ChatGPT AI Guide: Analyse Two कुर्ता Try-On Logs

By TryBuy Editorial

Use ChatGPT to compare two kurta try-on logs and identify which observations need repeating. The output should distinguish known changes in the outfit or setup from differences whose cause is unknown. It should not decide that the garment stretched, the wearer changed size or a seam is faulty simply because the second session felt different.

Prepare two dated logs

For each session, record the exact garment and selected size, underlayer, bottoms, waistband position and relevant accessories. Include the ordinary actions observed and the measurement methods used. Keep numerical values with their units and endpoints. If something was not recorded during the first session, label it unknown rather than reconstructing it from memory as a confirmed fact.

Add photographs only when they are relevant and appropriately used. State any change in camera or stance that affects the comparison. A differently angled mirror photo should not be treated as a controlled measurement. Separate what the wearer felt from what an observer could see, and keep preferences such as a desired hem position distinct from a claim that the garment changed.

Copy the two-log analysis prompt

Compare my labelled try-on logs without diagnosing causes. Return a table of observation, session-one record, session-two record, known changed variable, unresolved variable and smallest useful repeat check. Preserve garment identifiers, measurements and units. Do not infer shrinkage, stretching, body change or construction faults. Mark comparisons invalid when the method or setup differs. Keep successful repeated observations as well as unresolved ones, and end with no more than three focused next actions.

Ask the model to cite the log field behind each known variable. A statement such as accessory changed should be traceable to an entry, not inferred because it would explain the result conveniently.

Filled fictional example

Imagine session one records comfortable sleeve placement during typing. Session two reports the cuff catching at the wrist and explicitly notes that a watch was added. The fictional output can identify the watch as a known setup difference and propose repeating the ordinary action without it. It cannot conclude that the watch is the only cause or that the sleeve requires alteration.

Another row may show the same comfortable shoulder observation in both sessions. Keep that stable evidence in the table. A helpful comparison does not turn every entry into a problem merely because one issue prompted the analysis. The final decision can preserve what works while isolating the unanswered point.

Revise the analysis carefully

• “Show which proposed explanations are hypotheses rather than supplied facts, and move them out of the known-change column.”
• “Remove repeat checks unrelated to the decision I identified.”
• “Update only the affected rows after this new observation; retain the original session logs unchanged.”

If a revised answer changes a source measurement or turns an unrecorded underlayer into a confirmed one, correct the record handling before using the result. A well-formatted comparison is not reliable when its inputs have been quietly rewritten.

Review and run the repeat check yourself

Verify that the selected checks are ordinary, relevant and practical for you. Repeat the observation under the stated setup, then add your result as a new log rather than overwriting an earlier session. If a concern remains, the next action may be a specific seller question or an in-person professional assessment. The model cannot inspect hidden construction or establish a physical cause.

Use the original practical guide for keeping the two-session fit record and the men’s kurtas collection for exact product context when comparing another design. Your deliverable is a changed-variable ledger and a short repeat plan. It should make your evidence easier to interpret without converting a personal observation into a universal claim about a kurta or fabric.

For general instructions and follow-up techniques, see official OpenAI prompting guidance.