-
Implementation of ResNet34 and ResNet18 + model trained on cifar-10 linked article : https://arxiv.org/pdf/1512.03385.pdf
-
Implementation of a convolutionnal neural network with numpy
- Add ResNe50 and +
- Add vectorizaion of computation for convolution (numpy)
linked article : https://arxiv.org/pdf/1501.07338.pdf
- Add BatchNormalization layer and transpose convolution (numpy)
- Add numpy implementation in C++
- Add numpy implementation in VHDL on a FPGA
-
Implementation of Auto-Encoder and VAE (Keras) linked article : https://arxiv.org/pdf/1312.6114.pdf
-
Implementation of VanillaGAN linked article : https://arxiv.org/pdf/1406.2661.pdf
-
Implementation of DCGAN linked article : https://arxiv.org/pdf/1511.06434.pdf
-
Implementation of WGAN-GP linked article : https://arxiv.org/pdf/1704.00028.pdf
- Add FCC-GAN
linked article : https://arxiv.org/pdf/1905.02417.pdf
- Add Adversarial VAE
linked article : https://arxiv.org/pdf/1701.04722.pdf
- Generate new pokemon with VAE and FCC-GAN
- Generate mask for thresholding with Auto Encoder
-
Implementation of Proximal policy optimisation + model trained on openAI gym Bipedal-Walker and pendulum linked article : https://arxiv.org/pdf/1707.06347.pdf
-
Implementation of REINFORCE (Vanilla Policy Gradient method)
-
Implementation of actor-critic
- Fix DDPG : https://arxiv.org/pdf/1509.02971.pdf
- Add SAC
- Add DQN : https://arxiv.org/pdf/1312.5602.pdf
- Train an agent on tetris with DQN
- Add GridWorld with Tabular Q-Learning