content

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "HyperLSTM",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AYV_dMVDxyc2"
      },
      "source": [
        "[![Github](https://img.shields.io/github/stars/labmlai/annotated_deep_learning_paper_implementations?style=social)](https://github.com/labmlai/annotated_deep_learning_paper_implementations)\n",
        "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/labmlai/annotated_deep_learning_paper_implementations/blob/master/labml_nn/hypernetworks/experiment.ipynb)                    \n",
        "\n",
        "## HyperLSTM\n",
        "\n",
        "This is an experiment training Shakespear dataset with HyperLSTM from paper HyperNetworks."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZCzmCrAIVg0L"
      },
      "source": [
        "!pip install labml-nn"
      ],
      "outputs": [],
      "execution_count": null
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0hJXx_g0wS2C"
      },
      "source": [
        "from labml import experiment\n",
        "from labml_nn.hypernetworks.experiment import Configs"
      ],
      "outputs": [],
      "execution_count": null
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WQ8VGpMGwZuj"
      },
      "source": [
        "# Create experiment\n",
        "experiment.create(name=\"hyper_lstm\", comment='')\n",
        "# Create configs\n",
        "conf = Configs()\n",
        "# Load configurations\n",
        "experiment.configs(conf,\n",
        "                    # A dictionary of configurations to override\n",
        "                    {'tokenizer': 'character',\n",
        "                    'text': 'tiny_shakespeare',\n",
        "                    'optimizer.learning_rate': 2.5e-4,\n",
        "                    'optimizer.optimizer': 'Adam',\n",
        "                    'prompt': 'It is',\n",
        "                    'prompt_separator': '',\n",
        "\n",
        "                    'rnn_model': 'hyper_lstm',\n",
        "\n",
        "                    'train_loader': 'shuffled_train_loader',\n",
        "                    'valid_loader': 'shuffled_valid_loader',\n",
        "\n",
        "                    'seq_len': 512,\n",
        "                    'epochs': 128,\n",
        "                    'batch_size': 2,\n",
        "                    'inner_iterations': 25})\n",
        "\n",
        "\n",
        "# Set models for saving and loading\n",
        "experiment.add_pytorch_models({'model': conf.model})\n",
        "\n",
        "conf.init()"
      ],
      "outputs": [],
      "execution_count": null
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f07vAOaHwumr"
      },
      "source": [
        "# Start the experiment\n",
        "with experiment.start():\n",
        "    # `TrainValidConfigs.run`\n",
        "    conf.run()"
      ],
      "outputs": [],
      "execution_count": null
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "crH6MzKmw-SY"
      },
      "source": [],
      "outputs": [],
      "execution_count": null
    },
    {
      "cell_type": "code",
      "source": [
        "/content/sample_data"
      ],
      "metadata": {
        "id": "gSTod2tdmxJq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "VSGzg0kzm1QH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "VgJUq7Wmm28Y"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}