Mahmoud Mabrouk
Mahmoud Mabrouk
Currently only the outputs for spans are shown. From [SyncLinear.com](https://synclinear.com) | [AGE-154](https://linear.app/agenta/issue/AGE-154/[fea]-show-the-inputs-for-spans)
Creating an app from code. If using the line *model_1*=ag.GroupedMultipleChoiceParam( *default*="gpt-3.5-turbo", *choices*=get_all_supported_llm_models()), We get the warning when running the app from shell (or in the lambda/docker logs) From [SyncLinear.com](https://synclinear.com) |...
To reproduce: - Create a test set with a correct_answer - Run an evaluation The evaluation will fail and the result will not be seen Expected behavior: The evaluation will...
I think we should disable this for cloud and keep it only for oss. @aakrem Tracking issue for: - [x] https://github.com/Agenta-AI/agenta/security/code-scanning/26
In the playground, each time we add a datapoint to a test set, the default selected option in the multi-select down below is the "+Add new". For users who are...
In entity recognition tasks, users often need to evaluate multiple outputs. For example, if the task is to extract the author and date from a PDF, the user might create...
**Is your feature request related to a problem? Please describe.** Using llms without streaming is slow. **Describe the solution you'd like** Add a feature to stream outputs to the SDK...
We would like to enable the users to run the evaluation workflow from the CLI (without going through the UI). The UI in this case would be mostly used to...
We would like to the table for evaluation comparison view at the header of each output the data about the configuration that is used for that evaluation. Tasks: - [...
- [ ] #1045 (requires #1044) - [x] #1066 - [ ] #941 - [x] #940 - [ ] Create template with images - [ ] Create link to code...