intent classification

Notebook content (JSON format):

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AcooU1NoTWl4"
      },
      "source": [
        "# Intent Detection: Identify user intent efficiently with a custom classifier\n",
        "\n",
        "In this cookbook, we will explore classification for intent detection and classification using our Classifier Factory.\n",
        "\n",
        "To keep things straightforward, we will concentrate on a particular example that involves single-target classification."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7lZo_t--T1FV"
      },
      "source": [
        "## Dataset\n",
        "We will use a subset of the [mteb/amazon_massive_intent](https://huggingface.co/datasets/mteb/amazon_massive_intent) dataset. This subset includes an intent for different user requests.\n",
        "\n",
        "### Subset\n",
        "Let's download and prepare the subset. We will install the `datasets` library and load the dataset."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "YAwwzneFhci9"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "!pip install datasets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Rd4Lo0ixXXhg",
        "cellView": "form"
      },
      "outputs": [],
      "source": [
        "# @title Loading and preparing subset\n",
        "%%capture\n",
        "# Import necessary libraries\n",
        "from datasets import load_dataset\n",
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Load the entire amazon_massive_intent dataset\n",
        "dataset = load_dataset(\"mteb/amazon_massive_intent\")\n",
        "\n",
        "# Filter for English language samples\n",
        "train_samples = dataset[\"train\"].filter(lambda x: x[\"lang\"] == \"en\")\n",
        "\n",
        "# Select only the required columns\n",
        "train_samples = train_samples.select_columns([\"text\", \"label_text\"])\n",
        "\n",
        "# Convert to pandas DataFrame\n",
        "train_df = pd.DataFrame(train_samples)\n",
        "\n",
        "# Function to remove labels with less than 200 samples and limit each label to 600 samples\n",
        "def process_labels(df, min_samples=200, max_samples=600):\n",
        "    label_counts = df[\"label_text\"].value_counts()\n",
        "    labels_to_keep = label_counts[label_counts >= min_samples].index\n",
        "    df = df[df[\"label_text\"].isin(labels_to_keep)]\n",
        "\n",
        "    # Limit each label to max_samples\n",
        "    balanced_df = pd.DataFrame()\n",
        "    for label in labels_to_keep:\n",
        "        label_samples = df[df[\"label_text\"] == label].sample(\n",
        "            n=min(len(df[df[\"label_text\"] == label]), max_samples), random_state=42\n",
        "        )\n",
        "        balanced_df = pd.concat([balanced_df, label_samples])\n",
        "\n",
        "    return balanced_df\n",
        "\n",
        "# Process labels in the training dataset\n",
        "train_df = process_labels(train_df)\n",
        "\n",
        "# Split the training data into train, validation, and test sets\n",
        "train_df, temp_df = train_test_split(\n",
        "    train_df, test_size=0.2, random_state=42, stratify=train_df[\"label_text\"]\n",
        ")\n",
        "validation_df, test_df = train_test_split(\n",
        "    temp_df, test_size=0.5, random_state=42, stratify=temp_df[\"label_text\"]\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "oodmsDzPeN9j",
        "outputId": "dd3e34fd-071c-4527-d6c7-76f521eacc55"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>label_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>888</th>\n",
              "      <td>what's the weather forecast for today</td>\n",
              "      <td>weather_query</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11290</th>\n",
              "      <td>send the email to this new email address</td>\n",
              "      <td>email_sendemail</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6194</th>\n",
              "      <td>hey please provide information about events by...</td>\n",
              "      <td>calendar_query</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5663</th>\n",
              "      <td>schedule my meeting with mr. john hopkins tomo...</td>\n",
              "      <td>calendar_set</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2496</th>\n",
              "      <td>please play some lady gaga</td>\n",
              "      <td>play_music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3674</th>\n",
              "      <td>do i need to take an umbrella today</td>\n",
              "      <td>weather_query</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4610</th>\n",
              "      <td>what are all the upcoming events</td>\n",
              "      <td>calendar_query</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2646</th>\n",
              "      <td>let's listen to a playlist</td>\n",
              "      <td>play_music</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4493</th>\n",
              "      <td>hey i just lost my wallet today</td>\n",
              "      <td>general_quirky</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4401</th>\n",
              "      <td>what is the story of the movie titanic</td>\n",
              "      <td>general_quirky</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>665 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                    text       label_text\n",
              "888                what's the weather forecast for today    weather_query\n",
              "11290           send the email to this new email address  email_sendemail\n",
              "6194   hey please provide information about events by...   calendar_query\n",
              "5663   schedule my meeting with mr. john hopkins tomo...     calendar_set\n",
              "2496                          please play some lady gaga       play_music\n",
              "...                                                  ...              ...\n",
              "3674                 do i need to take an umbrella today    weather_query\n",
              "4610                    what are all the upcoming events   calendar_query\n",
              "2646                          let's listen to a playlist       play_music\n",
              "4493                     hey i just lost my wallet today   general_quirky\n",
              "4401              what is the story of the movie titanic   general_quirky\n",
              "\n",
              "[665 rows x 2 columns]"
            ]
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Display the test DataFrame to verify\n",
        "test_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "IduVHQod6peK",
        "outputId": "b041b87b-303f-4c3e-dd63-85bbdd9409b6"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1200x1800 with 3 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# @title Data distribution\n",
        "import matplotlib.pyplot as plt\n",
        "from collections import Counter\n",
        "\n",
        "# Function to count the number of samples for each label\n",
        "def count_labels(samples):\n",
        "    labels = [sample[\"label_text\"] for sample in samples.to_dict(\"records\")]\n",
        "    return Counter(labels)\n",
        "\n",
        "# Count labels for each dataset\n",
        "train_label_counts = count_labels(train_df)\n",
        "validation_label_counts = count_labels(validation_df)\n",
        "test_label_counts = count_labels(test_df)\n",
        "\n",
        "# Create a single figure with subplots for bar charts\n",
        "fig, axes = plt.subplots(3, 1, figsize=(12, 18))\n",
        "\n",
        "# Plot the bar charts for label distribution\n",
        "def plot_label_distribution(ax, label_counts, title):\n",
        "    # Sort labels by count in descending order\n",
        "    sorted_labels, sorted_counts = zip(\n",
        "        *sorted(label_counts.items(), key=lambda x: x[1], reverse=True)\n",
        "    )\n",
        "    ax.bar(sorted_labels, sorted_counts, color=\"skyblue\")\n",
        "    ax.set_ylabel(\"Count\")\n",
        "    ax.set_title(title)\n",
        "    ax.tick_params(axis=\"x\", rotation=45)\n",
        "\n",
        "# Plot label distribution for each dataset\n",
        "plot_label_distribution(\n",
        "    axes[0], train_label_counts, \"Train Samples (Label Distribution)\"\n",
        ")\n",
        "plot_label_distribution(\n",
        "    axes[1], validation_label_counts, \"Validation Samples (Label Distribution)\"\n",
        ")\n",
        "plot_label_distribution(axes[2], test_label_counts, \"Test Samples (Label Distribution)\")\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9HKm6rNHUF_n"
      },
      "source": [
        "## Format Data\n",
        "\n",
        "Now that we have loaded our dataset, we will convert it to the proper desired format to upload for training.\n",
        "\n",
        "The data will be converted to a **JSONL** format as follows:\n",
        "```json\n",
        "{\"text\": \"place a birthday party with ale ross and amy in my calendar\", \"labels\": {\"intent\": \"calendar_set\"}}\n",
        "{\"text\": \"new music tracks\", \"labels\": {\"intent\": \"play_music\"}}\n",
        "{\"text\": \"get me the details of upcoming oscar two thousand and seventeen\", \"labels\": {\"intent\": \"calendar_query\"}}\n",
        "{\"text\": \"is there any event today in my calendar\", \"labels\": {\"intent\": \"calendar_query\"}}\n",
        "{\"text\": \"send email to mommy that i'll be going the party\", \"labels\": {\"intent\": \"email_sendemail\"}}\n",
        "...\n",
        "```\n",
        "With an example of a label being:\n",
        "```json\n",
        "\"labels\": {\n",
        "  \"intent\": \"email_sendemail\"\n",
        "}\n",
        "```\n",
        "For **single-target** classification."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ltHoh_QXJh0n",
        "outputId": "b5ab05ce-92b9-4b0d-c8b6-0bbe51388094"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 5313/5313 [00:00<00:00, 66090.92it/s]\n",
            "100%|██████████| 664/664 [00:00<00:00, 71039.12it/s]\n",
            "100%|██████████| 665/665 [00:00<00:00, 74006.00it/s]\n"
          ]
        }
      ],
      "source": [
        "from tqdm import tqdm\n",
        "import json\n",
        "\n",
        "def dataset_to_jsonl(dataset, output_file):\n",
        "    # Extract the unique labels from the dataset\n",
        "    unique_labels = dataset[\"label_text\"].unique()\n",
        "\n",
        "    # Open the output file in write mode\n",
        "    with open(output_file, \"w\") as f:\n",
        "        # Iterate over each row in the dataset\n",
        "        for _, row in tqdm(dataset.iterrows(), total=dataset.shape[0]):\n",
        "            # Extract the text and label from the row\n",
        "            text = row[\"text\"]\n",
        "            intent = row[\"label_text\"]\n",
        "\n",
        "            # Create the JSON object with the desired structure\n",
        "            json_object = {\"text\": text, \"labels\": {\"intent\": intent}}\n",
        "\n",
        "            # Write the JSON object to the file as a JSON line\n",
        "            f.write(json.dumps(json_object) + \"\\n\")\n",
        "\n",
        "# Save files\n",
        "dataset_to_jsonl(train_df, \"train.jsonl\")\n",
        "dataset_to_jsonl(validation_df, \"validation.jsonl\")\n",
        "dataset_to_jsonl(test_df, \"test.jsonl\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "82fzZ0kJV0VS"
      },
      "source": [
        "The data was converted and saved properly. We can now train our model.\n",
        "\n",
        "## Training\n",
        "There are two methods to train the model: either upload and train via [la platforme](https://console.mistral.ai/build/finetuned-models) or via the [API](https://classifier-factory.platform-docs-9m1.pages.dev/capabilities/finetuning/classifier_factory/).\n",
        "\n",
        "First, we need to install `mistralai`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Td3wp01pWJkC"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "!pip install mistralai"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QGiC7aod01AD"
      },
      "source": [
        "And setup our client, you can create an API key [here](https://console.mistral.ai/api-keys/)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Unv2sTn200PM"
      },
      "outputs": [],
      "source": [
        "from mistralai import Mistral\n",
        "import os\n",
        "\n",
        "# Set the API key for Mistral\n",
        "api_key = \"API_KEY\"\n",
        "\n",
        "# Set your Weights and Biases key\n",
        "wandb_key = \"WANDB_KEY\"\n",
        "\n",
        "# Initialize the Mistral client\n",
        "client = Mistral(api_key=api_key)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "52AJ-7Jo1Jus"
      },
      "source": [
        "We will upload 2 files, the training set and the validation set ( optional ) that will be used for validation loss."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "cWbVjdDju0RR"
      },
      "outputs": [],
      "source": [
        "# Upload the training data\n",
        "training_data = client.files.upload(\n",
        "    file={\n",
        "        \"file_name\": \"train.jsonl\",\n",
        "        \"content\": open(\"train.jsonl\", \"rb\"),\n",
        "    }\n",
        ")\n",
        "\n",
        "# Upload the validation data\n",
        "validation_data = client.files.upload(\n",
        "    file={\n",
        "        \"file_name\": \"validation.jsonl\",\n",
        "        \"content\": open(\"validation.jsonl\", \"rb\"),\n",
        "    }\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BwdgEY5v1Ut5"
      },
      "source": [
        "With the data uploaded, we can create a job.\n",
        "\n",
        "We allow users to keep track of aconsiderable amount of metrics via our Weights and Biases integration that we strongly recommend, you can make use of it by providing the project name and your key."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "x98lTm6Vu62p",
        "outputId": "e7ec1e01-10c6-4ead-d914-589fd2de8781"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "{\n",
            "    \"id\": \"db22ea7e-1895-4309-92d9-9ac881b1b117\",\n",
            "    \"auto_start\": false,\n",
            "    \"model\": \"ministral-3b-latest\",\n",
            "    \"status\": \"QUEUED\",\n",
            "    \"created_at\": 1744814706,\n",
            "    \"modified_at\": 1744814706,\n",
            "    \"training_files\": [\n",
            "        \"c36d490e-d2f8-4f98-9679-1c55625e09d5\"\n",
            "    ],\n",
            "    \"hyperparameters\": {\n",
            "        \"training_steps\": 100,\n",
            "        \"learning_rate\": 4e-05,\n",
            "        \"weight_decay\": 0.1,\n",
            "        \"warmup_fraction\": 0.05,\n",
            "        \"epochs\": null,\n",
            "        \"seq_len\": 16384\n",
            "    },\n",
            "    \"validation_files\": [\n",
            "        \"4b62db5b-47b8-42fb-a782-89457766cff7\"\n",
            "    ],\n",
            "    \"fine_tuned_model\": null,\n",
            "    \"suffix\": null,\n",
            "    \"integrations\": [\n",
            "        {\n",
            "            \"project\": \"intent-classifier\",\n",
            "            \"name\": null,\n",
            "            \"run_name\": null,\n",
            "            \"url\": null\n",
            "        }\n",
            "    ],\n",
            "    \"trained_tokens\": null,\n",
            "    \"metadata\": {\n",
            "        \"expected_duration_seconds\": null,\n",
            "        \"cost\": 0.0,\n",
            "        \"cost_currency\": null,\n",
            "        \"train_tokens_per_step\": null,\n",
            "        \"train_tokens\": null,\n",
            "        \"data_tokens\": null,\n",
            "        \"estimated_start_time\": null\n",
            "    }\n",
            "}\n"
          ]
        }
      ],
      "source": [
        "# Create a fine-tuning job\n",
        "created_job = client.fine_tuning.jobs.create(\n",
        "    model=\"ministral-3b-latest\",\n",
        "    job_type=\"classifier\",\n",
        "    training_files=[{\"file_id\": training_data.id, \"weight\": 1}],\n",
        "    validation_files=[validation_data.id],\n",
        "    hyperparameters={\"training_steps\": 100, \"learning_rate\": 0.00004},\n",
        "    auto_start=False,\n",
        "    integrations=[\n",
        "        {\n",
        "            \"project\": \"intent-classifier\",\n",
        "            \"api_key\": wandb_key,\n",
        "        }\n",
        "    ]\n",
        ")\n",
        "print(json.dumps(created_job.model_dump(), indent=4))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eBN3I8CE1eNT"
      },
      "source": [
        "Once the job is created, we can review details such as the number of epochs and other relevant information. This allows us to make informed decisions before initiating the job.\n",
        "\n",
        "We'll retrieve the job and wait for it to complete the validation process before starting. This validation step ensures the job is ready to begin."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kXtqDouNd_40",
        "outputId": "d47be517-4b34-4c0f-b607-a45e5316e5c0"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "{\n",
            "    \"id\": \"db22ea7e-1895-4309-92d9-9ac881b1b117\",\n",
            "    \"auto_start\": false,\n",
            "    \"model\": \"ministral-3b-latest\",\n",
            "    \"status\": \"VALIDATED\",\n",
            "    \"created_at\": 1744814706,\n",
            "    \"modified_at\": 1744814709,\n",
            "    \"training_files\": [\n",
            "        \"c36d490e-d2f8-4f98-9679-1c55625e09d5\"\n",
            "    ],\n",
            "    \"hyperparameters\": {\n",
            "        \"training_steps\": 100,\n",
            "        \"learning_rate\": 4e-05,\n",
            "        \"weight_decay\": 0.1,\n",
            "        \"warmup_fraction\": 0.05,\n",
            "        \"epochs\": 47.74101432172152,\n",
            "        \"seq_len\": 16384\n",
            "    },\n",
            "    \"classifier_targets\": [\n",
            "        {\n",
            "            \"name\": \"intent\",\n",
            "            \"labels\": [\n",
            "                \"social_post\",\n",
            "                \"email_sendemail\",\n",
            "                \"datetime_query\",\n",
            "                \"play_music\",\n",
            "                \"email_query\",\n",
            "                \"news_query\",\n",
            "                \"weather_query\",\n",
            "                \"calendar_query\",\n",
            "                \"general_quirky\",\n",
            "                \"qa_factoid\",\n",
            "                \"play_radio\",\n",
            "                \"calendar_set\",\n",
            "                \"qa_definition\",\n",
            "                \"calendar_remove\",\n",
            "                \"transport_query\",\n",
            "                \"cooking_recipe\"\n",
            "            ]\n",
            "        }\n",
            "    ],\n",
            "    \"validation_files\": [\n",
            "        \"4b62db5b-47b8-42fb-a782-89457766cff7\"\n",
            "    ],\n",
            "    \"fine_tuned_model\": null,\n",
            "    \"suffix\": null,\n",
            "    \"integrations\": [\n",
            "        {\n",
            "            \"project\": \"intent-classifier\",\n",
            "            \"name\": null,\n",
            "            \"run_name\": null,\n",
            "            \"url\": null\n",
            "        }\n",
            "    ],\n",
            "    \"trained_tokens\": null,\n",
            "    \"metadata\": {\n",
            "        \"expected_duration_seconds\": 400,\n",
            "        \"cost\": 3.28,\n",
            "        \"cost_currency\": \"EUR\",\n",
            "        \"train_tokens_per_step\": 65536,\n",
            "        \"train_tokens\": 6553600,\n",
            "        \"data_tokens\": 137274,\n",
            "        \"estimated_start_time\": null\n",
            "    },\n",
            "    \"events\": [\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814706,\n",
            "            \"data\": {\n",
            "                \"status\": \"QUEUED\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814709,\n",
            "            \"data\": {\n",
            "                \"status\": \"VALIDATING\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814709,\n",
            "            \"data\": {\n",
            "                \"status\": \"VALIDATED\"\n",
            "            }\n",
            "        }\n",
            "    ],\n",
            "    \"checkpoints\": []\n",
            "}\n"
          ]
        }
      ],
      "source": [
        "# Retrieve the job details\n",
        "retrieved_job = client.fine_tuning.jobs.get(job_id=created_job.id)\n",
        "print(json.dumps(retrieved_job.model_dump(), indent=4))\n",
        "\n",
        "import time\n",
        "from IPython.display import clear_output\n",
        "\n",
        "# Wait for the job to be validated\n",
        "while retrieved_job.status not in [\"VALIDATED\"]:\n",
        "    retrieved_job = client.fine_tuning.jobs.get(job_id=created_job.id)\n",
        "\n",
        "    clear_output(wait=True)  # Clear the previous output (User Friendly)\n",
        "    print(json.dumps(retrieved_job.model_dump(), indent=4))\n",
        "    time.sleep(1)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YaNg0v9s0kk7"
      },
      "source": [
        "We can now run the job."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "q6k7hW5cfEG-",
        "outputId": "82e233a0-c5d1-4ce1-d048-ac1049231284"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "{\n",
            "    \"id\": \"db22ea7e-1895-4309-92d9-9ac881b1b117\",\n",
            "    \"auto_start\": false,\n",
            "    \"model\": \"ministral-3b-latest\",\n",
            "    \"status\": \"QUEUED\",\n",
            "    \"created_at\": 1744814706,\n",
            "    \"modified_at\": 1744814712,\n",
            "    \"training_files\": [\n",
            "        \"c36d490e-d2f8-4f98-9679-1c55625e09d5\"\n",
            "    ],\n",
            "    \"hyperparameters\": {\n",
            "        \"training_steps\": 100,\n",
            "        \"learning_rate\": 4e-05,\n",
            "        \"weight_decay\": 0.1,\n",
            "        \"warmup_fraction\": 0.05,\n",
            "        \"epochs\": 47.74101432172152,\n",
            "        \"seq_len\": 16384\n",
            "    },\n",
            "    \"classifier_targets\": [\n",
            "        {\n",
            "            \"name\": \"intent\",\n",
            "            \"labels\": [\n",
            "                \"social_post\",\n",
            "                \"email_sendemail\",\n",
            "                \"datetime_query\",\n",
            "                \"play_music\",\n",
            "                \"email_query\",\n",
            "                \"news_query\",\n",
            "                \"weather_query\",\n",
            "                \"calendar_query\",\n",
            "                \"general_quirky\",\n",
            "                \"qa_factoid\",\n",
            "                \"play_radio\",\n",
            "                \"calendar_set\",\n",
            "                \"qa_definition\",\n",
            "                \"calendar_remove\",\n",
            "                \"transport_query\",\n",
            "                \"cooking_recipe\"\n",
            "            ]\n",
            "        }\n",
            "    ],\n",
            "    \"validation_files\": [\n",
            "        \"4b62db5b-47b8-42fb-a782-89457766cff7\"\n",
            "    ],\n",
            "    \"fine_tuned_model\": null,\n",
            "    \"suffix\": null,\n",
            "    \"integrations\": [\n",
            "        {\n",
            "            \"project\": \"intent-classifier\",\n",
            "            \"name\": null,\n",
            "            \"run_name\": null,\n",
            "            \"url\": null\n",
            "        }\n",
            "    ],\n",
            "    \"trained_tokens\": null,\n",
            "    \"metadata\": {\n",
            "        \"expected_duration_seconds\": 400,\n",
            "        \"cost\": 3.28,\n",
            "        \"cost_currency\": \"EUR\",\n",
            "        \"train_tokens_per_step\": 65536,\n",
            "        \"train_tokens\": 6553600,\n",
            "        \"data_tokens\": 137274,\n",
            "        \"estimated_start_time\": null\n",
            "    },\n",
            "    \"events\": [\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814706,\n",
            "            \"data\": {\n",
            "                \"status\": \"QUEUED\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814709,\n",
            "            \"data\": {\n",
            "                \"status\": \"VALIDATING\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814709,\n",
            "            \"data\": {\n",
            "                \"status\": \"VALIDATED\"\n",
            "            }\n",
            "        }\n",
            "    ],\n",
            "    \"checkpoints\": []\n",
            "}\n"
          ]
        }
      ],
      "source": [
        "# Start the fine-tuning job\n",
        "client.fine_tuning.jobs.start(job_id=created_job.id)\n",
        "\n",
        "# Retrieve the job details again\n",
        "retrieved_job = client.fine_tuning.jobs.get(job_id=created_job.id)\n",
        "print(json.dumps(retrieved_job.model_dump(), indent=4))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "46TXC-Jf2Mdd"
      },
      "source": [
        "The job is now starting. Let's keep track of the status and print the information.\n",
        "\n",
        "We highly recommend making use of our Weights and Biases integration to keep track of multiple metrics.\n",
        "\n",
        "### WANDB\n",
        "\n",
        "**Training:**\n",
        "\n",
        "![intent-train-loss.png](https://raw.githubusercontent.com/mistralai/cookbook/refs/heads/main/mistral/classifier_factory/intent-train-loss.png)\n",
        "\n",
        "**Eval/Validation:**\n",
        "\n",
        "![intent-validation-loss.png](https://raw.githubusercontent.com/mistralai/cookbook/refs/heads/main/mistral/classifier_factory/intent-validation-loss.png)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2Edzh8DWvw5V",
        "outputId": "5820cc88-b536-4dbe-a447-ee5e40787388"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "{\n",
            "    \"id\": \"db22ea7e-1895-4309-92d9-9ac881b1b117\",\n",
            "    \"auto_start\": false,\n",
            "    \"model\": \"ministral-3b-latest\",\n",
            "    \"status\": \"SUCCESS\",\n",
            "    \"created_at\": 1744814706,\n",
            "    \"modified_at\": 1744814940,\n",
            "    \"training_files\": [\n",
            "        \"c36d490e-d2f8-4f98-9679-1c55625e09d5\"\n",
            "    ],\n",
            "    \"hyperparameters\": {\n",
            "        \"training_steps\": 100,\n",
            "        \"learning_rate\": 4e-05,\n",
            "        \"weight_decay\": 0.1,\n",
            "        \"warmup_fraction\": 0.05,\n",
            "        \"epochs\": 47.74101432172152,\n",
            "        \"seq_len\": 16384\n",
            "    },\n",
            "    \"classifier_targets\": [\n",
            "        {\n",
            "            \"name\": \"intent\",\n",
            "            \"labels\": [\n",
            "                \"social_post\",\n",
            "                \"email_sendemail\",\n",
            "                \"datetime_query\",\n",
            "                \"play_music\",\n",
            "                \"email_query\",\n",
            "                \"news_query\",\n",
            "                \"weather_query\",\n",
            "                \"calendar_query\",\n",
            "                \"general_quirky\",\n",
            "                \"qa_factoid\",\n",
            "                \"play_radio\",\n",
            "                \"calendar_set\",\n",
            "                \"qa_definition\",\n",
            "                \"calendar_remove\",\n",
            "                \"transport_query\",\n",
            "                \"cooking_recipe\"\n",
            "            ]\n",
            "        }\n",
            "    ],\n",
            "    \"validation_files\": [\n",
            "        \"4b62db5b-47b8-42fb-a782-89457766cff7\"\n",
            "    ],\n",
            "    \"fine_tuned_model\": \"ft:classifier:ministral-3b-latest:8e2706f0:20250416:db22ea7e\",\n",
            "    \"suffix\": null,\n",
            "    \"integrations\": [\n",
            "        {\n",
            "            \"project\": \"intent-classifier\",\n",
            "            \"name\": null,\n",
            "            \"run_name\": null,\n",
            "            \"url\": \"https://wandb.ai/mistral-ai/intent-classifier/runs/tpyu9twr\"\n",
            "        }\n",
            "    ],\n",
            "    \"trained_tokens\": 1638400,\n",
            "    \"metadata\": {\n",
            "        \"expected_duration_seconds\": 400,\n",
            "        \"cost\": 3.28,\n",
            "        \"cost_currency\": \"EUR\",\n",
            "        \"train_tokens_per_step\": 65536,\n",
            "        \"train_tokens\": 6553600,\n",
            "        \"data_tokens\": 137274,\n",
            "        \"estimated_start_time\": null\n",
            "    },\n",
            "    \"events\": [\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814706,\n",
            "            \"data\": {\n",
            "                \"status\": \"QUEUED\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814709,\n",
            "            \"data\": {\n",
            "                \"status\": \"VALIDATING\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814709,\n",
            "            \"data\": {\n",
            "                \"status\": \"VALIDATED\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814716,\n",
            "            \"data\": {\n",
            "                \"status\": \"RUNNING\"\n",
            "            }\n",
            "        },\n",
            "        {\n",
            "            \"name\": \"status-updated\",\n",
            "            \"created_at\": 1744814940,\n",
            "            \"data\": {\n",
            "                \"status\": \"SUCCESS\"\n",
            "            }\n",
            "        }\n",
            "    ],\n",
            "    \"checkpoints\": [\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000191,\n",
            "                \"valid_loss\": 0.000625,\n",
            "                \"valid_mean_token_accuracy\": 1.000433\n",
            "            },\n",
            "            \"step_number\": 100,\n",
            "            \"created_at\": 1744814919\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000181,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 99,\n",
            "            \"created_at\": 1744814909\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000221,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 98,\n",
            "            \"created_at\": 1744814909\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000189,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 97,\n",
            "            \"created_at\": 1744814909\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000186,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 96,\n",
            "            \"created_at\": 1744814909\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000192,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 95,\n",
            "            \"created_at\": 1744814909\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000174,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 94,\n",
            "            \"created_at\": 1744814899\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000196,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 93,\n",
            "            \"created_at\": 1744814899\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.000199,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 92,\n",
            "            \"creat\n",
            "[...]\n",
            "n_loss\": 0.001359,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 20,\n",
            "            \"created_at\": 1744814769\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001401,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 19,\n",
            "            \"created_at\": 1744814769\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001442,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 18,\n",
            "            \"created_at\": 1744814769\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001453,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 17,\n",
            "            \"created_at\": 1744814769\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001489,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 16,\n",
            "            \"created_at\": 1744814769\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001482,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 15,\n",
            "            \"created_at\": 1744814759\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001528,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 14,\n",
            "            \"created_at\": 1744814759\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001536,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 13,\n",
            "            \"created_at\": 1744814759\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001585,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 12,\n",
            "            \"created_at\": 1744814759\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001623,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 11,\n",
            "            \"created_at\": 1744814759\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001646,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 10,\n",
            "            \"created_at\": 1744814749\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001688,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 9,\n",
            "            \"created_at\": 1744814749\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001721,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 8,\n",
            "            \"created_at\": 1744814749\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001749,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 7,\n",
            "            \"created_at\": 1744814749\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001787,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 6,\n",
            "            \"created_at\": 1744814749\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001819,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 5,\n",
            "            \"created_at\": 1744814749\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001887,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 4,\n",
            "            \"created_at\": 1744814739\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001917,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 3,\n",
            "            \"created_at\": 1744814739\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.001893,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 2,\n",
            "            \"created_at\": 1744814739\n",
            "        },\n",
            "        {\n",
            "            \"metrics\": {\n",
            "                \"train_loss\": 0.00189,\n",
            "                \"valid_loss\": 0.0,\n",
            "                \"valid_mean_token_accuracy\": 0.0\n",
            "            },\n",
            "            \"step_number\": 1,\n",
            "            \"created_at\": 1744814739\n",
            "        }\n",
            "    ]\n",
            "}\n"
          ]
        }
      ],
      "source": [
        "# Wait for the job to complete\n",
        "while retrieved_job.status in [\"QUEUED\", \"RUNNING\"]:\n",
        "    retrieved_job = client.fine_tuning.jobs.get(job_id=created_job.id)\n",
        "\n",
        "    clear_output(wait=True)  # Clear the previous output (User Friendly)\n",
        "    job_info = json.dumps(retrieved_job.model_dump(), indent=4)\n",
        "    if len(job_info) > 10000:\n",
        "        print(job_info[:5000] + \"\\n[...]\\n\" + job_info[-5000:])\n",
        "    else:\n",
        "        print(job_info)\n",
        "    time.sleep(5)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DG5kiDgV2fkC"
      },
      "source": [
        "### Inference\n",
        "Our model is trained and ready for use! Let's test it on a sample from our test set!"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OU7_6LsEwdDL",
        "outputId": "d71bbf48-f647-4695-f462-3472bdb515be"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Text: what's the weather forecast for today\n",
            "Classifier Response: {\n",
            "    \"id\": \"4c521c6507674c7da7fb8c2fa78bf8ae\",\n",
            "    \"model\": \"ft:classifier:ministral-3b-latest:8e2706f0:20250416:db22ea7e\",\n",
            "    \"results\": [\n",
            "        {\n",
            "            \"intent\": {\n",
            "                \"scores\": {\n",
            "                    \"social_post\": 3.0807409530098084e-06,\n",
            "                    \"email_sendemail\": 1.2064355132679339e-06,\n",
            "                    \"datetime_query\": 0.0008676558500155807,\n",
            "                    \"play_music\": 4.579112555802567e-07,\n",
            "                    \"email_query\": 5.323042842064751e-06,\n",
            "                    \"news_query\": 0.00020608631893992424,\n",
            "                    \"weather_query\": 0.9971635937690735,\n",
            "                    \"calendar_query\": 0.00047916508628986776,\n",
            "                    \"general_quirky\": 0.0006549410172738135,\n",
            "                    \"qa_factoid\": 0.0001449998380849138,\n",
            "                    \"play_radio\": 1.3382128599914722e-05,\n",
            "                    \"calendar_set\": 9.489350304647814e-06,\n",
            "                    \"qa_definition\": 9.003245213534683e-05,\n",
            "                    \"calendar_remove\": 2.98595659842249e-06,\n",
            "                    \"transport_query\": 0.0002646200591698289,\n",
            "                    \"cooking_recipe\": 9.289039007853717e-05\n",
            "                }\n",
            "            }\n",
            "        }\n",
            "    ]\n",
            "}\n"
          ]
        }
      ],
      "source": [
        "# Load the test samples\n",
        "with open(\"test.jsonl\", \"r\") as f:\n",
        "    test_samples = [json.loads(l) for l in f.readlines()]\n",
        "\n",
        "# Classify the first test sample\n",
        "classifier_response = client.classifiers.classify(\n",
        "    model=retrieved_job.fine_tuned_model,\n",
        "    inputs=[test_samples[0][\"text\"]],\n",
        ")\n",
        "print(\"Text:\", test_samples[0][\"text\"])\n",
        "print(\"Classifier Response:\", json.dumps(classifier_response.model_dump(), indent=4))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qLzgYdUmzE1u"
      },
      "source": [
        "The score with the highest result is `weather_query`, with an over 99% score!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "473my4Uv2zAQ"
      },
      "source": [
        "There you have it: a simple guide on how to train your own classifier and use our batch inference.\n",
        "\n",
        "For a more specific multi-label classifier, visit this [cookbook](https://colab.research.google.com/github/mistralai/cookbook/blob/main/mistral/classifier_factory/moderation_classifier.ipynb).\n",
        "\n",
        "For a more product focused in-depth guide on both multi-target, with an evaluation comparison between LLMs and our classifier, visit this [cookbook](https://colab.research.google.com/github/mistralai/cookbook/blob/main/mistral/classifier_factory/product_classification.ipynb)."
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.12.3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}