Update ORT format v5 change docs to cover limited backwards compatibility in 1.14. (#14413)
This commit is contained in:
Родитель
85d7e9c596
Коммит
3bc092b1ea
|
@ -1,12 +1,27 @@
|
|||
# ORT Format Update in 1.13
|
||||
|
||||
In ONNX Runtime 1.13, there was a breaking change to the
|
||||
[ORT format](https://onnxruntime.ai/docs/reference/ort-format-models.html) in order to enable additional execution
|
||||
providers with statically registered kernels in a minimal build.
|
||||
[ORT format](https://onnxruntime.ai/docs/reference/ort-format-models.html) (version 5) in order to enable additional
|
||||
execution providers with statically registered kernels in a minimal build.
|
||||
More details can be found [here](../onnxruntime/core/flatbuffers/schema/README.md#version-5).
|
||||
|
||||
Unfortunately, this means that any older models (prior to ORT format version 5) will no longer work with ONNX Runtime
|
||||
1.13 or later and must be re-converted.
|
||||
## Backwards Compatibility
|
||||
|
||||
### ONNX Runtime 1.13
|
||||
Any older models (prior to ORT format version 5) will no longer work with ONNX Runtime 1.13 and must be re-converted.
|
||||
|
||||
### ONNX Runtime 1.14+
|
||||
ONNX Runtime 1.14+ provides limited backwards compatibility for loading older models (prior to ORT format version 5).
|
||||
- In a full build, older models may be loaded but any saved runtime optimizations will be ignored.
|
||||
- In a minimal build, older models cannot be loaded.
|
||||
|
||||
An older model may be re-converted.
|
||||
|
||||
It is also possible to load an older ORT format model in a full build and then save it back out as an ORT format model.
|
||||
This process may be used to upgrade an ORT format model. However, any saved runtime optimizations from the older model
|
||||
will be ignored.
|
||||
|
||||
## Re-converting an ORT format model
|
||||
Please refer
|
||||
[here](https://onnxruntime.ai/docs/reference/ort-format-models.html#convert-onnx-models-to-ort-format) for instructions
|
||||
on how to convert an ONNX model to ORT format.
|
||||
|
|
|
@ -50,10 +50,10 @@ Support for storing `graph_doc_string` field in Model (ORT FlatBuffers format).
|
|||
Update kernel def hashing to not depend on ordering of type constraint types (NOT BACKWARDS COMPATIBLE).
|
||||
|
||||
## Version 5
|
||||
Deprecate kernel def hashes and add KernelTypeStrResolver info to replace them (NOT BACKWARDS COMPATIBLE).
|
||||
The change to the ORT format itself is not backwards compatibility-breaking, but ORT does not provide backwards
|
||||
Remove kernel def hashes and add KernelTypeStrResolver info to replace them (LIMITED BACKWARDS COMPATIBILITY).
|
||||
The change to the ORT format itself is not backwards compatibility-breaking, but ORT provides limited backwards
|
||||
compatibility for processing older models with missing KernelTypeStrResolver info.
|
||||
|
||||
The motivation for this update is to support additional execution providers with statically registered kernels.
|
||||
The original approach of using kernel def hashes is not so extensible as it requires the execution provider providing
|
||||
The original approach of using kernel def hashes was not so extensible as it required the execution provider providing
|
||||
hashes to be enabled at model conversion time.
|
||||
|
|
|
@ -1015,7 +1015,7 @@ Status InferenceSession::LoadOrtModelWithLoader(std::function<Status()> load_ort
|
|||
// TODO This change was introduced in 1.13. Remove this note a few releases later, e.g., 1.15.
|
||||
constexpr auto* kOrtFormatVersion5BreakingChangeNote =
|
||||
"This build doesn't support ORT format models older than version 5. "
|
||||
"See: https://github.com/microsoft/onnxruntime/blob/rel-1.13.0/docs/ORT_Format_Update_in_1.13.md";
|
||||
"See: https://github.com/microsoft/onnxruntime/blob/rel-1.14.0/docs/ORT_Format_Update_in_1.13.md";
|
||||
|
||||
ORT_RETURN_IF(!is_supported,
|
||||
"The ORT format model version [", fbs_ort_model_version->string_view(),
|
||||
|
|
Загрузка…
Ссылка в новой задаче