steamxd-google-colab-training.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "d67XGHBpoNGe"
      },
      "outputs": [],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive', force_remount=True)\n",
        "!ln -s '/content/gdrive/My Drive/' /mydrive\n",
        "%cd /mydrive"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/terminalai/STEAMxD.git\n",
        "!pip install -qr STEAMxD/yolov5/requirements.txt\n",
        "\n",
        "from IPython.core.magic import register_line_cell_magic\n",
        "\n",
        "@register_line_cell_magic\n",
        "def writetemplate(line, cell):\n",
        "    with open(line, 'w') as f:\n",
        "        f.write(cell.format(**globals()))"
      ],
      "metadata": {
        "id": "yPR5-IeFRks8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E4cINwo3ocTk"
      },
      "outputs": [],
      "source": [
        "%%writetemplate STEAMxD/yolov5/data/baseline.yaml\n",
        "path: ../datasets/baseline2/\n",
        "train: train/images\n",
        "val: valid/images\n",
        "test:\n",
        "\n",
        "names:\n",
        "  0: humans\n",
        "  1: pets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nJ40UGKAovb_"
      },
      "outputs": [],
      "source": [
        "%%writetemplate STEAMxD/yolov5/models/baseline.yaml\n",
        "\n",
        "# Parameters\n",
        "nc: 2  # number of classes\n",
        "depth_multiple: 0.67  # model depth multiple\n",
        "width_multiple: 0.75  # layer channel multiple\n",
        "anchors:\n",
        "  - [10,13, 16,30, 33,23]  # P3/8\n",
        "  - [30,61, 62,45, 59,119]  # P4/16\n",
        "  - [116,90, 156,198, 373,326]  # P5/32\n",
        "\n",
        "# YOLOv5 v6.0 backbone\n",
        "backbone:\n",
        "  # [from, number, module, args]\n",
        "  [[-1, 1, Conv, [64, 6, 2, 2]],  # 0-P1/2\n",
        "   [-1, 1, Conv, [128, 3, 2]],  # 1-P2/4\n",
        "   [-1, 3, C3, [128]],\n",
        "   [-1, 1, Conv, [256, 3, 2]],  # 3-P3/8\n",
        "   [-1, 6, C3, [256]],\n",
        "   [-1, 1, Conv, [512, 3, 2]],  # 5-P4/16\n",
        "   [-1, 9, C3, [512]],\n",
        "   [-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32\n",
        "   [-1, 3, C3, [1024]],\n",
        "   [-1, 1, SPPF, [1024, 5]],  # 9\n",
        "  ]\n",
        "\n",
        "# YOLOv5 v6.0 head\n",
        "head:\n",
        "  [[-1, 1, Conv, [512, 1, 1]],\n",
        "   [-1, 1, nn.Upsample, [None, 2, 'nearest']],\n",
        "   [[-1, 6], 1, Concat, [1]],  # cat backbone P4\n",
        "   [-1, 3, C3, [512, False]],  # 13\n",
        "\n",
        "   [-1, 1, Conv, [256, 1, 1]],\n",
        "   [-1, 1, nn.Upsample, [None, 2, 'nearest']],\n",
        "   [[-1, 4], 1, Concat, [1]],  # cat backbone P3\n",
        "   [-1, 3, C3, [256, False]],  # 17 (P3/8-small)\n",
        "\n",
        "   [-1, 1, Conv, [256, 3, 2]],\n",
        "   [[-1, 14], 1, Concat, [1]],  # cat head P4\n",
        "   [-1, 3, C3, [512, False]],  # 20 (P4/16-medium)\n",
        "\n",
        "   [-1, 1, Conv, [512, 3, 2]],\n",
        "   [[-1, 10], 1, Concat, [1]],  # cat head P5\n",
        "   [-1, 3, C3, [1024, False]],  # 23 (P5/32-large)\n",
        "\n",
        "   [[17, 20, 23], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)\n",
        "  ]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "j8ebqN4jPVLj"
      },
      "outputs": [],
      "source": [
        "hsv_h =  0.015  # image HSV-Hue augmentation (fraction)\n",
        "hsv_s = 0.7  # image HSV-Saturation augmentation (fraction)\n",
        "hsv_v = 0.4  # image HSV-Value augmentation (fraction)\n",
        "degrees = 20.0  # image rotation (+/- deg)\n",
        "translate = 0.1  # image translation (+/- fraction)\n",
        "scale = 0.6  # image scale (+/- gain)\n",
        "shear = 0.0  # image shear (+/- deg)\n",
        "perspective = 0.000  # image perspective (+/- fraction), range 0-0.001\n",
        "flipud = 0.0  # image flip up-down (probability)\n",
        "fliplr = 0.0  # image flip left-right (probability)\n",
        "mosaic = 0.5  # image mosaic (probability)\n",
        "mixup = 0.0  # image mixup (probability)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "h3dCinQ1o5Fv"
      },
      "outputs": [],
      "source": [
        "%%writetemplate STEAMxD/yolov5/data/hyps/hyp.scratch-low.yaml\n",
        "\n",
        "# YOLOv5 🚀 by Ultralytics, GPL-3.0 license\n",
        "# Hyperparameters for low-augmentation COCO training from scratch\n",
        "# python train.py --batch 64 --cfg yolov5n6.yaml --weights '' --data coco.yaml --img 640 --epochs 300 --linear\n",
        "# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials\n",
        "\n",
        "#################################\n",
        "### DO NOT ALTER THIS PORTION ###\n",
        "#################################\n",
        "lr0: 0.01  # initial learning rate (SGD=1E-2, Adam=1E-3)\n",
        "lrf: 0.01  # final OneCycleLR learning rate (lr0 * lrf)\n",
        "momentum: 0.937  # SGD momentum/Adam beta1\n",
        "weight_decay: 0.0005  # optimizer weight decay 5e-4\n",
        "warmup_epochs: 3.0  # warmup epochs (fractions ok)\n",
        "warmup_momentum: 0.8  # warmup initial momentum\n",
        "warmup_bias_lr: 0.1  # warmup initial bias lr\n",
        "box: 0.05  # box loss gain\n",
        "cls: 0.5  # cls loss gain\n",
        "cls_pw: 1.0  # cls BCELoss positive_weight\n",
        "obj: 1.0  # obj loss gain (scale with pixels)\n",
        "obj_pw: 1.0  # obj BCELoss positive_weight\n",
        "iou_t: 0.20  # IoU training threshold\n",
        "anchor_t: 4.0  # anchor-multiple threshold\n",
        "# anchors: 3  # anchors per output layer (0 to ignore)\n",
        "fl_gamma: 0.0  # focal loss gamma (efficientDet default gamma=1.5)\n",
        "#################################\n",
        "### DO NOT ALTER THIS PORTION ###\n",
        "#################################\n",
        "###########################################\n",
        "### Image Augmentation Parameters Start ###\n",
        "###########################################\n",
        "hsv_h: {hsv_h}  # image HSV-Hue augmentation (fraction)\n",
        "hsv_s: {hsv_s}  # image HSV-Saturation augmentation (fraction)\n",
        "hsv_v: {hsv_v}  # image HSV-Value augmentation (fraction)\n",
        "degrees: {degrees}  # image rotation (+/- deg)\n",
        "translate: {translate}  # image translation (+/- fraction)\n",
        "scale: {scale}  # image scale (+/- gain)\n",
        "shear: {shear} # image shear (+/- deg)\n",
        "perspective: {perspective}  # image perspective (+/- fraction), range 0-0.001\n",
        "flipud: {flipud}  # image flip up-down (probability)\n",
        "fliplr: {fliplr}  # image flip left-right (probability)\n",
        "mosaic: {mosaic}  # image mosaic (probability)\n",
        "mixup: {mixup}  # image mixup (probability)\n",
        "#########################################\n",
        "### Image Augmentation Parameters End ###\n",
        "#########################################\n",
        "# copy_paste is for image segmentation leave this augmentation alone\n",
        "copy_paste: 0.0  # segment copy-paste (probability)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "DMCK2U-Co7Gr",
        "outputId": "64afc070-c11c-401d-aa95-a06b560acd02"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[34m\u001b[1mtrain: \u001b[0mweights=yolov5s.pt, cfg=yolov5/models/baseline.yaml, data=yolov5/data/baseline.yaml, hyp=yolov5/data/hyps/hyp.scratch-low.yaml, epochs=10, batch_size=16, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, bucket=, cache=ram, image_weights=False, device=, multi_scale=False, single_cls=False, optimizer=SGD, sync_bn=False, workers=8, project=yolov5/runs/train, name=sample, exist_ok=False, quad=False, cos_lr=False, label_smoothing=0.0, patience=100, freeze=[0], save_period=-1, seed=0, local_rank=-1, entity=None, upload_dataset=False, bbox_interval=-1, artifact_alias=latest\n",
            "\u001b[34m\u001b[1mgithub: \u001b[0mskipping check (not a git repository), for updates see https://github.com/ultralytics/yolov5\n",
            "YOLOv5 🚀 2023-1-16 Python-3.8.16 torch-1.13.0+cu116 CPU\n",
            "\n",
            "\u001b[34m\u001b[1mhyperparameters: \u001b[0mlr0=0.01, lrf=0.01, momentum=0.937, weight_decay=0.0005, warmup_epochs=3.0, warmup_momentum=0.8, warmup_bias_lr=0.1, box=0.05, cls=0.5, cls_pw=1.0, obj=1.0, obj_pw=1.0, iou_t=0.2, anchor_t=4.0, fl_gamma=0.0, hsv_h=0.015, hsv_s=0.7, hsv_v=0.4, degrees=20.0, translate=0.1, scale=0.6, shear=0.0, perspective=0.0, flipud=0.0, fliplr=0.0, mosaic=0.5, mixup=0.0, copy_paste=0.0\n",
            "\u001b[34m\u001b[1mClearML: \u001b[0mrun 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 🚀 in ClearML\n",
            "\u001b[34m\u001b[1mComet: \u001b[0mrun 'pip install comet_ml' to automatically track and visualize YOLOv5 🚀 runs in Comet\n",
            "\u001b[34m\u001b[1mTensorBoard: \u001b[0mStart with 'tensorboard --logdir yolov5/runs/train', view at http://localhost:6006/\n",
            "Downloading https://ultralytics.com/assets/Arial.ttf to /root/.config/Ultralytics/Arial.ttf...\n",
            "100% 755k/755k [00:00<00:00, 15.3MB/s]\n",
            "Downloading https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s.pt to yolov5s.pt...\n",
            "100% 14.1M/14.1M [00:00<00:00, 103MB/s] \n",
            "\n",
            "\n",
            "                 from  n    params  module                                  arguments                     \n",
            "  0                -1  1      5280  models.common.Conv                      [3, 48, 6, 2, 2]              \n",
            "  1                -1  1     41664  models.common.Conv                      [48, 96, 3, 2]                \n",
            "  2                -1  2     65280  models.common.C3                        [96, 96, 2]                   \n",
            "  3                -1  1    166272  models.common.Conv                      [96, 192, 3, 2]               \n",
            "  4                -1  4    444672  models.common.C3                        [192, 192, 4]                 \n",
            "  5                -1  1    664320  models.common.Conv                      [192, 384, 3, 2]              \n",
            "  6                -1  6   2512896  models.common.C3                        [384, 384, 6]                 \n",
            "  7                -1  1   2655744  models.common.Conv                      [384, 768, 3, 2]              \n",
            "  8                -1  2   4134912  models.common.C3                        [768, 768, 2]                 \n",
            "  9                -1  1   1476864  models.common.SPPF                      [768, 768, 5]                 \n",
            " 10                -1  1    295680  models.common.Conv                      [768, 384, 1, 1]              \n",
            " 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 12           [-1, 6]  1         0  models.common.Concat                    [1]                           \n",
            " 13                -1  2   1182720  models.common.C3                        [768, 384, 2, False]          \n",
            " 14                -1  1     74112  models.common.Conv                      [384, 192, 1, 1]              \n",
            " 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          \n",
            " 16           [-1, 4]  1         0  models.common.Concat                    [1]                           \n",
            " 17                -1  2    296448  models.common.C3                        [384, 192, 2, False]          \n",
            " 18                -1  1    332160  models.common.Conv                      [192, 192, 3, 2]              \n",
            " 19          [-1, 14]  1         0  models.common.Concat                    [1]                           \n",
            " 20                -1  2   1035264  models.common.C3                        [384, 384, 2, False]          \n",
            " 21                -1  1   1327872  models.common.Conv                      [384, 384, 3, 2]              \n",
            " 22          [-1, 10]  1         0  models.common.Concat                    [1]                           \n",
            " 23                -1  2   4134912  models.common.C3                        [768, 768, 2, False]          \n",
            " 24      [17, 20, 23]  1     28287  models.yolo.Detect                      [2, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [192, 384, 768]]\n",
            "baseline summary: 291 layers, 20875359 parameters, 20875359 gradients, 48.2 GFLOPs\n",
            "\n",
            "Transferred 57/481 items from yolov5s.pt\n",
            "\u001b[34m\u001b[1moptimizer:\u001b[0m SGD(lr=0.01) with parameter groups 79 weight(decay=0.0), 82 weight(decay=0.0005), 82 bias\n",
            "\u001b[34m\u001b[1malbumentations: \u001b[0mBlur(p=0.01, blur_limit=(3, 7)), MedianBlur(p=0.01, blur_limit=(3, 7)), ToGray(p=0.01), CLAHE(p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mScanning /content/STEAMxD/STEAMxD/datasets/baseline2/train/labels... 783 images, 0 backgrounds, 0 corrupt: 100% 783/783 [00:00<00:00, 1505.74it/s]\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mNew cache created: /content/STEAMxD/STEAMxD/datasets/baseline2/train/labels.cache\n",
            "\u001b[34m\u001b[1mtrain: \u001b[0mCaching images (0.5GB ram): 100% 783/783 [00:10<00:00, 72.92it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mScanning /content/STEAMxD/STEAMxD/datasets/baseline2/valid/labels... 118 images, 0 backgrounds, 0 corrupt: 100% 118/118 [00:00<00:00, 502.51it/s]\n",
            "\u001b[34m\u001b[1mval: \u001b[0mNew cache created: /content/STEAMxD/STEAMxD/datasets/baseline2/valid/labels.cache\n",
            "\u001b[34m\u001b[1mval: \u001b[0mCaching images (0.1GB ram): 100% 118/118 [00:01<00:00, 66.54it/s]\n",
            "\n",
            "\u001b[34m\u001b[1mAutoAnchor: \u001b[0m4.95 anchors/target, 0.999 Best Possible Recall (BPR). Current anchors are a good fit to dataset ✅\n",
            "Plotting labels to yolov5/runs/train/sample/labels.jpg... \n",
            "Image sizes 640 train, 640 val\n",
            "Using 2 dataloader workers\n",
            "Logging results to \u001b[1myolov5/runs/train/sample\u001b[0m\n",
            "Starting training for 10 epochs...\n",
            "\n",
            "      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size\n",
            "        0/9         0G     0.1174    0.03791    0.02893         64        640:   2% 1/49 [01:20<1:04:09, 80.21s/it]"
          ]
        }
      ],
      "source": [
        "!python STEAMxD/yolov5/train.py --img 640 --batch 16 --epochs 200 --data STEAMxD/yolov5/data/baseline.yaml --cfg STEAMxD/yolov5/models/baseline.yaml --weights yolov5s.pt --cache --name baseline1 --project 'STEAMxD/yolov5/runs/train'"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}