Experiment-Notebooks/LGR_Robot_Experiments_Final_Tests.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "LGR-Robot-Experiments-Final-Tests.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "Sux7XfAIoOKb",
        "99mbOXu47GA7",
        "ZYpYb2Mq7Uf1",
        "ML2ltx5u6bfK"
      ],
      "toc_visible": true,
      "machine_shape": "hm"
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "92ff2b3d46be41c687b6ebaacc3a2b12": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_1698ce603eee4841be074e544dddee81",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_ccdc404d2dd94f6d922da1bb48a2f293",
              "IPY_MODEL_cfddde2ba20f4f5e814d86d7ad64b7f1"
            ]
          }
        },
        "1698ce603eee4841be074e544dddee81": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": "row wrap",
            "width": "100%",
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": "inline-flex",
            "left": null
          }
        },
        "ccdc404d2dd94f6d922da1bb48a2f293": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_6ab68b6995834b45a2a434247cd88a8c",
            "_dom_classes": [],
            "description": "Validation sanity check: ",
            "_model_name": "FloatProgressModel",
            "bar_style": "info",
            "max": 0,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 0,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_5d7783a4bb114a0db29478f42d32203e"
          }
        },
        "cfddde2ba20f4f5e814d86d7ad64b7f1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_5b74cd7125184a3c9e824004f3a26f4f",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 0/0 [00:00&lt;?, ?it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_d5a439b9b5f8442884a67965faf2c90e"
          }
        },
        "6ab68b6995834b45a2a434247cd88a8c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "5d7783a4bb114a0db29478f42d32203e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": "2",
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "5b74cd7125184a3c9e824004f3a26f4f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "d5a439b9b5f8442884a67965faf2c90e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "25c68beb388b48e8a62dc343ac2a4c40": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_5a0d8fd2f5f34b0998d5db3f5cfa1362",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_4b0c1c4ce64f4bada89801cb97ace4a3",
              "IPY_MODEL_e41cb5932cd8485e88243093f0315129"
            ]
          }
        },
        "5a0d8fd2f5f34b0998d5db3f5cfa1362": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": "row wrap",
            "width": "100%",
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": "inline-flex",
            "left": null
          }
        },
        "4b0c1c4ce64f4bada89801cb97ace4a3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_0c41caea796a417a93b930aad85da0a7",
            "_dom_classes": [],
            "description": "Epoch 8:  13%",
            "_model_name": "FloatProgressModel",
            "bar_style": "info",
            "max": 5386,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 3775,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_12debfc9dfe0406baa734378a2fcd10d"
          }
        },
        "e41cb5932cd8485e88243093f0315129": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_4d53b23ec6424220a174bed6c9369467",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 3775/5386 [1:05:58&lt;28:09,  1.05s/it, loss=0.419, v_num=-146]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_7f066db9cf044877a5d70a29f8ba1caf"
          }
        },
        "0c41caea796a417a93b930aad85da0a7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "12debfc9dfe0406baa734378a2fcd10d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": "2",
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "4d53b23ec6424220a174bed6c9369467": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "7f066db9cf044877a5d70a29f8ba1caf": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Sux7XfAIoOKb"
      },
      "source": [
        "# COPY MASSIVE DEMO DATA : ONLY DO THIS ONCE!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WbVxwVKO88He"
      },
      "source": [
        "## Partitioned Saved Daata"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UI51vG-OtnJq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fdb729bb-2ff0-4c46-9c79-b7accdcb16c2"
      },
      "source": [
        "!ls /content/drive/MyDrive/ASU/Language-Rewards/NeurIPS-Dataset-Shortcut/NeurIPS-Dataset/"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0gnAQ6879MG_UR5.json  Nxr7390WBGL_UR5.json\n",
            "0gnAQKWAgmK_UR5.json  Nxr7398BKL2_UR5.json\n",
            "0gnAQQM0VEv_UR5.json  Nxr74AWlqoz_UR5.json\n",
            "0gnAQz3gBDv_UR5.json  Nxr74N3WGnp_UR5.json\n",
            "1joVJw0KZyo_UR5.json  Nxr74zNgvJp_UR5.json\n",
            "2RpGK4wjLQ1_UR5.json  Nxr74zZ6gJ8_UR5.json\n",
            "2RpGKR4XJNP_UR5.json  nZAGKpxp8MY_UR5.json\n",
            "2RpGVjNAZPv_UR5.json  nZAGKy09k1l_UR5.json\n",
            "2RpGVzDvM7v_UR5.json  nZAGw5B3YjE_UR5.json\n",
            "3lqYW4MMzpx_UR5.json  nZAGwGE6nY7_UR5.json\n",
            "3lqYWMoM1On_UR5.json  nZAGwgMPvx7_UR5.json\n",
            "48rWMR4KxO0_UR5.json  nZAGwGw18Wp_UR5.json\n",
            "48rWMzgQ5O1_UR5.json  O7vJ44GG0kp_UR5.json\n",
            "48rWXXD880V_UR5.json  O7vJ4B9Jz6G_UR5.json\n",
            "48rWXZj3rwg_UR5.json  OC-LGR-Robo-Pure-Contrastive-dataset-3\n",
            "58vANLyKMNX_UR5.json  oYBAL3A8WmB_UR5.json\n",
            "58vAYNZqxQq_UR5.json  oYBALDGORjL_UR5.json\n",
            "58vAYYKA9GZ_UR5.json  oYBALNVNwmK_UR5.json\n",
            "66wYORYJXpR_UR5.json  oYBALpnQYy3_UR5.json\n",
            "66wYZQ331zR_UR5.json  oYBALqVExXz_UR5.json\n",
            "76xD44M3VyG_UR5.json  oYBALxr5MnX_UR5.json\n",
            "76xD45pQDR8_UR5.json  oYBAx63VW3j_UR5.json\n",
            "76xD4KJ3Oo1_UR5.json  oYBAxQvJyQY_UR5.json\n",
            "76xDPnxPGWr_UR5.json  oYBAxv9y53z_UR5.json\n",
            "8qyX45M7xoj_UR5.json  P1w635XZxOA_UR5.json\n",
            "8qyX4GB6BZj_UR5.json  P1w6379v81W_UR5.json\n",
            "8qyX4k4XAW5_UR5.json  P1w63jD65g4_UR5.json\n",
            "8qyX4pN2o2l_UR5.json  P1w63MVMV1l_UR5.json\n",
            "8qyX4VRRyx5_UR5.json  P1w642ZBVz6_UR5.json\n",
            "8qyXQBwKW9r_UR5.json  P1w64ywD9w2_UR5.json\n",
            "8qyXQE1ER0W_UR5.json  pgD9M1443y2_UR5.json\n",
            "98z301vE1VP_UR5.json  pgD9MApYynm_UR5.json\n",
            "98z3023NzOY_UR5.json  pgD9y8nnYE6_UR5.json\n",
            "98z30rqxQBP_UR5.json  pgD9y90PG8Q_UR5.json\n",
            "98z340Mwv38_UR5.json  pgD9y91x1lQ_UR5.json\n",
            "98z34JgKJOD_UR5.json  pgD9ywAJnEX_UR5.json\n",
            "98z34RLkzYz_UR5.json  pgD9yzVNROQ_UR5.json\n",
            "98z34xD6q13_UR5.json  pgD9yzzN1NQ_UR5.json\n",
            "AnL838kGjG3_UR5.json  Q1xA4MD96Z5_UR5.json\n",
            "AnL83qZ6wRl_UR5.json  Q1xA96ZxQPG_UR5.json\n",
            "AnL84JN16qp_UR5.json  Q1xA9WVWVyq_UR5.json\n",
            "BgM13nOm06Q_UR5.json  qjEAzAy06P2_UR5.json\n",
            "BgM14kKDOpX_UR5.json  qjEAzQ3oY93_UR5.json\n",
            "BgM14ZAR7YX_UR5.json  r0GgA2OJ5OK_UR5.json\n",
            "chunk_meta\t      r0GgA980DDE_UR5.json\n",
            "DROw31l0K3x_UR5.json  r0GgAEvKELK_UR5.json\n",
            "DROw3qgx7Pq_UR5.json  r0GgAkQAVG4_UR5.json\n",
            "DROw4z4XX2n_UR5.json  r0GgAZGv6x6_UR5.json\n",
            "ElgK4A4Yxy4_UR5.json  r0GgAZxYY34_UR5.json\n",
            "ElgK4JL687K_UR5.json  r0GgOKy57wW_UR5.json\n",
            "g5q7mJmwn7D_UR5.json  Rgy83l9Ervw_UR5.json\n",
            "g5q7mplmVg6_UR5.json  Rgy83ooQPY0_UR5.json\n",
            "g5q7mY41M53_UR5.json  Rgy893A01ZE_UR5.json\n",
            "g5q7z1ygOJ9_UR5.json  Rgy89Zv4XP0_UR5.json\n",
            "g5q7zAEq8Ak_UR5.json  test_demonstration.py\n",
            "g5q7zmjlGw3_UR5.json  ValayTest_1.tar.xz\n",
            "g5q7zRQ2QQr_UR5.json  vgMmGlrYNKV_UR5.json\n",
            "g5q7zzDYml9_UR5.json  vgMmV0MK2BV_UR5.json\n",
            "GZkw3vrpDNJ_UR5.json  vgMmVWrBX7n_UR5.json\n",
            "GZkw41QgnW5_UR5.json  Video-OC-LGR-Robo-Pure-Contrastive-dataset-3\n",
            "GZkw41YXMB0_UR5.json  Video-OC-LGR-Robo-Pure-Contrastive-dataset-4\n",
            "GZkw44x8nD5_UR5.json  Video-OC-LGR-Robo-Pure-Contrastive-dataset-5\n",
            "GZkw4y70PK7_UR5.json  Video-OC-LGR-Robo-Pure-Contrastive-dataset-7\n",
            "j2wAD1zz0Mv_UR5.json  VmDJ3Og3zPz_UR5.json\n",
            "j2wAD46o0K4_UR5.json  VmDJ63DJk8W_UR5.json\n",
            "j2wAD7glzr4_UR5.json  VmDJ662k76W_UR5.json\n",
            "j2wAp8R8V3P_UR5.json  VmDJ6BQp54O_UR5.json\n",
            "j2wApAXDmK5_UR5.json  VmDJ6MZp9AX_UR5.json\n",
            "JZnE3Ykxk42_UR5.json  wjNnJkZqBpX_UR5.json\n",
            "k5x6q63MzpE_UR5.json  wjNnWLNELRr_UR5.json\n",
            "k5x6qojk79Y_UR5.json  WnEW43r3Qon_UR5.json\n",
            "k5x6qQ4Om1K_UR5.json  X6GY4WgxBAg_UR5.json\n",
            "KOoA3Y2PjvY_UR5.json  X6GY8L0jBAv_UR5.json\n",
            "KOoA40p7GlY_UR5.json  xG0oKlovpkz_UR5.json\n",
            "LgpA4jnqEgX_UR5.json  xG0oXgROAV9_UR5.json\n",
            "LgpA4o8mr4X_UR5.json  xG0oXRqLXVn_UR5.json\n",
            "lOy8G4EW8B7_UR5.json  Y6JV373kWrA_UR5.json\n",
            "lOy8GqZorxr_UR5.json  Y6JV3vQYO0W_UR5.json\n",
            "lOy8rM6ZK5J_UR5.json  Y6JVgPq17vW_UR5.json\n",
            "lOy8rvW9VnM_UR5.json  y84pLJ0zopz_UR5.json\n",
            "main_meta.csv\t      y84pLJQO83n_UR5.json\n",
            "Mjq73ZL65JQ_UR5.json  y84pLoW8xyn_UR5.json\n",
            "Mjq74LDW8MR_UR5.json  y84pLrM8087_UR5.json\n",
            "Mjq74R65pAA_UR5.json  y84pYLg3j5W_UR5.json\n",
            "Mjq74vqA3om_UR5.json  Z4KZ7mk25qE_UR5.json\n",
            "mwz7J189899_UR5.json  Z4KZ7WXYOJJ_UR5.json\n",
            "mwz7Jl0OWLp_UR5.json  Z4KZjlVL70v_UR5.json\n",
            "mwz7Jlkn403_UR5.json  zm8qZ15rjYO_UR5.json\n",
            "mwz7JLYz8XA_UR5.json  zm8qZ2pvl47_UR5.json\n",
            "mwz7v2gP6N9_UR5.json  zm8qZGvWKMY_UR5.json\n",
            "mwz7vMAJGnE_UR5.json  zm8qZL04Eg8_UR5.json\n",
            "mwz7vR50JqO_UR5.json  zm8qZROxzLY_UR5.json\n",
            "Nxr730xmYG6_UR5.json\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DiWyrGkT84r1"
      },
      "source": [
        "!cp -r /content/drive/MyDrive/ASU/Language-Rewards/NeurIPS-Dataset-Shortcut/NeurIPS-Dataset/Video-OC-LGR-Robo-Pure-Contrastive-dataset-7 Video-OC-LGR-Robo-Pure-Contrastive-dataset-7"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x0EImrDi66af"
      },
      "source": [
        "# WALL OF TEXT TO AVOIDD CLICKING COPY AGAIN\n",
        "# WALL OF TEXT TO AVOIDD CLICKING COPY AGAIN\n",
        "# WALL OF TEXT TO AVOIDD CLICKING COPY AGAIN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "99mbOXu47GA7"
      },
      "source": [
        "# Install Core Deps"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NiXCGNqq56KE"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LV45QPPpUcoc",
        "outputId": "46d9e1d1-7a27-46ae-d395-22d6a0d66467"
      },
      "source": [
        "# Instaall repository\n",
        "# ! pip install --upgrade git+https://github.com/valayDave/language-conditioned-irl@robotics-experiments"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting git+https://****@github.com/valayDave/language-conditioned-irl@robotics-experiments\n",
            "  Cloning https://****@github.com/valayDave/language-conditioned-irl (to revision robotics-experiments) to /tmp/pip-req-build-g9clw4ve\n",
            "  Running command git clone -q 'https://****@github.com/valayDave/language-conditioned-irl' /tmp/pip-req-build-g9clw4ve\n",
            "  Running command git checkout -b robotics-experiments --track origin/robotics-experiments\n",
            "  Switched to a new branch 'robotics-experiments'\n",
            "  Branch 'robotics-experiments' set up to track remote branch 'robotics-experiments' from 'origin'.\n",
            "Requirement already satisfied, skipping upgrade: transformers in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (4.6.1)\n",
            "Requirement already satisfied, skipping upgrade: pytorch-lightning==1.2.5 in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (1.2.5)\n",
            "Requirement already satisfied, skipping upgrade: einops in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (0.3.0)\n",
            "Requirement already satisfied, skipping upgrade: neptune-client in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (0.9.16)\n",
            "Requirement already satisfied, skipping upgrade: neptune-contrib in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (0.27.1)\n",
            "Requirement already satisfied, skipping upgrade: moviepy in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (0.2.3.5)\n",
            "Requirement already satisfied, skipping upgrade: h5py in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (3.1.0)\n",
            "Requirement already satisfied, skipping upgrade: gym in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (0.17.3)\n",
            "Requirement already satisfied, skipping upgrade: opencv-python in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (4.1.2.30)\n",
            "Requirement already satisfied, skipping upgrade: sklearn in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (0.0)\n",
            "Requirement already satisfied, skipping upgrade: click in /usr/local/lib/python3.7/dist-packages (from language-conditioned-rl==0.0.5) (7.1.2)\n",
            "Requirement already satisfied, skipping upgrade: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (1.19.5)\n",
            "Requirement already satisfied, skipping upgrade: requests in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (2.23.0)\n",
            "Requirement already satisfied, skipping upgrade: packaging in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (20.9)\n",
            "Requirement already satisfied, skipping upgrade: tokenizers<0.11,>=0.10.1 in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (0.10.3)\n",
            "Requirement already satisfied, skipping upgrade: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (2019.12.20)\n",
            "Requirement already satisfied, skipping upgrade: sacremoses in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (0.0.45)\n",
            "Requirement already satisfied, skipping upgrade: importlib-metadata; python_version < \"3.8\" in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (4.0.1)\n",
            "Requirement already satisfied, skipping upgrade: huggingface-hub==0.0.8 in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (0.0.8)\n",
            "Requirement already satisfied, skipping upgrade: filelock in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (3.0.12)\n",
            "Requirement already satisfied, skipping upgrade: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers->language-conditioned-rl==0.0.5) (4.41.1)\n",
            "Requirement already satisfied, skipping upgrade: torchmetrics>=0.2.0 in /usr/local/lib/python3.7/dist-packages (from pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.3.2)\n",
            "Requirement already satisfied, skipping upgrade: tensorboard>=2.2.0 in /usr/local/lib/python3.7/dist-packages (from pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (2.5.0)\n",
            "Requirement already satisfied, skipping upgrade: PyYAML!=5.4.*,>=5.1 in /usr/local/lib/python3.7/dist-packages (from pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (5.3.1)\n",
            "Requirement already satisfied, skipping upgrade: future>=0.17.1 in /usr/local/lib/python3.7/dist-packages (from pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.18.2)\n",
            "Requirement already satisfied, skipping upgrade: torch>=1.4 in /usr/local/lib/python3.7/dist-packages (from pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (1.8.1+cu101)\n",
            "Requirement already satisfied, skipping upgrade: fsspec[http]>=0.8.1 in /usr/local/lib/python3.7/dist-packages (from pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (2021.5.0)\n",
            "Requirement already satisfied, skipping upgrade: requests-oauthlib>=1.0.0 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (1.3.0)\n",
            "Requirement already satisfied, skipping upgrade: websocket-client!=1.0.0,>=0.35.0 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (1.0.1)\n",
            "Requirement already satisfied, skipping upgrade: Pillow>=1.1.6 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (7.1.2)\n",
            "Requirement already satisfied, skipping upgrade: PyJWT in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (2.1.0)\n",
            "Requirement already satisfied, skipping upgrade: bravado in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (11.0.3)\n",
            "Requirement already satisfied, skipping upgrade: oauthlib>=2.1.0 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (3.1.0)\n",
            "Requirement already satisfied, skipping upgrade: pandas in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (1.1.5)\n",
            "Requirement already satisfied, skipping upgrade: GitPython>=2.0.8 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (3.1.17)\n",
            "Requirement already satisfied, skipping upgrade: six>=1.12.0 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (1.15.0)\n",
            "Requirement already satisfied, skipping upgrade: urllib3 in /usr/local/lib/python3.7/dist-packages (from neptune-client->language-conditioned-rl==0.0.5) (1.24.3)\n",
            "Requirement already satisfied, skipping upgrade: attrdict>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from neptune-contrib->language-conditioned-rl==0.0.5) (2.0.1)\n",
            "Requirement already satisfied, skipping upgrade: matplotlib in /usr/local/lib/python3.7/dist-packages (from neptune-contrib->language-conditioned-rl==0.0.5) (3.2.2)\n",
            "Requirement already satisfied, skipping upgrade: joblib>=0.13 in /usr/local/lib/python3.7/dist-packages (from neptune-contrib->language-conditioned-rl==0.0.5) (1.0.1)\n",
            "Requirement already satisfied, skipping upgrade: decorator<5.0,>=4.0.2 in /usr/local/lib/python3.7/dist-packages (from moviepy->language-conditioned-rl==0.0.5) (4.4.2)\n",
            "Requirement already satisfied, skipping upgrade: imageio<3.0,>=2.1.2 in /usr/local/lib/python3.7/dist-packages (from moviepy->language-conditioned-rl==0.0.5) (2.4.1)\n",
            "Requirement already satisfied, skipping upgrade: cached-property; python_version < \"3.8\" in /usr/local/lib/python3.7/dist-packages (from h5py->language-conditioned-rl==0.0.5) (1.5.2)\n",
            "Requirement already satisfied, skipping upgrade: scipy in /usr/local/lib/python3.7/dist-packages (from gym->language-conditioned-rl==0.0.5) (1.4.1)\n",
            "Requirement already satisfied, skipping upgrade: pyglet<=1.5.0,>=1.4.0 in /usr/local/lib/python3.7/dist-packages (from gym->language-conditioned-rl==0.0.5) (1.5.0)\n",
            "Requirement already satisfied, skipping upgrade: cloudpickle<1.7.0,>=1.2.0 in /usr/local/lib/python3.7/dist-packages (from gym->language-conditioned-rl==0.0.5) (1.3.0)\n",
            "Requirement already satisfied, skipping upgrade: scikit-learn in /usr/local/lib/python3.7/dist-packages (from sklearn->language-conditioned-rl==0.0.5) (0.22.2.post1)\n",
            "Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers->language-conditioned-rl==0.0.5) (2020.12.5)\n",
            "Requirement already satisfied, skipping upgrade: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers->language-conditioned-rl==0.0.5) (2.10)\n",
            "Requirement already satisfied, skipping upgrade: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers->language-conditioned-rl==0.0.5) (3.0.4)\n",
            "Requirement already satisfied, skipping upgrade: pyparsing>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging->transformers->language-conditioned-rl==0.0.5) (2.4.7)\n",
            "Requirement already satisfied, skipping upgrade: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata; python_version < \"3.8\"->transformers->language-conditioned-rl==0.0.5) (3.4.1)\n",
            "Requirement already satisfied, skipping upgrade: typing-extensions>=3.6.4; python_version < \"3.8\" in /usr/local/lib/python3.7/dist-packages (from importlib-metadata; python_version < \"3.8\"->transformers->language-conditioned-rl==0.0.5) (3.7.4.3)\n",
            "Requirement already satisfied, skipping upgrade: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.4.4)\n",
            "Requirement already satisfied, skipping upgrade: google-auth<2,>=1.6.3 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (1.30.0)\n",
            "Requirement already satisfied, skipping upgrade: werkzeug>=0.11.15 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (1.0.1)\n",
            "Requirement already satisfied, skipping upgrade: absl-py>=0.4 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.12.0)\n",
            "Requirement already satisfied, skipping upgrade: grpcio>=1.24.3 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (1.34.1)\n",
            "Requirement already satisfied, skipping upgrade: setuptools>=41.0.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (57.0.0)\n",
            "Requirement already satisfied, skipping upgrade: tensorboard-plugin-wit>=1.6.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (1.8.0)\n",
            "Requirement already satisfied, skipping upgrade: markdown>=2.6.8 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (3.3.4)\n",
            "Requirement already satisfied, skipping upgrade: tensorboard-data-server<0.7.0,>=0.6.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.6.1)\n",
            "Requirement already satisfied, skipping upgrade: protobuf>=3.6.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (3.12.4)\n",
            "Requirement already satisfied, skipping upgrade: wheel>=0.26; python_version >= \"3\" in /usr/local/lib/python3.7/dist-packages (from tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.36.2)\n",
            "Requirement already satisfied, skipping upgrade: aiohttp; extra == \"http\" in /usr/local/lib/python3.7/dist-packages (from fsspec[http]>=0.8.1->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (3.7.4.post0)\n",
            "Requirement already satisfied, skipping upgrade: simplejson in /usr/local/lib/python3.7/dist-packages (from bravado->neptune-client->language-conditioned-rl==0.0.5) (3.17.2)\n",
            "Requirement already satisfied, skipping upgrade: monotonic in /usr/local/lib/python3.7/dist-packages (from bravado->neptune-client->language-conditioned-rl==0.0.5) (1.6)\n",
            "Requirement already satisfied, skipping upgrade: msgpack in /usr/local/lib/python3.7/dist-packages (from bravado->neptune-client->language-conditioned-rl==0.0.5) (1.0.2)\n",
            "Requirement already satisfied, skipping upgrade: bravado-core>=5.16.1 in /usr/local/lib/python3.7/dist-packages (from bravado->neptune-client->language-conditioned-rl==0.0.5) (5.17.0)\n",
            "Requirement already satisfied, skipping upgrade: python-dateutil in /usr/local/lib/python3.7/dist-packages (from bravado->neptune-client->language-conditioned-rl==0.0.5) (2.8.1)\n",
            "Requirement already satisfied, skipping upgrade: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas->neptune-client->language-conditioned-rl==0.0.5) (2018.9)\n",
            "Requirement already satisfied, skipping upgrade: gitdb<5,>=4.0.1 in /usr/local/lib/python3.7/dist-packages (from GitPython>=2.0.8->neptune-client->language-conditioned-rl==0.0.5) (4.0.7)\n",
            "Requirement already satisfied, skipping upgrade: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->neptune-contrib->language-conditioned-rl==0.0.5) (0.10.0)\n",
            "Requirement already satisfied, skipping upgrade: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->neptune-contrib->language-conditioned-rl==0.0.5) (1.3.1)\n",
            "Requirement already satisfied, skipping upgrade: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.7/dist-packages (from google-auth<2,>=1.6.3->tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.2.8)\n",
            "Requirement already satisfied, skipping upgrade: cachetools<5.0,>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from google-auth<2,>=1.6.3->tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (4.2.2)\n",
            "Requirement already satisfied, skipping upgrade: rsa<5,>=3.1.4; python_version >= \"3.6\" in /usr/local/lib/python3.7/dist-packages (from google-auth<2,>=1.6.3->tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (4.7.2)\n",
            "Requirement already satisfied, skipping upgrade: attrs>=17.3.0 in /usr/local/lib/python3.7/dist-packages (from aiohttp; extra == \"http\"->fsspec[http]>=0.8.1->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (21.2.0)\n",
            "Requirement already satisfied, skipping upgrade: multidict<7.0,>=4.5 in /usr/local/lib/python3.7/dist-packages (from aiohttp; extra == \"http\"->fsspec[http]>=0.8.1->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (5.1.0)\n",
            "Requirement already satisfied, skipping upgrade: async-timeout<4.0,>=3.0 in /usr/local/lib/python3.7/dist-packages (from aiohttp; extra == \"http\"->fsspec[http]>=0.8.1->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (3.0.1)\n",
            "Requirement already satisfied, skipping upgrade: yarl<2.0,>=1.0 in /usr/local/lib/python3.7/dist-packages (from aiohttp; extra == \"http\"->fsspec[http]>=0.8.1->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (1.6.3)\n",
            "Requirement already satisfied, skipping upgrade: jsonref in /usr/local/lib/python3.7/dist-packages (from bravado-core>=5.16.1->bravado->neptune-client->language-conditioned-rl==0.0.5) (0.2)\n",
            "Requirement already satisfied, skipping upgrade: swagger-spec-validator>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from bravado-core>=5.16.1->bravado->neptune-client->language-conditioned-rl==0.0.5) (2.7.3)\n",
            "Requirement already satisfied, skipping upgrade: jsonschema[format]>=2.5.1 in /usr/local/lib/python3.7/dist-packages (from bravado-core>=5.16.1->bravado->neptune-client->language-conditioned-rl==0.0.5) (2.6.0)\n",
            "Requirement already satisfied, skipping upgrade: smmap<5,>=3.0.1 in /usr/local/lib/python3.7/dist-packages (from gitdb<5,>=4.0.1->GitPython>=2.0.8->neptune-client->language-conditioned-rl==0.0.5) (4.0.0)\n",
            "Requirement already satisfied, skipping upgrade: pyasn1<0.5.0,>=0.4.6 in /usr/local/lib/python3.7/dist-packages (from pyasn1-modules>=0.2.1->google-auth<2,>=1.6.3->tensorboard>=2.2.0->pytorch-lightning==1.2.5->language-conditioned-rl==0.0.5) (0.4.8)\n",
            "Requirement already satisfied, skipping upgrade: rfc3987; extra == \"format\" in /usr/local/lib/python3.7/dist-packages (from jsonschema[format]>=2.5.1->bravado-core>=5.16.1->bravado->neptune-client->language-conditioned-rl==0.0.5) (1.3.8)\n",
            "Requirement already satisfied, skipping upgrade: strict-rfc3339; extra == \"format\" in /usr/local/lib/python3.7/dist-packages (from jsonschema[format]>=2.5.1->bravado-core>=5.16.1->bravado->neptune-client->language-conditioned-rl==0.0.5) (0.7)\n",
            "Requirement already satisfied, skipping upgrade: webcolors; extra == \"format\" in /usr/local/lib/python3.7/dist-packages (from jsonschema[format]>=2.5.1->bravado-core>=5.16.1->bravado->neptune-client->language-conditioned-rl==0.0.5) (1.11.1)\n",
            "Building wheels for collected packages: language-conditioned-rl\n",
            "  Building wheel for language-conditioned-rl (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for language-conditioned-rl: filename=language_conditioned_rl-0.0.5-cp37-none-any.whl size=120961 sha256=c4e3cee5088c524ca09a2360879c994427e79bdf28776ee83c626a9b0c928e73\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-ntbez_tk/wheels/ef/8d/39/26d4d0b37f86f1424b32edef638034ce38eff36fbbd662b901\n",
            "Successfully built language-conditioned-rl\n",
            "Installing collected packages: language-conditioned-rl\n",
            "  Found existing installation: language-conditioned-rl 0.0.5\n",
            "    Uninstalling language-conditioned-rl-0.0.5:\n",
            "      Successfully uninstalled language-conditioned-rl-0.0.5\n",
            "Successfully installed language-conditioned-rl-0.0.5\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ogoGsYQXxO79",
        "outputId": "e4ae9898-0900-45a0-dcac-428a395fbb6b"
      },
      "source": [
        "!pip uninstall -y torchtext"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[33mWARNING: Skipping torchtext as it is not installed.\u001b[0m\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4Sd5eM1yZhYr",
        "outputId": "e2e1b041-fb87-47b9-b6e2-059b71309370"
      },
      "source": [
        "!pip install metaflow h5py"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: metaflow in /usr/local/lib/python3.7/dist-packages (2.3.0)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.7/dist-packages (3.1.0)\n",
            "Requirement already satisfied: boto3 in /usr/local/lib/python3.7/dist-packages (from metaflow) (1.17.88)\n",
            "Requirement already satisfied: click>=7.0 in /usr/local/lib/python3.7/dist-packages (from metaflow) (7.1.2)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from metaflow) (2.23.0)\n",
            "Requirement already satisfied: pylint in /usr/local/lib/python3.7/dist-packages (from metaflow) (2.8.3)\n",
            "Requirement already satisfied: numpy>=1.14.5; python_version == \"3.7\" in /usr/local/lib/python3.7/dist-packages (from h5py) (1.19.5)\n",
            "Requirement already satisfied: cached-property; python_version < \"3.8\" in /usr/local/lib/python3.7/dist-packages (from h5py) (1.5.2)\n",
            "Requirement already satisfied: botocore<1.21.0,>=1.20.88 in /usr/local/lib/python3.7/dist-packages (from boto3->metaflow) (1.20.88)\n",
            "Requirement already satisfied: s3transfer<0.5.0,>=0.4.0 in /usr/local/lib/python3.7/dist-packages (from boto3->metaflow) (0.4.2)\n",
            "Requirement already satisfied: jmespath<1.0.0,>=0.7.1 in /usr/local/lib/python3.7/dist-packages (from boto3->metaflow) (0.10.0)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->metaflow) (3.0.4)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->metaflow) (2.10)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->metaflow) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->metaflow) (2020.12.5)\n",
            "Requirement already satisfied: mccabe<0.7,>=0.6 in /usr/local/lib/python3.7/dist-packages (from pylint->metaflow) (0.6.1)\n",
            "Requirement already satisfied: toml>=0.7.1 in /usr/local/lib/python3.7/dist-packages (from pylint->metaflow) (0.10.2)\n",
            "Requirement already satisfied: isort<6,>=4.2.5 in /usr/local/lib/python3.7/dist-packages (from pylint->metaflow) (5.8.0)\n",
            "Requirement already satisfied: astroid==2.5.6 in /usr/local/lib/python3.7/dist-packages (from pylint->metaflow) (2.5.6)\n",
            "Requirement already satisfied: python-dateutil<3.0.0,>=2.1 in /usr/local/lib/python3.7/dist-packages (from botocore<1.21.0,>=1.20.88->boto3->metaflow) (2.8.1)\n",
            "Requirement already satisfied: lazy-object-proxy>=1.4.0 in /usr/local/lib/python3.7/dist-packages (from astroid==2.5.6->pylint->metaflow) (1.6.0)\n",
            "Requirement already satisfied: wrapt<1.13,>=1.11 in /usr/local/lib/python3.7/dist-packages (from astroid==2.5.6->pylint->metaflow) (1.12.1)\n",
            "Requirement already satisfied: typed-ast<1.5,>=1.4.0; implementation_name == \"cpython\" and python_version < \"3.8\" in /usr/local/lib/python3.7/dist-packages (from astroid==2.5.6->pylint->metaflow) (1.4.3)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil<3.0.0,>=2.1->botocore<1.21.0,>=1.20.88->boto3->metaflow) (1.15.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZYpYb2Mq7Uf1"
      },
      "source": [
        "# Import Core Modules"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mw2CWJS9MoxX"
      },
      "source": [
        "from language_conditioned_rl.tests.mountaincar import MountainCarTestCase"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_6f3JYmA7Wq0"
      },
      "source": [
        "from language_conditioned_rl.models.transformer import VanillaTransformer,MultiModalTransformer,SinusoidalPositionalEmbedding\n",
        "from language_conditioned_rl.models.cross_modal_attention import CrossModalAttentionWithMask, MultiModalAttentionBlock\n",
        "from language_conditioned_rl.models.reward_model import LGRPureContrastiveRewardLearner,DataAndOptimizerConf,LGRRewardOnlyHeadLearner\n",
        "from language_conditioned_rl.dataloaders.mountaincar.dataset import BertEmbedContrastiveTokenizedDataset,BertEmbedContrastiveTokenizedDatasetWithSentences\n",
        "import torch \n",
        "import pytorch_lightning as pl\n",
        "import torch.nn as nn \n",
        "import torch.nn.functional as F \n",
        "from torch.utils.data import Dataset,DataLoader\n",
        "import einops\n",
        "import zlib"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8uzOKOWLOsa2"
      },
      "source": [
        "import random \n",
        "from dataclasses import dataclass,field\n",
        "from typing import List\n",
        "from functools import partial\n",
        "import os\n",
        "import matplotlib.pyplot as plt\n",
        "import os\n",
        "import json\n",
        "import numpy as np\n",
        "import zlib\n",
        "import base64"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z7H-yHVudidf"
      },
      "source": [
        "from transformers import AutoModel,AutoTokenizer"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pC-o-ihvfVBW"
      },
      "source": [
        "import itertools\n",
        "import math\n",
        "import json\n",
        "from metaflow import parallel_map\n",
        "import pandas\n",
        "from matplotlib import pyplot as plt\n",
        "import numpy\n",
        "import os"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PAHspDPm9Cds"
      },
      "source": [
        "from language_conditioned_rl.models.transformer import OmniChannelTransformer,OmniTransformerCoreConfig,ChannelConfiguration,ChannelEmbeddingContinous,ChannelEmbeddingDiscrete,ChannelData\n",
        "from language_conditioned_rl.models.reward_model import LGROmniChannelRewardOnlyHeadLearner,RewardHeadWithOnlyBackbone"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NZlEUnwJPQ2J"
      },
      "source": [
        "from language_conditioned_rl.models.transformer import ImagePatchEmbedding,TextEmbeddingsPretrain"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZOqsqHi8Mi7d"
      },
      "source": [
        "import torch\n",
        "from torchvision import transforms\n",
        "from PIL import Image\n",
        "from torch.nn.utils.rnn import pad_sequence"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9LEnjz1Ea26z"
      },
      "source": [
        "from language_conditioned_rl.dataloaders.robotics.datamaker import H5DataCreatorMainDataCreator,HDF5ContrastiveSetCreator\n",
        "from language_conditioned_rl.models.transformer import PRETRAINED_MODEL\n",
        "from transformers import AutoTokenizer\n",
        "import random\n",
        "from language_conditioned_rl.dataloaders.robotics.dataset import RandomPairwiseDataset,ContrastiveCollateFn,ContrastiveSampleGeneratedDataset\n",
        "from language_conditioned_rl.dataloaders.channel import ContrastiveGenerator\n",
        "from language_conditioned_rl.models.robotics.reward_model import LGRRoboRewardLearner"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VrjqQNcaOFiu"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4VfrsGvmCzku"
      },
      "source": [
        "# Configuration"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U410AWKDACOF"
      },
      "source": [
        "import gym\n",
        "PRETRAINED_MODEL = 'bert-base-uncased'\n",
        "MAX_TEXT_LENGTH = 25\n",
        "\n",
        "\n",
        "CONTENT_FRAC = 0.20\n",
        "BATCH_SIZE = 55\n",
        "WARMUP = 3000\n",
        "MAX_CYCLES = 10\n",
        "MAX_EPOCHS = 25\n",
        "LEARNING_RATE = 1e-4\n",
        "TOTAL_STEPS= MAX_EPOCHS * int(10000/BATCH_SIZE)\n",
        "NO_LR_SCHEDULER=True\n",
        "LR_SCHEDULER_INTERVAL='step'\n",
        "\n",
        "IMAGE_SIZE = (256,256)\n",
        "PATCH_SIZE = 64\n",
        "PATCH_EMBEDDING_DIMS = 128\n",
        "\n",
        "DISCRETE_EMBEDDING_SIZE=128\n",
        "MAX_TEXT_LENGTH = 30\n",
        "\n",
        "EPISODE_SAMPLES=30\n",
        "SENTENCE_PER_EPISODE=5\n",
        "SENTENCE_SAMPLE_BUFFER_SIZE = 100\n",
        "\n",
        "CONTINOUS_VALUE_DIMS = {\n",
        " 'joint_gripper_velocity': 1,\n",
        " 'joint_robot_position': 6,\n",
        " 'joint_robot_velocity': 6,\n",
        " 'tcp_angular_veloctiy': 3,\n",
        " 'tcp_linear_velocity': 3,\n",
        " 'tcp_orientation': 3,\n",
        " 'tcp_position': 3,\n",
        " 'tcp_target_orientation': 3,\n",
        " 'tcp_target_position': 3,\n",
        " }\n",
        "\n",
        "DISCRETE_CHANNELS = {\n",
        "     'joint_gripper' : True\n",
        " }\n",
        "NO_PAD_CHANNELS = {\n",
        "     'image': True,\n",
        "     'image_sequence':True\n",
        " }\n",
        "\n",
        "ALL_CHANNELS = [\n",
        "  'image_sequence',\n",
        "  'joint_gripper_velocity',\n",
        "  'joint_robot_position',\n",
        "  'joint_robot_velocity',\n",
        "  'tcp_angular_veloctiy',\n",
        "  'tcp_linear_velocity',\n",
        "  'tcp_orientation',\n",
        "  'tcp_position',\n",
        "  'tcp_target_orientation',\n",
        "  'tcp_target_position',\n",
        "  'image',\n",
        "  'text',\n",
        "  'joint_gripper',\n",
        "  'joint_combined_vector'\n",
        "  'final_target_coordinates',\n",
        "]\n",
        "\n",
        "USE_CHANNELS = [\n",
        "  # 'joint_robot_position',\n",
        "  # 'image_sequence',\n",
        "  'final_target_coordinates',\n",
        "  'text',\n",
        "  'tcp_position',\n",
        "  # 'joint_combined_vector'\n",
        "]\n",
        "EXPERIMENT_NAME= f'Video-OC-LGR-Robo-Pure-Contrastive'\n",
        "\n",
        "DATASET_NAME = f'{EXPERIMENT_NAME}-dataset-7'\n",
        "\n",
        "\n",
        "#NTXentLoss condds\n",
        "TEMPRATURE = 1.0\n",
        "EXPNUM = 1\n",
        "PROJECT_NAME = 'Language-Grounded-Rewards-Robo'\n",
        "\n",
        "\n",
        "TEST_EPISODE_SAMPLES=10\n",
        "TEST_SENTENCE_PER_EPISODE = 2\n",
        "\n",
        "# PAATH TO DAATA COMES HEERE> \n",
        "DATA_PTH = '/content/drive/MyDrive/ASU/Language-Rewards/NeurIPS-Dataset-Shortcut/NeurIPS-Dataset'\n",
        "DEMO_DS_FILE = 'demonstration.hdf5'\n",
        "META_SAVE_PTH = os.path.join(DATA_PTH,'chunk_meta')\n",
        "MAIN_META_PTH = os.path.join(DATA_PTH,'main_meta.csv')\n",
        "CORE_TRANSFORMER_PARAMS = dict(\n",
        "  #  if pooling_strategy == 'mean' then mean of all\n",
        "  pooling_strategy = 'mean',  # Can be 'cls' or 'mean'\n",
        "  num_layers=1,\\\n",
        "  transformer_embedding_size=512,\\\n",
        "  layer_norm_epsilon=0.00001,\\\n",
        "  scale=0.01,\\\n",
        "  resid_pdrop=0.1,\\\n",
        "  attn_pdrop=0.1,\\\n",
        "  num_heads = 2,\\\n",
        "  embd_pdrop=0.1,\n",
        ")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-yNKA1dzNGQI",
        "outputId": "d1316537-c114-4df6-b55c-7b4bcba557f7"
      },
      "source": [
        "DATASET_NAME"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'Video-OC-LGR-Robo-Pure-Contrastive-dataset-7'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R9CSoeRU-svP"
      },
      "source": [
        "# Pytorch Dataloader"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AM_W0lgvPvjq"
      },
      "source": [
        "from language_conditioned_rl.dataloaders.robotics.datamaker import SampleContrastingRule,PickingNoisyContrastRule,SameObjectPouringIntensityRule,PouringShapeSizeContrast,PickingObjectContrastingRule,ContrastingActionsRule,OBJECT_IDENTIFIERS,FORBIDDEN_SENTENCES"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zuCkWZ9zQSDe"
      },
      "source": [
        "from language_conditioned_rl.dataloaders.robotics.contrastive_dataset import SentenceContrastiveDataset,TaskBasedSentenceContrastiveDataset,JointsChannelsConcatDataset"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y75EMyPmQ_a2"
      },
      "source": [
        "from typing import List,Tuple"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AtSFwyetTFh5"
      },
      "source": [
        "# RULES = [SameObjectPouringIntensityRule(),PouringShapeSizeContrast(),PickingObjectContrastingRule(),ContrastingActionsRule()]\n",
        "RULES = [PickingNoisyContrastRule(),PickingObjectContrastingRule()]\n",
        "# Set ruules over here to create the contrastive samples. "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u1_bAOCslEp6"
      },
      "source": [
        "RULE_DISTRIBUTION = [80,20]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g6r-dyivbg2T"
      },
      "source": [
        "# TRAIN_RULES = [PouringShapeSizeContrast(),PickingObjectContrastingRule()]\n",
        "TRAIN_RULES = RULES#[PickingObjectContrastingRule()]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5iPl_a4IfmOH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b80b0970-0a29-4439-c892-ec05bc60e8ea"
      },
      "source": [
        "USE_CHANNELS"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['final_target_coordinates', 'text', 'tcp_position']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zTYJAP8yQcGP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6614f70f-fa51-4ac9-afd8-b5ca51ffb786"
      },
      "source": [
        "dataset = JointsChannelsConcatDataset(\n",
        "    DATASET_NAME,\\\n",
        "    normalize_images=True,\\\n",
        "    use_channels=USE_CHANNELS,\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "            Dataset Loaded \n",
            "\n",
            "            Data Set Contains information from the Following Rules : \n",
            "\n",
            "            PickingObjectContrastingRule, PickingNoisyContrastRule\n",
            "            Used Originally Stored Indices : True\n",
            "            \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bfDVkeSTRO84"
      },
      "source": [
        "dataset.remake_indices(300000,\\\n",
        "    rules=TRAIN_RULES,\\\n",
        "    rule_distribution=RULE_DISTRIBUTION,\\\n",
        ")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fPa4XNdXej9q",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5267eb61-80bd-4b87-b090-d61eee8aff40"
      },
      "source": [
        "BATCH_SIZE"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "55"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tFa4WqL0R1xo"
      },
      "source": [
        "robo_ds = dataset"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yST16GgOJKxj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4ac41e5c-fe03-468a-8f42-fbc86b8bbe21"
      },
      "source": [
        "len(robo_ds)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "296237"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dGQkB1f1_3_8"
      },
      "source": [
        "# rrds = RandomPairwiseDataset(DEMO_DS_FILE)\n",
        "# rrdsloader = DataLoader(rrds,batch_size=BATCH_SIZE,collate_fn=ContrastiveCollateFn(),num_workers=0,shuffle=True)\n",
        "roboloader = DataLoader(robo_ds,batch_size=BATCH_SIZE,collate_fn=dataset.collate_fn(),num_workers=0,shuffle=True,drop_last=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ywenNKH6uyBc",
        "outputId": "f5b51d10-8b6a-424b-b65f-43beff9ef4de"
      },
      "source": [
        "%time zt = next(iter(roboloader))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 62.3 ms, sys: 5.02 ms, total: 67.3 ms\n",
            "Wall time: 38.2 ms\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qdFDMLDWfSiU"
      },
      "source": [
        "# Multi-task Transformer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lrCTmRf6ksHV"
      },
      "source": [
        "from language_conditioned_rl.models.robotics.reward_model import LGRMulitTaskRoboRewardLearner"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BM9Ykz7bfUxm"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9uTcoSs-QAjK"
      },
      "source": [
        "# Tranformer Initialisation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Smqxb2smSDCU"
      },
      "source": [
        "\n",
        "def NEPUTUNE_JSON_FIXER(json_str):\n",
        "    return json.loads(json_str.replace(\"'\",'\"').replace('None','null').replace('True','true').replace('False','false'))\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GqgEv0iDR9wy"
      },
      "source": [
        "def from_neptune(project_name,\n",
        "                experiment_name,\n",
        "                checkpoint_path, base_path='model_checkpoints/',\n",
        "                api_token=None,):\n",
        "  import neptune\n",
        "  import os\n",
        "  import json\n",
        "  if api_token is None:\n",
        "      raise Exception(\"API Token Missing\")\n",
        "\n",
        "  project = neptune.init(project_name,\n",
        "                          api_token=api_token\n",
        "                          )\n",
        "  my_exp = project.get_experiments(id=experiment_name)[0]\n",
        "  my_exp.download_artifact(checkpoint_path, destination_dir=base_path)\n",
        "  ckck_name = checkpoint_path.split('/')[1]\n",
        "  checkpoint = torch.load(os.path.join(\n",
        "      base_path, ckck_name), map_location=torch.device('cpu'))\n",
        "\n",
        "  return checkpoint"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kGUc91BlSHrG"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_o7uyeDJOO_f"
      },
      "source": [
        "from language_conditioned_rl.models.robotics.reward_model import make_model as MakeRobotModel"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tR3tBYlrQY0M"
      },
      "source": [
        "data_params = DataAndOptimizerConf(\n",
        "    NUM_TRAIN_SAMPLES = len(robo_ds),\n",
        "    BATCH_SIZE = BATCH_SIZE,\n",
        "    MAX_CYCLES = MAX_CYCLES,\n",
        "    MAX_EPOCHS = MAX_EPOCHS,\n",
        "    LEARNING_RATE = LEARNING_RATE,\n",
        "    NO_LR_SCHEDULER=NO_LR_SCHEDULER,\n",
        "    LR_SCHEDULER_FREQUENCY=LR_SCHEDULER_INTERVAL,\n",
        "    WARMUP=WARMUP,\n",
        "    TOTAL_STEPS =TOTAL_STEPS\n",
        ")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ViQ77A_VTZWQ"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C-Xaf3ndQoit",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c93c9566-c053-4686-ce23-9e44195efbdf"
      },
      "source": [
        "# trans = LGRRoboRewardLearner(transformer_config,data_params=data_params)\n",
        "trans = MakeRobotModel(\n",
        "    use_channels=USE_CHANNELS,\n",
        "    CORE_TRANSFORMER_PARAMETERS=CORE_TRANSFORMER_PARAMS,\n",
        "    dataparams=data_params,\n",
        "    is_multitask=False,\n",
        "    is_pc_grad=False,\n",
        "    channel_embed_size=DISCRETE_EMBEDDING_SIZE,\n",
        "    patch_embed_size=PATCH_EMBEDDING_DIMS,\n",
        "    patch_size=PATCH_SIZE,\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertModel: ['cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias', 'cls.predictions.bias', 'cls.predictions.transform.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.decoder.weight', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.transform.LayerNorm.bias']\n",
            "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
            "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EgPysjWJijiE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e00e26ea-51ef-4dab-bb2c-43b85c20c874"
      },
      "source": [
        "DISCRETE_EMBEDDING_SIZE"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "128"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_tSVM7MfDa5r",
        "outputId": "4e33626f-6c4a-4e4d-a7b5-831907c39bad"
      },
      "source": [
        "trans.model.channel_routes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'final_target_coordinates': ['text', 'tcp_position'],\n",
              " 'tcp_position': ['text'],\n",
              " 'text': ['tcp_position', 'final_target_coordinates']}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1Is1bf_sQfKQ"
      },
      "source": [
        "# ptx = trans.model.tcp_position__1dcov(ttx.pos_channel_holder.tcp_position.sequence) #image_sequence__embedding.to_patch_embedding(ttx.pos_channel_holder.image_sequence.sequence)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LCkDILsUF9tf"
      },
      "source": [
        "# ptx.shape"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rT0ae6ZUZqTH"
      },
      "source": [
        "# Training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4GV2NmrIYT5K"
      },
      "source": [
        "MAX_EPOCHS = 400"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ev3NwblTTx0W"
      },
      "source": [
        "# Add Neptune AAPI Key Here\n",
        "API_KEY = None\n",
        "# proj = f'valay/{PROJECT_NAME}'\n",
        "proj = None"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7uIdEBmdEUR_",
        "outputId": "45cd9cff-d8fc-428f-b96a-6009bbfb7a58"
      },
      "source": [
        "from pytorch_lightning.loggers import NeptuneLogger\n",
        "\n",
        "NOTE = f'''\n",
        "Using Video Dataset\n",
        "\n",
        "Rules : {', '.join(list(set(dataset.indices_rules))) }\n",
        "\n",
        "Normalized Images and Multi-task based loss setup. \n",
        "\n",
        "Smaller Trajectories. \n",
        "\n",
        "Using Direct Objects and tcp trajectory\n",
        "\n",
        "Added Noisy Trajectories too\n",
        "'''\n",
        "logger_config_added_options = dict(\n",
        "    BATCH_SIZE = BATCH_SIZE,\n",
        "    WARMUP = data_params.WARMUP,\n",
        "    MAX_CYCLES = MAX_CYCLES,\n",
        "    MAX_EPOCHS = MAX_EPOCHS,\n",
        "    LEARNING_RATE = LEARNING_RATE,\n",
        "    TRAIN_SAMPLE_SIZE=str(data_params.NUM_TRAIN_SAMPLES),\n",
        "    MODEL_NAME=trans._get_name(),\n",
        "    NO_LR_SCHEDULER=NO_LR_SCHEDULER,\n",
        "    TOTAL_STEPS=TOTAL_STEPS,\n",
        "    channel_configurations = [c.to_json() for c in trans.model.config.channel_configurations]\n",
        ")\n",
        "\n",
        "neptune_logger = NeptuneLogger(experiment_name=EXPERIMENT_NAME,\n",
        "                             project_name=proj,\n",
        "                             api_key = API_KEY,\n",
        "                             params={**CORE_TRANSFORMER_PARAMS,**logger_config_added_options,'transformer_params':CORE_TRANSFORMER_PARAMS,'data_params':logger_config_added_options},\n",
        "                             description=NOTE)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "NeptuneLogger will work in online mode\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 53
        },
        "id": "6eVp4i5Kb2X5",
        "outputId": "bae74b02-ca2b-4967-ab2c-9595ea9bdcbe"
      },
      "source": [
        "NOTE"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'\\nUsing Video Dataset\\n\\nRules : PickingObjectContrastingRule, PickingNoisyContrastRule\\n\\nNormalized Images and Multi-task based loss setup. \\n\\nSmaller Trajectories. \\n\\nUsing Direct Objects and tcp trajectory\\n\\nAdded Noisy Trajectories too\\n'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Kgx229zsZKqC",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "aff008e5-7ecb-4bca-8f23-2a6b4a5f85e6"
      },
      "source": [
        "# neptune_logger.experiment_id\n",
        "pl.trainer.seed_everything(42)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Global seed set to 42\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "42"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "92ff2b3d46be41c687b6ebaacc3a2b12",
            "1698ce603eee4841be074e544dddee81",
            "ccdc404d2dd94f6d922da1bb48a2f293",
            "cfddde2ba20f4f5e814d86d7ad64b7f1",
            "6ab68b6995834b45a2a434247cd88a8c",
            "5d7783a4bb114a0db29478f42d32203e",
            "5b74cd7125184a3c9e824004f3a26f4f",
            "d5a439b9b5f8442884a67965faf2c90e",
            "25c68beb388b48e8a62dc343ac2a4c40",
            "5a0d8fd2f5f34b0998d5db3f5cfa1362",
            "4b0c1c4ce64f4bada89801cb97ace4a3",
            "e41cb5932cd8485e88243093f0315129",
            "0c41caea796a417a93b930aad85da0a7",
            "12debfc9dfe0406baa734378a2fcd10d",
            "4d53b23ec6424220a174bed6c9369467",
            "7f066db9cf044877a5d70a29f8ba1caf"
          ]
        },
        "id": "fy5wiwmLQUXC",
        "outputId": "5f8e4705-80f1-4d20-9181-2885ccc98795"
      },
      "source": [
        "# NUM_TRAIN_SAMPLES = len(tr)\n",
        "from pytorch_lightning import Trainer\n",
        "from pytorch_lightning.callbacks import ModelCheckpoint\n",
        "from pytorch_lightning.callbacks import LearningRateMonitor\n",
        "lr_logger = LearningRateMonitor(logging_interval='step')\n",
        "model_checkpoint = ModelCheckpoint(filename='model/checkpoints/{epoch:02d}-{loss:.2f}',\n",
        "                                   save_weights_only=True,\n",
        "                                   save_top_k=3,\n",
        "                                   monitor='loss',\n",
        "                                   period=1)\n",
        "trainer = Trainer(\n",
        "    max_epochs=MAX_EPOCHS,\\\n",
        "    progress_bar_refresh_rate=25,\\\n",
        "    gpus=1,\\\n",
        "    logger=neptune_logger,\n",
        "    callbacks=[lr_logger],\n",
        "    checkpoint_callback=model_checkpoint\n",
        ")\n",
        "\n",
        "trainer.fit(trans, roboloader)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "GPU available: True, used: True\n",
            "TPU available: None, using: 0 TPU cores\n",
            "/usr/local/lib/python3.7/dist-packages/pytorch_lightning/utilities/distributed.py:52: UserWarning: you defined a validation_step but have no val_dataloader. Skipping validation loop\n",
            "  warnings.warn(*args, **kwargs)\n",
            "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "https://ui.neptune.ai/valay/Language-Grounded-Rewards-Robo/e/LRO-146\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "  | Name             | Type                       | Params\n",
            "----------------------------------------------------------------\n",
            "0 | model            | OmniChannelTransformer     | 81.1 M\n",
            "1 | sfmx             | Softmax                    | 0     \n",
            "2 | log_sigmoid      | LogSigmoid                 | 0     \n",
            "3 | reward_predictor | RewardHeadWithOnlyBackbone | 1.3 M \n",
            "4 | cross_entropy    | CrossEntropyLoss           | 0     \n",
            "----------------------------------------------------------------\n",
            "58.9 M    Trainable params\n",
            "23.4 M    Non-trainable params\n",
            "82.4 M    Total params\n",
            "329.474   Total estimated model params size (MB)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "92ff2b3d46be41c687b6ebaacc3a2b12",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validation sanity check', layout=Layout…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\r"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/pytorch_lightning/utilities/distributed.py:52: UserWarning: The dataloader, train dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 4 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
            "  warnings.warn(*args, **kwargs)\n",
            "/usr/local/lib/python3.7/dist-packages/pytorch_lightning/utilities/distributed.py:52: RuntimeWarning: You are using `LearningRateMonitor` callback with models that have no learning rate schedulers. Please see documentation for `configure_optimizers` method.\n",
            "  warnings.warn(*args, **kwargs)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "25c68beb388b48e8a62dc343ac2a4c40",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Training', layout=Layout(flex='2'), max…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/language_conditioned_rl/models/embeddings.py:20: UserWarning: The number of elements in the out tensor of shape [26] is 26 which does not match the computed number of elements 481. Note that this may occur as a result of rounding error. The out tensor will be resized to a tensor of shape (481,). (Triggered internally at  /pytorch/aten/src/ATen/native/cuda/RangeFactories.cu:263.)\n",
            "  torch.arange(padding_idx + 1, max_pos, out=getattr(make_positions, buf_name))\n",
            "Unexpected error in ping thread.\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/neptune/internal/threads/ping_thread.py\", line 37, in run\n",
            "    self.__backend.ping_experiment(self.__experiment)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/neptune/internal/api_clients/hosted_api_clients/hosted_alpha_leaderboard_api_client.py\", line 391, in ping_experiment\n",
            "    self.leaderboard_swagger_client.api.ping(experimentId=str(experiment.internal_id)).response().result\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 239, in response\n",
            "    six.reraise(*sys.exc_info())\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/six.py\", line 703, in reraise\n",
            "    raise value\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 200, in response\n",
            "    swagger_result = self._get_swagger_result(incoming_response)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 124, in wrapper\n",
            "    return func(self, *args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 303, in _get_swagger_result\n",
            "    self.request_config.response_callbacks,\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 334, in unmarshal_response\n",
            "    raise_on_unexpected(incoming_response)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 408, in raise_on_unexpected\n",
            "    raise make_http_exception(response=http_response)\n",
            "bravado.exception.HTTPServiceUnavailable: 503 : {\"code\":503,\"errorType\":\"SERVICE_UNAVAILABLE\",\"title\":\"Service is not available. Try again later.\"}\n",
            "Unexpected error in ping thread.\n",
            "Traceback (most recent call last):\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/neptune/internal/threads/ping_thread.py\", line 37, in run\n",
            "    self.__backend.ping_experiment(self.__experiment)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/neptune/internal/api_clients/hosted_api_clients/hosted_alpha_leaderboard_api_client.py\", line 391, in ping_experiment\n",
            "    self.leaderboard_swagger_client.api.ping(experimentId=str(experiment.internal_id)).response().result\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 239, in response\n",
            "    six.reraise(*sys.exc_info())\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/six.py\", line 703, in reraise\n",
            "    raise value\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 200, in response\n",
            "    swagger_result = self._get_swagger_result(incoming_response)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 124, in wrapper\n",
            "    return func(self, *args, **kwargs)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 303, in _get_swagger_result\n",
            "    self.request_config.response_callbacks,\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 334, in unmarshal_response\n",
            "    raise_on_unexpected(incoming_response)\n",
            "  File \"/usr/local/lib/python3.7/dist-packages/bravado/http_future.py\", line 408, in raise_on_unexpected\n",
            "    raise make_http_exception(response=http_response)\n",
            "bravado.exception.HTTPServiceUnavailable: 503 : {\"errorType\":\"SERVICE_UNAVAILABLE\",\"code\":503,\"title\":\"Service is not available. Try again later.\"}\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gS5QqoxCsKFg"
      },
      "source": [
        "# Testing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ze1o1OQ60Bjn"
      },
      "source": [
        "RULES = [PickingNoisyContrastRule(),PickingObjectContrastingRule()]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1nojgdOW1TxM"
      },
      "source": [
        "from language_conditioned_rl.tests.robotics import save_test_data,run_test_pipeline,plot_test_case_results,save_json_to_file,RoboTestDataset"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KO0GNymLPriO",
        "outputId": "32988abc-4d4f-43d1-a0ec-f95dcbf27996"
      },
      "source": [
        "rtds = RoboTestDataset(DATASET_NAME,use_channels=USE_CHANNELS,size=100)\n",
        "rtds.remake_indices(size=100,rules=RULES)\n",
        "return_object = run_test_pipeline(\n",
        "        trans,\n",
        "        dataset=rtds,\n",
        "        batch_size=20,\n",
        "    )"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "            Dataset Loaded \n",
            "\n",
            "            Data Set Contains information from the Following Rules : \n",
            "\n",
            "            PickingObjectContrastingRule, PickingNoisyContrastRule\n",
            "            Used Originally Stored Indices : False\n",
            "            \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zknKeuFZz7wM"
      },
      "source": [
        "# return_object"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SSW1t1yhgXEE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "adab34f8-820c-4efa-a995-a9ca6ce35627"
      },
      "source": [
        "PLT_NAME = 'robo-reward-dist.pdf'\n",
        "FILE_SAVE_NAME = 'model_results.csv'\n",
        "plot_test_case_results(\n",
        "    return_object,show=True,rule_map={r.rule_name:r.rule_map for r in RULES},plt_name=PLT_NAME\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 2160x2160 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lmFu3a7Bq5Ep"
      },
      "source": [
        "neptune_logger.experiment.log_artifact(PLT_NAME)\n",
        "pandas.DataFrame(return_object).to_csv(FILE_SAVE_NAME)\n",
        "neptune_logger.experiment.log_artifact(FILE_SAVE_NAME)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6hPBp5UUq68_"
      },
      "source": [
        "last_model_save_pth =os.path.join(model_checkpoint.best_model_path,'last.pt')\n",
        "trainer.save_checkpoint(last_model_save_pth)\n",
        "model_name = 'checkpoints/last.pt'\n",
        "neptune_logger.experiment.log_artifact(last_model_save_pth, model_name)\n"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}