Untitled3.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1-SyK7Hu3aPA",
        "outputId": "be535df4-7f9b-41f0-ad3d-3968e80c02e5"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "First output value: 0.7920\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0847  pr:0.7910 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0836  pr:0.7891 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0819  pr:0.7862 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0797  pr:0.7823 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0769  pr:0.7773 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0736  pr:0.7713 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0697  pr:0.7641 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0654  pr:0.7557 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0605  pr:0.7460 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0552  pr:0.7350 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0495  pr:0.7226 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0436  pr:0.7087 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0374  pr:0.6933 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0311  pr:0.6764 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0250  pr:0.6580 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0191  pr:0.6382 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0137  pr:0.6169 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0089  pr:0.5944 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0050  pr:0.5708 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0021  pr:0.5463 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0005  pr:0.5213 - target:0.5000\n",
            "\n",
            "\n",
            "X y W =>>>  tensor([-1., -2.]) tensor([ 2., -3., -3.])\n",
            "X x W =>>  tensor(1.)\n",
            "current loss: 0.0000  pr:0.4960 - target:0.5000\n",
            "\n",
            "\n",
            "Last output value: 0.4960\n"
          ]
        }
      ],
      "source": [
        "import torch\n",
        "from untitled2 import Perceptron\n",
        "\n",
        "def main():\n",
        "  pytorch_model = Perceptron()\n",
        "  pytorch_model.load_state_dict(torch.load('model.pt'))\n",
        "  pytorch_model.eval()\n",
        "  dummy_input = torch.zeros(1,2)\n",
        "  torch.onnx.export(pytorch_model, dummy_input, 'model.onnx', verbose=True)\n",
        "\n",
        "if __name__ == '__main__':\n",
        "  main()"
      ]
    }
  ]
}