-
Hi everyone! In a recent issue, I discovered that Deepchem seems to lack native support for evaluating GANs. I wonder if there are steps that we can take so that functions like `.evaluate()` are prese…
-
Hello! I've found a performance issue in /awesome_gans/pggan/pggan_train.py: `with tf.Session(config=config) as s`[(here)](https://github.com/kozistr/Awesome-GANs/blob/20062fb5347921af4f2053465dc02677…
-
- https://arxiv.org/abs/2104.12476
- 2021
Generative Adversarial Network(GAN)に関する最近の研究では、生成的CNNの異なる層が合成画像の異なるセマンティクスを保持していることが明らかになった。
しかし、特定の層で表現される意味的属性を制御するための明示的な次元を持つGANモデルはほとんどない。
本論文では、異な…
e4exp updated
3 years ago
-
- https://arxiv.org/abs/2103.16748
- 2021
Generative Adversarial Networks (GAN)は、大規模な画像データセットを用いて無条件に画像を生成する際に素晴らしい結果を出す。
しかし、生成された画像は、特に分散性の高いデータセット(寝室や教会など)では、まだ見分けがつきにくい。
本論文では、画像生成の限界をさらに押し広…
e4exp updated
3 years ago
-
- https://arxiv.org/abs/2103.12685
- 2021
機械学習における多くのアプリケーションは、最小化問題として組み立てられ、勾配ベースの技術を用いて効率的に解くことができる。
しかし、最近の生成モデル、特にGANの応用により、標準的な最適化手法が適さないことが多いミニマックスゲームを解くことに関心が集まっている。
実務者が経験する既知の問題の中には、収束…
e4exp updated
3 years ago
-
G(z) is redundant in the formula, in which I added a text to point out.
-
Contributors: @margaretmz, @sayakpaul
GitHub repo: https://github.com/margaretmz/esrgan-e2e-tflite-tutorial/
-
My problem is that I want to train a network to do the following: input a set of parameters and output a image. which gans can i use or at least refer to. anybody can help me?
-
I'm trying to create a new model by myself by using the following command:
``` shell
python train.py --outdir=training-runs --data=https://data-efficient-gans.mit.edu/datasets/100-shot-obama.zip --g…
-
### Deep Learning Simplified Repository (Proposing new issue)
:red_circle: **Project Title** : Sudoku Solver using CNN
:red_circle: **Aim** :
Network was able to solve the puzzles with 95% accurac…