unit06-dl-tips/6.2-learning-rates/6.2-part5-1-scheduler-step.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d41a97fb-bbd4-4850-b4ca-d590c47b77af",
   "metadata": {},
   "source": [
    "# Unit 6. Essential Deep Learning Tips & Tricks\n",
    "\n",
    "# 6.2 Learning Rates and Learning Rate Schedulers\n",
    "\n",
    "## Part 5-0. A Baseline for Learning Rate Schedulers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "eec68953",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch    : 1.13.1\n",
      "lightning: 2.0.0rc1\n",
      "\n",
      "conda environment: dl-fundamentals\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -p torch,lightning --conda"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e80fef60",
   "metadata": {},
   "outputs": [],
   "source": [
    "import lightning as L\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import torchmetrics\n",
    "from lightning.pytorch.loggers import CSVLogger\n",
    "\n",
    "from shared_utilities import CustomDataModule, PyTorchMLP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "26845f11-4a0d-48f3-818a-22e4093b79ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "class LightningModel(L.LightningModule):\n",
    "    def __init__(self, model, learning_rate):\n",
    "        super().__init__()\n",
    "\n",
    "        self.learning_rate = learning_rate\n",
    "        self.model = model\n",
    "\n",
    "        self.save_hyperparameters(ignore=[\"model\"])\n",
    "\n",
    "        self.train_acc = torchmetrics.Accuracy(task=\"multiclass\", num_classes=2)\n",
    "        self.val_acc = torchmetrics.Accuracy(task=\"multiclass\", num_classes=2)\n",
    "        self.test_acc = torchmetrics.Accuracy(task=\"multiclass\", num_classes=2)\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.model(x)\n",
    "\n",
    "    def _shared_step(self, batch):\n",
    "        features, true_labels = batch\n",
    "        logits = self(features)\n",
    "\n",
    "        loss = F.cross_entropy(logits, true_labels)\n",
    "        predicted_labels = torch.argmax(logits, dim=1)\n",
    "        return loss, true_labels, predicted_labels\n",
    "\n",
    "    def training_step(self, batch, batch_idx):\n",
    "        loss, true_labels, predicted_labels = self._shared_step(batch)\n",
    "\n",
    "        self.log(\"train_loss\", loss)\n",
    "        self.train_acc(predicted_labels, true_labels)\n",
    "        self.log(\n",
    "            \"train_acc\", self.train_acc, prog_bar=True, on_epoch=True, on_step=False\n",
    "        )\n",
    "        return loss\n",
    "\n",
    "    def validation_step(self, batch, batch_idx):\n",
    "        loss, true_labels, predicted_labels = self._shared_step(batch)\n",
    "\n",
    "        self.log(\"val_loss\", loss, prog_bar=True)\n",
    "        self.val_acc(predicted_labels, true_labels)\n",
    "        self.log(\"val_acc\", self.val_acc, prog_bar=True)\n",
    "\n",
    "    def test_step(self, batch, batch_idx):\n",
    "        loss, true_labels, predicted_labels = self._shared_step(batch)\n",
    "        self.test_acc(predicted_labels, true_labels)\n",
    "        self.log(\"test_acc\", self.test_acc)\n",
    "\n",
    "    def configure_optimizers(self):\n",
    "        opt = torch.optim.SGD(self.parameters(), lr=self.learning_rate)\n",
    "        sch = torch.optim.lr_scheduler.StepLR(opt, step_size=10, gamma=0.5)\n",
    "\n",
    "        return [opt], [sch]\n",
    "        \n",
    "        \"\"\" \n",
    "        return {\n",
    "            \"optimizer\": opt,\n",
    "            \"lr_scheduler\": {\n",
    "                \"scheduler\": sch,\n",
    "                \"monitor\": \"train_loss\",\n",
    "                \"interval\": \"epoch\",\n",
    "                \"frequency\": 10\n",
    "            },\n",
    "        }\n",
    "        \"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "aa7bcf14-8046-4213-accf-544b67e74883",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Global seed set to 123\n",
      "GPU available: True (mps), used: False\n",
      "TPU available: False, using: 0 TPU cores\n",
      "IPU available: False, using: 0 IPUs\n",
      "HPU available: False, using: 0 HPUs\n",
      "/Users/sebastianraschka/Desktop/lightning-master/src/lightning/pytorch/trainer/setup.py:201: UserWarning: MPS available but not used. Set `accelerator` and `devices` using `Trainer(accelerator='mps', devices=1)`.\n",
      "  rank_zero_warn(\n"
     ]
    }
   ],
   "source": [
    "L.seed_everything(123)\n",
    "dm = CustomDataModule()\n",
    "\n",
    "pytorch_model = PyTorchMLP(num_features=100, num_classes=2)\n",
    "lightning_model = LightningModel(model=pytorch_model, learning_rate=0.1)\n",
    "\n",
    "trainer = L.Trainer(\n",
    "    max_epochs=100,\n",
    "    accelerator=\"cpu\",\n",
    "    devices=\"auto\",\n",
    "    logger=CSVLogger(save_dir=\"logs/\", name=\"my-model\"),\n",
    "    deterministic=True,\n",
    ")\n",
    "\n",
    "# Create a Tuner\n",
    "#tuner = Tuner(trainer)\n",
    "\n",
    "# finds learning rate automatically\n",
    "# sets hparams.lr or hparams.learning_rate to that learning rate\n",
    "#lr_finder = tuner.lr_find(lightning_model, datamodule=dm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9abe1027-3ec8-4597-9188-1068ee91df3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# update hparams of the model\n",
    "#lightning_model.hparams.learning_rate = lr_finder.suggestion()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "14d23b49-4c47-4bb1-bf6c-b87aabf0b22d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sebastianraschka/Desktop/lightning-master/src/lightning/pytorch/callbacks/model_checkpoint.py:612: UserWarning: Checkpoint directory logs/my-model/version_0/checkpoints exists and is not empty.\n",
      "  rank_zero_warn(f\"Checkpoint directory {dirpath} exists and is not empty.\")\n",
      "\n",
      "  | Name      | Type               | Params\n",
      "-------------------------------------------------\n",
      "0 | model     | PyTorchMLP         | 15.3 K\n",
      "1 | train_acc | MulticlassAccuracy | 0     \n",
      "2 | val_acc   | MulticlassAccuracy | 0     \n",
      "3 | test_acc  | MulticlassAccuracy | 0     \n",
      "-------------------------------------------------\n",
      "15.3 K    Trainable params\n",
      "0         Non-trainable params\n",
      "15.3 K    Total params\n",
      "0.061     Total estimated model params size (MB)\n",
      "/Users/sebastianraschka/Desktop/lightning-master/src/lightning/fabric/loggers/csv_logs.py:188: UserWarning: Experiment logs directory logs/my-model/version_0 exists and is not empty. Previous log files in this directory will be deleted when the new ones are saved!\n",
      "  rank_zero_warn(\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Sanity Checking: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sebastianraschka/Desktop/lightning-master/src/lightning/pytorch/trainer/connectors/data_connector.py:208: PossibleUserWarning: The dataloader, val_dataloader 0, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 10 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  rank_zero_warn(\n",
      "/Users/sebastianraschka/Desktop/lightning-master/src/lightning/pytorch/trainer/connectors/data_connector.py:208: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 10 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  rank_zero_warn(\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "94212428fb4a4c8f816f521a9f15cca8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Training: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "71ee9be4fffd4476b8b9dfb3e1b9639a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ce15ba1948b64dfea90c3a7f743161a4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "234ecadc28554e28a11dc102b67aa8ef",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b51ad2489d2648de8963d08056f5dcb3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9717e870c02f4991a0744d13260c0aaf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6e8c8fdb961a4a0eb41e18e93184296e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "42094a1f29ec4210857068db3a9c21cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4651185260de4286b44c9d1cdceff33d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88c0764c68d845af9db3b61af00a42ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7d766c72fb3c4ae88f2fa46bd88f9763",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d8ec306cd2ce4a9fa5aae86a95433576",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8583a2c7057143a39d1482f7df8a77ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5b137951b5be46e196f3a2c47211cbd9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "38323a559ef14d7b961e94ae61c16f00",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f4b390cba2b94cfe8ec1d7fd488869db",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a327c7ea4c1f4b7498ea40e64be08f84",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0790ce5d0a4c4bdcb67a0c05fc577e47",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6481b0a861ef4b27bcd35edb43b805c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "11afb68d8823464c8fe62eb8bdc558f1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3ceb9bb2da8e4a4a8e67053adfcf866d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "430aa6eda15d4264a3ada0b2f3067d93",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88123f2a455e4234bd8c8f06846fa6a7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1cb7f5ae20ec4371a8f862e096345708",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f9869229aba4689b35102ebca21bd70",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9f11cd804c44976bef866bc12950917",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b2450627408541229057beb6c8cc1e21",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c31537df4d78410b92f5f8dc7f4a29bf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3d8023f438bb4c74a30e5122ab3a948a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bc565076b6db49fa9899c35106316c1e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1547016664e247bb8ad2b7bd80c25ce6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5b8dbb7f67154b3f85fc0c2da197666e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8a016c3ccfaa4e159c1081ab50c8cb80",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5048ffc9cf1d49a8a44280b7f75770b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "01ba15ab4fc44c61b165c94e8b6de2ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f2a0fdbaed4a40d1a46dd200c58b4bd6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a2ec1ee57704b8db5f1c5dae5e06b66",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "27e13425e3f9493cb076823c982b9a1d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "34ac0fb38518487c9cf092693b2c9790",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f71fa52852ea467fb419938b6267ac51",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "08351e97f3d34d51b60829c6939bf548",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4104b8695fde45ed9bd423adcc3a260e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c2fe8bf7140d4cc0b92b24d2005e4337",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1609f478d0094706aa3c047169c78d3d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "10e631d8e1aa41f48dc9d3d9904333f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd3729d17fc1498e82c65a7f9c7be5c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "90b1accae7d54f0398e8d602f01f27b9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0b15137d858d45498d92566c6f5afda9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e0acabb29c91498b825081bbb2469004",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f8867264e864014b279d87ffa347ead",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eb6bd3406d2947978988afffe49980e7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5f0db49db177404588a184d2f60898af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e2252d069521459088459e38c951ab2b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3b75ddc5fba5443d90c6301f6345d329",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ab58e75afd2c4fa38debf83011cdd669",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Validation: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`Trainer.fit` stopped: `max_epochs=100` reached.\n"
     ]
    }
   ],
   "source": [
    "trainer.fit(model=lightning_model, datamodule=dm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3d4b7eb3-9f3c-4095-ad85-b37acf4557dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Plot\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "metrics = pd.read_csv(f\"{trainer.logger.log_dir}/metrics.csv\")\n",
    "\n",
    "aggreg_metrics = []\n",
    "agg_col = \"epoch\"\n",
    "for i, dfg in metrics.groupby(agg_col):\n",
    "    agg = dict(dfg.mean())\n",
    "    agg[agg_col] = i\n",
    "    aggreg_metrics.append(agg)\n",
    "\n",
    "df_metrics = pd.DataFrame(aggreg_metrics)\n",
    "df_metrics[[\"train_loss\", \"val_loss\"]].plot(\n",
    "    grid=True, legend=True, xlabel=\"Epoch\", ylabel=\"Loss\"\n",
    ")\n",
    "\n",
    "plt.ylim([0.0, 0.9])\n",
    "plt.savefig(\"step_loss.pdf\")\n",
    "\n",
    "df_metrics[[\"train_acc\", \"val_acc\"]].plot(\n",
    "    grid=True, legend=True, xlabel=\"Epoch\", ylabel=\"ACC\"\n",
    ")\n",
    "\n",
    "plt.ylim([0.7, 1.0])\n",
    "plt.savefig(\"step_acc.pdf\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "27a7e4b8-a243-4a2b-99a9-35b32dbf0398",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/sebastianraschka/Desktop/lightning-master/src/lightning/pytorch/trainer/connectors/data_connector.py:208: PossibleUserWarning: The dataloader, test_dataloader 0, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 10 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  rank_zero_warn(\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3aa4e21f4225462e8e2620e7e4f9cd5a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Testing: 0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\">        Test metric        </span>┃<span style=\"font-weight: bold\">       DataLoader 0        </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
       "│<span style=\"color: #008080; text-decoration-color: #008080\">         test_acc          </span>│<span style=\"color: #800080; text-decoration-color: #800080\">    0.8870000243186951     </span>│\n",
       "└───────────────────────────┴───────────────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1m       Test metric       \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      DataLoader 0       \u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
       "│\u001b[36m \u001b[0m\u001b[36m        test_acc         \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m   0.8870000243186951    \u001b[0m\u001b[35m \u001b[0m│\n",
       "└───────────────────────────┴───────────────────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "[{'test_acc': 0.8870000243186951}]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trainer.test(model=lightning_model, datamodule=dm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "59624e66-9aaa-4b8e-ac62-91c6fcf11583",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "opt = torch.optim.SGD(pytorch_model.parameters(), lr=0.1)\n",
    "sch = torch.optim.lr_scheduler.StepLR(opt, step_size=10, gamma=0.5)\n",
    "\n",
    "\n",
    "lrs = []\n",
    "max_epochs = 100\n",
    "\n",
    "for epoch in range(max_epochs):\n",
    "    opt.step()\n",
    "    lrs.append(opt.param_groups[0][\"lr\"])\n",
    "    sch.step()\n",
    "\n",
    "plt.plot(range(max_epochs), lrs)\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Learning rate')\n",
    "\n",
    "#plt.savefig('steps.pdf')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}