Add FP64 precision support in intel_npu (openvinotoolkit#24284)

CailiWuIntel · ArtemySkrebkov · web-flow · commit 20d91fd6a824 · 2024-06-18T16:40:49.000Z
### Details:
Add FP64 precision pipeline support in intel_npu。

### Tickets:
 - E-121878

Co-authored-by: Artemy Skrebkov &lt;artemy.skrebkov@intel.com&gt;
diff --git a/src/plugins/intel_npu/src/backend/src/zero_infer_request.cpp b/src/plugins/intel_npu/src/backend/src/zero_infer_request.cpp
@@ -409,9 +409,11 @@ void ZeroInferRequest::check_network_precision(const ov::element::Type_t precisi
         break;
     case ov::element::Type_t::i64:
         break;
+    case ov::element::Type_t::f64:
+        break;
     default:
         OPENVINO_THROW("Unsupported tensor precision: " + ov::element::Type(precision).get_type_name() +
-                       "! Supported precisions: FP32, FP16, U4, I4, U8, I8, U16, I16, U32, I32, U64, I64");
+                       "! Supported precisions: FP32, FP16, U4, I4, U8, I8, U16, I16, U32, I32, U64, I64, FP64");
     }
 }
 

Original file line number	Diff line number	Diff line change
`@@ -409,9 +409,11 @@ void ZeroInferRequest::check_network_precision(const ov::element::Type_t precisi`
`409`	`409`	`break;`
`410`	`410`	`case ov::element::Type_t::i64:`
`411`	`411`	`break;`
	`412`	`+ case ov::element::Type_t::f64:`
	`413`	`+ break;`
`412`	`414`	`default:`
`413`	`415`	`OPENVINO_THROW("Unsupported tensor precision: " + ov::element::Type(precision).get_type_name() +`
`414`		`- "! Supported precisions: FP32, FP16, U4, I4, U8, I8, U16, I16, U32, I32, U64, I64");`
	`416`	`+ "! Supported precisions: FP32, FP16, U4, I4, U8, I8, U16, I16, U32, I32, U64, I64, FP64");`
`415`	`417`	`}`
`416`	`418`	`}`
`417`	`419`