throughput in progress bar

[roywei]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Hi from recent update we found the throughput info is missing in progress bar, any idea why it's removed? it's quite useful for benchmarking. Thanks!
it's used to be like: with 5.26 it/s info

        0/2      8.36G     0.1027     0.1332    0.09538          6        640: 100%|██████████| 463/463 [01:29<00:00,  5.19it/s]
        1/2      8.36G    0.08447    0.08337    0.08307          6        640: 100%|██████████| 463/463 [01:28<00:00,  5.26it/s]
        2/2      8.36G    0.07142     0.0816    0.07307          3        640: 100%|██████████| 463/463 [01:28<00:00,  5.23it/s]

now it's

[1,0]<stderr>:      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
       0/99      14.3G      0.105     0.2458    0.09572        118        640: 100%|██████████| 58/58 00:31[1,0]<stderr>:
[1,0]<stderr>:
[1,0]<stderr>:      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
       1/99      14.3G     0.1005     0.0963    0.09465        187        640: 100%|██████████| 58/58 00:22[1,0]<stderr>:
[1,0]<stderr>:
[1,0]<stderr>:      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
       2/99      14.3G    0.09316    0.08498    0.08702        206        640: 100%|██████████| 58/58 00:22[1,0]<stderr>:
[1,0]<stderr>:
[1,0]<stderr>:      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
       3/99      14.3G    0.08193    0.08364    0.07913        148        640: 100%|██████████| 58/58 00:24[1,0]<stderr>:
[1,0]<stderr>:
[1,0]<stderr>:      Epoch    GPU_mem   box_loss   obj_loss   cls_loss  Instances       Size
       4/99      14.3G    0.07587    0.08373    0.07301        156        640: 100%|██████████| 58/58 00:24[1,0]<stderr>:

Additional

No response

[github-actions]

👋 Hello @roywei, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email support@ultralytics.com.

Requirements

Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

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):

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training, validation, inference, export and benchmarks on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@roywei thanks for the feedback! I got rid of the iteration speed to save some space, but it appears my implementation is bad based on your console output showing that shouldn't be there. The PR is #10211, which created a new global TQDM_BAR_FORMAT.