0 Fork 0

Files

T

Glenn Jocher 9ac9ec3449 Created using Colaboratory

2020-11-05 18:34:15 +01:00

380 KiB

Vendored

Raw Blame History

This notebook was written by Ultralytics LLC, and is freely available for redistribution under the GPL-3.0 license. For more information please visit https://github.com/ultralytics/yolov5 and https://www.ultralytics.com.

Setup

Clone repo, install dependencies and check PyTorch and GPU.

In [1]:

!git clone https://github.com/ultralytics/yolov5  # clone repo
%cd yolov5
%pip install -qr requirements.txt  # install dependencies

import torch
from IPython.display import Image, clear_output  # to display images
from utils.google_utils import gdrive_download  # to download models/datasets

clear_output()
print('Setup complete. Using torch %s %s' % (torch.__version__, torch.cuda.get_device_properties(0) if torch.cuda.is_available() else 'CPU'))

Setup complete. Using torch 1.7.0+cu101 _CudaDeviceProperties(name='Tesla V100-SXM2-16GB', major=7, minor=0, total_memory=16130MB, multi_processor_count=80)

1. Inference

detect.py runs inference on a variety of sources, downloading models automatically from the latest YOLOv5 release.

In [38]:

!python detect.py --weights yolov5s.pt --img 640 --conf 0.25 --source inference/images/
Image(filename='inference/output/zidane.jpg', width=600)

Out [38]:

Namespace(agnostic_nms=False, augment=False, classes=None, conf_thres=0.25, device='', img_size=640, iou_thres=0.45, save_conf=False, save_dir='inference/output', save_txt=False, source='inference/images/', update=False, view_img=False, weights=['yolov5s.pt'])
Using CUDA device0 _CudaDeviceProperties(name='Tesla V100-SXM2-16GB', total_memory=16130MB)

Fusing layers... 
Model Summary: 140 layers, 7.45958e+06 parameters, 0 gradients
image 1/2 /content/yolov5/inference/images/bus.jpg: 640x480 4 persons, 1 buss, 1 skateboards, Done. (0.012s)
image 2/2 /content/yolov5/inference/images/zidane.jpg: 384x640 2 persons, 2 ties, Done. (0.012s)
Results saved to inference/output
Done. (0.113s)

Available inference sources:

2. Test

Test a model on COCO val or test-dev dataset to evaluate trained accuracy. Models are downloaded automatically from the latest YOLOv5 release. To show results by class use the --verbose flag. Note that pycocotools metrics may be 1-2% better than the equivalent repo metrics, as is visible below, due to slight differences in mAP computation.

2.1 COCO val2017

Download COCO val 2017 dataset (1GB - 5000 images), and test model accuracy.

In [16]:

# Download COCO val2017
torch.hub.download_url_to_file('https://github.com/ultralytics/yolov5/releases/download/v1.0/coco2017val.zip', 'tmp.zip')
!unzip -q tmp.zip -d ../ && rm tmp.zip

HBox(children=(FloatProgress(value=0.0, max=819257867.0), HTML(value='')))

In [17]:

# Run YOLOv5x on COCO val2017
!python test.py --weights yolov5x.pt --data coco.yaml --img 640

Namespace(augment=False, batch_size=32, conf_thres=0.001, data='./data/coco.yaml', device='', img_size=640, iou_thres=0.65, save_conf=False, save_dir='runs/test', save_json=True, save_txt=False, single_cls=False, task='val', verbose=False, weights=['yolov5x.pt'])
Using CUDA device0 _CudaDeviceProperties(name='Tesla V100-SXM2-16GB', total_memory=16130MB)

Downloading https://github.com/ultralytics/yolov5/releases/download/v3.1/yolov5x.pt to yolov5x.pt...
100% 170M/170M [00:05<00:00, 32.2MB/s]

Fusing layers... 
Model Summary: 284 layers, 8.89222e+07 parameters, 0 gradients
Scanning labels ../coco/labels/val2017.cache (4952 found, 0 missing, 48 empty, 0 duplicate, for 5000 images): 5000it [00:00, 16239.02it/s]
               Class      Images     Targets           P           R      mAP@.5  mAP@.5:.95: 100% 157/157 [01:22<00:00,  1.89it/s]
                 all       5e+03    3.63e+04       0.409       0.754       0.672       0.483
Speed: 5.9/2.0/7.9 ms inference/NMS/total per 640x640 image at batch-size 32

COCO mAP with pycocotools... saving runs/test/detections_val2017_yolov5x_results.json...
loading annotations into memory...
Done (t=0.73s)
creating index...
index created!
Loading and preparing results...
DONE (t=4.48s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=86.55s).
Accumulating evaluation results...
DONE (t=13.15s).
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.492
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.676
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.534
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.318
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.541
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.633
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.376
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.617
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.670
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.493
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.723
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.812
Results saved to runs/test

2.2 COCO test-dev2017

Download COCO test2017 dataset (7GB - 40,000 images), to test model accuracy on test-dev set (20,000 images). Results are saved to a *.json file which can be submitted to the evaluation server at https://competitions.codalab.org/competitions/20794.

In [ ]:

# Download COCO test-dev2017
gdrive_download('1cXZR_ckHki6nddOmcysCuuJFM--T-Q6L','coco2017labels.zip')  # annotations
!f="test2017.zip" && curl http://images.cocodataset.org/zips/$f -o $f && unzip -q $f && rm $f  # 7GB,  41k images
!mv ./test2017 ./coco/images && mv ./coco ../  # move images into /coco and move /coco alongside /yolov5

In [ ]:

# Run YOLOv5s on COCO test-dev2017 using --task test
!python test.py --weights yolov5s.pt --data coco.yaml --task test

3. Train

Download COCO128, a small 128-image tutorial dataset, start tensorboard and train YOLOv5s from a pretrained checkpoint for 3 epochs (actual training is typically much longer, around 300-1000 epochs, depending on your dataset).

In [22]:

# Download COCO128
torch.hub.download_url_to_file('https://github.com/ultralytics/yolov5/releases/download/v1.0/coco128.zip', 'tmp.zip')
!unzip -q tmp.zip -d ../ && rm tmp.zip

HBox(children=(FloatProgress(value=0.0, max=22090455.0), HTML(value='')))

Train a YOLOv5s model on COCO128 with dataset --data coco128.yaml, starting from pretrained --weights yolov5s.pt, or from randomly initialized --weights '' --cfg yolov5s.yaml. Models are downloaded automatically from the latest YOLOv5 release, and COCO, COCO128, and VOC datasets are downloaded automatically on first use.

All training results are saved to runs/exp0 for the first experiment, then runs/exp1, runs/exp2 etc. for subsequent experiments.

In [ ]:

# Start tensorboard (optional)
%load_ext tensorboard
%tensorboard --logdir runs

In [23]:

# Train YOLOv5s on COCO128 for 3 epochs
!python train.py --img 640 --batch 16 --epochs 3 --data coco128.yaml --weights yolov5s.pt --nosave --cache

Using CUDA device0 _CudaDeviceProperties(name='Tesla V100-SXM2-16GB', total_memory=16130MB)

Namespace(adam=False, batch_size=16, bucket='', cache_images=True, cfg='', data='./data/coco128.yaml', device='', epochs=3, evolve=False, global_rank=-1, hyp='data/hyp.scratch.yaml', image_weights=False, img_size=[640, 640], local_rank=-1, log_imgs=10, logdir='runs/', multi_scale=False, name='', noautoanchor=False, nosave=True, notest=False, rect=False, resume=False, single_cls=False, sync_bn=False, total_batch_size=16, weights='yolov5s.pt', workers=8, world_size=1)
Start Tensorboard with "tensorboard --logdir runs/", view at http://localhost:6006/
2020-11-05 16:39:40.555423: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.1
Install Weights & Biases for experiment logging via 'pip install wandb' (recommended)
Hyperparameters {'lr0': 0.01, 'lrf': 0.2, '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': 0.0, 'translate': 0.1, 'scale': 0.5, 'shear': 0.0, 'perspective': 0.0, 'flipud': 0.0, 'fliplr': 0.5, 'mosaic': 1.0, 'mixup': 0.0}

                 from  n    params  module                                  arguments                     
  0                -1  1      3520  models.common.Focus                     [3, 32, 3]                    
  1                -1  1     18560  models.common.Conv                      [32, 64, 3, 2]                
  2                -1  1     19904  models.common.BottleneckCSP             [64, 64, 1]                   
  3                -1  1     73984  models.common.Conv                      [64, 128, 3, 2]               
  4                -1  1    161152  models.common.BottleneckCSP             [128, 128, 3]                 
  5                -1  1    295424  models.common.Conv                      [128, 256, 3, 2]              
  6                -1  1    641792  models.common.BottleneckCSP             [256, 256, 3]                 
  7                -1  1   1180672  models.common.Conv                      [256, 512, 3, 2]              
  8                -1  1    656896  models.common.SPP                       [512, 512, [5, 9, 13]]        
  9                -1  1   1248768  models.common.BottleneckCSP             [512, 512, 1, False]          
 10                -1  1    131584  models.common.Conv                      [512, 256, 1, 1]              
 11                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          
 12           [-1, 6]  1         0  models.common.Concat                    [1]                           
 13                -1  1    378624  models.common.BottleneckCSP             [512, 256, 1, False]          
 14                -1  1     33024  models.common.Conv                      [256, 128, 1, 1]              
 15                -1  1         0  torch.nn.modules.upsampling.Upsample    [None, 2, 'nearest']          
 16           [-1, 4]  1         0  models.common.Concat                    [1]                           
 17                -1  1     95104  models.common.BottleneckCSP             [256, 128, 1, False]          
 18                -1  1    147712  models.common.Conv                      [128, 128, 3, 2]              
 19          [-1, 14]  1         0  models.common.Concat                    [1]                           
 20                -1  1    313088  models.common.BottleneckCSP             [256, 256, 1, False]          
 21                -1  1    590336  models.common.Conv                      [256, 256, 3, 2]              
 22          [-1, 10]  1         0  models.common.Concat                    [1]                           
 23                -1  1   1248768  models.common.BottleneckCSP             [512, 512, 1, False]          
 24      [17, 20, 23]  1    229245  models.yolo.Detect                      [80, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]
Model Summary: 191 layers, 7.46816e+06 parameters, 7.46816e+06 gradients

Transferred 370/370 items from yolov5s.pt
Optimizer groups: 62 .bias, 70 conv.weight, 59 other
Scanning images: 100% 128/128 [00:00<00:00, 5375.86it/s]
Scanning labels ../coco128/labels/train2017.cache (126 found, 0 missing, 2 empty, 0 duplicate, for 128 images): 128it [00:00, 15114.61it/s]
Caching images (0.1GB): 100% 128/128 [00:00<00:00, 141.99it/s]
Scanning labels ../coco128/labels/train2017.cache (126 found, 0 missing, 2 empty, 0 duplicate, for 128 images): 128it [00:00, 12572.50it/s]
Caching images (0.1GB): 100% 128/128 [00:01<00:00, 79.68it/s] 
NumExpr defaulting to 2 threads.

Analyzing anchors... anchors/target = 4.26, Best Possible Recall (BPR) = 0.9946
Image sizes 640 train, 640 test
Using 2 dataloader workers
Logging results to runs/exp0
Starting training for 3 epochs...

     Epoch   gpu_mem       box       obj       cls     total   targets  img_size
       0/2     3.22G   0.04202   0.06746   0.01503    0.1245       194       640: 100% 8/8 [00:03<00:00,  2.51it/s]
               Class      Images     Targets           P           R      mAP@.5  mAP@.5:.95: 100% 8/8 [00:03<00:00,  2.42it/s]
                 all         128         929       0.405       0.762       0.701       0.449

     Epoch   gpu_mem       box       obj       cls     total   targets  img_size
       1/2     3.17G   0.04461   0.05873   0.01689    0.1202       142       640: 100% 8/8 [00:01<00:00,  4.14it/s]
               Class      Images     Targets           P           R      mAP@.5  mAP@.5:.95: 100% 8/8 [00:01<00:00,  6.49it/s]
                 all         128         929       0.402       0.772       0.703       0.452

     Epoch   gpu_mem       box       obj       cls     total   targets  img_size
       2/2     3.17G   0.04445   0.06545   0.01666    0.1266       149       640: 100% 8/8 [00:01<00:00,  4.33it/s]
               Class      Images     Targets           P           R      mAP@.5  mAP@.5:.95: 100% 8/8 [00:02<00:00,  2.78it/s]
                 all         128         929       0.395       0.766       0.701       0.455
Optimizer stripped from runs/exp0/weights/last.pt, 15.2MB
Optimizer stripped from runs/exp0/weights/best.pt, 15.2MB
3 epochs completed in 0.005 hours.

4. Visualize

4.1 Weights & Biases Logging (🚀 NEW)

Weights & Biases (W&B) is now integrated with YOLOv5 for real-time visualization and cloud logging of training runs. This allows for better run comparison and introspection, as well improved visibility and collaboration among team members. To enable W&B logging install wandb, and then train normally (you will be guided setup on first use).

$ pip install wandb

During training you will see live updates at https://www.wandb.com/, and you can create and share detailed Reports of your results. For more information see the YOLOv5 Weights & Biases Tutorial.

4.2 Local Logging

All results are logged by default to the runs/exp0 directory, with a new directory created for each new training as runs/exp1, runs/exp2, etc. View train_batch*.jpg to see training images, labels and augmentation effects. A Mosaic Dataloader is used for training (shown below), a new concept developed by Ultralytics and first featured in YOLOv4.

In [ ]:

Image(filename='runs/exp0/train_batch0.jpg', width=800)  # train batch 0 mosaics and labels
Image(filename='runs/exp0/test_batch0_gt.jpg', width=800)  # test batch 0 ground truth
Image(filename='runs/exp0/test_batch0_pred.jpg', width=800)  # test batch 0 predictions

test_batch0_gt.jpg train batch 0 mosaics and labels

test_batch0_gt.jpg shows test batch 0 ground truth

test_batch0_pred.jpg shows test batch 0 predictions

Training losses and performance metrics are also logged to Tensorboard and a custom runs/exp0/results.txt logfile. results.txt is plotted as results.png (below) after training completes. Here we show YOLOv5s trained on COCO128 to 300 epochs, starting from scratch (blue), and from pretrained yolov5s.pt (orange).

In [ ]:

from utils.utils import plot_results 
plot_results(save_dir='runs/exp0')  # plot results.txt as results.png
Image(filename='results.png', width=800)

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Google Colab Notebook with free GPU:
Kaggle Notebook with free GPU: https://www.kaggle.com/ultralytics/yolov5
Google Cloud Deep Learning VM. See GCP Quickstart Guide
Docker Image https://hub.docker.com/r/ultralytics/yolov5. See Docker Quickstart Guide

Appendix

Optional extras below. Unit tests validate repo functionality and should be run on any PRs submitted.

In [ ]:

# Re-clone repo
%cd ..
%rm -rf yolov5 && git clone https://github.com/ultralytics/yolov5
%cd yolov5

In [ ]:

# Test all models
%%shell
for x in s m l x
do
  python test.py --weights yolov5$x.pt --data coco.yaml --img 640
done

In [ ]:

# Run unit tests
%%shell
#git clone https://github.com/ultralytics/yolov5
#cd yolov5
pip install -qr requirements.txt onnx

export PYTHONPATH="$PWD" # to run *.py. files in subdirectories
for x in yolov5s #yolov5m yolov5l yolov5x # models
do
  rm -rf runs
  python train.py --weights $x.pt --cfg $x.yaml --epochs 3 --img 320 --device 0  # train
  for di in 0 cpu # inference devices
  do
    python detect.py --weights $x.pt --device $di  # detect official
    python detect.py --weights runs/exp0/weights/last.pt --device $di  # detect custom
    python test.py --weights $x.pt --device $di # test official
    python test.py --weights runs/exp0/weights/last.pt --device $di # test custom
  done
  python models/yolo.py --cfg $x.yaml # inspect
  python models/export.py --weights $x.pt --img 640 --batch 1 # export
done

380 KiB Vendored Raw Blame History

Setup

1. Inference

2. Test

2.1 COCO val2017

2.2 COCO test-dev2017

3. Train

4. Visualize

4.1 Weights & Biases Logging (🚀 NEW)

4.2 Local Logging

Environments

Appendix

380 KiB

Vendored

Raw Blame History