Test a Pretrained Model

min read - Updated on Oct 28, 2026

You can test a pretrained model before you add it to a Report, to see which labels it gives your own customer feedback. Every pretrained model has a page with a test console and two ways to test: Text classifies a sentence or a paragraph you type and shows its predictions, and Document classifies a file of feedback by creating a Report. A few texts from your own data are enough to see whether the model's labels fit your feedback, before it reads thousands of records.

Getting Ready

Pretrained models and their test console are available in every package, including the free Starter package.

Open a Model's Page

Each pretrained model has a page in the Kimola app and a page on kimola.com. Both show the model's description, a test console and its Labels, each with a description of what it covers. Read the labels before you test: their descriptions are what you judge a prediction against.

In the app, you open a model's page from the Classifiers step of the Report wizard. Under PRE-BUILT MODELS, click the icon next to a model's name, and its page opens in a new tab. At the bottom of every model page, More on Gallery lists the other pretrained models, so you can move from one model's page to the next. The app's console has both Document and Text, and it opens on Document.

On kimola.com, every model has a page under kimola.com/pretrained-ai-models, with a console under See it in Action! that tests text only. List Pretrained Models lists all 15 models with a link to each page.

Test with Text

A text test classifies one text at a time and shows the result in a moment. It creates nothing in your account. Use it to try the kinds of feedback you will analyze: a typical complaint, a text that mentions two topics, a short text, a text with mixed feelings.

Step 1: Open the Text Tab

In the app, click Text in the test console. The console splits into two boxes: the text on the left, and Predictions on the right. When the tab opens, Kimola loads a sample text and classifies it, so you see the model at work right away. Load a sample text under the box loads another one.

On kimola.com, the console already holds a sample text. Click Analyze! to classify it, or Load a sample text to load another one.

Step 2: Type or Paste a Text

Replace the sample with a piece of feedback of your own. In the app, you don't need to press a button: once you stop typing, the model classifies the text and its predictions appear on the right. Each change you make is classified again, so you can rewrite a sentence and see how the labels follow it. On kimola.com, click Analyze! after each change.

Step 3: Read the Predictions

The predictions are a short list in a structured format, and what each entry holds depends on the Multi-Label switch at the top right of the console (Multi-label on kimola.com).

With Multi-Label on, the model gives every label the text matches, and each label comes with the sentiment of the text toward it, Positive or Negative. This is how a multi-label Report labels feedback: one review can mention several topics, and each of them gets its own sentiment.

With Multi-Label off, the model gives the single label that fits the text best, with its probability, a number between 0 and 1: the closer it is to 1, the more certain the model is. This is how a single-label Report labels feedback, under one main theme. See Multi-label vs. Single-label Classification.

The switch is on when the console opens. In the app, turning it off or on classifies the same text again; on kimola.com, click Analyze! again. The Sentiment Classifier and the Entity Extractor are the exceptions: on their pages in the app, the switch is off and can't be turned on.

Example

A bank tests the Banking Feedback Classifier with an app review: "The app logged me out three times while I was sending money, but the branch staff fixed my card limit in five minutes." With Multi-Label on, the predictions are Mobile & Online Banking and Transactions & Payments, both Negative, and Customer Service and Loans & Credit, both Positive. With Multi-Label off, the prediction is Mobile & Online Banking alone, with a probability of 0.94.

Test with a Document

A document test classifies a whole file of feedback with the model and is available in the app. It creates a Report, so you see the model's labels, their sentiment and their share of the feedback on real data, in the same screens you will use later. It goes through the same steps as creating a Report with pretrained models.

Step 1: Upload Your File

In the test console, keep Document selected. Drag and drop your file onto the upload area, or click it to choose a file. The console accepts one file in .xls, .xlsx, .csv or .tsv format, smaller than 30 MB. Kimola checks the file's format, size and structure, then opens the Report wizard.

Step 2: Choose the Columns

The Sheets & Columns step shows a preview of your file. Choose the Text Column, the column that holds the feedback to analyze. The Url Column and the Date Column are optional: a Date Column lets you see how the labels move over time.

Then click Advanced. Continue skips the Classifiers step and creates the Report with the Automatic Classifier and the Sentiment Classifier, the two models every new Report starts with; the model you are testing isn't among them.

Note

Click Advanced, not Continue: with Continue, the Report is classified without the model you are testing.

Step 3: Choose the Model and the Structure

In Classifiers, the Automatic Classifier and the Sentiment Classifier are selected under PRE-BUILT MODELS, as in every new Report. Open PRE-BUILT MODELS (if your account has custom models, the step opens on CUSTOM MODELS), type the model's name in the filter and select it. Clear the Automatic Classifier and the Sentiment Classifier to see the model's labels alone, or keep them to compare its labels with the themes the Automatic Classifier finds on its own.

In Structure, Enable Multi-label Classification decides whether a piece of feedback can get several labels, as the Multi-Label switch does in a text test. It is on by default.

Step 4: Choose the Dimensions

In Dimensions, choose the research questions you want asked of every piece of feedback, such as Pain Points or Unmet Needs, or continue without any: Dimensions are optional. The ones you choose are listed under My List. See What is a Dimension?

Step 5: Review and Create the Report

The Review step shows your choices. Report Title is optional, and Report Language is the language the Report is written in. Dataset shows the number of rows and the file they come from, followed by Required Query. Click Create Report! to start the analysis.

A document test uses queries like any Report: one analyzed row, multiplied by the number of models and the number of Dimensions (queries = rows × models × Dimensions). Every model you keep in the Classifiers step and every Dimension you choose adds to the queries of each row. See How to Calculate Queries.

Note

A document test uses queries like any Report: queries = rows × models × Dimensions.

The new Report appears in Reports like any other. Once the analysis is complete, you read the model's labels on its Overview and in its other screens.

A small file is enough for a test: a few hundred rows your team knows well use few queries and can be checked row by row.

Tip

Test with a few hundred rows your team knows well: they use few queries and can be checked row by row.

After the Test

When the predictions match what your team expects, add the model to your Reports in the Classifiers step of the Report wizard. See Create a Report Using Pretrained Models.

When they don't, look for a pattern rather than a single wrong answer. If many texts get no fitting label, or the labels are too broad for your questions, try a model closer to your feedback: an industry classifier for your sector, or the Consumer Feedback Classifier for feedback about a product, a service or a brand in general. A pretrained model's labels are set by Kimola and can't be changed. When you need labels of your own, create a custom model, for example from the Report your test created (see Create a Custom Model with a Report), and test it in its own console (see Test a Custom Model).

Was this article helpful?

Tell us about your thoughts and experiences regarding the article.