Created
January 29, 2020 20:35
-
-
Save petewarden/180ea281bbe09a6a7cbc8b2558097fc7 to your computer and use it in GitHub Desktop.
Example of using the new Flatbuffers Python API to modify a TensorFlow Lite file
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def StripTfliteFile(tflite_input, tflite_output): | |
| """Given 'tflite_input`, write stripped version, 'tflite_output'.""" | |
| if not os.path.exists(tflite_input): | |
| raise RuntimeError("Invalid filename %r" % tflite_input) | |
| with open(tflite_input, "rb") as file_handle: | |
| file_data = bytearray(file_handle.read()) | |
| model_obj = schema_fb.Model.GetRootAsModel(file_data, 0) | |
| model = schema_fb.ModelT.InitFromObj(model_obj) | |
| model.description = "" | |
| for subgraph in model.subgraphs: | |
| subgraph.name = "" | |
| for tensor in subgraph.tensors: | |
| tensor.name = "" | |
| builder = flatbuffers.Builder(1024) | |
| model_offset = model.Pack(builder) | |
| builder.Finish(model_offset) | |
| model_data = builder.Output() | |
| with open(tflite_output, "wb") as out_file: | |
| out_file.write(model_data) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hmmm, Netron says "RangeError: Offset is outside the bounds of the DataView ". Stripped model doesn't load in TFLite aswell: "ERROR: The model is not a valid Flatbuffer buffer"