Fp16 Int8 Quantization

Cuantización de modelos ML a FP16/INT8 para reducir memoria y acelerar inferencia en el pipeline KYC

davidcastagnetoa bd610e7 2 files · 4.0 KB Updated

File contents

davidcastagnetoa/skills/tree/main/skills/fp16_int8_quantization commit bd610e7673

Frequently asked questions

npx skillmds@latest add davidcastagnetoa/fp16-int8-quantization