@spec image_classification(
  Bumblebee.model_info(),
  Bumblebee.Featurizer.t(),
  keyword()
) :: Nx.Serving.t()

Builds serving for image classification.

The serving accepts image_classification_input/0 and returns image_classification_output/0. A list of inputs is also supported.

options

Options

• :top_k - the number of top predictions to include in the output. If the configured value is higher than the number of labels, all labels are returned. Defaults to 5

• :compile - compiles all computations for predefined input shapes during serving initialization. Should be a keyword list with the following keys:

  • :batch_size - the maximum batch size of the input. Inputs are optionally padded to always match this batch size

  It is advised to set this option in production and also configure a defn compiler using :defn_options to maximally reduce inference time.

• :defn_options - the options for JIT compilation. Defaults to []

examples

Examples

{:ok, resnet} = Bumblebee.load_model({:hf, "microsoft/resnet-50"})
{:ok, featurizer} = Bumblebee.load_featurizer({:hf, "microsoft/resnet-50"})

serving = Bumblebee.Vision.image_classification(resnet, featurizer)

image = ...
Nx.Serving.run(serving, image)
#=> %{
#=>   predictions: [
#=>     %{label: "Egyptian cat", score: 0.979233980178833},
#=>     %{label: "tabby, tabby cat", score: 0.00679466687142849},
#=>     %{label: "tiger cat", score: 0.005290505941957235},
#=>     %{label: "lynx, catamount", score: 0.004550771787762642},
#=>     %{label: "Siamese cat, Siamese", score: 1.1611092486418784e-4}
#=>   ]
#=> }