Upload GPT_Neo_125M_Grug_Training.ipynb

GPT_Neo_125M_Grug_Training.ipynb

@@ -0,0 +1,116 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "provenance": []
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "language_info": {
+      "name": "python"
+    },
+    "accelerator": "GPU",
+    "gpuClass": "standard"
+  },
+  "cells": [
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "u7RAqjzj4ylm"
+      },
+      "outputs": [],
+      "source": [
+        "!pip install happytransformer"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "from happytransformer import HappyGeneration\n",
+        "\n",
+        "happy_gen = HappyGeneration(\"GPT-NEO\", \"EleutherAI/gpt-neo-125M\")"
+      ],
+      "metadata": {
+        "id": "4V9hd8bQ41HD"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "!wget https://huggingface.co/datasets/DarwinAnim8or/grug/resolve/main/grug-training.txt"
+      ],
+      "metadata": {
+        "id": "hz3fzo5W9ppf"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "from happytransformer import GENTrainArgs \n",
+        "\n",
+        "args = GENTrainArgs(learning_rate =1e-5, num_train_epochs = 2)\n",
+        "happy_gen.train(\"grug-training.txt\", args=args)"
+      ],
+      "metadata": {
+        "id": "Yl4wNVvK5Bex"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "from happytransformer import GENSettings\n",
+        "args_top_k = GENSettings(no_repeat_ngram_size=3, do_sample=True,top_k=50, temperature=0.7, max_length=50, early_stopping=False)"
+      ],
+      "metadata": {
+        "id": "LXi7hXFtBLpN"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "result = happy_gen.generate_text(\"\"\"Person: \"Hello grug\"\n",
+        "Grug: \"hello person\"\n",
+        "###\n",
+        "Person: \"how are you grug\"\n",
+        "Grug: \"grug doing ok. grug find many berry. good for tribe.\"\n",
+        "###\n",
+        "Person: \"what does grug think of new spear weapon?\"\n",
+        "Grug: \"grug no like new spear weapon. grug stick bigger. spear too small, break easy\"\n",
+        "###\n",
+        "Person: \"what does grug think of football?\"\n",
+        "Grug: \\\"\"\"\", args=args_top_k)\n",
+        "#print(result)\n",
+        "print(result.text)"
+      ],
+      "metadata": {
+        "id": "ih4KihPy_U_h"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "#To save the model, run this cell.\n",
+        "happy_gen.save(\"gpt-grug-125m-epoch4/\")"
+      ],
+      "metadata": {
+        "id": "LFUPtXAo_dTz"
+      },
+      "execution_count": null,
+      "outputs": []
+    }
+  ]
+}