{"id":12792,"date":"2022-02-16T08:34:12","date_gmt":"2022-02-16T16:34:12","guid":{"rendered":"https:\/\/www.couchbase.com\/blog\/?p=12792"},"modified":"2025-06-13T21:25:06","modified_gmt":"2025-06-14T04:25:06","slug":"leveraging-your-ml-models-with-couchbase-analytics-user-defined-functions-udf","status":"publish","type":"post","link":"https:\/\/www.couchbase.com\/blog\/ko\/leveraging-your-ml-models-with-couchbase-analytics-user-defined-functions-udf\/","title":{"rendered":"\uce74\uc6b0\uce58\ubca0\uc774\uc2a4 \uc560\ub110\ub9ac\ud2f1\uc2a4 \uc0ac\uc6a9\uc790 \uc815\uc758 \ud568\uc218(UDF)\ub85c ML \ubaa8\ub378 \ud65c\uc6a9\ud558\uae30"},"content":{"rendered":"

With Couchbase 7.0 you’re now capable of allowing integration of Python UDFs with Couchbase Analytics. In Part 1 of this blog series<\/a>, we covered the essentials for getting Couchbase and the Analytics for Machine Learning (ML) set up.<\/span><\/p>\n

ML has radically transformed the ways in which organizations understand their customer\u2019s needs. Advanced analytics domains like predictive analytics (customer churn, customer sentiment, etc.) and financial modeling are growing more reliant on processing data at scale, near real-time and extracting valuable insights from it.<\/span><\/p>\n

To help our customers obtain analytical insights in real time, we have created a seamless pipeline from Python-based machine learning models to Couchbase Analytics. In this post, I walk through the following steps to show you how to apply external algorithms to data that is resident in Couchbase.<\/span><\/p>\n

Six steps for applying ML models against your NoSQL data:<\/span><\/h4>\n
    \n
  1. Train the Model<\/span><\/li>\n
  2. Codify the Model<\/span><\/li>\n
  3. Package and Deploy the code<\/span><\/li>\n
  4. Import the needed data for this project<\/span><\/li>\n
  5. Writing the UDF<\/span><\/li>\n
  6. Using the UDF in your instance for CB (DP Mode)<\/span><\/li>\n<\/ol>\n

    Before we dive in, let’s find a dataset that will make an interesting demonstration of the capabilities we are building. There are movie reviews on several different websites, but to have a holistic understanding of critic reviews there is no better place than Rotten Tomatoes. This website allows you to compare the ratings given by regular users (audience score) and the ratings or reviews given by critics (tomatometer<\/em>) who are certified members of various writing guilds or film critic associations.<\/span><\/p>\n

    The two datasets used for this blog can be found on <\/span>kaggle.com<\/span><\/a>. These are rather large files so a link to them is provided so you can download them when you follow along.<\/span><\/p>\n

    In the <\/span>movies<\/span><\/em> dataset, each record represents a movie available on Rotten Tomatoes, with the URL used for scraping the movie title, description, genres, duration, director, actors, user ratings, and critic ratings. In the <\/span>movie_reviews<\/span><\/em> dataset, each record represents a critic review published on Rotten Tomatoes, with the URL used for scraping the critic name, review publication, date, score, and content.<\/span><\/p>\n

    Training the ML model<\/span><\/h4>\n

    Before you start exploring the power of integration between ML and NoSQL, you\u2019ll need to develop and train a machine learning model in Python.<\/span><\/p>\n

    For the purposes of this blog, we will use a simple logistic regression model that utilizes the scikit-learn library. At its core, the model takes in data and analyzes their sentiments regarding movie reviews. You can follow along with the steps outlined below or you can download all of the necessary files from our <\/span>GitHub repo<\/span><\/a>.\u00a0<\/span><\/p>\n

    For this blog, we are using an open-source predictive algorithm on the movie reviews dataset to determine sentiment, i.e., to determine if the reviews are positive or negative for a given movie. In today\u2019s examples, we have already trained the model using a subset from the file you downloaded earlier. For the purposes of this blog we utilize a CSV (comma-separated values) file to import our data.\u00a0<\/span><\/p>\n

    Below is a sample of the code for the model itself:<\/span><\/p>\n

    \"\"<\/p>\n

    The entirety of the code sample can be found in the <\/span>GitHub repository.<\/span><\/a><\/p>\n

    When you run the model Python script shown above, you should get the following result:<\/span><\/p>\n

    \"\"<\/p>\n

    You can read more about scikit-learn metrics like precision, recall, f1-score, and support <\/span>here<\/span><\/a>. We now have a functional, well-performing machine learning model fully trained in Python.<\/span><\/p>\n

    Creating a Python library<\/span><\/h4>\n

    In order to reference the machine learning model, you will need to create a Python library. Below is the library for this particular example:<\/span><\/p>\n

    \"\"<\/p>\n

    There are two primary components of the library:<\/span><\/p>\n

    Model constructor<\/b>\u2014This constructor creates a file called <\/span>sentiment_model<\/span><\/em> in the <\/span>pipelines<\/em> <\/span>folder of our Jupyter environment.<\/span><\/p>\n

    getSentiment method<\/b>\u2014This method predicts the customer sentiment associated with the parameter (or argument) passed into it.<\/span><\/p>\n

    Save the file as <\/span>sentiment.py<\/span><\/i> within the <\/span>pipelines<\/span><\/i> folder with the file <\/span>sentiment_model<\/span><\/i>.<\/span><\/p>\n

    Packaging and deploying the library<\/span><\/h4>\n

    This is a critical step in what will come next\u2014unlocking the power of Python user-defined functions! Please pay attention to detail as it is more syntax-dependent than any of the others. Be sure to read the appropriate documentation closely. Follow the link <\/span>to learn more about User-Defined Functions<\/span><\/a>.<\/span><\/p>\n

    To package the model and library we created in the previous steps, we will use the shiv utility. If shiv is not already installed, use the command <\/span>pip install shiv<\/span><\/i> (or <\/span><\/i>pip3 install shiv<\/span><\/i> depending on your environment). Additionally, if you\u2019re interested in reading the documentation for this command line utility, you can find it <\/span>here<\/span><\/a>.\u00a0<\/span><\/p>\n

    Steps to package the model:<\/b><\/p>\n