Fix embedding_4bit out variant

larryliu0820 · facebook-github-bot · commit ded3d257051d · 2024-04-19T08:52:06.000-07:00
Summary: In #3095 there's an issue with the embedding_4bit schema which causes mismatch between functional and out variant. P1217884556 Differential Revision: D56357762
diff --git a/exir/passes/_quant_patterns_and_replacements.py b/exir/passes/_quant_patterns_and_replacements.py
@@ -179,7 +179,7 @@ def embedding_byte_dtype_out_meta(
 
 quantized_decomposed_lib.define(
     "embedding_4bit.dtype(Tensor weight, Tensor weight_scales, Tensor? weight_zero_points, "
-    "int weight_quant_min, int weight_quant_max, Tensor indices, ScalarType? dtype=None) -> Tensor",
+    "int weight_quant_min, int weight_quant_max, Tensor indices, *, ScalarType? dtype=None) -> Tensor",
 )
 
 quantized_decomposed_lib.define(

Original file line number	Diff line number	Diff line change
`@@ -179,7 +179,7 @@ def embedding_byte_dtype_out_meta(`
`179`	`179`
`180`	`180`	`quantized_decomposed_lib.define(`
`181`	`181`	`"embedding_4bit.dtype(Tensor weight, Tensor weight_scales, Tensor? weight_zero_points, "`
`182`		`- "int weight_quant_min, int weight_quant_max, Tensor indices, ScalarType? dtype=None) -> Tensor",`
	`182`	`+ "int weight_quant_min, int weight_quant_max, Tensor indices, *, ScalarType? dtype=None) -> Tensor",`
`183`	`183`	`)`
`184`	`184`
`185`	`185`	`quantized_decomposed_lib.define(`