dbg

2021-12-09 18:08:16 +08:00 · 2021-12-09 18:08:16 +08:00 · 7c6567cc6b
parent 6c5d1ebc28
commit 7c6567cc6b
1 changed files with 4 additions and 4 deletions
--- a/ppcls/engine/slim/quant.py
+++ b/ppcls/engine/slim/quant.py
@ -46,10 +46,10 @@ def quantize_model(config, model):
        assert config["Slim"]["quant"]["name"].lower(
        ) == 'pact', 'Only PACT quantization method is supported now'
        QUANT_CONFIG["activation_preprocess_type"] = "PACT"
-        model.quanted = QAT(config=QUANT_CONFIG)
-        model.quanted.quantize_model(model)
+        model.quanter = QAT(config=QUANT_CONFIG)
+        model.quanter.quantize(model)
        logger.info("QAT model summary:")
        paddle.summary(model, (1, 3, 224, 224))
    else:
-        model.quanted = None
-    return model.quanted
+        model.quanter = None
+    return