Mahi-Mai

Testing on rectangular images?

1

So, it seems to me that UNet can only be trained and tested on perfect square images. If I ask my model to predict on a rectangular image, the prediction...

So, I'd like to use the skew() operator. However, even after lots of testing, I can't quite figure out how magnitude affects the output. Even when I put it as...

So I noticed that when I apply random erasing, the changes get applied to the annotation. ``` p = Augmentor.Pipeline(source_directory='/dir/samples/images', save_format='png') p.random_erasing(probability=0.5, rectangle_area=0.11) p.ground_truth('/dir/samples/annotations') p.random_distortion(probability=0.6, grid_width=5, grid_height=5, magnitude=4) p.skew(probability=1, magnitude=10)...

ValueError: operands could not be broadcast together with remapped shapes [original->remapped]: (3,2) and requested shape (4,2)

2

I was so excited when I stumbled across this code, and I'm so grateful that its creator chose to share it with us. However, I'm struggling to use it. Here's...

COCO sample previews show multiple sample masks

2

I'm not sure what the deal is, but I set up my dataset as described in your guide. The json file generates fine, but when I generate several random previews,...

channel_shift_range not affecting images?

10

So, I've decided adding random channel shift would be an important image augmentation for my training set. But when I test the output, I'm not seeing any changes no matter...

It's my understanding that one of the keys to success for the first uNet was the implementation of weighted loss. This is something I've always struggled to implement myself- my...

cannot import color_palette

2

``` import os import sys sys.path.append('/usr/local/lib/python3.6/site-packages') sys.path.append('/Users/me/plantcv') #sys.path.append('/Users/me/plantcv/plantcv/plantcv') import numpy as np import cv2 from matplotlib import pyplot as plt %matplotlib inline from plantcv import plantcv as pcv from plantcv...