Need help - C++ ONNX Failing

[kleingeo]

Describe the issue

I am trying to get a sklearn model from Python into C++. On the C++ side, I have no idea what I am doing. The Python can be seen below and runs without any issues.

import numpy as np
np.random.seed(0)
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC

iris = load_iris()
X, y = iris.data, iris.target
X = X.astype(np.float32)
X_train, X_test, y_train, y_test = train_test_split(X, y)
clr = SVC()
clr.fit(X_train, y_train)

# Convert into ONNX format.
from skl2onnx import to_onnx, convert_sklearn
from skl2onnx.common.data_types import FloatTensorType, Int64TensorType


model_name = "test_model.onnx"

initial_types = [("input", FloatTensorType(X.shape[1:]))]

onx = convert_sklearn(clr, 'test_svc',
                      initial_types=initial_types)

with open(model_name, "wb") as f:
    f.write(onx.SerializeToString())

import onnxruntime as rt

sess = rt.InferenceSession(model_name, providers=["CPUExecutionProvider"])
input_name = sess.get_inputs()[0].name
label_name = [sess.get_outputs()[0].name, sess.get_outputs()[1].name]
pred_onx = sess.run(label_name, {input_name: X_test[0]})
print(pred_onx)
>>> [array([2]), array([[-0.24219441,  1.1082064 ,  2.228405  ]], dtype=float32)]

The C++ code fails and I truly have no idea why.

#include <iostream>
#include <sstream>
#include <string>
#include <vector>
#include <onnxruntime_cxx_api.h>


int main(int argc) {

    std::array<float, 4> input_arr { 5.8f, 2.8f, 5.1f, 2.4f };
    std::array<int64_t, 1> input_shape{ 4, };
    std::vector<const char*> output_node_names = { "label", "probabilities" };

    std::vector<int> input_dims = { 4, };

    constexpr int32_t input_size = 4;

    const wchar_t* model_file = L"C:/test_sklearn_onnx/test_model.onnx";

    // onnxruntime setup
    Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "test_onnx");
    Ort::SessionOptions session_options;
    Ort::Session session = Ort::Session(env, model_file, session_options);

    auto memory_info = Ort::MemoryInfo::CreateCpu(OrtArenaAllocator, OrtMemTypeDefault);

    Ort::Value input_tensor = Ort::Value::CreateTensor<float>(memory_info,
        input_arr.data(), input_arr.size(),
        input_shape.data(), input_shape.size());

    std::vector<const char*> input_node_names = { "input" };

    auto output_tensors = session.Run(Ort::RunOptions{ nullptr }, input_node_names.data(), &input_tensor, 1, output_node_names.data(), 4);

}

To reproduce

Run the same code again.

Urgency

No response

Platform

Windows

OS Version

Windows 11

ONNX Runtime Installation

Released Package

ONNX Runtime Version or Commit ID

1.21.0

ONNX Runtime API

C++

Architecture

X64

Execution Provider

Default CPU

Execution Provider Library Version

No response