HN user

thepythiccoder1

60 karma
Posts26
Comments2
View on HN
github.com 6y ago

A Transformer-Based Approach for Source ACL 2020

thepythiccoder1
2pts0
medium.com 6y ago

Advanced Tips for Production ML

thepythiccoder1
1pts0
github.com 7y ago

Show HN: Homoglyph Attack Prevention with OCR

thepythiccoder1
5pts0
github.com 7y ago

Show HN: Object Detection for Complex Classification Scenarios

thepythiccoder1
2pts0
medium.com 7y ago

Making Petabyte Scale Data Actionable Part 1

thepythiccoder1
1pts0
towardsdatascience.com 7y ago

Introducing Semantic Structure to Neural NLP

thepythiccoder1
3pts0
towardsdatascience.com 7y ago

4 Common Pitfalls of Neural NLP

thepythiccoder1
2pts0
medium.com 7y ago

NLP Primer from Bag of Words to BERT

thepythiccoder1
3pts0
towardsdatascience.com 7y ago

Top Milestones in Neural NLP

thepythiccoder1
2pts0
medium.com 7y ago

An Overview of Neural NLP Milestones

thepythiccoder1
4pts0
towardsdatascience.com 7y ago

Silver, Gold and Electrum: 3 Data Techniques for Multi-Task Deep Learning

thepythiccoder1
1pts0
medium.com 7y ago

5 steps to more interactive deep learning

thepythiccoder1
2pts0
github.com 7y ago

Show HN: TagAnomaly anomaly detection made easy

thepythiccoder1
1pts0
www.usatoday.com 8y ago

Israel hopes to be 4th country on the moon by end of the year

thepythiccoder1
1pts0
medium.com 8y ago

One Click Tensorflow.js

thepythiccoder1
5pts0
medium.com 8y ago

What does it mean to be the first Microsoft CDA in Israel?

thepythiccoder1
1pts0
github.com 8y ago

Show HN: One Click Pose Estimation

thepythiccoder1
84pts16
medium.com 8y ago

8mins to Production ML

thepythiccoder1
2pts0
www.microsoft.com 8y ago

Azure Event Hub Ingestion at Scale with Python and Kubernetes

thepythiccoder1
2pts0
www.microsoft.com 8y ago

Python Event Hub processing at scale

thepythiccoder1
2pts0
github.com 8y ago

Train FB Detectron on your own data using Microsoft VoTT

thepythiccoder1
2pts0
github.com 8y ago

Show HN: Scalable Python Eventhub Client

thepythiccoder1
1pts0
github.com 8y ago

Show HN: Pytorch RNN autobatcher

thepythiccoder1
2pts0
github.com 8y ago

Microsofts Object Detection Tool has a new home

thepythiccoder1
4pts0
github.com 8y ago

Microsoft releases open source tool for object detection

thepythiccoder1
4pts0
www.microsoft.com 8y ago

How to simply use object detection for complex image classification scenarios

thepythiccoder1
2pts0

Hmm the issue is that there might be some correlation between the syntactic and lexical similarity and the actual subject matter the authors are talking about.

I've been working on a similar project.

Three things that I would suggest would be

- add documents definitively written by the authors (Maddison, John jay, etc) from outside the federalist papers to your train set.

http://oll.libertyfund.org/titles/jay-the-correspondence-and...

http://www.gutenberg.org/ebooks/author/14, etc

- Add another feature which looks at the frequency of the function (closed class words) such as articles, prepositions etc these are very stylistic and hard for an author to control, they are also independent of the content, this is a classical feature in forensics.

- Add a distractor case to your train and validation set I.e documents written by a non federalist such as Thomas Jefferson and confirm that they don't get clustered into one of the other federalist authors.

If you have questions feel free to tweet me it seems like a cool project @pythiccoder

I'm not sure I agree, live coding, when polished and done right grounds a presentation. Without a live coding element "technical" presentations can quickly devolve into marketing where grandiose claims are often made that are devoid any merit. I think the real gap is most people underestimate how much preparation time must be invested in a live coding demo.