Machine-Learning-Collection
Machine-Learning-Collection copied to clipboard
aladdinpersson
A resource for learning about Machine learning & Deep Learning
In the original paper Upsample is done prior to the convolutions in the Generator, while Downsample is done after the convolutions within the ConvBlock. I believe the following changes should...
Hey Aladdin, thanks for your tutorials! I've been implementing the Transformer architecture and learning about einsum. Following your implementation (einsum) against one without einsum I found differences in the final...
# Description: In the Pytorch/GANs /CycleGAN/ generator_model.py file, there is a minor typo in the Generator class initialization. The second argument `9` refers to num_features, but it should be num_residuals...
`object_loss = self.mse( torch.flatten(exists_box * pred_box), torch.flatten(exists_box * target[..., 20:21]), ) # ======================= # # FOR NO OBJECT LOSS # # ======================= # #max_no_obj = torch.max(predictions[..., 20:21], predictions[..., 25:26]) #no_object_loss...
Hi, ``` import random import cv2 from matplotlib import pyplot as plt import matplotlib.patches as patches import numpy as np import albumentations as A def visualize(image): plt.figure(figsize=(10, 10)) plt.axis("off") plt.imshow(image)...
While i was playing with WGAN-gp i faced up RunTime Error that appeared in gradient_penalty function. It turns out transforms.Resize(64) sometimes produced images like 64x78, so change it to transforms.Resize((64,...
Is there a pre-trained weight for the image segmentation task? Starting from zero results in an undesirable accuracy for me:(
1 - Clearer description of the problem being solved 2 - how to install tqdm as it is not a native library in python
Metadata
Owner
Metadata
A resource for learning about Machine learning & Deep Learning