# AI Lab

<figure><img src="https://lh7-rt.googleusercontent.com/slidesz/AGV_vUfLWfmitXvc_5f3cQvaYHyRwBnAceIfrTGTRAQWhUhvz9SU-UMfaM6XSQsILpAPYUJWbY85flVdY4lSbXoHgFQcjcvO8woL8RQE--KbEFBsd03vd3b9EzlaMWRh1T0HTdnlRDgx6IC5LRhYPsl3rlaiRUDsyjfj=s2048?key=0Vuv-rJx-rGDStWDzAwBDw" alt=""><figcaption><p>AI Lab page</p></figcaption></figure>

Limitations for the Knowledge Management (KM) System:

1. **CSV File Validation**: Imported **.CSV files** must undergo validation to ensure they meet the specified header requirements and data integrity before being accepted into the system.
2. **Single Knowledge Base**: The system does not support the use of assistants with **more than one Knowledge Base** at a time, limiting users to managing knowledge within a single base for streamlined operations.

<figure><img src="https://lh7-rt.googleusercontent.com/slidesz/AGV_vUcbeKGhc3Bi5LypI6BChvIp-L0SAJdK7m6-ewrA5p9zRAAD3mmg4emaBNuHzxFegZ19vKUWYiPI2wzM5Mx8KLQ_uq47FvzX9fwb7rlphvC6ncwghuO3Rsrprp7GsXgdocXgosZyntrAb9k6KTz7SWI_2A6dBCbN=s2048?key=0Vuv-rJx-rGDStWDzAwBDw" alt="" width="563"><figcaption><p>AI Lab CSV Format</p></figcaption></figure>

In order to upload a test case file, you need to have these columns as follow:

**Input Column (\* Mandatory)**

* **no.\***: Test case number (Format: case-\[case]-\[step])
* **question\***: The test case prepared to assess Generative AI.
* **expected chunk id:** The knowledge chunk expected to be used for answering the question.
* **expected answer:** The answer anticipated from Generative AI.

<figure><img src="https://lh7-rt.googleusercontent.com/slidesz/AGV_vUesUAuBBoMCpNw_DijMVKCqdo0hDbU7veLL3m7SkG59bMSn-AY0jAv4N76Jrmjvg6fmlxSGKvFQnCxXWP-av9nclY8yqTDcUrHffWBiZ_EJT9Y5K38PMMyu6qb9uyngZu7xkeyS_0uuROk7QJ7_0oU3zZT6cJgY=s2048?key=0Vuv-rJx-rGDStWDzAwBDw" alt=""><figcaption><p>AI Lab Option</p></figcaption></figure>

The action of testing AI Lab consists of 3 options, as follows:

1. **Search KM Answer:** Tests the selected chunk to generate an answer and automatically verifies document match between the expected and actual chunk.&#x20;
2. **Generate Answer with Actual KM:** Tests the entire process of answer generation by the Generative AI.
3. **Generate Answer with Expected KM:** Tests only generating answer with using expected chunk from input file.

<figure><img src="https://lh7-rt.googleusercontent.com/slidesz/AGV_vUflUkm9xr0biyturY6ESZKMzvyvxaGAEqyd8h2lZgwfGXq-lapMLAWD7UCFv9pgcOPN20hckpssuFKNPsmYP1ECbNds36Zj8t5fYb0PTioAhN_EXbAG6vKN2i8QXm7YNJt-6oph8bbZiyYMLzVIBw1iTCCK5VHI=s2048?key=0Vuv-rJx-rGDStWDzAwBDw" alt=""><figcaption><p>AI Lab Result</p></figcaption></figure>

After finish executing, the test history will appear. User can click download as a zip file, consists of:

1. **Summary:** Show the overall summary of test case.<br>

   <figure><img src="https://lh7-rt.googleusercontent.com/slidesz/AGV_vUegqQgzhlGyZeC24pObXNKEUqhEZ15FserGdUWTSUN-PMg6Nd4we0ZZ5fa_cZPy9hPkxTK2HKsTC-HAG-mX9rVHYXCOsmSbsRTibKSNsWhx0OlShmY6cY_5GHUUAd7DFeSHLFxUi6L6VnDTmvVozB7O2rEAKl29=s2048?key=0Vuv-rJx-rGDStWDzAwBDw" alt="" width="375"><figcaption><p>Sample of downloaded Summary file,</p></figcaption></figure>
2. **Test-cases:** Show the result of each test case.<br>

   <figure><img src="https://lh7-rt.googleusercontent.com/slidesz/AGV_vUeQw3OULLVpKpBq-mFtsm01ZZm8ckZngrUv5IOHDQKbG624-H1KbxyporeJOzNC2Op6vyDrbQe0QlmlnflDN1wc6sEfhJ0mguxK23b63pC99gqDi9DaLSUrZVe2wGLIMANl0uJJSfMdfz1Suyog2a17DYw4UQVj=s2048?key=0Vuv-rJx-rGDStWDzAwBDw" alt="" width="563"><figcaption><p>Sample of downloaded Test-cases file</p></figcaption></figure>

**Warning:**

The result will be displayed only once. Leaving the page will reset the test history and the test result.

<br>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.amitysolutions.com/amity-solutions/amity-bots-gen-ai-platform/ai-lab.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
