kindle 0.2.4

PyTorch no-code model builder.

Kindle - PyTorch no-code model builder

Documentation

Kindle is an easy model build package for PyTorch. Building a deep learning model became so simple that almost all model can be made by copy and paste from other existing model codes. So why code? when we can simply build a model with yaml markup file.

Kindle builds a model with no code but yaml file which its method is inspired from YOLOv5.

Contents

• Installation
  • Install with pip
  • Install from source
  • For contributors
• AutoML with Kindle
• Usage
• Supported modules
• Custom module support
  • Custom module with yaml
  • Custom module from source

Installation

Install with pip

PyTorch is required prior to install. Please visit PyTorch installation guide to install.

You can install kindle by pip.

$ pip install kindle

Install from source

Please visit Install from source wiki page

For contributors

Please visit For contributors wiki page

Usage

Build a model

1. Make model yaml file

• Example model https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html

input_size: [32, 32]
input_channel: 3

depth_multiple: 1.0
width_multiple: 1.0

backbone:
    [
        [-1, 1, Conv, [6, 5, 1, 0], {activation: LeakyReLU}],
        [-1, 1, MaxPool, [2]],
        [-1, 1, nn.Conv2d, [16, 5, 1, 2], {bias: False}],
        [-1, 1, nn.BatchNorm2d, []],
        [-1, 1, nn.ReLU, []],
        [-1, 1, MaxPool, [2]],
        [-1, 1, Flatten, []],
        [-1, 1, Linear, [120, ReLU]],
        [-1, 1, Linear, [84, ReLU]],
    ]

head:
  [
      [-1, 1, Linear, [10]]
  ]

2. Build the model with kindle

from kindle import Model

model = Model("model.yaml"), verbose=True)

idx |       from |   n |   params |          module |                           arguments | in_channel | out_channel |        in shape |       out shape |
----------------------------------------------------------------------------------------------------------------------------------------------------------
  0 |         -1 |   1 |      616 |            Conv | [6, 5, 1, 0], activation: LeakyReLU |          3 |           8 |     [3, 32, 32] |     [8, 32, 32] |
  1 |         -1 |   1 |        0 |         MaxPool |                                 [2] |          8 |           8 |       [8 32 32] |     [8, 16, 16] |
  2 |         -1 |   1 |    3,200 |       nn.Conv2d |          [16, 5, 1, 2], bias: False |          8 |          16 |       [8 16 16] |    [16, 16, 16] |
  3 |         -1 |   1 |       32 |  nn.BatchNorm2d |                                  [] |         16 |          16 |      [16 16 16] |    [16, 16, 16] |
  4 |         -1 |   1 |        0 |         nn.ReLU |                                  [] |         16 |          16 |      [16 16 16] |    [16, 16, 16] |
  5 |         -1 |   1 |        0 |         MaxPool |                                 [2] |         16 |          16 |      [16 16 16] |      [16, 8, 8] |
  6 |         -1 |   1 |        0 |         Flatten |                                  [] |         -1 |        1024 |        [16 8 8] |          [1024] |
  7 |         -1 |   1 |  123,000 |          Linear |                       [120, 'ReLU'] |       1024 |         120 |          [1024] |           [120] |
  8 |         -1 |   1 |   10,164 |          Linear |                        [84, 'ReLU'] |        120 |          84 |           [120] |            [84] |
  9 |         -1 |   1 |      850 |          Linear |                                [10] |         84 |          10 |            [84] |            [10] |
Model Summary: 20 layers, 137,862 parameters, 137,862 gradients

AutoML with Kindle

• Kindle offers the easiest way to build your own deep learning architecture. Beyond building a model, AutoML became easier with Kindle and Optuna or other optimization frameworks.
• For further information, please refer to https://limjk.ai/kindle/usages/#automl-with-optuna

Supported modules

• Detailed documents can be found https://limjk.ai/kindle/modules/

Module	Components	Arguments
Conv	Conv -> BatchNorm -> Activation	[out_channels, kernel_size, stride, padding, groups, activation]
DWConv	DWConv -> BatchNorm -> Activation	[out_channels, kernel_size, stride, padding, activation]
Bottleneck	Expansion ConvBNAct -> ConvBNAct	[out_channels, shortcut, groups, expansion, activation]
AvgPool	Average pooling	[kernel_size, stride, padding]
MaxPool	Max pooling	[kernel_size, stride, padding]
GlobalAvgPool	Global Average Pooling	[]
Flatten	Flatten	[]
Concat	Concatenation	[dimension]
Linear	Linear	[out_channels, activation]
Add	Add	[]
UpSample	UpSample	[]
Identity	Identity	[]
YamlModule	Custom module from yaml file	['yaml/file/path', arg0, arg1, ...]
nn.{module_name}	PyTorch torch.nn.* module	Please refer to https://pytorch.org/docs/stable/nn.html

• nn.{module_name} is currently experimental. This might change in the future release. Use with caution.

Custom module support

Custom module with yaml

You can make your own custom module with yaml file.

1. custom_module.yaml

args: [96, 32]

module:
    # [from, repeat, module, args]
    [
        [-1, 1, Conv, [arg0, 1, 1]],
        [0, 1, Conv, [arg1, 3, 1]],
        [0, 1, Conv, [arg1, 5, 1]],
        [0, 1, Conv, [arg1, 7, 1]],
        [[1, 2, 3], 1, Concat, [1]],
        [[0, 4], 1, Add, []],
    ]

• Arguments of yaml module can be defined as arg0, arg1 ...

2. model_with_custom_module.yaml

input_size: [32, 32]
input_channel: 3

depth_multiple: 1.0
width_multiple: 1.0

backbone:
    [
        [-1, 1, Conv, [6, 5, 1, 0]],
        [-1, 1, MaxPool, [2]],
        [-1, 1, YamlModule, ["custom_module.yaml", 48, 16]],
        [-1, 1, MaxPool, [2]],
        [-1, 1, Flatten, []],
        [-1, 1, Linear, [120, ReLU]],
        [-1, 1, Linear, [84, ReLU]],
        [-1, 1, Linear, [10]]
    ]

• Note that argument of yaml module can be provided.

3. Build model

from kindle import Model

model = Model("model_with_custom_module.yaml"), verbose=True)

idx |       from |   n |     params |          module |            arguments |                       in shape |       out shape |
---------------------------------------------------------------------------------------------------------------------------------
  0 |         -1 |   1 |        616 |            Conv |         [6, 5, 1, 0] |                    [3, 32, 32] |     [8, 32, 32] |
  1 |         -1 |   1 |          0 |         MaxPool |                  [2] |                      [8 32 32] |     [8, 16, 16] |
  2 |         -1 |   1 |     10,832 |      YamlModule |    ['custom_module'] |                      [8 16 16] |    [24, 16, 16] |
  3 |         -1 |   1 |          0 |         MaxPool |                  [2] |                     [24 16 16] |      [24, 8, 8] |
  4 |         -1 |   1 |          0 |         Flatten |                   [] |                       [24 8 8] |          [1536] |
  5 |         -1 |   1 |    184,440 |          Linear |        [120, 'ReLU'] |                         [1536] |           [120] |
  6 |         -1 |   1 |     10,164 |          Linear |         [84, 'ReLU'] |                          [120] |            [84] |
  7 |         -1 |   1 |        850 |          Linear |                 [10] |                           [84] |            [10] |
Model Summary: 36 layers, 206,902 parameters, 206,902 gradients

Custom module from source

You can make your own custom module from the source.

1. custom_module_model.yaml

input_size: [32, 32]
input_channel: 3

depth_multiple: 1.0
width_multiple: 1.0

custom_module_paths: ["tests.test_custom_module"]  # Paths to the custom modules of the source

backbone:
    # [from, repeat, module, args]
    [
        [-1, 1, MyConv, [6, 5, 3]],
        [-1, 1, MaxPool, [2]],
        [-1, 1, MyConv, [16, 3, 5, SiLU]],
        [-1, 1, MaxPool, [2]],
        [-1, 1, Flatten, []],
        [-1, 1, Linear, [120, ReLU]],
        [-1, 1, Linear, [84, ReLU]],
        [-1, 1, Linear, [10]]
    ]

2. Write PyTorch module and ModuleGenerator

tests/test_custom_module.py

from typing import List, Union, Dict, Any

import numpy as np
import torch
from torch import nn

from kindle.generator import GeneratorAbstract
from kindle.torch_utils import Activation, autopad


class MyConv(nn.Module):
    def __init__(
        self,
        in_channels: int,
        out_channels: int,
        kernel_size: int,
        n: int,
        activation: Union[str, None] = "ReLU",
    ) -> None:
        super().__init__()
        convs = []
        for i in range(n):
            convs.append(
                nn.Conv2d(
                    in_channels,
                    in_channels if (i + 1) != n else out_channels,
                    kernel_size,
                    padding=autopad(kernel_size),
                    bias=False,
                )
            )

        self.convs = nn.Sequential(*convs)
        self.batch_norm = nn.BatchNorm2d(out_channels)
        self.activation = Activation(activation)()

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        return self.activation(self.batch_norm(self.convs(x)))


class MyConvGenerator(GeneratorAbstract):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)

    @property
    def out_channel(self) -> int:
        return self._get_divisible_channel(self.args[0] * self.width_multiply)

    @property
    def in_channel(self) -> int:
        if isinstance(self.from_idx, list):
            raise Exception("from_idx can not be a list.")
        return self.in_channels[self.from_idx]

    @property
    def kwargs(self) -> Dict[str, Any]:
        args = [self.in_channel, self.out_channel, *self.args[1:]]
        return self._get_kwargs(MyConv, args)

    @torch.no_grad()
    def compute_out_shape(self, size: np.ndarray, repeat: int = 1) -> List[int]:
        module = self(repeat=repeat)
        module.eval()
        module_out = module(torch.zeros([1, *list(size)]))
        return list(module_out.shape[-3:])

    def __call__(self, repeat: int = 1) -> nn.Module:
        if repeat > 1:
            module = [MyConv(**self.kwargs) for _ in range(repeat)]
        else:
            module = MyConv(**self.kwargs)

        return self._get_module(module)

3. Build a model

from kindle import Model

model = Model("custom_module_model.yaml"), verbose=True)

idx |       from |   n |   params |          module |                           arguments | in_channel | out_channel |        in shape |       out shape |
----------------------------------------------------------------------------------------------------------------------------------------------------------
  0 |         -1 |   1 |    1,066 |          MyConv |                           [6, 5, 3] |          3 |           8 |     [3, 32, 32] |     [8, 32, 32] |
  1 |         -1 |   1 |        0 |         MaxPool |                                 [2] |          8 |           8 |       [8 32 32] |     [8, 16, 16] |
  2 |         -1 |   1 |    3,488 |          MyConv |                  [16, 3, 5, 'SiLU'] |          8 |          16 |       [8 16 16] |    [16, 16, 16] |
  3 |         -1 |   1 |        0 |         MaxPool |                                 [2] |         16 |          16 |      [16 16 16] |      [16, 8, 8] |
  4 |         -1 |   1 |        0 |         Flatten |                                  [] |         -1 |        1024 |        [16 8 8] |          [1024] |
  5 |         -1 |   1 |  123,000 |          Linear |                       [120, 'ReLU'] |       1024 |         120 |          [1024] |           [120] |
  6 |         -1 |   1 |   10,164 |          Linear |                        [84, 'ReLU'] |        120 |          84 |           [120] |            [84] |
  7 |         -1 |   1 |      850 |          Linear |                                [10] |         84 |          10 |            [84] |            [10] |
Model Summary: 29 layers, 138,568 parameters, 138,568 gradients

Planned features

• Custom module support
• Custom module with yaml support
• Graphical model file generator
• Use pre-trained model
• More modules!