Skip to content

A lightweight yet powerful audio-to-MIDI converter with pitch bend detection

Move from model-type to model-serialization

David Rubinstein3ce2617b2e94
diff --git a/README.md b/README.md
index 6357ba8..5155e59 100644
--- a/README.md
+++ b/README.md
@@ -94,7 +94,7 @@ Optionally, you may append any of the following flags to your prediction command
- `--save-model-outputs` to additionally save raw model outputs as an NPZ file
- `--save-note-events` to additionally save the predicted note events as a CSV file
-If you want to use a non-default model type (e.g., use CoreML instead of TF), use the `--model-type` argument. The CLI will change the loaded model to the type you prefer.
+If you want to use a non-default model type (e.g., use CoreML instead of TF), use the `--model-serialization` argument. The CLI will change the loaded model to the type you prefer.
To discover more parameter control, run:
```bash
diff --git a/basic_pitch/predict.py b/basic_pitch/predict.py
index 725af8b..d4b40df 100644
--- a/basic_pitch/predict.py
+++ b/basic_pitch/predict.py
@@ -50,7 +50,7 @@ def main() -> None:
"[tensorflow, coreml, tensorflow-lite, onnx]",
)
parser.add_argument(
- "--model-type",
+ "--model-serialization",
type=str,
choices=["tf", "coreml", "tflite", "onnx"],
help="If used, --model-path is ignored and instead the model serialization type" "specified is used.",
@@ -159,8 +159,8 @@ def main() -> None:
for audio_path in audio_path_list:
verify_input_path(audio_path)
- if args.model_type:
- model = Model(build_icassp_2022_model_path(FilenameSuffix[args.model_type]))
+ if args.model_serialization:
+ model = Model(build_icassp_2022_model_path(FilenameSuffix[args.model_serialization]))
else:
model = Model(args.model_path)