aws-samples / aws-samples/mask-rcnn-tensorflow
TypeError: Conv2D() got an unexpected keyword argument 'seed'
- Dominant language
- Python
- Stars
- 39
- Forks
- 24
- PR merge metrics
- No merged PRs in 30d
Description
In [tensorpack.models.conv2d]
when I try: `python3 train.py --load=data/coco/pretrained-models/ImageNet-R50-AlignPadding.npz`
I got a error: `TypeError: Conv2D() got an unexpected keyword argument 'seed'`
located at: `File "/home/cv/mask-rcnn-tensorflow/MaskRCNN/model/backbone.py", line 239, in resnet_fpn_backbone`
` l = Conv2D('conv0', l, 64, 7, strides=2, padding='VALID', seed=seed_gen.next())`
Is it because there is no `seed` parameter in conv2D?
How can I fix it？
Thanks！
Contributor guide
Research direction
Start with model/backbone.py at resnet_fpn_backbone and reproduce the reported command. Inspect the installed tensorpack Conv2D signature and compare it with the seed argument used there. Done means the command no longer raises the unexpected-keyword TypeError.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100