-
Hi Tianhong, thank you for your inspiring work! While reading the paper, I had some questions regarding the term “MAR.” Aside from the difference mentioned in the paper—where the next set of tokens in…
-
Hello, while reading the paper, I couldn't find details on how the loss function is calculated for the feature and prediction tokens. In Section 3.3, it mentions that 'While GTL employs a next-token p…
-
### Deep Learning Simplified Repository (Proposing new issue)
:red_circle: **Project Title** : Time Series Model on Counter Strike Market Sale Dataset
:red_circle: **Aim** : To develop a time series…
arpy8 updated
2 weeks ago
-
This research area is developing very fast in the past two years, which could improve latency significantly with quality on par. Will huggingface team be interested into the implementation of it? Tha…
-
Hi, thank you for the insightful work!
I have some concerns regarding the classifier-free guidance (CFG) in auto-regressive models.
CFG in this work is implemented as follows:
https://github.…
-
#### Is your feature request related to a problem? Please describe
I've recently come across a situation when I needed to do a lot of AR model searches for a considerable number of samples. There are…
-
再请教一个问题,训练好的sat模型如何再两张GPU上加载并推理? 我目前只有A100(40G)的版本,有时推理会报内u才能溢出。
请问如何设置多GPU推理?
下面模型加载部分如何设置呢?谢谢
`# load model
model, model_args = AutoModel.from_pretrained(
args.from_pretra…
-
## Description:
I am planning to implement various causal discovery methods for time series data. The methods I am particularly interested in include CDAN, ACD, TiMINo, and NTS-NOTEARS. Each of these…
-
`TimesFMForecaster` should set `context_len` and `horizon` len to a reasonable value, automatically per default.
The `horizon_len` can be obtained from `fh`, but it makes sense to give the user the…
-
Since emulating SST from SAT doesn't seem that challenging, we would like to try to autoregressively emulate ACCESS-OM2’s vertically integrated ocean heat content evolution given surface forcing (basi…