Noms plus clairs : snippets/, scripts de test numérotés, verifier_ollama.py
- base_connaissances/ → snippets/ (corpus/ → textes/, instructions.md → consignes_redaction.md, decoupage.py → fichiers.py, importer.py → envoyer.py) ; python3 -m snippets.envoyer. - instance-de-test/ : 1_demarrer.sh, 2_creer_boite_et_cle.py, 3_brancher_ia.py, 4_essayer_brouillons.py, voir_empreintes.sh, connexion_admin.py. - ollama/verifier.py → ollama/verifier_ollama.py ; tests, tâches VS Code et docs alignés. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
1 parent
3e0b5d03a3
commit
10ed887417
142 files changed
+89
-85
No files matched your search
@@ -0,0 +1,10 @@
|
||||
# What is the Error Threshold?
|
||||
|
||||
The Error Threshold defines the recognition-error value above which a measurement is
|
||||
considered invalid.
|
||||
By default, the threshold is 0.7. This mechanism is primarily used to reject measurements when a
|
||||
target is obstructed or its appearance has changed too much for the measurement to be
|
||||
considered reliable.
|
||||
The threshold can be overridden with a user-defined value when required by the application.
|
||||
|
||||
Source : AIMsight FAQ, rév. 1.5
|
||||
Reference in new issue
Block a user