{
  "schemaVersion": 1,
  "lessonId": "optimization-01",
  "contentHash": "c9130b38274b14e717a5c3ce998deefe45cf5e21cc528c20c45872f0424a7771",
  "executedAt": "2026-10-20T10:45:35.408082+00:00",
  "environment": {
    "backend": "cpu",
    "deviceCount": 1,
    "python": "3.14.3",
    "jax": "0.9.2",
    "numpy": "2.4.4",
    "matplotlib": "3.10.8"
  },
  "stdout": "Predictions: [3. 0. 2.]\nLoss: 0.0\nPredictions: [3. 0. 2.]\nLoss: 0.0\nAnalytic gradients: [-1.3333334 -1.3333334] -2.0\nWrong residual shape: (3, 3)\nPASS: optimization-01\n",
  "notebookCells": [
    {
      "sourceHash": "42c775783751377ba2aab9d68b2feea3931a9de4fc5f8973dd94df663415b734",
      "outputs": []
    },
    {
      "sourceHash": "0bc1fcb4bca04817241cf5bacc66c153dc73a55001a1ac3606b60882a49d6b48",
      "outputs": []
    },
    {
      "sourceHash": "a651e36743bd73073b977a336a8ad7fcee0131a89c384fbb143594263bd10510",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Predictions: [3. 0. 2.]\n",
            "Loss: 0.0\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "867213f7b75f29ba81311e69929403d564d8bfb51ff3811f93540e2eb1b8aa91",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Predictions: [3. 0. 2.]\n",
            "Loss: 0.0\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "a1da12341c8ec13f76febac10f98b68570eba2877baad740750eae618b74e07a",
      "outputs": []
    },
    {
      "sourceHash": "25df6d662f9d3f0c4e984c6a201f79fa0cfd7a2633dc4b0773dfba6c2dd4ee6c",
      "outputs": [
        {
          "output_type": "display_data",
          "metadata": {},
          "data": {
            "image/png": "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",
            "text/plain": [
              "Executed Matplotlib figure"
            ]
          }
        }
      ]
    },
    {
      "sourceHash": "2063bf92ed8a139c55f86c6c1b4e159b672c52c3b45f8c6210c8f58c81633ac3",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Analytic gradients: [-1.3333334 -1.3333334] -2.0\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "58f1a7cf026e28a5bcbf53bdd0ebe15df35e61f0afce0e1b9340ab7153b8da5b",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Wrong residual shape: (3, 3)\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "f01578515c2fbe9b88aaccad68e8f599cd79d8f6542c473ce8486189aec8b077",
      "outputs": []
    },
    {
      "sourceHash": "042700c9241738ac87643d4fc21914c3c65c7023dcf063b97a8994a11cf040c0",
      "outputs": []
    },
    {
      "sourceHash": "a9069fe4bd9b785337e0506cd18caf77a8aa136474101d1a902414471996de6b",
      "outputs": []
    },
    {
      "sourceHash": "0c1f4dc5d0608e97d3a78f9a4d4d7afa3f82bde3399eb6415d7afaacb08b4343",
      "outputs": []
    },
    {
      "sourceHash": "9a947a712c6037178b96971251d620f344ea495008975fe8f9abcbc0bcf6933c",
      "outputs": []
    },
    {
      "sourceHash": "72846778954535c25828a21cb19a067b4ff06273b49bd00ccca81aca46bbd47e",
      "outputs": []
    },
    {
      "sourceHash": "fc53496fe8be7ea32fbada5ff14467954efbe0b7857f55f5fe1937e48bc1494b",
      "outputs": []
    }
  ],
  "scope": "Recorded instructor reference run; all script assertions and notebook code cells passed. No accelerator performance claim."
}