haoline 0.3.0__py3-none-any.whl

haoline/tests/__init__.py

@@ -0,0 +1,4 @@
+# Copyright (c) 2025 HaoLine Contributors
+# SPDX-License-Identifier: MIT
+
+"""Tests for HaoLine module."""

haoline/tests/conftest.py

@@ -0,0 +1,123 @@
+# Copyright (c) 2025 HaoLine Contributors
+# SPDX-License-Identifier: MIT
+
+"""
+Pytest configuration and shared fixtures for HaoLine tests.
+"""
+
+from __future__ import annotations
+
+import logging
+import tempfile
+from pathlib import Path
+
+import numpy as np
+import onnx
+import pytest
+from onnx import TensorProto, helper
+
+
+@pytest.fixture
+def logger():
+    """Create a test logger."""
+    return logging.getLogger("haoline.test")
+
+
+@pytest.fixture
+def simple_conv_model():
+    """Create a simple Conv model for testing."""
+    X = helper.make_tensor_value_info("X", TensorProto.FLOAT, [1, 3, 224, 224])
+    W = helper.make_tensor(
+        "W",
+        TensorProto.FLOAT,
+        [64, 3, 7, 7],
+        np.random.randn(64, 3, 7, 7).astype(np.float32).flatten().tolist(),
+    )
+    Y = helper.make_tensor_value_info("Y", TensorProto.FLOAT, [1, 64, 218, 218])
+
+    conv = helper.make_node("Conv", ["X", "W"], ["Y"], kernel_shape=[7, 7])
+
+    graph = helper.make_graph([conv], "simple_conv", [X], [Y], [W])
+    model = helper.make_model(graph, opset_imports=[helper.make_opsetid("", 17)])
+
+    with tempfile.NamedTemporaryFile(suffix=".onnx", delete=False) as f:
+        onnx.save(model, f.name)
+        yield Path(f.name)
+
+    # Cleanup
+    Path(f.name).unlink(missing_ok=True)
+
+
+@pytest.fixture
+def conv_bn_relu_model():
+    """Create a Conv-BN-ReLU model for testing."""
+    X = helper.make_tensor_value_info("X", TensorProto.FLOAT, [1, 3, 224, 224])
+    W = helper.make_tensor(
+        "W",
+        TensorProto.FLOAT,
+        [64, 3, 7, 7],
+        np.random.randn(64, 3, 7, 7).astype(np.float32).flatten().tolist(),
+    )
+    scale = helper.make_tensor(
+        "scale", TensorProto.FLOAT, [64], np.ones(64, dtype=np.float32).tolist()
+    )
+    bias = helper.make_tensor(
+        "bias", TensorProto.FLOAT, [64], np.zeros(64, dtype=np.float32).tolist()
+    )
+    mean = helper.make_tensor(
+        "mean", TensorProto.FLOAT, [64], np.zeros(64, dtype=np.float32).tolist()
+    )
+    var = helper.make_tensor("var", TensorProto.FLOAT, [64], np.ones(64, dtype=np.float32).tolist())
+    Y = helper.make_tensor_value_info("Y", TensorProto.FLOAT, [1, 64, 112, 112])
+
+    conv = helper.make_node(
+        "Conv",
+        ["X", "W"],
+        ["conv_out"],
+        kernel_shape=[7, 7],
+        strides=[2, 2],
+        pads=[3, 3, 3, 3],
+    )
+    bn = helper.make_node(
+        "BatchNormalization",
+        ["conv_out", "scale", "bias", "mean", "var"],
+        ["bn_out"],
+    )
+    relu = helper.make_node("Relu", ["bn_out"], ["Y"])
+
+    graph = helper.make_graph(
+        [conv, bn, relu], "conv_bn_relu", [X], [Y], [W, scale, bias, mean, var]
+    )
+    model = helper.make_model(graph, opset_imports=[helper.make_opsetid("", 17)])
+
+    with tempfile.NamedTemporaryFile(suffix=".onnx", delete=False) as f:
+        onnx.save(model, f.name)
+        yield Path(f.name)
+
+    # Cleanup
+    Path(f.name).unlink(missing_ok=True)
+
+
+@pytest.fixture
+def matmul_model():
+    """Create a simple MatMul model for testing."""
+    X = helper.make_tensor_value_info("X", TensorProto.FLOAT, [1, 512])
+    W = helper.make_tensor(
+        "W",
+        TensorProto.FLOAT,
+        [512, 1000],
+        np.random.randn(512, 1000).astype(np.float32).flatten().tolist(),
+    )
+    Y = helper.make_tensor_value_info("Y", TensorProto.FLOAT, [1, 1000])
+
+    matmul = helper.make_node("MatMul", ["X", "W"], ["Y"])
+
+    graph = helper.make_graph([matmul], "matmul", [X], [Y], [W])
+    model = helper.make_model(graph, opset_imports=[helper.make_opsetid("", 17)])
+
+    with tempfile.NamedTemporaryFile(suffix=".onnx", delete=False) as f:
+        onnx.save(model, f.name)
+        yield Path(f.name)
+
+    # Cleanup
+    Path(f.name).unlink(missing_ok=True)