Posts

• Semantic search with embedding representation from python

  Build and interact with a MS MARCO search application

  Dec 4, 2020

• Timing Vespa queries

  Disconsidering network and other non-search related costs

  Dec 2, 2020

• The Probabilistic Relevance Framework

  Working in progress ...

  Dec 1, 2020

• Fine-tuning a BERT model for search applications

  How to ensure training and serving encoding compatibility

  Nov 25, 2020

• Proposal for a simplified pyvespa API

  Simplification, embeddings and consistency

  Nov 17, 2020

• Fine-tuning a BERT model with transformers

  Setup a custom Dataset, fine-tune BERT with Transformers Trainer and export the model via ONNX.

  Nov 12, 2020

• Debug first iteration of BERT models

  Recall, positive and null NDCG and wrong model output issues.

  Nov 8, 2020

• Fine-tuning a BERT model for search applications with Optuna tuning

  Fine-tune BERT models using Optuna for hyperparameter tuning

  Nov 7, 2020

• Feed, get, update and delete Vespa data with pyvespa

  Nov 6, 2020

• How to evaluate Vespa ranking functions from python

  Using pyvespa to evaluate cord19 search application ranking functions currently in production.

  Nov 5, 2020

• Fine-tuning a BERT model for search applications

  Create training and serving compatible encodings, custom dataset and a basic setup based on the Transformers library.

  Oct 29, 2020

• How to connect and interact with search applications from python

  A pyvespa library overview. Connect, query, collect data and evaluate query models.

  Oct 28, 2020

• Sequential feature selection applied to ranking features

  Jun 19, 2020

• Reducing the number of initial features to select

  Can we use univariate feature prediction to eliminate useless features?

  Jun 11, 2020

• Best subset selection of Vespa ranking features

  May 27, 2020