was discussed.
Week 9 Progress
-
Clutter removal for the other three cases was implemented.
-
A custom training dataset was created using a list of names of organisations.
-
Spacy "en_core_web_sm" was trained on that custom dataset used.
-
Training was done in a way:
1- This training was more of an instantaneous one i.e. You train it for google, It will work only for google.
2- Need to figure out the way to keep the trained model while pushing the code. "nlp.to_disk()" internal property was used for saving the trained model to the local system.
-
More reduction and updation to code was done by removing redundancy and making the code more function-oriented.
-
Wiki has been Updated
Conclusion and Further Plans
Integration of the algorithm with the FOSSOLOGY