colorization

colorization copied to clipboard

Model architecture does not have H*W*Q size tensors for (a,b) probability

1

As mentioned in the [paper](https://arxiv.org/pdf/1603.08511.pdf), the loss function uses **Z, Z^** which are of shape `H*W*Q`. However the model architecture computes the `(a,b)` probability distribution in a tensor of shape...

tathagatv

Update for requirements.txt to correct package names

2

PIL is installed via Pillow scimage is install via scikit-image

behnfeldta

Is there nay change about the model that the repo download from the amazon server? it is not working properly, output is not good. I used this model earlier, it...

celikmustafa89

How to find the 313 bins

1

First of all i'm new to machine learning, could someone explain to me how to quantize the ab color space and the one-hot encoding scheme.

Vaishnav-P

Hi, I have a question regarding the annealed mean implementation (PyTorch code). If I understand correctly this step is implemented, for testing, as an additional convolutional layer applied to the...

laaRaa

get colored images from my pretrained caffemodel

13

hello,thanks for your work. I want to use the pre-trained caffemodel that training from my own dataset to output the colored images.First I find colorize.py( Test-Time Python Script) and replace...

ZiqiYe77

Can I use official caffe code?

1

Hi, rich, I compiled your modified caffe in windows successfully, But I got a runtime error: TypeError: No to_python (by-value) converter found for C++ type: class caffe::LayerParameter. then I try...

bisonliao

import caffe

1

hi, i have this error  any ideas?

iRushdii

Hello, I wanted to resume training my model. As the README file stated, the command to resume training is: `./train/train_resume.sh ./train/models/colornet_iter_[ITERNUMBER].solverstate [GPU_ID]` However, if you look at the **train_resume.sh** file,...

amoazeni75

Hi is it possible or is there a example to train the model with PyTorch with my own dataset?

luke-777