Dirichlet Likelihood TFT model

Description

`ValueError: Expected value argument (Tensor of shape (32, 12, 1)) to be within the support (Simplex()) of the distribution Dirichlet(concentration: torch.Size([12, 1])), but found invalid values:
tensor([[[0.5691],
         [0.8910],
         [0.6211],
         [0.0000],
         [0.3685],
         [0.3022],
         [0.5626],
         [0.3967],
         [0.6753],
         [0.6197],
         [0.4106],
         [0.6041]],
...
        [[0.0000],
         [0.0000],
         [0.0000],
         [0.0000],
         [0.0000],
         [0.0000],
         [0.4241],
         [0.3950],
         [0.3232],
         [0.2716],
         [0.3947],
         [0.2650]]])`

Contributor guide

Tech stack

pythonpytorch

Domain

datamachine learning

Issue type

bug

DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.

3

Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.

1-3 hours

Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.

stale

ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.

needs investigation

Prerequisites

PyTorch basicsTemporal Fusion TransformerDirichlet distribution

Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.

40

Research direction

The error indicates that the TFT model output tensor of shape (32, 12, 1) is being passed as the value argument to Dirichlet, but Dirichlet expects a simplex (values summing to 1 along the last dimension). The issue likely stems from the TFT model's output not being normalized. Investigate the darts source code for the TFT model and its integration with DirichletLikelihood, specifically how the output is transformed before being passed to the distribution. Look at the comments in the issue for any additional context or attempted fixes. Check if there is a missing softmax or normalization step in the model head.