You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Generates an `ETRecord` from the given objects, serializes it and saves it to the given path.
204
214
The objects that will be serialized to an `ETRecord` are all the graph modules present
205
-
in the `export_modules` dict, the graph module present in the edge dialect program object,
215
+
in the `extra_recorded_export_modules` dict, the graph module present in the edge dialect program object,
206
216
and also the graph module present in the ExecuTorch program object, which
207
217
is the closest graph module representation of what is eventually run on the device.
208
218
In addition to all the graph modules, we also serialize the program buffer, which the users
@@ -213,7 +223,8 @@ def generate_etrecord(
213
223
et_record: Path to where the `ETRecord` file will be saved to.
214
224
edge_dialect_program: `EdgeProgramManager` for this model returned by the call to to_edge()
215
225
executorch_program: The ExecuTorch program for this model returned by the call to `to_executorch()` or the `BundledProgram` of this model
216
-
export_modules [Optional]: **Should be ignored by OSS users**. A dictionary of graph modules with the key being the user provided name and the
226
+
exported_program: Optional graph module for this model returned by the call to `torch.export` from nn.Module.
227
+
extra_recorded_export_modules [Optional]: **Should be ignored by OSS users**. A dictionary of graph modules with the key being the user provided name and the
217
228
value being the corresponding exported module. The exported graph modules can be either the
218
229
output of `torch.export()` or `exir.to_edge()`.
219
230
@@ -229,15 +240,32 @@ def generate_etrecord(
229
240
# is an etrecord when it's used later in the Developer Tools.
0 commit comments