.. _examples:

========
Examples
========

Basic Level
^^^^^^^^^^^

.. raw:: html

   <div>
     <img src="../_static/basic_fire_magic.png" style="float: right; padding-left: 24px;" />
     <p>Create instance from config file with params.</p>
     <blockquote>Please note that <b>Basic Fire Magic</b> also allows your hero to cast fire spells at reduced cost.</blockquote>
   </div>

.. code-block:: yaml

   # transform.yaml
   _target_: torchvision.transforms.Normalize
   mean: [0.5, 0.5, 0.5]
   std: [0.5, 0.5, 0.5]

.. code-block:: python

   import hydra_slayer
   import yaml

   registry = hydra_slayer.Registry()
   with open("dataset.yaml") as stream:
       raw_config = yaml.safe_load(stream)

   transform = registry.get_from_params(**raw_config)
   transform
   # Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])


Advanced Level
^^^^^^^^^^^^^^

.. raw:: html

   <div>
     <img src="../_static/advanced_fire_magic.png" style="float: right; padding-left: 24px;" />
     <p>Create <a href="https://www.cs.toronto.edu/~kriz/cifar.html">CIFAR100</a> dataset from config file with params.</p>
     <blockquote>Please note that <b>Advanced Fire Magic</b> also allows your hero to cast fire spells at reduced cost and increased effectiveness.</blockquote>
   </div>

.. code-block:: yaml

   # dataset.yaml
   _target_: torchvision.datasets.CIFAR100
   root: ./data
   train: false
   transform:
     _target_: torchvision.transforms.Compose
     transforms:
       - _target_: torchvision.transforms.ToTensor
       - _target_: torchvision.transforms.Normalize
         mean: [0.5, 0.5, 0.5]
         std: [0.5, 0.5, 0.5]
   download: true

.. code-block:: python

   import hydra_slayer
   import yaml

   registry = hydra_slayer.Registry()
   with open("dataset.yaml") as stream:
       config = yaml.safe_load(stream)

   dataset = registry.get_from_params(**config)
   dataset
   # Dataset CIFAR100
   #     Number of datapoints: 10000
   #     Root location: ./data
   #     Split: Test
   #     StandardTransform
   # Transform: Compose(
   #     ToTensor()
   #     Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])
   # )


Expert level
^^^^^^^^^^^^

.. raw:: html

   <div>
     <img src="../_static/expert_fire_magic.png" style="float: right; padding-left: 24px;" />
     <p>Read multiple CSV files as pandas dataframes and merge them.</p>
     <blockquote>Please note that <b>Expert Fire Magic</b> also allows your hero to cast fire spells at reduced cost and maximum effectiveness.</blockquote>
   </div>

.. code-block:: yaml

   # dataset.yaml
   dataframe:
     _target_: pandas.merge
     left:
       _target_: pandas.read_csv
       filepath_or_buffer: dataset/dataset_part1.csv

       # By default, hydra-slayer use partial fit for functions
       # (what is useful with activation functions in neural networks).
       # But if we want to call ``pandas.read_csv`` function instead,
       # then we should pass ``call_meta_factory`` manually.
       meta_factory: &call_function
         _target_: catalyst.tools.registry.call_meta_factory
     right:
       _target_: pandas.read_csv
       filepath_or_buffer: dataset/dataset_part2.csv
       meta_factory: *call_function
     how: inner
     on: user
     meta_factory: *call_function

.. code-block:: python

   import hydra_slayer
   import yaml

   registry = hydra_slayer.Registry()
   with open("config.yaml") as stream:
       raw_config = yaml.safe_load(stream)

   config = registry.get_from_params(**raw_config)

   dataset = config["dataframe"]
   dataset
   # <class 'pandas.core.frame.DataFrame'>
   #    user country  premium  ...
   # 0     1     USA     True  ...
   # 1     2     USA    False  ...
   #     ...     ...      ...  ...


..
  Master level
  ^^^^^^^^^^^^

  .. raw:: html

    <div>
      <img src="../_static/expert_fire_magic.png" style="float: right; padding-left: 24px;" />
      <p>Sorry, the person who is responsible for the expert level example was eaten by hydras last week.</p>
      <blockquote>Please note that <b>Maximum Fire Magic</b> also allows your hero to cast fire spells at reduced cost and maximum effectiveness.</blockquote>
    </div>

  ..
    .. code-block:: yaml

      # dataset.yaml
      _target_: torchvision.datasets.CIFAR100
      root: ./data
      train: false
      transform:
        _target_: torchvision.transforms.Compose
        transforms:
          - _target_: torchvision.transforms.ToTensor
          - _target_: torchvision.transforms.Normalize
            mean: [0.5,0.5,0.5]
            std: [0.5,0.5,0.5]
      download: true

    .. code-block:: yaml

      # config.yaml
      dataset:
        _target_: torch.utils.data.DataLoader
        # TODO: will not work as dict will be retuned, not dataset
        dataset:
          # Read dataset from "dataset.yaml", roughly equivalent to
          # with open("dataset.yaml") as stream:
          #     params = yaml.safe_load(stream)
          _target_: yaml.safe_load
          stream:
            _target_: open
            file: dataset_config.yaml
          meta_factory:
            _target_: hydra_slayer.call_meta_factory
        batch_size: 32
        shuffle: false

      model:
        _target_: torchvision.models.resnet18
        pretrained: true
        meta_factory:
          _target_: hydra_slayer.call_meta_factory

    .. code-block:: python

      import hydra_slayer
      import torch
      import yaml

      registry = hydra_slayer.Registry()
      with open("config.yaml") as stream:
          raw_config = yaml.safe_load(stream)

      config = registry.get_from_params(**raw_config)
      model, dataset = config["model"], config["dataset"]

      model.eval()
      with torch.no_grad():
          for batch, y_true in dataset:
              y_preds = model(batch)