.. _sec_object_detection_quick: Object Detection - Quick Start ============================== Object detection is the process of identifying and localizing objects in an image and is an important task in computer vision. Follow this tutorial to learn how to use AutoGluon for object detection. **Tip**: If you are new to AutoGluon, review :ref:`sec_imgquick` first to learn the basics of the AutoGluon API. Our goal is to detect motorbike in images by `YOLOv3 model `__. A tiny dataset is collected from VOC dataset, which only contains the motorbike category. The model pretrained on the COCO dataset is used to fine-tune our small dataset. With the help of AutoGluon, we are able to try many models with different hyperparameters automatically, and return the best one as our final model. To start, import ObjectDetector: .. code:: python from autogluon.vision import ObjectDetector .. parsed-literal:: :class: output /var/lib/jenkins/miniconda3/envs/autogluon-tutorial-object-detection-v3/lib/python3.9/site-packages/gluoncv/__init__.py:40: UserWarning: Both `mxnet==1.9.1` and `torch==1.10.2+cu102` are installed. You might encounter increased GPU memory footprint if both framework are used at the same time. warnings.warn(f'Both `mxnet=={mx.__version__}` and `torch=={torch.__version__}` are installed. ' Tiny\_motorbike Dataset ----------------------- We collect a toy dataset for detecting motorbikes in images. From the VOC dataset, images are randomly selected for training, validation, and testing - 120 images for training, 50 images for validation, and 50 for testing. This tiny dataset follows the same format as VOC. Using the commands below, we can download this dataset, which is only 23M. The name of unzipped folder is called ``tiny_motorbike``. Anyway, the task dataset helper can perform the download and extraction automatically, and load the dataset according to the detection formats. .. code:: python url = 'https://autogluon.s3.amazonaws.com/datasets/tiny_motorbike.zip' dataset_train = ObjectDetector.Dataset.from_voc(url, splits='trainval') .. parsed-literal:: :class: output tiny_motorbike/ ├── Annotations/ ├── ImageSets/ └── JPEGImages/ Fit Models by AutoGluon ----------------------- In this section, we demonstrate how to apply AutoGluon to fit our detection models. We use mobilenet as the backbone for the YOLOv3 model. Two different learning rates are used to fine-tune the network. The best model is the one that obtains the best performance on the validation dataset. You can also try using more networks and hyperparameters to create a larger searching space. We ``fit`` a classifier using AutoGluon as follows. In each experiment (one trial in our searching space), we train the model for 5 epochs to avoid bursting our tutorial runtime. .. code:: python time_limit = 60*30 # at most 0.5 hour detector = ObjectDetector() hyperparameters = {'epochs': 5, 'batch_size': 8} hyperparameter_tune_kwargs={'num_trials': 2} detector.fit(dataset_train, time_limit=time_limit, hyperparameters=hyperparameters, hyperparameter_tune_kwargs=hyperparameter_tune_kwargs) .. parsed-literal:: :class: output ============================================================================= WARNING: ObjectDetector is deprecated as of v0.4.0 and may contain various bugs and issues! In a future release ObjectDetector may be entirely reworked to use Torch as a backend. This future change will likely be API breaking.Users should ensure they update their code that depends on ObjectDetector when upgrading to future AutoGluon releases. For more information, refer to ObjectDetector refactor GitHub issue: https://github.com/awslabs/autogluon/issues/1559 ============================================================================= The number of requested GPUs is greater than the number of available GPUs.Reduce the number to 1 Randomly split train_data into train[144]/validation[26] splits. Starting HPO experiments .. parsed-literal:: :class: output 0%| | 0/2 [00:00 != ): { root.dataset voc_tiny != auto root.ssd.base_network vgg16_atrous != resnet50_v1 root.ssd.data_shape 300 != 512 root.valid.batch_size 16 != 8 root.gpus (0, 1, 2, 3) != (0,) root.num_workers 4 != 8 root.dataset_root ~/.mxnet/datasets/ != auto root.train.early_stop_patience -1 != 10 root.train.early_stop_max_value 1.0 != inf root.train.early_stop_baseline 0.0 != -inf root.train.seed 233 != 54 root.train.batch_size 16 != 8 root.train.epochs 20 != 5 } Saved config to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-object-detection-v3/docs/_build/eval/tutorials/object_detection/f8efb1c7/.trial_0/config.yaml No gpu detected, fallback to cpu. You can ignore this warning if this is intended. Using transfer learning from ssd_512_resnet50_v1_coco, the other network parameters are ignored. Start training from [Epoch 0] [Epoch 0] Training cost: 65.236850, CrossEntropy=3.779503, SmoothL1=1.163054 [Epoch 0] Validation: car=0.691492464754497 motorbike=0.7169003253371304 dog=0.0 person=0.7003156712699655 bus=nan chair=nan cow=nan pottedplant=nan bicycle=nan boat=nan mAP=0.5271771153403982 [Epoch 0] Current best map: 0.527177 vs previous 0.000000, saved to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-object-detection-v3/docs/_build/eval/tutorials/object_detection/f8efb1c7/.trial_0/best_checkpoint.pkl [Epoch 1] Training cost: 64.812908, CrossEntropy=2.875078, SmoothL1=1.306778 [Epoch 1] Validation: car=0.5454545454545455 motorbike=0.7951259820123776 dog=1.0000000000000002 person=0.7436221684031445 bus=nan chair=nan cow=nan pottedplant=nan bicycle=nan boat=nan mAP=0.7710506739675169 [Epoch 1] Current best map: 0.771051 vs previous 0.527177, saved to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-object-detection-v3/docs/_build/eval/tutorials/object_detection/f8efb1c7/.trial_0/best_checkpoint.pkl [Epoch 2] Training cost: 64.975384, CrossEntropy=2.573965, SmoothL1=1.131789 [Epoch 2] Validation: car=0.5881118881118882 motorbike=0.808175459149485 dog=0.0 person=0.82401410042075 bus=nan chair=nan cow=nan pottedplant=nan bicycle=nan boat=nan mAP=0.5550753619205308 [Epoch 3] Training cost: 65.202270, CrossEntropy=2.351436, SmoothL1=1.070161 [Epoch 3] Validation: car=0.6181818181818183 motorbike=0.798377670668798 dog=0.0 person=0.8602015520559108 bus=nan chair=nan cow=nan pottedplant=nan bicycle=nan boat=nan mAP=0.5691902602266318 [Epoch 4] Training cost: 64.798859, CrossEntropy=2.294009, SmoothL1=1.045786 [Epoch 4] Validation: car=0.40449594890460056 motorbike=0.7893528319424593 dog=0.0 person=0.4996091798697604 bus=nan chair=nan cow=nan pottedplant=nan bicycle=nan boat=nan mAP=0.4233644901792051 Applying the state from the best checkpoint... Finished, total runtime is 376.92 s { 'best_config': { 'dataset': 'auto', 'dataset_root': 'auto', 'estimator': , 'gpus': [0], 'horovod': False, 'num_workers': 8, 'resume': '', 'save_interval': 1, 'ssd': { 'amp': False, 'base_network': 'resnet50_v1', 'data_shape': 512, 'filters': None, 'nms_thresh': 0.45, 'nms_topk': 400, 'ratios': ( [1, 2, 0.5], [1, 2, 0.5, 3, 0.3333333333333333], [1, 2, 0.5, 3, 0.3333333333333333], [1, 2, 0.5, 3, 0.3333333333333333], [1, 2, 0.5], [1, 2, 0.5]), 'sizes': (30, 60, 111, 162, 213, 264, 315), 'steps': (8, 16, 32, 64, 100, 300), 'syncbn': False, 'transfer': 'ssd_512_resnet50_v1_coco'}, 'train': { 'batch_size': 8, 'dali': False, 'early_stop_baseline': -inf, 'early_stop_max_value': inf, 'early_stop_min_delta': 0.001, 'early_stop_patience': 10, 'epochs': 5, 'log_interval': 100, 'lr': 0.001, 'lr_decay': 0.1, 'lr_decay_epoch': (160, 200), 'momentum': 0.9, 'seed': 54, 'start_epoch': 0, 'wd': 0.0005}, 'valid': { 'batch_size': 8, 'iou_thresh': 0.5, 'metric': 'voc07', 'val_interval': 1}}, 'total_time': 376.9148666858673, 'train_map': 0.5072572581736238, 'valid_map': 0.7710506739675169} .. parsed-literal:: :class: output Note that ``num_trials=2`` above is only used to speed up the tutorial. In normal practice, it is common to only use ``time_limit`` and drop ``num_trials``. Also note that hyperparameter tuning defaults to random search. After fitting, AutoGluon automatically returns the best model among all models in the searching space. From the output, we know the best model is the one trained with the second learning rate. To see how well the returned model performed on test dataset, call detector.evaluate(). .. code:: python dataset_test = ObjectDetector.Dataset.from_voc(url, splits='test') test_map = detector.evaluate(dataset_test) print("mAP on test dataset: {}".format(test_map[1][-1])) .. parsed-literal:: :class: output tiny_motorbike/ ├── Annotations/ ├── ImageSets/ └── JPEGImages/ mAP on test dataset: 0.34605385823820556 Below, we randomly select an image from test dataset and show the predicted class, box and probability over the origin image, stored in ``predict_class``, ``predict_rois`` and ``predict_score`` columns, respectively. You can interpret ``predict_rois`` as a dict of (``xmin``, ``ymin``, ``xmax``, ``ymax``) proportional to original image size. .. code:: python image_path = dataset_test.iloc[0]['image'] result = detector.predict(image_path) print(result) .. parsed-literal:: :class: output predict_class predict_score \ 0 motorbike 0.920063 1 person 0.757578 2 person 0.222493 3 person 0.164949 4 person 0.157512 .. ... ... 82 person 0.028850 83 person 0.028748 84 person 0.028559 85 person 0.028550 86 person 0.028549 predict_rois 0 {'xmin': 0.31125956773757935, 'ymin': 0.431519... 1 {'xmin': 0.37460440397262573, 'ymin': 0.279367... 2 {'xmin': 0.9970632791519165, 'ymin': 0.4861781... 3 {'xmin': 0.993183970451355, 'ymin': 0.60069912... 4 {'xmin': 0.6415563225746155, 'ymin': 0.0, 'xma... .. ... 82 {'xmin': 0.99004065990448, 'ymin': 0.639587581... 83 {'xmin': 0.601842999458313, 'ymin': 0.00850366... 84 {'xmin': 0.08381837606430054, 'ymin': 0.0, 'xm... 85 {'xmin': 0.38603806495666504, 'ymin': 0.380946... 86 {'xmin': 0.983705461025238, 'ymin': 0.03952397... [87 rows x 3 columns] Prediction with multiple images is permitted: .. code:: python bulk_result = detector.predict(dataset_test) print(bulk_result) .. parsed-literal:: :class: output predict_class predict_score \ 0 motorbike 0.920063 1 person 0.757578 2 person 0.222493 3 person 0.164949 4 person 0.157512 ... ... ... 4002 person 0.028438 4003 person 0.027862 4004 person 0.027711 4005 person 0.027630 4006 person 0.027448 predict_rois \ 0 {'xmin': 0.31125956773757935, 'ymin': 0.431519... 1 {'xmin': 0.37460440397262573, 'ymin': 0.279367... 2 {'xmin': 0.9970632791519165, 'ymin': 0.4861781... 3 {'xmin': 0.993183970451355, 'ymin': 0.60069912... 4 {'xmin': 0.6415563225746155, 'ymin': 0.0, 'xma... ... ... 4002 {'xmin': 0.9574559330940247, 'ymin': 0.6985545... 4003 {'xmin': 0.5104297399520874, 'ymin': 0.1801516... 4004 {'xmin': 0.5849680304527283, 'ymin': 0.3427146... 4005 {'xmin': 0.20679672062397003, 'ymin': 0.497470... 4006 {'xmin': 0.5176151990890503, 'ymin': 0.5120053... image 0 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 1 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 2 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 3 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 4 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... ... ... 4002 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 4003 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 4004 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 4005 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... 4006 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb... [4007 rows x 4 columns] We can also save the trained model, and use it later. .. code:: python savefile = 'detector.ag' detector.save(savefile) new_detector = ObjectDetector.load(savefile) .. parsed-literal:: :class: output /var/lib/jenkins/miniconda3/envs/autogluon-tutorial-object-detection-v3/lib/python3.9/site-packages/mxnet/gluon/block.py:1784: UserWarning: Cannot decide type for the following arguments. Consider providing them as input: data: None input_sym_arg_type = in_param.infer_type()[0]