translate-programming-languages.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Translate programming languages\n",
    "\n",
    "To translate from one programming language to another we can use the comments to specify the source and target languages.\n",
    "\n",
    "## Settings\n",
    "\n",
    "| Setting | Value |\n",
    "| --- | --- |\n",
    "| Model | gpt-35-turbo |\n",
    "| Max tokens | 200 |\n",
    "| Temperature | 0 | \n",
    "| Top p | 1 | \n",
    "| Frequency penalty | 0.0 |\n",
    "| Presence penalty | 0.0 |\n",
    "| Stop sequence | `###` |\n",
    "\n",
    "## Prompt\n",
    "\n",
    "### System\n",
    "\n",
    "```text\n",
    "You are a virtual agent that helps users translate code from one programming language to another.\n",
    "```\n",
    "\n",
    "### User\n",
    "```text\n",
    "##### Translate this function  from Python into C#\n",
    "### Python\n",
    "    \n",
    "    def predict_proba(X: Iterable[str]):\n",
    "        return np.array([predict_one_probas(tweet) for tweet in X])\n",
    "    \n",
    "### C#\n",
    "```"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Install Azure OpenAI Service .NET SDK\n",
    "\n",
    "[Azure.AI.OpenAI NuGet package](https://www.nuget.org/packages/Azure.AI.OpenAI/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div></div><div></div><div><strong>Installed Packages</strong><ul><li><span>Azure.AI.OpenAI, 1.0.0-beta.15</span></li></ul></div></div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#r \"nuget: Azure.AI.OpenAI, 1.0.0-beta.15\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "using System;\n",
    "using Azure.AI.OpenAI;"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Configure Azure OpenAI client"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var AOAI_ENDPOINT = Environment.GetEnvironmentVariable(\"AOAI_ENDPOINT\");\n",
    "var AOAI_KEY = Environment.GetEnvironmentVariable(\"AOAI_KEY\");\n",
    "var AOAI_DEPLOYMENTID = Environment.GetEnvironmentVariable(\"AOAI_DEPLOYMENTID\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var endpoint = new Uri(AOAI_ENDPOINT);\n",
    "var credentials = new Azure.AzureKeyCredential(AOAI_KEY);\n",
    "var openAIClient = new OpenAIClient(endpoint, credentials);"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define prompt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var systemPrompt = \"You are a virtual agent that helps users translate code from one programming language to another.\";\n",
    "\n",
    "var userPrompt = \n",
    "    \"\"\"\n",
    "    ##### Translate this function  from Python into C#\n",
    "    ### Python\n",
    "        \n",
    "        def predict_proba(X: Iterable[str]):\n",
    "            return np.array([predict_one_probas(tweet) for tweet in X])\n",
    "        \n",
    "    ### C#\n",
    "    \"\"\";"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Configure completion options"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var completionOptions = new ChatCompletionsOptions\n",
    "{\n",
    "    DeploymentName=AOAI_DEPLOYMENTID,\n",
    "    MaxTokens=200,\n",
    "    Temperature=0f,\n",
    "    FrequencyPenalty=0.0f,\n",
    "    PresencePenalty=0.0f,\n",
    "    NucleusSamplingFactor=1, // Top P\n",
    "    StopSequences={\"###\"}\n",
    "};"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate completions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "completionOptions.Messages.Add(new ChatRequestSystemMessage(systemPrompt));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "completionOptions.Messages.Add(new ChatRequestUserMessage(userPrompt));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "ChatCompletions response = await openAIClient.GetChatCompletionsAsync(completionOptions);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Here is the equivalent C# code for the given Python function:\n",
       "\n",
       "```csharp\n",
       "using System;\n",
       "using System.Collections.Generic;\n",
       "using System.Linq;\n",
       "\n",
       "public class MyClass\n",
       "{\n",
       "    public double[] PredictProba(IEnumerable<string> X)\n",
       "    {\n",
       "        return X.Select(tweet => PredictOneProbas(tweet)).ToArray();\n",
       "    }\n",
       "\n",
       "    public double PredictOneProbas(string tweet)\n",
       "    {\n",
       "        // Implement the logic for predict_one_probas(tweet) here\n",
       "        // You can replace this with your actual implementation\n",
       "        return 0.0;\n",
       "    }\n",
       "}\n",
       "```\n",
       "\n",
       "In the C# code above, I have created a class `MyClass` with two methods: `PredictProba` and `PredictOneProbas`. The `PredictProba` method takes an `IEnumerable<string>` as input and uses LINQ's `Select` method to apply the `PredictOneProbas` method to each element in the input collection. Finally, it converts the result to an array using `ToArray()`"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "response.Choices.First().Message.Content"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".NET (C#)",
   "language": "C#",
   "name": ".net-csharp"
  },
  "language_info": {
   "name": "polyglot-notebook"
  },
  "polyglot_notebook": {
   "kernelInfo": {
    "defaultKernelName": "csharp",
    "items": [
     {
      "aliases": [],
      "languageName": "csharp",
      "name": "csharp"
     }
    ]
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}