export RTDETR as coreML

[keisan1231]

Question

How can I use RTDETR coreML??

Additional

I am trying to use RTDETR in an iOS application.
Following this comment, I attempted to export the model in CoreML format, but I encountered the following error:

ValueError: In op, of type linear, named out_w, the named input `weight` must have the same data type as the named input `x`. However, weight has dtype fp32 whereas x has dtype int32.

code

from ultralytics import RTDETR

# Load your model
model = RTDETR('rtdetr-l.pt')

# Ensure model and inputs are in the same data type (e.g., float32)
model = model.float()

# Now try exporting again
model.export(format="coreml")

All detail log

``` [W NNPACK.cpp:64] Could not initialize NNPACK! Reason: Unsupported hardware. rt-detr-l summary: 494 layers, 32148140 parameters, 0 gradients, 103.8 GFLOPs

PyTorch: starting from 'rtdetr-l.pt' with input shape (1, 3, 640, 640) BCHW and output shape(s) (1, 300, 84) (63.4 MB)
scikit-learn version 1.5.1 is not supported. Minimum required version: 0.17. Maximum required version: 1.1.2. Disabling scikit-learn conversion API.
Torch version 2.2.2 has not been tested with coremltools. You may run into unexpected errors. Torch 2.2.0 is the most recent version that has been tested.

CoreML: starting export with coremltools 7.2...
Converting PyTorch Frontend ==> MIL Ops: 0%| | 0/2329 [00:00<?, ? ops/s]Saving value type of int64 into a builtin type of int32, might lose precision!
Saving value type of int64 into a builtin type of int32, might lose precision!
Converting PyTorch Frontend ==> MIL Ops: 25%|█████████▊ | 574/2329 [00:00<00:00, 5716.50 ops/s]Saving value type of float64 into a builtin type of fp32, might lose precision!

ERROR - converting 'matmul' op (located at: '11'):

Converting PyTorch Frontend ==> MIL Ops: 25%|██████████ | 584/2329 [00:00<00:00, 5540.22 ops/s]
CoreML: export failure ❌ 20.0s: In op, of type linear, named out_w, the named input weight must have the same data type as the named input x. However, weight has dtype fp32 whereas x has dtype int32.
Traceback (most recent call last):
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/ml_xcode_create.py", line 10, in
model.export(format="coreml")
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/ultralytics/engine/model.py", line 591, in export
return Exporter(overrides=args, _callbacks=self.callbacks)(model=self.model)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/ultralytics/engine/exporter.py", line 310, in call
f[4], _ = self.export_coreml()
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/ultralytics/engine/exporter.py", line 142, in outer_func
raise e
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/ultralytics/engine/exporter.py", line 137, in outer_func
f, model = inner_func(*args, **kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/ultralytics/engine/exporter.py", line 633, in export_coreml
ct_model = ct.convert(
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/_converters_entry.py", line 581, in convert
mlmodel = mil_convert(
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/converter.py", line 188, in mil_convert
return _mil_convert(model, convert_from, convert_to, ConverterRegistry, MLModel, compute_units, **kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/converter.py", line 212, in _mil_convert
proto, mil_program = mil_convert_to_proto(
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/converter.py", line 288, in mil_convert_to_proto
prog = frontend_converter(model, **kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/converter.py", line 108, in call
return load(*args, **kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/load.py", line 82, in load
return _perform_torch_convert(converter, debug)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/load.py", line 116, in _perform_torch_convert
prog = converter.convert()
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/converter.py", line 581, in convert
convert_nodes(self.context, self.graph)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/ops.py", line 86, in convert_nodes
raise e # re-raise exception
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/ops.py", line 81, in convert_nodes
convert_single_node(context, node)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/ops.py", line 134, in convert_single_node
add_op(context, node)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/frontend/torch/ops.py", line 879, in matmul
res = mb.linear(x=linear_x, weight=transposed_weight, name=node.name)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/mil/ops/registry.py", line 182, in add_op
return cls._add_op(op_cls_to_add, **kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/mil/builder.py", line 182, in _add_op
new_op = op_cls(**kwargs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/mil/operation.py", line 191, in init
self._validate_and_set_inputs(input_kv)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/mil/operation.py", line 504, in _validate_and_set_inputs
self.input_spec.validate_inputs(self.name, self.op_type, input_kvs)
File "/Users/yamaguchikeiichi/Downloads/smart-judge/ball_detect_coreml_create/testEnv/lib/python3.10/site-packages/coremltools/converters/mil/mil/input_type.py", line 137, in validate_inputs
raise ValueError(msg)
ValueError: In op, of type linear, named out_w, the named input weight must have the same data type as the named input x. However, weight has dtype fp32 whereas x has dtype int32.

</details>