showtens Documentation#

showtens is a package for displaying tensors EASILY. In one line, you can view image tensors, whether they are batched or not, grayscale or not. No need to spend several lines formatting, making into a grid, import matplotlib, etc. In just one line, your tensor shows up. In just one line, it is saved.

Installation instructions#

TODO

Function use :#

showtens.show_image(tensor: Tensor | ndarray, columns: int | None = None, rescale: bool = False, clamp_range: tuple[float] = (0.0, 1.0), colorbar: bool = False, max_width: int | None = None, padding: int = 3, pad_value: float = 0.0) → None[source]#

Shows tensor of shape (*,C,H,W) as an image using pyplot. Any extra dimensions are treated as batch dimensions, and displayed in a grid. By default, the image is simply clipped to the 0-1 range. Use rescale and clamp_range to modify this behaviour.

Parameters:

  • tensor – (H,W) or (C,H,W) or (*,C,H,W) tensor or numpy array to display.

  • columns – number of columns to use for the grid of images (default 8 or less)

  • rescale – whether to rescale the images to 0-1 range. Uses min-max scaling.

  • clamp_range – (min,max), tuple of values to clamp the images to, before rescaling. Use None to disable clamping

  • colorbar – whether to add a colorbar to the image, only works for grayscale images (default False)

  • max_width – maximum width of the image

  • padding – number of pixels between images in the grid

  • pad_value – inter-padding value for the grid of images

showtens.save_image(tensor: Tensor | ndarray, folder: str, name: str = 'imagetensor', columns: int | None = None, rescale: bool = False, clamp_range: tuple[float] = (0.0, 1.0), colorbar: bool = False, max_width: int | None = None, padding: int = 3, pad_value: float = 0.0, create_folder: bool = True) → None[source]#

Saves tensor of shape (*,C,H,W) as a png image. Any extra dimensions are treated as batch dimensions, and displayed in a grid. By default, the image is simply clipped to the 0-1 range. Use rescale and clamp_range to modify this behaviour.

Parameters:

  • tensor – (H,W) or (C,H,W) or (*,C,H,W) tensor or numpy array to display

  • folder – relative path of folder where to save the image

  • name – name of the image (do not include extension)

  • columns – number of columns to use for the grid of images (default 8 or less)

  • rescale – whether to rescale the images to 0-1 range

  • clamp_range – (min,max), tuple of values to clamp the images to, before rescaling. Use None to disable clamping

  • colorbar – whether to add a colorbar to the image, only works for grayscale images (default False)

  • max_width – maximum width of the image

  • padding – number of pixels between images in the grid

  • pad_value – inter-padding value for the grid of images

  • create_folder – whether to create the folder if it does not exist (default True)

showtens.save_video(tensor: Tensor | ndarray, folder: str, name: str = 'videotensor', fps: int = 30, columns: int | None = None, max_width: int | None = None, padding: int = 3, pad_value: float = 0.0, create_folder: bool = True, quality: float = 8.0) → None[source]#

Shows tensor as a video. Accepts both (T,H,W), (T,3,H,W) and (*,T,3,H,W) tensors. Tensor should be a float tensor with values in [0,1]. Clips the values otherwise.

Saved as H.264 (libx264) in an mp4 container.

Parameters:

  • tensor – (T,H,W) or (T,3,H,W) or (*,T,3,H,W) float tensor or numpy array

  • folder – path to save the video

  • name – name of the video

  • fps – fps of the video (default 30)

  • columns – number of columns to use for the grid of videos (default 8 or less)

  • max_width – maximum width of the image

  • padding – number of pixels between images in the grid

  • pad_value – inter-padding value for the grid of images

  • create_folder – whether to create the folder if it does not exist (default True)

  • quality – encoding quality between 1 and 10, 10 being lossless (default 8)

showtens.show_video(tensor: Tensor | ndarray, fps: int = 30, columns: int | None = None, max_width: int | None = None, padding: int = 3, pad_value: float = 0.0, quality: float = 8.0) → None[source]#

Shows tensor as a video. Accepts both (T,H,W), (T,3,H,W) and (*,T,3,H,W) tensors. Tensor should be a float tensor with values in [0,1]. Clips the values otherwise.

In a notebook, the video is displayed inline. Otherwise, it is opened in the default web browser. Both use the same html5 player, with play/pause and seeking.

Parameters:

  • tensor – (T,H,W) or (T,3,H,W) or (*,T,3,H,W) float tensor or numpy array

  • fps – fps of the video (default 30)

  • columns – number of columns to use for the grid of videos (default 8 or less)

  • max_width – maximum width of the video

  • padding – number of pixels between videos in the grid

  • pad_value – inter-padding value for the grid of videos

  • quality – encoding quality between 1 and 10, 10 being lossless (default 8)

showtens.gridify(tensor: Tensor, max_width: int | None = None, columns: int | None = None, padding: int = 3, pad_value: float = 0.0) → Tensor[source]#

Makes a grid of images/videos from a batch of images. Like torchvision’s make_grid, but more flexible. Accepts (B,*,H,W)

Parameters:

  • tensor – (B,*,H,W) tensor

  • max_width – max width of the output grid. Resizes images to fit the width

  • columns – number of columns of the grid. If None, uses 8 or less

  • padding – padding to add to the images

  • pad_value – color of the padding

Returns:

(*,H’,W’) tensor, representing the grid of images/videos