Yann-CV vintage-models issues

Yann-CV / vintage-models

Pytorch personal implementation of the most famous neural network models

GNU General Public License v3.0

0 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Update README.md

#31 Yann-CV closed 2 months ago
2
No more torch import

#30 Yann-CV closed 2 months ago
2
logging should be done at the end of the epoch or the step ?

#29 Yann-CV closed 3 months ago
2
Create test_to_be_removed

#28 Yann-CV closed 4 months ago
1
Create pull_request_template.md

#27 Yann-CV closed 4 months ago
1
Simplify the generator and classifier in experiments to not use list to forward to on_end methods

#26 Yann-CV closed 4 months ago
1
Rename AddPool2D as SumPool2D

#25 Yann-CV closed 4 months ago
0
Make sure every module is compatible with the to device conversion

#24 Yann-CV closed 4 months ago
1
AddPool2D should be name SumPool2D

#23 Yann-CV closed 4 months ago
1
The generation method should be in the lightning module

#22 Yann-CV closed 3 months ago
2
All experiments are instantiating the device to use with the same consistent way

#21 Yann-CV closed 5 months ago
0
All loss should be computed outside of the model in the LightningModule

#20 Yann-CV closed 3 months ago
2
make sure the import are consistent among all script/modules

#19 Yann-CV closed 3 months ago
2
in experiments, the usage of training_step_outputs, validation_step_outputs, test_step_outputs is weird

#18 Yann-CV closed 4 months ago
2
Missing type annotation in experiments

#17 Yann-CV closed 4 months ago
1
The model checkpoint is always saved with accuracy = 0.00 for generation models

#16 Yann-CV opened 5 months ago
1
All the created modules are not compatible with the the `to` instruction

#15 Yann-CV closed 4 months ago
2
Experiments - VAE on MNIST - The device of the experiments should be selected automatically from torch

#14 Yann-CV closed 4 months ago
1
Add the generative adversarial model (GAN) as vintage model

#13 Yann-CV closed 2 months ago
3
Add the variational autoencoder to vintage models

#12 Yann-CV closed 8 months ago
0
PositionalEncoding1D is not adapted to minibatches processing

#11 Yann-CV closed 8 months ago
1
add docstrings everywhere

#10 Yann-CV closed 12 months ago
0
Add already implemented models and future candidate in the README.md

#9 Yann-CV closed 11 months ago
0
Add lenet-5 as new vintage cnn

#8 Yann-CV closed 12 months ago
0
Add linting check in CI

#7 Yann-CV closed 12 months ago
0
Train and Infer image classification of a ViT model on the MNIST dataset

#6 Yann-CV closed 12 months ago
0
Add test in the CI workflow (and fix bugs)

#5 Yann-CV closed 12 months ago
0
first full version of Vit for image classification

#4 Yann-CV closed 12 months ago
0
add residual with specific behavior for self attention

#3 Yann-CV closed 1 year ago
0
add positional encoding (learnable or not) and mlp

#2 Yann-CV closed 1 year ago
0
add attention mechanisms

#1 Yann-CV closed 1 year ago
0