> ## Documentation Index
> Fetch the complete documentation index at: https://nixtla-feat-posthog-analytics.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Optimization Objectives

NeuralForecast is a highly modular framework capable of augmenting a
wide variety of robust neural network architectures with different point
or probability outputs as defined by their optimization objectives.

## Point losses

| Scale-Dependent                             | Percentage-Errors                             | Scale-Independent                           | Robust                                                    |
| :------------------------------------------ | :-------------------------------------------- | :------------------------------------------ | :-------------------------------------------------------- |
| [**MAE**](../../losses.pytorch#class-mae)   | [**MAPE**](../../losses.pytorch#class-mape)   | [**MASE**](../../losses.pytorch#class-mase) | [**Huber**](../losses.pytorch#class-huberloss)            |
| [**MSE**](../../losses.pytorch#class-mse)   | [**sMAPE**](../../losses.pytorch#class-smape) |                                             | [**Tukey**](../../losses.pytorch#class-tukeyloss)         |
| [**RMSE**](../../losses.pytorch#class-rmse) |                                               |                                             | [**HuberMQLoss**](../../losses.pytorch#class-hubermqloss) |

## Probabilistic losses

| Parametric Probabilities                                             | Non-Parametric Probabilities                                |
| :------------------------------------------------------------------- | :---------------------------------------------------------- |
| [**Normal**](../../losses.pytorch#class-distributionloss)            | [**QuantileLoss**](../../losses.pytorch#class-quantileloss) |
| [**StudenT**](../../losses.pytorch#class-distributionloss)           | [**MQLoss**](../../losses.pytorch#class-mqloss)             |
| [**Poisson**](../../losses.pytorch#class-distributionloss)           | [**HuberQLoss**](../../losses.pytorch#class-huberiqloss)    |
| [**Negative Binomial**](../../losses.pytorch#class-distributionloss) | [**HuberMQLoss**](../../losses.pytorch#class-hubermqloss)   |
| [**Tweedie**](../../losses.pytorch#class-distributionloss)           | [**IQLoss**](../../losses.pytorch#class-iqloss)             |
| [**PMM**](../../losses.pytorch#class-pmm)                            | [**HuberIQLoss**](../../losses.pytorch#class-huberiqloss)   |
| [**GMM**](../../losses.pytorch#class-gmm)                            | [**ISQF**](../../losses.pytorch#class-isqf)                 |
| [**NBMM**](../../losses.pytorch#class-nbmm)                          |                                                             |