AttributeError: module 'wandb' has no attribute 'run'

[HypnosWei]

❔Question

When I use ipython, I just run this
[run train.py --epochs 2 --batch-size 32 --data ./data/train.yaml --weights ./runs/train/exp5/weights/best.pt --cfg ./models/yolov5s.yaml ]

I already create new environment ,but it doesn't work.
I also download the whole yolov5 files again, but it doesn't work too.

but the command window show that

Additional context

[

[github-actions]

👋 Hello @HypnosWei, 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://www.ultralytics.com or email Glenn Jocher at [email protected].

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

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/models/ultralytics/yolov5
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Docker Image https://hub.docker.com/r/ultralytics/yolov5. 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 (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@HypnosWei you code may be out of date. Is this reproducible with current master? If so please submit a colab notebook we can us to reproduce and debug. Thank you!

[HypnosWei]

I try to pip install wanb and it work ,the code keep update ? It is very awesome!
I am a student and try to finish the final project about object detect, but the best mAP of our training model is just 0.458 . I want to try the data augmentation , but find the hyp.scratch.yaml has flip and some of trick, so I don't know how can improve the result . if you have some better recommendation ,I will very thank you!!

sorry about my pool english .

[glenn-jocher]

@HypnosWei great, so everything is working correctly now?

For training tips you may want to upload your results. 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.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[mattpopovich]

Just to clarify for anyone else with this issue, you get it probably because you do not have wandb installed. Installing it via pip install wandb fixed the issue for me!

[glenn-jocher]

@mattpopovich @AyushExel this is likely due to wandb creating a wandb directory, which is then not differentiable from a package import. I've raised this issue with W&B before.

[AyushExel]

@glenn-jocher this can be solved if we include the library components in check

    try:
        import wandb
        from wandb import init, log, join  # test that these are available
    except ImportError:
       print("msg")

[glenn-jocher]

Update: This may be fixed now in PR # 2574.

[mattpopovich]

Looks good to me. As expected, now gives a notice to install wandb if it's not already installed and no longer crashes before training.

github: skipping check (Docker image)                                                                                                                                            
YOLOv5 🚀 v4.0-142-g1bf9365 torch 1.8.0a0+52ea372 CUDA:2 (GeForce GTX 1080, 8119.5625MB)                                                                                         
                                                                                                                                                                                 
Namespace(adam=False, artifact_alias='latest', batch_size=16, bbox_interval=-1, bucket='', cache_images=False, cfg='', data='./data/coco128.yaml', device='2', entity=None, epoch
s=5, evolve=False, exist_ok=False, global_rank=-1, hyp='data/hyp.scratch.yaml', image_weights=False, img_size=[640, 640], linear_lr=False, local_rank=-1, multi_scale=False, name
='test', noautoanchor=False, nosave=False, notest=False, project='runs/train', quad=False, rect=False, resume=False, save_dir='runs/train/test8', save_period=-1, single_cls=Fals
e, sync_bn=False, total_batch_size=16, upload_dataset=False, weights='yolov5s.pt', workers=8, world_size=1)                                                                      
Start Tensorboard with "tensorboard --logdir runs/train", view at http://localhost:6006/                                                                                         
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, mosa
ic=1.0, mixup=0.0                                                                                                                                                                
wandb: Install Weights & Biases for YOLOv5 logging with 'pip install wandb' (recommended)                                                                                        
                                                                                                                                                                                 
                 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]