diff --git a/tests/spd_forward_solve_notebook.ipynb b/tests/spd_forward_solve_notebook.ipynb index b6eb702..6b4a747 100644 --- a/tests/spd_forward_solve_notebook.ipynb +++ b/tests/spd_forward_solve_notebook.ipynb @@ -4,7 +4,7 @@ "metadata": { "colab": { "provenance": [], - "authorship_tag": "ABX9TyPlHXxhQkC8kGTNfxN/oKe/", + "authorship_tag": "ABX9TyN0JFpiKDeXK9tTd/a7xn9o", "include_colab_link": true }, "kernelspec": { @@ -47,7 +47,7 @@ "base_uri": "https://localhost:8080/" }, "id": "3hHTa-xCjIQv", - "outputId": "482cc3c9-43e2-451b-af42-a745f88b528e" + "outputId": "cf255d05-b0ab-4a79-9765-1ed23fbff784" }, "outputs": [ { @@ -59,8 +59,8 @@ "Running on cuda\n", "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n", "Collecting git+https://github.com/cai4cai/torchsparsegradutils\n", - " Cloning https://github.com/cai4cai/torchsparsegradutils to /tmp/pip-req-build-6zkqcw6n\n", - " Running command git clone -q https://github.com/cai4cai/torchsparsegradutils /tmp/pip-req-build-6zkqcw6n\n", + " Cloning https://github.com/cai4cai/torchsparsegradutils to /tmp/pip-req-build-plwawj9f\n", + " Running command git clone -q https://github.com/cai4cai/torchsparsegradutils /tmp/pip-req-build-plwawj9f\n", " Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n", " Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n", " Preparing wheel metadata ... \u001b[?25l\u001b[?25hdone\n", @@ -144,7 +144,7 @@ "base_uri": "https://localhost:8080/" }, "id": "0CzeubR_kUVA", - "outputId": "199b16e4-526f-4b44-c99e-be1c933e4458" + "outputId": "dc3ff1db-1425-4a82-cf54-0ad60f81963c" }, "execution_count": 2, "outputs": [ @@ -152,8 +152,6 @@ "output_type": "stream", "name": "stdout", "text": [ - "Downloading https://suitesparse-collection-website.herokuapp.com/MM/Rothberg/cfd2.tar.gz\n", - "untarring cfd2.tar.gz\n", "Loaded suitesparse matrix Rothberg/cfd2: type=, shape=(123440, 123440)\n", "Created random RHS with shape=(123440,)\n" ] @@ -198,7 +196,7 @@ { "cell_type": "code", "source": [ - "run_scipy = False\n", + "run_scipy = True\n", "if run_scipy:\n", " t, n = scipy_test(A_np_coo, b_np, scipy.sparse.linalg.spsolve, 'scipy.spsolve COO')\n", " t, n = scipy_test(A_np_csr, b_np, scipy.sparse.linalg.spsolve, 'scipy.spsolve CSR')\n", @@ -216,15 +214,30 @@ "base_uri": "https://localhost:8080/" }, "id": "4KaNyZpDogS4", - "outputId": "ab5a18a5-4f73-4b27-d4a5-b71350c68842" + "outputId": "34e0dc0b-392b-4485-ac55-1233dde26aaf" }, "execution_count": 4, "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.8/dist-packages/scipy/sparse/linalg/dsolve/linsolve.py:144: SparseEfficiencyWarning: spsolve requires A be CSC or CSR matrix format\n", + " warn('spsolve requires A be CSC or CSR matrix format',\n" + ] + }, { "output_type": "stream", "name": "stdout", "text": [ - "Skipping scipy tests\n" + "scipy.spsolve COO took 108.62 seconds - resnorm=2.15e-10\n", + "scipy.spsolve CSR took 98.29 seconds - resnorm=2.49e-10\n", + "scipy.cg COO took 72.09 seconds - resnorm=3.50e-03\n", + "scipy.cg CSR took 55.59 seconds - resnorm=3.50e-03\n", + "scipy.bicgstab COO took 112.65 seconds - resnorm=2.31e-03\n", + "scipy.bicgstab CSR took 88.79 seconds - resnorm=2.31e-03\n", + "scipy.minres COO took 0.54 seconds - resnorm=3.05e+01\n", + "scipy.minres CSR took 0.38 seconds - resnorm=3.05e+01\n" ] } ] @@ -250,10 +263,10 @@ ], "metadata": { "id": "DHqwjtVOrRyH", - "outputId": "20aa9e46-1e9d-4b55-b70c-88ff6eb1ccc5", "colab": { "base_uri": "https://localhost:8080/" - } + }, + "outputId": "307faf94-ee4a-4943-d483-ad084246ee0d" }, "execution_count": 5, "outputs": [ @@ -346,10 +359,10 @@ ], "metadata": { "id": "3HH-nGDSsIXy", - "outputId": "793408b1-955b-4f8b-b184-bf8f24ea188b", "colab": { "base_uri": "https://localhost:8080/" - } + }, + "outputId": "7a8dacff-efaa-416a-f6d5-9a87a2b84157" }, "execution_count": 7, "outputs": [ @@ -358,15 +371,15 @@ "name": "stdout", "text": [ "GPU memory allocated: 0.04Gb - max allocated: 0.04Gb\n", - "tsgu.sparse_generic_solve GPU CSR took 0.65 seconds - resnorm=1.33e+00\n", + "tsgu.sparse_generic_solve GPU CSR took 0.62 seconds - resnorm=1.93e+00\n", "GPU memory allocated: 0.04Gb - max allocated: 0.05Gb\n", - "tsgu.sparse_generic_solve minres GPU CSR took 0.48 seconds - resnorm=1.33e+00\n", + "tsgu.sparse_generic_solve minres GPU CSR took 0.46 seconds - resnorm=1.93e+00\n", "GPU memory allocated: 0.04Gb - max allocated: 0.05Gb\n", - "tsgu.sparse_generic_solve cg GPU CSR took 0.06 seconds - resnorm=2.39e+02\n", + "tsgu.sparse_generic_solve cg GPU CSR took 0.06 seconds - resnorm=2.43e+02\n", "GPU memory allocated: 0.04Gb - max allocated: 0.05Gb\n", - "tsgu.sparse_generic_solve bicgstab GPU CSR took 10.42 seconds - resnorm=2.15e-04\n", + "tsgu.sparse_generic_solve bicgstab GPU CSR took 9.35 seconds - resnorm=2.60e-04\n", "GPU memory allocated: 0.04Gb - max allocated: 0.05Gb\n", - "tsgu.sparse_solve_j4t cg GPU CSR took 8.32 seconds - resnorm=3.51e-03\n", + "tsgu.sparse_solve_j4t cg GPU CSR took 6.39 seconds - resnorm=2.74e-03\n", "GPU memory allocated: 0.04Gb - max allocated: 0.05Gb\n" ] }