cheatsheet.Rmd

---
title: "torch cheatsheet"
output: 
  flexdashboard::flex_dashboard:
    theme:
      version: 4
      bg: "white"
      fg: "black" 
      primary: "#ED79F9"
      base_font:
        google: Prompt
      code_font:
        google: JetBrains Mono
    orientation: columns
    vertical_layout: fill
---

```{r setup, include=FALSE}
library(flexdashboard)
# Install thematic and un-comment for themed static plots (i.e., ggplot2)
# thematic::thematic_rmd()
```

Column {data-width=250}
-----------------------------------------------------------------------

### Torch for R {data-height=500}

<div class="text-center">
```{r out.width="30%"}
knitr::include_graphics("https://torch.mlverse.org/css/images/hex/torch.png")
```
</div>

torch is a machine learning framework that helps accelerates the path from research prototyping to production deployment.

It is part of an ecosystem of packages to interface with specific dataset like 
torchaudio for audio-like and torchvision for image-like data.

### Installation

The torch R package uses the C++ libtorch library.  You can install the prerequisites directly from R.

```{r echo=TRUE, eval = FALSE}
install.packages("torch")
library(torch)
install_torch()
```

See `?install_torch` for GPU installation instructions.
See also the installation [page](https://torch.mlverse.org/docs/articles/installation.html).

### Components in torch {data-height=200}

torch has a modular API including the following compoenents:

- nn_modules
- datasets and dataloaders

Column {data-width=250}
-----------------------------------------------------------------------

### Chart C

```{r}

```

### Chart D

```{r}

```

Column {data-width=250}
-----------------------------------------------------------------------

### Chart C

```{r}

```

### Chart D

```{r}

```


Column {data-width=250}
-----------------------------------------------------------------------


### Chart C

```{r}

```

### Chart D

```{r}

```