Run experiments
Iterate on prompts and experiment against curated datasets without code
Last updated
Was this helpful?
Iterate on prompts and experiment against curated datasets without code
Last updated
Was this helpful?
Running experiments help you systematically test and validate changes in your LLM applications against a curated dataset. You can create a dataset, test a prompt, and save it as an experiment without code using Arize.
If you are looking to run experiments using our SDK, use the link below.
Load the dataset you created into prompt playground, and run it to see your results. Once you've finished the run, you can save it as an experiment to track your changes.
Download this and upload it into the UI. The CSV must have an id
column. See example CSV below:
If you want to use your application data, you can also .
Each prompt iteration is stored separately and Arize makes is very easy to compare experiment results against each other. You can quickly iterate on models, prompts, LLM parameters, and tools and have confidence that your changes will improve performance.
Create a task to run evaluations on your experiment results. We will run the evaluator task in the background as soon as you create the task.