Abdul Fatir

Pretrained Discriminators

4

Hi @kwotsin! Awesome work with this repo! Would it be possible for you to release the pre-trained discriminators too?

Add REDSDS code to nursery

9

*Issue #, if available:* *Description of changes:* Adds the code for our NeurIPS 2021 paper *Deep Explicit Duration Switching Models for Time Series* to the nursery. By submitting this pull...

Tidy up preprocess.py with pandas

8

In `preprocess_csv` function in `preprocess.py` ([link](https://github.com/abdulfatir/twitter-sentiment-analysis/blob/master/preprocess.py#L71)), pandas can be used to parse the csv more efficiently and with way less code. The machine I was using while developing the project...

Possible bug in the code

3

https://github.com/eelxpeng/UnsupervisedDeepLearning-Pytorch/blob/6ea7b7151ae62bf0130b56cc023f2be068aa87f0/udlp/clustering/vade.py#L119 Hi! In this line, why did you replace `-` (minus) with `+` at the end?

Use observation mask as feature in DeepAR

2

*Description of changes:* This PR adds the observation mask as an input feature to DeepAR. Previously lags were fed into the model but the model had no information on whether...

## Description Currently predictors are serialized into directories which are a bit difficult to move around. It would be nice to save the predictor as (and load it from) an...

DeepState crashes when trained on solar dataset

1

## Description `DeepStateEstimator` crashes when trained on the `solar_nips` dataset. It looks like the reason is that the PSD constraint is not being satisfied for one of the matrices. ##...

Baseline models do not support Multivariate Data

1

## Description Several baseline models (e.g., trivial predictors, seasonal naive) do not support multivariate data. ## Environment - Operating system: MacOS 12.6 - Python version: 3.8 - GluonTS version: 0.12.3...

Opening this as a FAQ. The [`pipeline.predict`](https://github.com/amazon-science/chronos-forecasting/blob/main/src/chronos/chronos.py#L309) interface accepts either a 1D/2D tensor or a list of tensors. If you want to perform inference on a large dataset, you can...

Opening this as a FAQ. - `num_samples` specifies the number of sample forecast paths that will be generated to construct the probabilistic forecast. We used 20 samples in the paper...