Query Classification on Steroids with BERT
Download as PPTX, PDF5 likes1,397 views
The document outlines a process for building a sophisticated deep learning model to classify user intent using machine learning, specifically leveraging the Ludwig framework and the ATIS dataset. It provides a step-by-step agenda for setting up the model, including utilizing Google Colab, TensorFlow, and ngrok to serve predictions via a public URL. Additionally, it emphasizes the importance of machine learning in discovering connections between business goals and customer behavior.
![!pip install ludwig
!pip install
ludwig[serve]
Installing
Ludwig](https://image.slidesharecdn.com/queryclassificationonsteroidswithbert-200402154128/85/Query-Classification-on-Steroids-with-BERT-17-320.jpg)
![Python code
snippet to test
the model
import pandas as pd
test_df = pd.read_csv("atis.test.csv",
index_col=0)
model =
LudwigModel.load("results/experiment_r
un/model")
predictions = model.predict(test_df)
test_df.reset_index().join(predictions)[["to
kens", "intent_predictions",
"slots_predictions"]]](https://image.slidesharecdn.com/queryclassificationonsteroidswithbert-200402154128/85/Query-Classification-on-Steroids-with-BERT-28-320.jpg)
![!curl -s
http://localhost:4040/a
pi/tunnels | python3 -c
"import sys, json;
print(json.load(sys.stdi
n)['tunnels'][0]['public_
url'])"
Extracting the
Ngrok public
URL](https://image.slidesharecdn.com/queryclassificationonsteroidswithbert-200402154128/85/Query-Classification-on-Steroids-with-BERT-35-320.jpg)
Query Classification on Steroids with BERT
- 1. Query Classification on Steroids with BERT Hamlet Batista @hamletbatista
- 5. “Machine learning can help you understand and predict intent in ways that simply aren’t possible manually. It can also help you find missed or unexpected connections between business goals and the habits of your key customer segments.” https://bit.ly/2UVebc1
- 7. Agenda • We will build a sophisticated deep learning model that can parse not just intent but also specific actions like the ones you give to Google Assistant, Siri and Alexa. • We will access the predictive model using a simple custom function in Google Sheets. • We will review the concepts that make this possible. • I will share resources you can use to learn more.
- 8. Our Toolkit https://colab.research.google.com/ We will type our commands in a Google Colab notebook.
- 9. Our Toolkit https://uber.github.io/ludwig/ Ludwig will help us build our deep learning model without writing code.
- 10. Our Toolkit https://github.com/howl-anderson/ATIS_dataset We will train our model using the ATIS dataset.
- 11. Our Toolkit https://ngrok.com/ Ngrok will help us expose the service URL inside our notebook.
- 12. Our Toolkit https://developers.google.com/apps-script We will code our Google Sheets function using Apps Script.
- 13. Setting up the notebook Downloading all required packages
- 16. Installing Tensorflow %tensorflow_version 1.x !pip install tensorflow- gpu==1.15.2 import tensorflow as tf; print(tf.__version__)
- 17. !pip install ludwig !pip install ludwig[serve] Installing Ludwig
- 19. Connect Google Drive from google.colab import drive drive.mount("/bert") %cd '/bert/My Drive/Writing/Query Classification on Steroids with BERT'
- 21. Preparing the training data Setting up the Ludwig configuration file
- 24. Building the predictive model Running Ludwig with our configuration file
- 25. !ludwig experiment --data_csv atis.train.csv -- model_definition_file model_definition.yaml Running Ludwig
- 27. Test the predictive model Running Ludwig serve
- 28. Python code snippet to test the model import pandas as pd test_df = pd.read_csv("atis.test.csv", index_col=0) model = LudwigModel.load("results/experiment_r un/model") predictions = model.predict(test_df) test_df.reset_index().join(predictions)[["to kens", "intent_predictions", "slots_predictions"]]
- 30. Serving the predictive model Running Ludwig serve
- 31. %%bash --bg nohup ludwig serve -m results/experiment_run /model > debug.log 2>&1 !tail debug.log Running Ludwig serve
- 33. !curl http://0.0.0.0:8000/pre dict -X POST -F 'tokens=BOS give me a flight from baltimore to newark EOS' Testing the service
- 34. %%bash --bg ./ngrok http 8000 2> ngrok.log !tail ngrok.log Running Ngrok
- 35. !curl -s http://localhost:4040/a pi/tunnels | python3 -c "import sys, json; print(json.load(sys.stdi n)['tunnels'][0]['public_ url'])" Extracting the Ngrok public URL
- 36. !curl https://e769db6e.ngro k.io/predict -X POST -F 'tokens=BOS explain meal codes sd d EOS' Testing the public service
- 37. Our journey so far 1.We built a travel specific intent prediction model 2.We can access the model using a temporary public URL 3.We need to access the URL from Apps Script and Google Sheets
- 38. Accessing the predictive model from Apps Script Creating a Google Sheets Custom Function
- 44. Testing in Google Sheets Calling our custom function fetchPrediction
- 47. Testing in Google Sheets
- 48. Where is BERT?
- 52. Python and deep learning resources 1. https://www.searchenginejournal.c om/author/hamlet-batista/ 1. Automated Intent Classification Using Deep Learning in Google Sheets 2. Automated Intent Classification Using Deep Learning 3. Automated Intent Classification Using Deep Learning (Part 2) 2. https://www.jcchouinard.com/pyth on-for-seo/ 1. Python for SEO: Complete Guide (in 5 Chapters) 3. https://moz.com/blog/intro-to- python 1. Intro to Python - Whiteboard Friday