-
I tried increasing `n_iter` to 500, but it still failed to generate valid samples. If it has passed the tests in the main branch, maybe it is due to some changes I made.
![图片](https://user-images.git…
-
Hi, congratulations, this is a nice job! I have a question about Appendix C3. In the paper, it can use the continuous estimation to scale the raw attention weights that bypass the problem of the non-…
-
Hello, me again !
I read in your paper that your infoloss should be based on the distribution of the subgraphs knowing the original graph and the parameters.
However, in your code, in order, you…
-
There is Jax/Flax based script available for pretraining wav2vec2 [here](https://github.com/huggingface/transformers/tree/main/examples/research_projects/jax-projects/wav2vec2). I have been trying to …
-
-
Hi, thanks for your inspiring work!
I notice that you used the default temperature=1 in all your F.gumbel_softmax implementations, and it didn't anneal to 0. Do you have any suggestions on why sho…
-
# 【PaddlePaddle Hackathon 4】核心框架开源贡献数据类型扩展任务合集
(此 ISSUE 为 PaddlePaddle Hackathon 第四期活动的任务 ISSUE,更多详见 [【PaddlePaddle Hackathon 第四期】任务总览](https://github.com/PaddlePaddle/Paddle/issues/50629))
注:为飞…
-
**Is your feature request related to a problem? Please describe.**
There is an issue with MacridVAE implementation in prototype clustering. The training procedure considers sampling a probability sim…
-
This is not a bug report.
I am confused about the PPO2 sample action strategy:
`def sample(self):`
`u = tf.random_uniform(tf.shape(self.logits))`
` return tf.argmax(self.logits - tf.log(-t…
-
(此 ISSUE 为 PaddlePaddle Hackathon 第三期活动的任务 ISSUE,更多详见 [【PaddlePaddle Hackathon 第三期】任务总览](https://github.com/PaddlePaddle/Paddle/issues/43938))
为飞桨框架新增一系列 API,提交流程请参考 [新增API 开发&提交流程](https://www.pad…