*   The new backend is used for 8 bits full integer post-training
            quantization
        *   The new backend removes the redundant rescales and fixes some bugs
            (shared weight/bias, extremely small scales, etc)
        *   Set `experimental_new_quantizer` in tf.lite.TFLiteConverter to False
            to disable this change
*   `tf.keras`
    *   `tf.keras.metrics.AUC` now support logit predictions.

* Remove gender bias in docs for handling errors. PR [#780](https://github.com/tiangolo/fastapi/pull/780). Original idea in PR [#761](https://github.com/tiangolo/fastapi/pull/761) by [@classywhetten](https://github.com/classywhetten).

AZJ,eACE,iBAAA,8BAGE,sBAAA,yCACE,MAAA,KACA,YAAA,EAAA,IAAA,EAAA,KAIJ,6BACE,iBAAA,qBACA,MAAA,KAZJ,kBACE,iBAAA,6BASA,gCACE,iBAAA,oBACA,MAAA,QAZJ,iBACE,iBAAA,6BAGE,wBAAA,2CACE,MAAA,KACA,YAAA,EAAA,IAAA,EAAA,KAIJ,+BACE,iBAAA,oBACA,MAAA,KAZJ,gBACE,iBAAA,+BASA,8BACE,iBAAA,sBACA,MAAA,QAZJ,eACE,iBAAA,4BAGE,sBAAA,yCACE,MAAA,KACA,YAAA,EAAA,IAAA,EAAA,KAIJ,6BACE,iBAAA,mBACA,MAAA,KAZJ,oBACE,iBAAA,8BAGE,2BAAA,8CACE,MAAA,KACA,YAAA,EAAA,IAAA,EAAA,KAIJ,kCACE,iBAAA,qBACA,MAAA,KAZJ,eACE,iBAAA,2BAGE,sBAAA,yCACE,MAAA,...