Train_YOLOv8.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7mGmQbAO5pQb"
      },
      "source": [
        "# Setup\n",
        "\n",
        "Pip install `ultralytics` and [dependencies](https://github.com/ultralytics/ultralytics/blob/main/requirements.txt) and check software and hardware."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wbvMlHd_QwMG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "27ca383c-0a97-4679-f1c5-ba843f033de7"
      },
      "source": [
        "%pip install ultralytics\n",
        "import ultralytics\n",
        "ultralytics.checks()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Ultralytics YOLOv8.0.145 🚀 Python-3.10.6 torch-2.0.1+cu118 CUDA:0 (Tesla T4, 15102MiB)\n",
            "Setup complete ✅ (2 CPUs, 12.7 GB RAM, 24.2/78.2 GB disk)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 1. Train\n",
        "\n",
        "Train YOLOv8 please edit attribute **data** to absolute path to **myfiles/yolo-export/yolo.yaml**.\n",
        "\n",
        "You can train yolo on several model sizes.\n",
        "\n",
        "\n",
        "For small model we select nano size yolov8n.pt. More info about models are well documented on the\n",
        "[documentation yolo](https://docs.ultralytics.com/tasks/detect/#models)\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "NgTmdvhh4v-L"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%cd -\n",
        "!yolo train model=yolov8n.pt data=myfiles/yolo-export/yolo.yaml epochs=150"
      ],
      "metadata": {
        "id": "S6yDLJ_R4e3R"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "At the end of training process, in home folder will be folder structure in **runs** folder. Best training weights is store in file **best.pt** which is locate in relative path in **runs/detect/train/weights/**."
      ],
      "metadata": {
        "id": "jIv1UZw75s7b"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 2. Export\n",
        "\n",
        "Export a YOLOv8 model to any supported format below with the `format` argument, i.e. `format=onnx`. See [YOLOv8 Export Docs](https://docs.ultralytics.com/modes/export/) for more information.\n",
        "\n",
        "- 💡 ProTip: Export to [ONNX](https://onnx.ai/) or [OpenVINO](https://docs.openvino.ai/latest/index.html) for up to 3x CPU speedup.  \n",
        "- 💡 ProTip: Export to [TensorRT](https://developer.nvidia.com/tensorrt) for up to 5x GPU speedup.\n",
        "\n",
        "\n",
        "| Format                                                             | `format` Argument | Model                     | Metadata | Arguments                                           |\n",
        "|--------------------------------------------------------------------|-------------------|---------------------------|----------|-----------------------------------------------------|\n",
        "| [PyTorch](https://pytorch.org/)                                    | -                 | `yolov8n.pt`              | ✅        | -                                                   |\n",
        "| [TorchScript](https://pytorch.org/docs/stable/jit.html)            | `torchscript`     | `yolov8n.torchscript`     | ✅        | `imgsz`, `optimize`                                 |\n",
        "| [ONNX](https://onnx.ai/)                                           | `onnx`            | `yolov8n.onnx`            | ✅        | `imgsz`, `half`, `dynamic`, `simplify`, `opset`     |\n",
        "| [OpenVINO](https://docs.openvino.ai/latest/index.html)             | `openvino`        | `yolov8n_openvino_model/` | ✅        | `imgsz`, `half`, `int8`                             |\n",
        "| [TensorRT](https://developer.nvidia.com/tensorrt)                  | `engine`          | `yolov8n.engine`          | ✅        | `imgsz`, `half`, `dynamic`, `simplify`, `workspace` |\n",
        "| [CoreML](https://github.com/apple/coremltools)                     | `coreml`          | `yolov8n.mlpackage`       | ✅        | `imgsz`, `half`, `int8`, `nms`                      |\n",
        "| [TF SavedModel](https://www.tensorflow.org/guide/saved_model)      | `saved_model`     | `yolov8n_saved_model/`    | ✅        | `imgsz`, `keras`, `int8`                            |\n",
        "| [TF GraphDef](https://www.tensorflow.org/api_docs/python/tf/Graph) | `pb`              | `yolov8n.pb`              | ❌        | `imgsz`                                             |\n",
        "| [TF Lite](https://www.tensorflow.org/lite)                         | `tflite`          | `yolov8n.tflite`          | ✅        | `imgsz`, `half`, `int8`                             |\n",
        "| [TF Edge TPU](https://coral.ai/docs/edgetpu/models-intro/)         | `edgetpu`         | `yolov8n_edgetpu.tflite`  | ✅        | `imgsz`                                             |\n",
        "| [TF.js](https://www.tensorflow.org/js)                             | `tfjs`            | `yolov8n_web_model/`      | ✅        | `imgsz`                                             |\n",
        "| [PaddlePaddle](https://github.com/PaddlePaddle)                    | `paddle`          | `yolov8n_paddle_model/`   | ✅        | `imgsz`                                             |\n",
        "| [ncnn](https://github.com/Tencent/ncnn)                            | `ncnn`            | `yolov8n_ncnn_model/`     | ✅        | `imgsz`, `half`                                     |\n"
      ],
      "metadata": {
        "id": "nPZZeNrLCQG6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!yolo export model=runs/detect/train/weights/ format=torchscript"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CYIjW4igCjqD",
        "outputId": "2b65e381-717b-4a6f-d6f5-5254c867f3a4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Ultralytics YOLOv8.0.145 🚀 Python-3.10.6 torch-2.0.1+cu118 CPU (Intel Xeon 2.30GHz)\n",
            "YOLOv8n summary (fused): 168 layers, 3151904 parameters, 0 gradients\n",
            "\n",
            "\u001b[34m\u001b[1mPyTorch:\u001b[0m starting from 'yolov8n.pt' with input shape (1, 3, 640, 640) BCHW and output shape(s) (1, 84, 8400) (6.2 MB)\n",
            "\n",
            "\u001b[34m\u001b[1mTorchScript:\u001b[0m starting export with torch 2.0.1+cu118...\n",
            "\u001b[34m\u001b[1mTorchScript:\u001b[0m export success ✅ 2.8s, saved as 'yolov8n.torchscript' (12.4 MB)\n",
            "\n",
            "Export complete (4.6s)\n",
            "Results saved to \u001b[1m/content\u001b[0m\n",
            "Predict:         yolo predict task=detect model=yolov8n.torchscript imgsz=640 \n",
            "Validate:        yolo val task=detect model=yolov8n.torchscript imgsz=640 data=None \n",
            "Visualize:       https://netron.app\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load YOLOv8n-seg, train it on COCO128-seg for 3 epochs and predict an image with it\n",
        "from ultralytics import YOLO\n",
        "\n",
        "model = YOLO('runs/detect/train/weights/best.pt')  # load a pretrained YOLOv8n segmentation model\n",
        "model('path/to/sample/images.jpg')  # predict on an image"
      ],
      "metadata": {
        "id": "WFPJIQl_L5HT"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}