tutorial/Tutorial9_Feature_Generation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Automatic Feature Generation for PPO\n",
        "This tutorial shows how automatic feature generation is conducted with OpenFE library."
      ],
      "metadata": {
        "id": "NYZ3GlhQfEPR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone -b 1.0.0 https://github.com/TradeMaster-NTU/TradeMaster.git"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JMChf7XWfkP3",
        "outputId": "5dce575d-f3e5-4a83-8ba0-858efa76eb1b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'TradeMaster'...\n",
            "remote: Enumerating objects: 6561, done.\u001b[K\n",
            "remote: Total 6561 (delta 0), reused 0 (delta 0), pack-reused 6561\u001b[K\n",
            "Receiving objects: 100% (6561/6561), 382.93 MiB | 22.83 MiB/s, done.\n",
            "Resolving deltas: 100% (3393/3393), done.\n",
            "Updating files: 100% (400/400), done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%cd '/content/TradeMaster'"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0R8DIwHaf7nc",
        "outputId": "a1ac1a62-db3e-4428-928b-415743d4bdcc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/TradeMaster\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -r requirements.txt"
      ],
      "metadata": {
        "id": "PvrgwhD3ge_l"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step1: Import Packages"
      ],
      "metadata": {
        "id": "DbuSrGMug5y5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "import sys\n",
        "from pathlib import Path\n",
        "import os\n",
        "import torch\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.model_selection import train_test_split\n",
        "from openfe import OpenFE, transform\n",
        "\n",
        "ROOT = os.path.dirname(os.path.abspath(\".\"))\n",
        "sys.path.append(ROOT)\n",
        "\n",
        "import argparse\n",
        "import os.path as osp\n",
        "from mmcv import Config\n",
        "from trademaster.utils import replace_cfg_vals\n",
        "from trademaster.nets.builder import build_net\n",
        "from trademaster.environments.builder import build_environment\n",
        "from trademaster.datasets.builder import build_dataset\n",
        "from trademaster.agents.builder import build_agent\n",
        "from trademaster.optimizers.builder import build_optimizer\n",
        "from trademaster.losses.builder import build_loss\n",
        "from trademaster.trainers.builder import build_trainer\n",
        "from trademaster.utils import plot\n",
        "from trademaster.utils import set_seed\n",
        "set_seed(2023)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8_SkBfP1g9yD",
        "outputId": "e712f8cf-eb9e-43cd-e4ea-72e4a4d58d8d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'\n",
            "2023-05-09 15:50:12,657\tINFO worker.py:973 -- Calling ray.init() again after it has already been called.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step2: Feature Generation"
      ],
      "metadata": {
        "id": "wfMWNIo7Nxuq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "path = 'data/portfolio_management/dj30/'\n",
        "df_train = pd.read_csv(path + \"train.csv\")\n",
        "df_valid = pd.read_csv(path + \"valid.csv\")\n",
        "df_test = pd.read_csv(path + \"test.csv\")\n",
        "tech_indicator_list = [\n",
        "    \"zopen\",\n",
        "    \"zhigh\",\n",
        "    \"zlow\",\n",
        "    \"zadjcp\",\n",
        "    \"zclose\",\n",
        "    \"zd_5\",\n",
        "    \"zd_10\",\n",
        "    \"zd_15\",\n",
        "    \"zd_20\",\n",
        "    \"zd_25\",\n",
        "    \"zd_30\"\n",
        "]\n",
        "\n",
        "train_x = pd.DataFrame(df_train[tech_indicator_list])\n",
        "valid_x = pd.DataFrame(df_valid[tech_indicator_list])\n",
        "test_x = pd.DataFrame(df_test[tech_indicator_list])\n",
        "\n",
        "X = pd.DataFrame(df_train[tech_indicator_list][:-1])\n",
        "Y = pd.DataFrame({\"close\": list(df_train['close'][1:])})\n",
        "\n",
        "n_jobs=4\n",
        "ofe = OpenFE()\n",
        "ofe.fit(data=X, label=Y, n_jobs=n_jobs)\n",
        "train_x, valid_x = transform(train_x, valid_x, ofe.new_features_list[:10], n_jobs=n_jobs) \n",
        "X, test_x = transform(X, test_x, ofe.new_features_list[:10], n_jobs=n_jobs)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qpaJXepGhv-a",
        "outputId": "f478e592-2604-4e7b-ca05-2f54722113c2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The number of candidate features is 418\n",
            "Start stage I selection.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:13<00:00,  1.20it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "35 same features have been deleted.\n",
            "Meet early-stopping in successive feature-wise halving.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:41<00:00,  2.57s/it]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The number of remaining candidate features is 352\n",
            "Start stage II selection.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 16/16 [00:14<00:00,  1.12it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Finish data processing.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_df_train = pd.read_csv(path + \"train.csv\", index_col=[0])\n",
        "new_df_valid = pd.read_csv(path + \"valid.csv\", index_col=[0])\n",
        "new_df_test = pd.read_csv(path + \"test.csv\", index_col=[0])\n",
        "\n",
        "new_features = ['autoFE_f_0', 'autoFE_f_1', 'autoFE_f_2',\n",
        "       'autoFE_f_3', 'autoFE_f_4', 'autoFE_f_5', 'autoFE_f_6', 'autoFE_f_7',\n",
        "       'autoFE_f_8', 'autoFE_f_9']\n",
        "\n",
        "new_df_train = pd.merge(new_df_train, train_x[new_features], left_index=True, right_index=True)\n",
        "new_df_train.to_csv(path + \"new_train.csv\", index=True)\n",
        "new_df_valid = pd.merge(new_df_valid, valid_x[new_features], left_index=True, right_index=True)\n",
        "new_df_valid.to_csv(path + \"new_valid.csv\", index=True)\n",
        "new_df_test = pd.merge(new_df_test, test_x[new_features], left_index=True, right_index=True)\n",
        "new_df_test.to_csv(path + \"new_test.csv\", index=True)"
      ],
      "metadata": {
        "id": "FOZDF6DFjnAD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step3: Import Configs"
      ],
      "metadata": {
        "id": "_HsMR-zglMxK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "parser = argparse.ArgumentParser(description='Download Alpaca Datasets')\n",
        "parser.add_argument(\"--config\", default=osp.join(ROOT, \"TradeMaster\", \"configs\", \"portfolio_management\", \"portfolio_management_dj30_ppo_ppo_adam_mse_fg.py\"),\n",
        "                    help=\"download datasets config file path\")\n",
        "parser.add_argument(\"--task_name\", type=str, default=\"train\")\n",
        "args, _ = parser.parse_known_args()\n",
        "\n",
        "cfg = Config.fromfile(args.config)\n",
        "task_name = args.task_name\n",
        "cfg = replace_cfg_vals(cfg)"
      ],
      "metadata": {
        "id": "6XjpFLjQlTIm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step4: Build Dataset"
      ],
      "metadata": {
        "id": "BTxZqdjQlWwC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "dataset = build_dataset(cfg)"
      ],
      "metadata": {
        "id": "GPtJBog1lYqp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step5: Build Trainer"
      ],
      "metadata": {
        "id": "SyGlBeZqlafr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "work_dir = os.path.join(ROOT, \"TradeMaster\", cfg.trainer.work_dir)\n",
        "\n",
        "if not os.path.exists(work_dir):\n",
        "    os.makedirs(work_dir)\n",
        "cfg.dump(osp.join(work_dir, osp.basename(args.config)))\n",
        "\n",
        "trainer = build_trainer(cfg, default_args=dict(dataset=dataset, device = device))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1SeD0nw7lePf",
        "outputId": "05d756e6-fc50-4358-f557-f306041938c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "2023-05-09 16:00:55,228\tWARNING compression.py:17 -- lz4 not available, disabling sample compression. This will significantly impact RLlib performance. To install lz4, run `pip install lz4`.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "| Arguments Keep work_dir: /content/drive/MyDrive/TradeMaster/work_dir/portfolio_management_dj30_ppo_ppo_adam_mse_fg\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step6: Train and Valid"
      ],
      "metadata": {
        "id": "UCzsbBm6lhHJ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import ray\n",
        "from ray.tune.registry import register_env\n",
        "from trademaster.environments.portfolio_management.environment import PortfolioManagementEnvironment\n",
        "def env_creator(env_name):\n",
        "    if env_name == 'portfolio_management':\n",
        "        env = PortfolioManagementEnvironment\n",
        "    else:\n",
        "        raise NotImplementedError\n",
        "    return env\n",
        "ray.init(ignore_reinit_error=True)\n",
        "register_env(\"portfolio_management\", lambda config: env_creator(\"portfolio_management\")(config))\n",
        "trainer.train_and_valid()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8QyVW9HgljgD",
        "outputId": "c820a96d-2c1f-4f1f-b56d-ef284bd396b4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "2023-05-09 16:00:56,598\tINFO worker.py:973 -- Calling ray.init() again after it has already been called.\n",
            "2023-05-09 16:00:56,617\tINFO logger.py:188 -- pip install \"ray[tune]\" to see TensorBoard files.\n",
            "2023-05-09 16:00:56,619\tWARNING logger.py:336 -- Could not instantiate TBXLogger: No module named 'tensorboardX'.\n",
            "2023-05-09 16:00:56,624\tINFO trainer.py:2321 -- Executing eagerly (framework='tf2'), with eager_tracing=False. For production workloads, make sure to set eager_tracing=True  in order to match the speed of tf-static-graph (framework='tf'). For debugging purposes, `eager_tracing=False` is the best choice.\n",
            "2023-05-09 16:00:56,634\tINFO trainer.py:903 -- Current log_level is WARN. For more information, set 'log_level': 'INFO' / 'DEBUG' or use the -v and -vv flags.\n",
            "2023-05-09 16:00:57,649\tWARNING env.py:135 -- Your env doesn't have a .spec.max_episode_steps attribute. This is fine if you have set 'horizon' in your config dictionary, or `soft_horizon`. However, if you haven't, 'horizon' will default to infinity, and your environment will not be reset.\n",
            "2023-05-09 16:00:58,684\tWARNING util.py:65 -- Install gputil for GPU system monitoring.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Episode: [1/2]\n",
            "+--------------+-------------+------------+--------------+\n",
            "| Total Return | Sharp Ratio | Volatility | Max Drawdown |\n",
            "+--------------+-------------+------------+--------------+\n",
            "| -86.547540%  |  -1.847291  | 0.827679%  |  87.620708%  |\n",
            "+--------------+-------------+------------+--------------+\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "2023-05-09 16:02:16,054\tWARNING deprecation.py:46 -- DeprecationWarning: `_get_slice_indices` has been deprecated. This will raise an error in the future!\n",
            "2023-05-09 16:02:16,058\tWARNING deprecation.py:46 -- DeprecationWarning: `slice` has been deprecated. Use `SampleBatch[start:stop]` instead. This will raise an error in the future!\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Valid Episode: [1/2]\n",
            "+--------------+-------------+------------+--------------+\n",
            "| Total Return | Sharp Ratio | Volatility | Max Drawdown |\n",
            "+--------------+-------------+------------+--------------+\n",
            "|  6.795733%   |   0.364199  | 2.155261%  |  32.487321%  |\n",
            "+--------------+-------------+------------+--------------+\n",
            "Train Episode: [2/2]\n",
            "+--------------+-------------+------------+--------------+\n",
            "| Total Return | Sharp Ratio | Volatility | Max Drawdown |\n",
            "+--------------+-------------+------------+--------------+\n",
            "| -89.195716%  |  -1.997400  | 0.850551%  |  89.898235%  |\n",
            "+--------------+-------------+------------+--------------+\n",
            "+--------------+-------------+------------+--------------+\n",
            "| Total Return | Sharp Ratio | Volatility | Max Drawdown |\n",
            "+--------------+-------------+------------+--------------+\n",
            "| -86.575989%  |  -1.829999  | 0.835840%  |  87.083558%  |\n",
            "+--------------+-------------+------------+--------------+\n",
            "Valid Episode: [2/2]\n",
            "+--------------+-------------+------------+--------------+\n",
            "| Total Return | Sharp Ratio | Volatility | Max Drawdown |\n",
            "+--------------+-------------+------------+--------------+\n",
            "|  7.712679%   |   0.389197  | 2.153141%  |  32.247085%  |\n",
            "+--------------+-------------+------------+--------------+\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Step7: Test"
      ],
      "metadata": {
        "id": "9U0io9pwlq1Q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import ray\n",
        "from ray.tune.registry import register_env\n",
        "from trademaster.environments.portfolio_management.environment import PortfolioManagementEnvironment\n",
        "def env_creator(env_name):\n",
        "    if env_name == 'portfolio_management':\n",
        "        env = PortfolioManagementEnvironment\n",
        "    else:\n",
        "        raise NotImplementedError\n",
        "    return env\n",
        "ray.init(ignore_reinit_error=True)\n",
        "register_env(\"portfolio_management\", lambda config: env_creator(\"portfolio_management\")(config))\n",
        "trainer.test()"
      ],
      "metadata": {
        "id": "w1Sw5ia6lst0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c3f33fa6-ff93-4f02-f087-f1c1ff40dc5d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "2023-05-09 16:06:40,156\tWARNING env.py:135 -- Your env doesn't have a .spec.max_episode_steps attribute. This is fine if you have set 'horizon' in your config dictionary, or `soft_horizon`. However, if you haven't, 'horizon' will default to infinity, and your environment will not be reset.\n",
            "2023-05-09 16:06:40,940\tWARNING util.py:65 -- Install gputil for GPU system monitoring.\n",
            "2023-05-09 16:06:41,025\tWARNING eager_tf_policy.py:724 -- Cannot restore an optimizer's state for tf eager! Keras is not able to save the v1.x optimizers (from tf.compat.v1.train) since they aren't compatible with checkpoints.\n",
            "2023-05-09 16:06:41,565\tINFO trainable.py:588 -- Restored on 172.28.0.12 from checkpoint: /root/ray_results/PPOTrainer_portfolio_management_2023-05-09_16-06-37k1vxwqse/tmptnde9i0zrestore_from_object/checkpoint-2\n",
            "2023-05-09 16:06:41,569\tINFO trainable.py:597 -- Current state after restoring: {'_iteration': 2, '_timesteps_total': None, '_time_total': 313.05774760246277, '_episodes_total': 3}\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test Best Episode\n",
            "+--------------+-------------+------------+--------------+\n",
            "| Total Return | Sharp Ratio | Volatility | Max Drawdown |\n",
            "+--------------+-------------+------------+--------------+\n",
            "|  16.367416%  |   1.381897  | 0.724094%  |  6.696040%   |\n",
            "+--------------+-------------+------------+--------------+\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 0.00000000e+00,  3.05970536e-03,  1.31899522e-02,  7.27648248e-03,\n",
              "        1.39758913e-03, -1.03331490e-03,  2.14846361e-03,  1.62698265e-03,\n",
              "       -4.60545304e-04, -6.31885346e-03,  3.28831921e-03,  7.88614397e-03,\n",
              "        6.22040096e-05, -8.51388267e-03, -7.17811191e-04, -4.23127059e-04,\n",
              "       -1.66106080e-02,  9.63072438e-03, -1.89121900e-02,  6.82263723e-03,\n",
              "        1.28993435e-02,  7.38708685e-04,  1.17664077e-02,  4.13381752e-03,\n",
              "        7.45895236e-03, -1.29415441e-03,  1.44970599e-03, -1.11693025e-04,\n",
              "        1.32227505e-03,  3.14629247e-04,  3.56695529e-03, -3.23848260e-03,\n",
              "       -1.33125246e-03,  4.87082169e-04,  8.89492177e-04,  1.18824585e-02,\n",
              "       -1.57107024e-02, -1.53079021e-02,  1.78989222e-02, -4.97593173e-03,\n",
              "       -3.00781952e-03, -1.08432142e-02,  1.88646063e-02,  9.36838398e-03,\n",
              "        1.54327368e-03,  1.35902662e-02,  4.67140772e-03,  7.04713378e-03,\n",
              "        7.48075233e-03, -2.67507102e-03,  3.96961531e-03, -6.89318142e-03,\n",
              "       -7.28553691e-03,  4.77728536e-03, -8.35215590e-03, -8.77986306e-04,\n",
              "        5.66598165e-03,  1.56760512e-02,  3.36420250e-03, -4.61072536e-03,\n",
              "       -1.51825210e-03,  4.58932356e-03,  1.17300969e-02, -2.90726151e-03,\n",
              "        1.12249346e-03, -4.80732519e-05,  6.79862491e-03, -2.71264658e-03,\n",
              "       -2.20915451e-03,  4.76085317e-04,  6.71878833e-03,  4.59442200e-03,\n",
              "       -3.01607940e-03, -6.58085962e-03,  8.88601938e-03, -9.60558443e-03,\n",
              "        3.89283875e-03, -2.19175593e-03, -9.86658888e-04, -3.72400897e-03,\n",
              "        6.44564590e-03, -4.86197229e-03,  6.99024183e-03, -7.79013120e-04,\n",
              "        2.76627867e-03,  8.62836241e-03,  5.72075008e-03, -1.12939754e-03,\n",
              "       -1.25017495e-02, -1.75780501e-02,  1.20962077e-02,  9.73229566e-03,\n",
              "       -1.02037773e-03, -7.58170867e-03, -4.26120136e-03,  5.72094572e-03,\n",
              "        2.14306728e-03,  5.76631181e-03, -3.29298568e-03, -1.00775095e-03,\n",
              "        3.12010226e-03,  1.31894479e-03,  4.64242711e-04,  2.66705259e-03,\n",
              "       -1.48504869e-04,  5.75161177e-03, -3.43401987e-03, -1.24489677e-03,\n",
              "       -3.67085008e-03,  3.19697786e-03,  4.88789433e-04, -2.90401207e-03,\n",
              "       -2.32095658e-03, -7.41330673e-03, -5.60174432e-03, -1.63267857e-02,\n",
              "        1.61832052e-02,  1.37679203e-03, -2.51730947e-03,  8.21129558e-03,\n",
              "        8.04907580e-03, -3.82509535e-03, -7.51865210e-04,  5.39625210e-03,\n",
              "        1.99355415e-03,  3.53428865e-03, -5.84193942e-03,  2.88312324e-03,\n",
              "       -6.83705686e-03,  1.18638277e-02,  2.56605577e-03, -2.18085937e-03,\n",
              "        2.15152553e-03,  2.38168071e-04, -7.79788766e-03, -1.85825981e-02,\n",
              "        1.32304801e-02,  8.13564159e-03, -6.50415318e-04,  4.73309700e-03,\n",
              "        3.75464023e-03, -2.25554544e-03, -3.42421639e-03,  3.70803846e-03,\n",
              "       -3.20779270e-03, -2.88730343e-03,  7.54900707e-03, -9.16299118e-03,\n",
              "        7.48226384e-03,  2.60118715e-03, -2.65559072e-03,  4.65075736e-03,\n",
              "        6.27127481e-03,  9.67032467e-04,  6.12578917e-04,  2.32826690e-03,\n",
              "       -6.17354997e-03, -1.14254182e-02, -2.62370723e-03,  5.54408029e-03,\n",
              "        6.25080824e-03,  5.87147113e-04,  1.40078299e-03, -6.18039586e-03,\n",
              "        6.04291340e-03, -5.72664979e-04,  1.84421012e-04, -1.70474832e-03,\n",
              "        4.22560016e-03, -2.15065831e-03, -7.96253538e-03, -8.48662854e-04,\n",
              "       -4.91859972e-03, -7.19050307e-03,  6.12346253e-03, -8.71186431e-03,\n",
              "        7.26107900e-03, -2.53985960e-03, -5.01937869e-03, -1.58763923e-02,\n",
              "       -1.98758251e-03,  9.25325206e-03,  1.39615752e-02,  5.78912402e-04,\n",
              "        1.37590042e-03, -1.36896406e-02,  3.48025076e-03, -1.44072776e-02,\n",
              "        1.45342167e-02, -7.06568378e-03,  7.04927817e-03,  1.90069289e-03,\n",
              "        8.43241846e-03, -1.04965075e-03, -6.86788046e-03, -4.66309979e-03,\n",
              "       -2.64319108e-04,  1.48165839e-02,  6.94268488e-03, -2.98215873e-03,\n",
              "        6.78706021e-03,  5.73603803e-03, -2.18755633e-03, -1.86273360e-03,\n",
              "        1.18022594e-03, -6.15436448e-04, -7.86939783e-03,  7.61713957e-03,\n",
              "        2.39889135e-03,  3.02519654e-03,  3.95245746e-03,  3.88179470e-03,\n",
              "       -8.49130143e-04,  5.27274908e-03,  1.89921026e-03, -3.59993604e-03,\n",
              "       -4.03241048e-03, -4.56400353e-03,  2.44770614e-03,  5.90766648e-04,\n",
              "       -5.33601873e-04, -6.28899650e-03, -4.49883470e-03, -7.58730716e-03,\n",
              "        2.70639198e-03,  4.93241445e-03, -3.24364862e-04, -2.37401968e-02,\n",
              "        4.79729940e-03, -1.79089330e-02, -1.10785168e-02,  1.69197996e-02,\n",
              "        9.98989909e-04,  1.78714620e-02,  1.19718614e-02,  8.07206994e-04,\n",
              "       -4.26574238e-04,  5.94748842e-03, -6.41418306e-03, -1.55845751e-03,\n",
              "        9.06008811e-03,  2.11448998e-03, -1.18777239e-02, -1.06341222e-02,\n",
              "        1.48906442e-02,  6.01810220e-03,  4.83646275e-03,  8.88457198e-03,\n",
              "        2.87382989e-03,  2.14202388e-03, -2.26165189e-03, -1.55505742e-03])"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot(trainer.test_environment.save_asset_memory(),alg=\"PPO\")"
      ],
      "metadata": {
        "id": "nDPGa-jblvUb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 505
        },
        "outputId": "d9ae91ef-7c89-44c7-e73e-1cd037225ec3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}