Skip to content

GitLab

Commit a960f549 authored by mibaumgartner's avatar mibaumgartner
Browse files

init prototype

parents
Show whitespace changes
Inline Side-by-side
Showing with 1302 additions and 0 deletions
.gitignore 0 → 100644
View file @ a960f549
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class
*.vscode
*.simg
*.sif
# C extensions
*.so
# Distribution / packaging
.Python
env/
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
*.egg-info/
.installed.cfg
*.egg
# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec
# Installer logs
pip-log.txt
pip-delete-this-directory.txt
# Unit test / coverage reports
htmlcov/
.tox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*,cover
.hypothesis/
# Translations
*.mo
*.pot
# Django stuff:
*.log
local_settings.py
# Flask stuff:
instance/
.webassets-cache
# Scrapy stuff:
.scrapy
# Sphinx documentation
docs/_build/
# PyBuilder
target/
# IPython Notebook
.ipynb_checkpoints
# pyenv
.python-version
# celery beat schedule file
celerybeat-schedule
# dotenv
.env
# virtualenv
venv/
ENV/
# Spyder project settings
.spyderproject
# Rope project settings
.ropeproject
*.memmap
*.png
*.zip
*.npz
*.npy
*.jpg
*.jpeg
.idea
*.txt
.idea/*
*.png
*.nii.gz
*.nii
*.tif
*.bmp
*.pkl
*.xml
*.pkl
*.pdf
*.png
*.jpg
*.jpeg
*.model
.DS_Store
Dockerfile 0 → 100644
View file @ a960f549
FROM nvidia/cuda:11.1.1-devel-ubuntu20.04
ARG env_det_num_threads=6
ARG env_det_verbose=1
# Setup environment variables
ENV TORCH_CUDA_ARCH_LIST=6.1+PTX;7.0+PTX;7.5+PTX FORCE_CUDA=1
ENV det_data=/opt/data det_models=/opt/models det_num_threads=$env_det_num_threads det_verbose=$env_det_verbose OMP_NUM_THREADS=1
# Install some tools
RUN apt-get update && export DEBIAN_FRONTEND=noninteractive && apt-get install -y \
git \
cmake \
make \
wget \
gnupg \
build-essential \
software-properties-common \
gdb \
ninja-build
# Setup miniconda and create a new python environment with python 3.7
RUN wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh \
&& chmod +x miniconda.sh \
&& ./miniconda.sh -b -p /opt/miniconda \
&& rm ./miniconda.sh \
&& ln -s /opt/miniconda/bin/activate /activate \
&& . /activate \
&& pip install numpy \
&& pip install torch==1.8.1+cu111 torchvision==0.9.1+cu111 torchaudio==0.8.1 -f https://download.pytorch.org/whl/torch_stable.html
# Install own code
COPY ./requirements.txt .
RUN mkdir ${det_data} \
&& mkdir ${det_models} \
&& mkdir -p /opt/code/nndet \
&& . /activate \
&& pip install -r requirements.txt \
&& pip install hydra-core --upgrade --pre \
&& pip install git+https://github.com/mibaumgartner/pytorch_model_summary.git
WORKDIR /opt/code/nndet
COPY . .
RUN . /activate && pip install -v -e .
README.md 0 → 100644
View file @ a960f549
<div align="center">
<img src=docs/source/nnDetection.svg width="600px">
![Version](https://img.shields.io/badge/nnDetection-v1.0-blue)
![Python](https://img.shields.io/badge/python-3.8-orange)
![CUDA](https://img.shields.io/badge/CUDA-10.1%2F10.2%2F11.0-green)
![license](https://img.shields.io/badge/License-Apache%202.0-red.svg)
</div>
# Installation
1. Install CUDA (>10.1) and cudnn (make sure to select [compatible versions](https://docs.nvidia.com/deeplearning/cudnn/support-matrix/index.html)!)
2. [Optional] Depending on your GPU you might need to set `TORCH_CUDA_ARCH_LIST`, check [compute capabilities](https://developer.nvidia.com/cuda-gpus) here.
3. Install [torch](https://pytorch.org/) (make sure to match the pytorch and CUDA versions!) (requires pytorch >1.7+)
4. Install [torchvision](https://github.com/pytorch/vision) (make sure to match the versions!)
5. Clone nnDetection, `cd [path_to_repo]` and `pip install -e .`
6. Upgrade hydra to next release: `pip install hydra-core --upgrade --pre`
7. Set environment variables (more info can be found below):
- `det_data`: [required] Path to the source directory where all the data will be located
- `det_models`: [required] Path to directory where all models will be saved
- `OMP_NUM_THREADS=1` : [required] Needs to be set! Otherwise bad things will happen... Refer to batchgenerators documentation.
- `det_num_threads`: [recommended] Number processes to use for augmentation (at least 6, default 12)
- `det_verbose`: [optional] Can be used to deactivate progress bars (activated by default)
- `MLFLOW_TRACKING_URI`: [optional] Specify the logging directory of mlflow. Refer to the [mlflow documentation](https://www.mlflow.org/docs/latest/tracking.html) for more information.
Note: nnDetection was developed on Linux => Windows is not supported.
<details close>
<summary>Test Installation</summary>
<br>
Run the following command in the terminal (!not! in pytorch root folder) to verify that the compilation of the C++/CUDA code was successfull:
```bash
python -c "import torch; import nndet._C; import nndet"
```
To test the whole installation please run the Toy Dataset example.
</details>
<details close>
<summary>Maximising Training Speed</summary>
<br>
To get the best possible performance we recommend using CUDA 11.0+ with cuDNN 8.1.X+ and a (!)locally compiled version(!) of Pytorch 1.7+
</details>
<details close>
<summary>Docker Container</summary>
<br>
The provided Dockerfile can be used to setup quick development environments or deploy nnDetection.
Please install [nvidia-docker](https://github.com/NVIDIA/nvidia-docker) before continuing.
All projects which are based on nnDetection assume that the base image was build with the following tagging scheme `nnDetection:[version]`.
To build a container (nnDetection Version 0.1) run the following command from the base directory:
```bash
docker build -t nndetection:0.1 .
```
or
```bash
docker build -t nndetection:0.1 --build-arg env_det_num_threads=6 --build-arg env_det_verbose=1 .
```
to overwrite the provided default parameters.
The docker container expects the data and models in `/opt/data` and `/opt/models` respectively.
The directories need to be mounted via docker commands e.g.
```bash
docker run --gpus all nndetection -v /path/to/data/on/pc:/opt/data -v /path/to/models/on/pc:/opt/models -it nndetection:0.1 /bin/bash
```
If nnDetection is already configured on the host PC the following command can be used to start the container with the correct paths.
```bash
docker run --gpus all -v ${det_data}:/opt/data -v ${det_models}:/opt/models -it nndetection:0.1 /bin/bash
```
After activating the environment via `. /activate` inside the container, training or inference scripts can be executed with the usual commands (see below).
Warning:
1. The current pytorch versions do not support the 3d conv speed up and thus compiling pytorch from source will run faster than this container.
2. When running a training inside the container it is necessary to [increase the shared memory](https://stackoverflow.com/questions/30210362/how-to-increase-the-size-of-the-dev-shm-in-docker-container).
I tested the following configuration on my local workstation:
```bash
docker run --gpus all -v ${det_data}:/opt/data -v ${det_models}:/opt/models -it nndetection:0.1 --shm-size=24gb /bin/bash
```
</details>
# nnDetection
<div align="center">
<img src=docs/source/nnDetectionFunctional.svg width="600px">
</div>
<details close>
<summary>nnDetection Module Overview</summary>
<br>
<div align="center">
<img src=docs/source/nnDetectionModule.svg width="600px">
</div>
</details>
<details close>
<summary>nnDetection Functional Details</summary>
<br>
<div align="center">
<img src=docs/source/nnDetectionFunctionalDetails.svg width="600px">
</div>
</details>
# Experiments & Data
The datasets used for our experiments are not hosted or maintained by us, please give credit to the authors of the datasets.
Some of the labels were corrected in datasets which we converted and can be downloaded.
The `Reproducing Experiments` section has an overview of multiple guides which explain the preparation of the datasets.
## Toy Dataset
Running `nndet_example` will automatically generate an example dataset with 3D squares and sqaures with holes which can be used to test the installation or experiment with prototype code.
The problem is very easy and the final results should be near perfect.
After running the generation script follow the `Planning`, `Training` and `Inference` instructions below to construct the whole nnDetection pipeline.
## Reproducing Experiments
<div align="center">
| <!-- --> | <!-- --> | <!-- --> |
|:--------------------------------:|:----------------------:|:----------------------------:|
| [Task 003 Liver](#TODO) | [Task 011 Kits](#TODO) | [Task 020 RibFrac](#TODO) |
| [Task 007 Pancreas](#TODO) | [Task 012 LIDC](#TODO) | [Task 021 ProstateX](#TODO) |
| [Task 008 Hepatic Vessel](#TODO) | [Task 017 CADA](#TODO) | [Task 025 LymphNodes](#TODO) |
| [Task 010 Colon](#TODO) | [Task 019 ADAM](#TODO) | [Task 016 Luna](#TODO) |
</div>
## Adding New Datasets
nnDetection relie on a standardized input format which is very similar to the [nnU-Net](https://github.com/MIC-DKFZ/nnUNet) format and allows easy integration of new datasets.
The format is explained below.
### Folders
All datasets should reside inside `Task[Number]_[Name]` folder inside the specified detection data folder (et the path to this folder with the `det_data` environment flag).
An overview is provided below ([Name] symbolise folder, `-` symbolise files, indents refer to substructures)
```text
${det_data}
[Task000_Example]
- dataset.yaml # dataset.json works too
[raw_splitted_data]
[imagesTr]
- case0000_0000.nii.gz # case0000 modality 0
- case0000_0001.nii.gz # case0000 modality 1
- case0001_0000.nii.gz # case0001 modality 0
- case0000_0001.nii.gz # case0001 modality 1
[labelsTr]
- case0000.nii.gz # instance segmentation case0000
- case0000.json # properties of case0000
- case0001.nii.gz # instance segmentation case0001
- case0001.json # properties of case0001
[imagesTs] # optional, same structure as imagesTr
...
[labelsTs] # optional, same structure as labelsTr
...
[Task001_Example1]
...
```
### Dataset Info
`dataset.yaml` or `dataset.json` provides general information about the dataset:
Note: [Important] Classes and modalities start with index 0!
```yaml
task: Task000D3_Example
name: "Example" # [Optional]
dim: 3 # number of spatial dimensions of the data
target_class: # define class of interest for patient level evaluations # TODO: check if this should be included
test_labels: True # manually splitted test set
labels: # classes of dataset; need to start at 0
"0": "Square"
"1": "SquareHole"
modalities: # modalities of dataset; need to start at 0
"0": "CT"
```
### Image Format
nnDetection uses the same image format as nnU-Net.
Each case consists of at least one 3D nifty file with one modalityand are saved in the `images` folders.
If multiple modalities are available, each modalities uses a separate file and the sequence at the end of the name indicates the modality (corresponds to the number specified in the dataset file).
An example with two modalities could look like this:
```text
- case001_0000.nii.gz # Case ID: case001; Modality: 0
- case001_0001.nii.gz # Case ID: case001; Modality: 1
- case002_0000.nii.gz # Case ID: case002; Modality: 0
- case002_0001.nii.gz # Case ID: case002; Modality: 1
```
If multiple modalities are available, please check beforehand if they need to be registered and perform registration befor nnDetection preprocessing. nnDetection does (!)not(!) include automatic registration of multiple modalities.
### Label Format
Labels are encoded with two files per case: one nifty file which contains the instance segmentation and one json file which includes the "meta" information of each instance.
The nifty file hould contain all annotated instances where each instance has a unique number and are in consecutive order (e.g. 0 ALWAYS refers to background, 1 refers to the first instance, 2 refers to the second instance ...)
`case[XXXX].json` label files need to provide the class of every instance in the segmentation. In this example the first isntance is assigned to class `0` and the second instance is assigned to class `1`:
```json
{
"instances": {
"1": 0,
"2": 1
}
}
```
Each label file needs a corresponding json file to define the classes.
## Using nnDetection
The following paragrah provides an high level overview of the functionality of nnDetection and which commands are available.
A typical flow of commands would look like this:
```text
nndet_prep -> nndet_unpack -> nndet_train -> nndet_consolidate -> nndet_predict
```
Eachs of this commands is explained below and more detailt information can be obtained by running `nndet_[command] -h` in the terminal.
### Planning & Preprocessing
Before training the networks, nnDetection needs to preprocess and analyze the data.
The preprocessing stage noramlizaes and resamples the data while the analyzed properties are used to create a plan which will be used for configuring the training.
nnDetectionV0 requires a GPU with approximately the same amount of VRAM you are planning to use for training (i.e. we used a completely freed RTX2080TI) to perform live estimation of the VRAM used by the network.
Future releases will improve this process...
```bash
nndet_prep [tasks] [-o / --overwrites]
# Example
nndet_prep 000
# Script
# /experiments/preprocess.py - main()
```
`-o` option can be used to overwrite parameters for planning and preprocessing (refer to the onfig files to see all parameters). A typical usecase is to increase or decrease `prep.num_processes` (number of processes used for cropping) and `prep.num_processes_processing` (number of processes used for resampling) depending on the size/number of modalities of the data and available RAM. The current values are fairly save if 64GB of RAM is available.
After planning and preprocessing the resulting data folder structure should look like this:
```text
[Task000_Example]
[raw_splitted]
[raw_cropped] # only needed for different resampling strategies
[imagesTr] # stores cropped image data; contains npz files
[labelsTr] # stores labels
[preprocessed]
[analysis]
[properties] # sufficient for new plans
[labelsTr] # labels in original format (original spacing)
[labelsTs] # optional
[Data identifier; e.g. D3V001_3d]
[imagesTr] # preprocessed data
[labelsTr] # preprocessed labels (resampled spacing)
- {name of plan}.pkl e.g. D3V001_3d.pkl
```
Befor starting the training copy the data (Task Folder, dataset info and preprocessed folder are needed) to a SSD (highly recommended) and unpack the image data with
TODO: update name after reafactoring planner name
```bash
nndet_unpack [path] [num_processes]
# Example (unpack example with 6 processes)
nndet_unpack ${det_data}/Task000D3_Example/preprocessed/D3C002_3d/imagesTr 6
# Script
# /experiments/utils.py - unpack()
```
### Training and Evaluation
After the planning and preprocessing stage is finished the training phase can be started.
The default setup of nnDetection is trained in a 5 fold cross-validation scheme.
First, check which plans were generated during planning by checken the preprocessing folder and looking for the pickled plan files. In most cases only the defaul plan will be generated (`D3V001_3d`) but there might be instances (e.g. Kits) where the low resolution plan will be generated too (`D3V001LR1_3d`).
```bash
nndet_train [task] [-o / --overwrites] [--sweep]
# Example (train default plan D3V001_3d and search best inference parameters)
nndet_train 000 --sweep
# Script
# /experiments/train.py - train()
```
Use `-o exp.fold=X` to overwrite the trained fold, this should be run for all folds `X = 0, 1, 2, 3, 4`!
The `--sweep` option tells nnDetection to look for the best hyparameters for inference by empirically evaluating them on the validation set.
Sweeping can also be performed later by running the following command:
```bash
nndet_sweep [task] [model] [fold]
# Example (sweep Task 000 of model RetinaUNetV001_D3V001_3d in fold 0)
nndet_sweep 000 RetinaUNetV001_D3V001_3d 0
# Script
# /experiments/train.py - sweep()
```
Evaluation can be invoked by the following command (requires access to model and preprocessed data):
```bash
nndet_eval [task] [model] [fold] [--test] [--case] [--boxes] [--seg] [--instances] [--analyze_boxes]
# Example (evaluate and analyze box predictions of default model)
nndet_eval 000 RetinaUNetV001_D3V001_3d 0 --boxes --analyze_boxes
# Script
# /experiments/train.py - evaluate()
# Note: --test invokes evaluation of the test set
# Note: --seg, --instances are placeholders for future versions and not working yet
```
### Inference
After running all fold it is time to collect the models and creat a unified inference plan.
The following command will copy all the models and predictions per fold and by adding the `sweep` options the empiricaly hyperparameter optimization across all fold can be started.
This will generate a unified plan for all models which will be used during inference.
```bash
nndet_consolidate [task] [model] [--overwrites] [--consolidate] [--num_folds] [--no_model] [--sweep_boxes] [--sweep_instances]
# Example
nndet_consolidate 000 RetinaUNetV001_D3V001_3d --sweep_boxes
# Script
# /experiments/consolidate.py - main()
```
Data which is located in `raw_splitted/imagesTs` will be automatically preprocessed and predicted by running the following command:
```bash
nndet_predict [task] [model] [--fold] [--num_models] [--num_tta] [--no_preprocess]
# Example
nndet_predict 000 RetinaUNetV001_D3V001_3d --fold -1
# Script
# /experiments/predict.py - main()
# Note: --num_models is not supported by default
```
If a self-made test set was used, evaluation can be performed by invoking `nndet_eval` as described above.
## nnU-Net for Detection
TODO
## Pretrained models
TODO
# FAQ
<details close>
<summary>GPU requirements</summary>
<br>
nnDetection v0.1 was developed for GPUs with at least 11GB of VRAM (e.g. RTX2080TI, TITAN RTX).
All of our experiments were conducted with a RTX2080TI.
While the memory can be adjusted by manipulating the correct setting we recommend using the default values for now.
Future releases will refactor the planning stage to improve the VRAM estimation and add support for different memory budgets.
</details>
<details close>
<summary>Error: Undefined CUDA symbols when importing `nndet._C`</summary>
<br>
Please double check CUDA version of your PC, pytorch, torchvision and nnDetection build!
Follow the installation instruction at the beginning!
</details>
<details close>
<summary>Error: No kernel image is available for execution"</summary>
<br>
You are probably executing the build on a machine with a GPU architecture which was not present/set during the build.
Please check [link](https://developer.nvidia.com/cuda-gpus) to find the correct SM architecture and set `TORCH_CUDA_ARCH_LIST`
approriately (e.g. check Dockefile for example).
Make sure to delete all caches before rebulding!
</details>
<details close>
<summary>Training with bounding boxes</summary>
<br>
The first release of nnDetection focuses on 3d medical images and Retina U-Net.
As a consequence training (specifically planning and augmentation) requrie segmentation annotations.
In many cases this limitation can be circumvented by converting the bounding boxes into segmentations.
</details>
<details close>
<summary>Mask RCNN and 2D Datasets</summary>
<br>
2D datasets and Mask R-CNN are not supported in the first release.
We hope to provide these sometime in the future.
</details>
<details close>
<summary>Multi GPU Training</summary>
<br>
Multi GPU training is not officially supported yet.
Inference and the metric computation are not properly designed to support these usecases!
</details>
<details close>
<summary>Prebuild package</summary>
<br>
We are planning to provide prebuild wheels in the future but no prebuild wheels are available right now.
Please use the provided Dockerfile or the installation instructions to run nnDetection.
</details>
# Cite
If you use nnDetection for your project/research/work please cite the following paper:
```text
TODO
```
# Acknowledgements
TODO
\ No newline at end of file
docs/source/nnDetection.svg 0 → 100644
View file @ a960f549
<?xml version="1.0" encoding="UTF-8"?>
<svg width="765px" height="289px" viewBox="0 0 765 289" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<title>Group</title>
<g id="Page-1" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
<g id="Group" transform="translate(1.007300, 0.397975)">
<g id="Group-53" transform="translate(7.626700, 0.000000)" font-family="Arial-BoldMT, Arial" font-weight="bold">
<text id="nnDet" font-size="126" letter-spacing="-0.0002" fill="#5E5E5E">
<tspan x="0" y="114">nnDet</tspan>
</text>
<text id="-" font-size="53"></text>
<text id="ection" font-size="126" letter-spacing="0.0252" fill="#5E5E5E">
<tspan x="371.6841" y="114">ec</tspan>
<tspan x="511.884891" y="114">t</tspan>
<tspan x="553.843875" y="114" letter-spacing="-0.0252">io</tspan>
<tspan x="665.766131" y="114">n</tspan>
</text>
</g>
<polygon id="Stroke-58" stroke="#5CC2B2" stroke-width="7" points="5.2865 11.185525 371.6251 11.185525 371.6251 135.677325 5.2865 135.677325"></polygon>
<g id="Group-63" transform="translate(192.065100, 145.287206)">
<path d="M68.4884,0 C65.666,0.0130191088 62.8409,1.05751911 62.8409,3.08651911 L62.8409,28.6236191 C62.8409,31.8406191 61.6053,34.7711191 59.5831,36.9716191 C59.5858,34.1675191 59.5883,31.8290191 59.5938,30.1346191 C59.661,9.49191911 45.7074,6.85761911 30.8311,22.6331191 C4.3138,50.7525191 0,81.3004191 0,121.931019 C0,128.033419 3.8963,133.311219 9.977,132.797419 C29.9023,131.114019 59.5938,121.760119 59.5938,99.8767191 C59.5938,92.2217191 59.5721,68.7012191 59.5724,50.4528191 C63.0363,49.0050191 66.0883,46.7554191 68.4884,43.9373191 C70.9602,46.8394191 74.112,49.1291191 77.7045,50.5707191 C77.7048,68.8016191 77.6938,92.2392191 77.6938,99.8767191 C77.6938,121.759819 107.3746,131.114019 127.2999,132.797419 C133.3806,133.311219 137.2875,128.033419 137.2875,121.931019 C137.2875,81.3004191 132.9738,50.7525191 106.4565,22.6331191 C91.5801,6.85791911 77.6266,9.49231911 77.6938,30.1346191 C77.6996,31.8948191 77.7018,34.3540191 77.7045,37.3038191 C75.4978,35.0711191 74.1252,32.0034191 74.1252,28.6236191 L74.1252,3.08651911 C74.1252,1.00611911 71.3108,-0.0126808912 68.4884,0 Z" id="Fill-60" fill="#90D481"></path>
<path d="M68.48836,1.91088852e-05 C65.66592,0.0130003688 62.84083,1.05754211 62.84083,3.08645511 L62.84083,28.6235791 C62.84083,31.8405391 61.60526,34.7710691 59.58305,36.9716291 C59.58579,34.1675391 59.58828,31.8289691 59.59377,30.1345791 C59.66098,9.49193211 45.7074,6.85758511 30.83101,22.6331191 C4.313727,50.7525391 0,81.3003991 0,121.931019 C0,128.033419 3.896223,133.311119 9.976946,132.797419 C29.9022,131.114019 59.59378,121.760119 59.59377,99.8767391 C59.59377,92.2216491 59.572,68.7012091 59.57233,50.4528291 C63.03621,49.0049991 66.0882,46.7554291 68.48836,43.9372691 C70.96015,46.8394391 74.112,49.1290591 77.70444,50.5707091 C77.70479,68.8015691 77.69372,92.2391591 77.69373,99.8767391 C77.69373,121.759719 107.3746,131.114019 127.2998,132.797419 C133.3806,133.311119 137.2875,128.033419 137.2875,121.931019 C137.2875,81.3003791 132.9738,50.7525391 106.4565,22.6331191 C91.58009,6.85793311 77.6265,9.49227811 77.69373,30.1345791 C77.69955,31.8948091 77.70173,34.3539991 77.70444,37.3038391 C75.49773,35.0710591 74.12518,32.0034391 74.12517,28.6235791 L74.12517,3.08645511 C74.12517,1.00610711 71.31079,-0.0127216712 68.48836,1.91088852e-05 Z" id="Stroke-62" stroke="#000000" stroke-width="2" stroke-linejoin="round"></path>
</g>
<g id="Group-67" transform="translate(0.000000, 149.659125)">
<path d="M57.4288,0.0524000016 C48.297,0.580900002 41.1156,4.9789 36.359,9.207 C34.7357,10.6416 33.4073,12.4512 32.6901,14.5276 C32.0105,16.5284 31.8595,18.5716 32.3126,20.497 C32.8411,22.6865 34.0869,24.6118 35.7102,25.7821 C36.1632,26.1219 36.8474,26.011 37.1494,25.558 C39.641,22.0094 43.6426,19.4706 47.3422,19.0931 C47.9085,19.0176 48.4417,19.3668 48.5927,19.8953 C48.7815,20.6126 48.319,21.285 47.6018,21.3228 C44.5062,21.6625 41.1534,23.8215 39.0016,26.8793 C37.567,28.9178 36.963,30.9541 37.3028,32.5773 C37.416,33.1436 37.1848,33.7429 36.6185,33.9694 C36.4675,34.0071 36.3566,34.0519 36.2056,34.052 C35.6771,34.052 35.2217,33.7075 35.1085,33.179 C34.8442,32.1219 34.8891,30.954 35.1911,29.746 C35.3798,28.991 35.0778,28.1958 34.4361,27.7051 C32.322,26.195 30.7365,23.7813 30.0947,21.0632 C29.9059,20.3082 29.7879,19.5532 29.7879,18.7982 C29.7502,17.8166 28.8796,17.1348 27.9358,17.3235 C24.2362,18.154 16.3486,21.3959 9.2136,31.7397 C4.1172,39.7429 1.93,48.7277 2.9493,53.4466 C3.0625,54.5414 3.3621,55.5607 3.7397,56.4667 C4.1172,57.335 5.1742,57.6771 6.0047,57.1863 C8.1566,55.9028 10.2659,55.3011 12.1157,55.0746 C15.8531,54.5838 19.0666,55.5276 21.1052,56.3959 C22.351,56.9244 23.5213,57.5662 24.5028,58.2835 C25.1446,58.7365 26.0459,58.5501 26.4611,57.8706 C27.4427,56.2473 28.6979,54.8882 30.0947,53.9067 C30.5477,53.6047 31.18,53.5976 31.5575,54.0129 C32.086,54.5414 31.9823,55.3814 31.416,55.7589 C28.8489,57.4954 26.8787,61.0015 26.4257,64.7011 C26.1237,67.155 26.5366,69.2714 27.5937,70.5172 C27.9712,70.9702 28.0538,71.6096 27.6762,72.0626 C27.4497,72.3269 27.143,72.4401 26.8033,72.4401 C26.5012,72.4401 26.204,72.3269 25.9774,72.0626 C24.4297,70.3261 23.7832,67.6033 24.1607,64.3944 C24.2362,63.7904 24.3518,63.2289 24.5028,62.6248 C24.6915,61.9076 24.4674,61.1101 23.9011,60.6193 C21.5983,58.6185 17.1012,56.7003 12.4578,57.3043 C12.2691,57.342 12.0803,57.3373 11.8916,57.3751 C7.7012,58.0923 4.1904,60.9992 2.4538,64.9253 C0.8682,68.5494 0,72.5204 0,76.7107 C0,93.5855 14.2723,107.2419 31.8643,107.2419 C40.8113,107.2419 48.9278,103.6933 54.7037,97.9929 C55.0434,97.6909 54.968,97.0892 54.5149,96.825 C53.7222,96.372 53.0002,95.8789 52.3206,95.3503 C51.9054,95.0106 51.7237,94.3782 52.0257,93.8875 C52.4032,93.2457 53.196,93.1301 53.7245,93.5454 C55.008,94.6401 59.8426,97.8137 64.2594,97.8513 C69.2426,97.8891 73.3197,93.6892 73.3197,88.5551 C73.3197,87.4981 73.1262,86.4788 72.8242,85.535 C71.7672,83.0435 69.731,81.2409 66.7486,80.1083 C60.2932,77.6545 51.122,79.1952 45.7614,81.9133 C40.4385,84.5936 36.963,89.5084 36.9253,89.5461 L36.7365,89.8056 C36.3968,90.2964 35.6393,90.3342 35.2619,89.8056 L35.0731,89.5461 C30.7317,83.4682 20.1968,81.0498 14.1566,84.6738 C13.6281,84.9758 12.9509,84.8579 12.6112,84.3671 C12.2337,83.8386 12.3871,83.0907 12.9533,82.7509 C19.1822,78.938 29.1462,80.592 34.8843,85.8772 C35.5261,86.4434 36.4652,86.4104 37.0314,85.8064 C38.6925,83.9943 41.3752,81.5357 44.735,79.837 C50.662,76.8546 60.4041,75.2337 67.5391,77.914 C69.5399,78.6691 71.1986,79.7308 72.4821,81.0521 C72.5576,80.8256 72.6779,80.559 72.7534,80.3325 C72.2627,69.8754 68.0722,63.7527 59.7293,61.1856 C59.1253,60.9968 58.7407,60.3952 58.8917,59.8289 C59.0427,59.1871 59.6892,58.8143 60.331,59.0031 C64.2571,60.1734 67.3456,62.0916 69.6862,64.7719 C70.1392,65.2627 70.9366,65.2674 71.3142,64.7011 C72.5222,63.0401 73.3197,60.6971 73.3197,58.1301 C73.3197,56.2803 72.9021,54.5886 72.2226,53.1163 C72.1848,53.0408 72.1423,52.9558 72.1046,52.8803 C72.1046,52.8803 72.1046,52.8827 72.1046,52.8449 C71.9536,52.5807 71.845,52.3188 71.6563,52.0545 C71.5053,51.7903 71.3566,51.5307 71.2434,51.2287 C70.1109,49.5677 68.2563,48.1685 65.8403,47.1115 C61.3479,45.1862 55.9093,44.9597 52.2853,46.5452 C48.0949,48.395 44.997,52.1678 44.2042,56.3959 C43.6756,59.1517 43.9446,63.3067 48.0972,67.4971 C48.5503,67.9501 48.588,68.7429 48.0972,69.1959 C47.8707,69.3846 47.6065,69.4554 47.3422,69.4554 C47.0402,69.4554 46.7783,69.3398 46.5518,69.1133 C42.7389,65.3004 41.1534,60.6642 42.0217,55.983 C42.2104,55.0015 42.4699,54.0577 42.8475,53.1517 C43.225,52.2456 43.1542,51.1862 42.5879,50.3557 C42.5124,50.2425 42.4699,50.2047 42.47,50.167 C41.0354,47.8264 33.6409,38.8794 18.8047,40.5404 C18.1252,40.6159 17.4787,40.0544 17.5542,39.3371 C17.592,38.7709 18.0497,38.3485 18.616,38.3108 C32.1686,36.8385 40.8443,43.8271 44.2042,48.4328 C44.5817,48.9613 45.3792,48.9943 45.8322,48.5036 C47.38,46.8048 49.2699,45.4127 51.4594,44.4689 C55.6121,42.6569 61.7631,42.8834 66.784,45.0352 C68.5961,45.7902 70.1038,46.7694 71.3496,47.9019 C71.4628,47.6754 71.5785,47.4088 71.6917,47.1823 C72.711,45.559 73.3197,43.4449 73.3197,41.1421 C73.3197,38.575 72.5599,36.2699 71.3142,34.571 C71.2764,34.5333 71.2812,34.5027 71.2434,34.4649 C70.6016,33.5588 69.4597,33.0987 68.3649,33.4385 C66.9303,33.8915 65.458,34.0874 64.0235,34.0874 C61.9094,34.0874 59.9086,33.6744 58.3608,33.1082 C57.8323,32.9194 57.4925,32.3886 57.6058,31.8223 C57.7191,31.105 58.4741,30.7228 59.1158,30.9493 C61.9849,32.0441 66.3357,32.3155 69.5446,30.4656 C69.7711,30.3146 69.9905,30.1565 70.217,30.0055 C70.9721,29.4392 71.5808,28.7244 72.0338,27.8938 C72.7888,26.3838 73.2725,24.5363 73.2725,22.5733 L73.2725,22.5379 C73.2725,20.9523 72.711,19.3573 71.6917,18.1493 C68.0676,13.7702 63.5752,12.0384 58.29,12.9821 C57.7238,13.0954 57.1575,12.8264 56.9687,12.2979 C56.7045,11.6184 57.1245,10.9034 57.8417,10.7524 C62.8626,9.8464 67.3503,11.0568 71.1254,14.3034 C71.6162,14.7564 72.4137,14.4969 72.5647,13.8551 C72.6779,13.2511 72.7912,12.4135 72.7534,11.3187 C72.6024,5.0898 70.5237,0.680000002 61.4635,0.0760000016 C60.0809,-0.0183999984 58.7334,-0.0230999984 57.4288,0.0524000016 Z M91.9947,0.0524000016 C90.6902,-0.0230999984 89.3427,-0.0183999984 87.9601,0.0760000016 C78.8998,0.680000002 76.8211,5.0898 76.6701,11.3187 C76.6324,12.4135 76.7456,13.2511 76.8589,13.8551 C77.0099,14.4969 77.8073,14.7564 78.2981,14.3034 C82.0732,11.0568 86.5609,9.8464 91.5818,10.7524 C92.2991,10.9034 92.719,11.6184 92.4548,12.2979 C92.266,12.8264 91.6998,13.0954 91.1335,12.9821 C85.8483,12.0383 81.3559,13.7702 77.7319,18.1493 C76.7126,19.3574 76.151,20.9523 76.151,22.5379 L76.151,22.5733 C76.1511,24.5363 76.6347,26.3838 77.3897,27.8938 C77.8427,28.7244 78.4515,29.4392 79.2065,30.0055 C79.433,30.1565 79.6525,30.3146 79.879,30.4656 C83.0878,32.3154 87.4386,32.0441 90.3077,30.9493 C90.9495,30.7228 91.7045,31.105 91.8177,31.8223 C91.931,32.3886 91.5913,32.9194 91.0627,33.1082 C89.5149,33.6744 87.5141,34.0874 85.4001,34.0874 C83.9655,34.0874 82.4932,33.8915 81.0587,33.4385 C79.9639,33.0987 78.8219,33.5588 78.1802,34.4649 C78.1424,34.5026 78.1471,34.5333 78.1094,34.571 C76.8636,36.2699 76.1038,38.575 76.1038,41.1421 C76.1038,43.4449 76.7126,45.559 77.7319,47.1823 C77.8451,47.4088 77.9607,47.6754 78.074,47.9019 C79.3198,46.7694 80.8274,45.7902 82.6395,45.0352 C87.6604,42.8834 93.8115,42.6569 97.9641,44.4689 C100.1537,45.4127 102.0436,46.8048 103.5914,48.5036 C104.0444,48.9943 104.8419,48.9613 105.2194,48.4328 C108.5793,43.8271 117.2549,36.8385 130.8076,38.3108 C131.3738,38.3485 131.8316,38.7709 131.8693,39.3371 C131.9448,40.0544 131.2984,40.616 130.6188,40.5404 C115.7826,38.8794 108.3882,47.8264 106.9536,50.167 C106.9536,50.2047 106.9111,50.2425 106.8356,50.3557 C106.2693,51.1862 106.1986,52.2456 106.5761,53.1517 C106.9536,54.0577 107.2131,55.0015 107.4019,55.983 C108.2701,60.6642 106.6846,65.3004 102.8717,69.1133 C102.6453,69.3398 102.3834,69.4554 102.0813,69.4554 C101.8171,69.4554 101.5528,69.3846 101.3263,69.1959 C100.8355,68.7429 100.8733,67.9501 101.3263,67.4971 C105.4789,63.3067 105.7479,59.1517 105.2194,56.3959 C104.4266,52.1678 101.3286,48.395 97.1383,46.5452 C93.5142,44.9597 88.0757,45.1862 83.5833,47.1115 C81.1672,48.1685 79.3127,49.5677 78.1802,51.2287 C78.0669,51.5307 77.9183,51.7903 77.7673,52.0545 C77.5785,52.3188 77.47,52.5807 77.319,52.8449 C77.319,52.8827 77.319,52.8803 77.319,52.8803 C77.2812,52.9558 77.2387,53.0408 77.201,53.1163 C76.5215,54.5886 76.1038,56.2803 76.1038,58.1301 C76.1038,60.6972 76.9014,63.0401 78.1094,64.7011 C78.4869,65.2674 79.2844,65.2627 79.7374,64.7719 C82.078,62.0916 85.1664,60.1734 89.0926,59.0031 C89.7343,58.8143 90.3808,59.1871 90.5318,59.8289 C90.6829,60.3952 90.2982,60.9968 89.6942,61.1856 C81.3513,63.7526 77.1609,69.8754 76.6701,80.3325 C76.7456,80.559 76.866,80.8256 76.9414,81.0521 C78.225,79.7308 79.8836,78.6691 81.8845,77.914 C89.0195,75.2337 98.7616,76.8546 104.6885,79.837 C108.0484,81.5358 110.731,83.9943 112.3921,85.8064 C112.9584,86.4104 113.8975,86.4434 114.5392,85.8772 C120.2774,80.592 130.2413,78.9381 136.4702,82.7509 C137.0365,83.0907 137.1899,83.8386 136.8124,84.3671 C136.4726,84.8579 135.7954,84.9759 135.2669,84.6738 C129.2267,81.0497 118.6918,83.4682 114.3504,89.5461 L114.1617,89.8056 C113.7842,90.3342 113.0268,90.2964 112.687,89.8056 L112.4983,89.5461 C112.4605,89.5084 108.985,84.5936 103.6622,81.9133 C98.3015,79.1952 89.1303,77.6545 82.6749,80.1083 C79.6926,81.2409 77.6564,83.0435 76.5993,85.535 C76.2973,86.4788 76.1038,87.4981 76.1038,88.5551 C76.1038,93.6892 80.181,97.8891 85.1641,97.8513 C89.581,97.8136 94.4155,94.6402 95.699,93.5454 C96.2276,93.1301 97.0203,93.2457 97.3978,93.8875 C97.6998,94.3783 97.5181,95.0105 97.1029,95.3503 C96.4234,95.8789 95.7014,96.372 94.9086,96.825 C94.4556,97.0892 94.3801,97.6909 94.7199,97.9929 C100.4958,103.6933 108.6123,107.2419 117.5593,107.2419 C135.1513,107.2419 149.4236,93.5855 149.4236,76.7107 C149.4236,72.5204 148.5553,68.5494 146.9698,64.9253 C145.2332,60.9992 141.7224,58.0924 137.532,57.3751 C137.3432,57.3373 137.1545,57.342 136.9657,57.3043 C132.3223,56.7003 127.8252,58.6185 125.5224,60.6193 C124.9561,61.1101 124.732,61.9076 124.9208,62.6248 C125.0718,63.2289 125.1874,63.7904 125.2629,64.3944 C125.6404,67.6033 124.9939,70.3261 123.4461,72.0626 C123.2196,72.3269 122.9223,72.4401 122.6203,72.4401 C122.2805,72.4401 121.9738,72.3269 121.7473,72.0626 C121.3698,71.6096 121.4524,70.9702 121.8299,70.5172 C122.8869,69.2714 123.2998,67.155 122.9978,64.7011 C122.5448,61.0015 120.5747,57.4954 118.0076,55.7589 C117.4413,55.3813 117.3375,54.5414 117.866,54.0129 C118.2436,53.5976 118.8759,53.6047 119.3289,53.9067 C120.7256,54.8882 121.9809,56.2473 122.9624,57.8706 C123.3777,58.5501 124.279,58.7365 124.9208,58.2835 C125.9023,57.5662 127.0725,56.9244 128.3184,56.3959 C130.3569,55.5276 133.5705,54.5838 137.3078,55.0746 C139.1577,55.3011 141.267,55.9028 143.4188,57.1863 C144.2493,57.6771 145.3064,57.335 145.6839,56.4667 C146.0614,55.5607 146.361,54.5414 146.4743,53.4466 C147.4935,48.7277 145.3064,39.7429 140.21,31.7397 C133.075,21.3959 125.1874,18.154 121.4878,17.3235 C120.544,17.1347 119.6733,17.8166 119.6356,18.7982 C119.6356,19.5532 119.5176,20.3082 119.3289,21.0632 C118.6871,23.7813 117.1015,26.195 114.9875,27.7051 C114.3457,28.1958 114.0437,28.991 114.2325,29.746 C114.5344,30.954 114.5793,32.122 114.315,33.179 C114.2018,33.7075 113.7465,34.052 113.2179,34.052 C113.0669,34.052 112.956,34.0071 112.805,33.9694 C112.2387,33.7429 112.0075,33.1436 112.1208,32.5773 C112.4606,30.954 111.8565,28.9178 110.422,26.8793 C108.2701,23.8214 104.9174,21.6625 101.8218,21.3228 C101.1045,21.285 100.6421,20.6126 100.8308,19.8953 C100.9818,19.3668 101.5151,19.0176 102.0813,19.0931 C105.7809,19.4706 109.7826,22.0094 112.2741,25.558 C112.5761,26.011 113.2604,26.1219 113.7134,25.7821 C115.3367,24.6118 116.5824,22.6865 117.111,20.497 C117.564,18.5717 117.413,16.5284 116.7335,14.5276 C116.0162,12.4512 114.6879,10.6416 113.0645,9.207 C108.3079,4.9789 101.1265,0.580900002 91.9947,0.0524000016 Z M74.7472,94.8313 C72.6709,98.3799 68.8934,100.7535 64.5898,100.7535 C63.0797,100.7535 61.572,100.4467 60.213,99.8805 C58.9672,99.3897 57.535,99.6233 56.5912,100.5293 C50.1358,106.645 41.3422,110.1912 32.2064,110.1912 C30.1301,110.1912 28.0844,110.0001 26.0836,109.6604 C31.2555,118.3054 40.698,124.0884 51.4948,124.0884 C59.8001,124.0884 67.0507,120.6483 72.4113,115.0989 C73.0531,114.4194 73.9142,114.0843 74.7826,114.0843 C75.6508,114.0843 76.5239,114.4193 77.1656,115.0989 C82.5263,120.6483 89.7768,124.0884 98.0821,124.0884 C108.8789,124.0884 118.3095,118.3054 123.4815,109.6604 C121.5184,110.0001 119.4846,110.1913 117.3705,110.1912 C108.1215,110.1912 99.3208,106.645 92.9031,100.5293 C91.9216,99.5855 90.5272,99.3519 89.2813,99.8805 C87.8846,100.4467 86.4147,100.7535 84.9046,100.7535 C80.6009,100.7535 76.8235,98.3799 74.7472,94.8313 Z" id="Fill-64" fill="#D478B8"></path>
<path d="M57.42888,0.0523533116 C48.29702,0.580868502 41.11568,4.978877 36.35904,9.206998 C34.73574,10.64154 33.40737,12.45121 32.6901,14.52754 C32.01058,16.52835 31.85959,18.57163 32.31259,20.49694 C32.84111,22.6865 34.08691,24.61179 35.71019,25.78209 C36.1632,26.12186 36.84744,26.01098 37.14945,25.55795 C39.64103,22.00934 43.64262,19.47057 47.34225,19.09307 C47.90852,19.01757 48.44175,19.36674 48.59276,19.89528 C48.78151,20.61255 48.31906,21.28503 47.60179,21.32274 C44.5062,21.66251 41.15343,23.82144 39.00162,26.87924 C37.56707,28.9178 36.96306,30.95404 37.30282,32.57729 C37.41607,33.14356 37.18483,33.74285 36.61858,33.96937 C36.46757,34.00712 36.35667,34.05192 36.20568,34.05195 C35.67716,34.05195 35.22176,33.70748 35.10853,33.17895 C34.84428,32.12192 34.88911,30.95403 35.19111,29.74596 C35.37987,28.99094 35.07786,28.19579 34.43609,27.70504 C32.32203,26.195 30.7365,23.7813 30.09471,21.06321 C29.90596,20.30818 29.78799,19.55316 29.78799,18.79814 C29.75024,17.81661 28.87961,17.13475 27.93582,17.32349 C24.23621,18.15401 16.34862,21.39592 9.213631,31.73969 C4.117231,39.74293 1.930025,48.7277 2.949306,53.44658 C3.062559,54.54136 3.362182,55.56067 3.739719,56.46667 C4.117231,57.33494 5.174253,57.67706 6.004786,57.1863 C8.1566,55.90276 10.26596,55.30109 12.11575,55.0746 C15.85311,54.58384 19.06668,55.52761 21.10523,56.39589 C22.35102,56.9244 23.5213,57.56619 24.50283,58.28345 C25.1446,58.73646 26.04591,58.55008 26.46117,57.87054 C27.4427,56.24724 28.6979,54.8882 30.09471,53.90668 C30.54773,53.60467 31.18006,53.59758 31.55757,54.01285 C32.08609,54.54136 31.98229,55.38135 31.416,55.75884 C28.84893,57.49539 26.87878,61.0015 26.42578,64.70113 C26.12377,67.15496 26.53668,69.27138 27.5937,70.51716 C27.97121,70.97018 28.0538,71.60958 27.67628,72.0626 C27.44978,72.32686 27.14306,72.44011 26.80329,72.44011 C26.50128,72.44011 26.20401,72.32685 25.97748,72.0626 C24.42969,70.32605 23.78321,67.60327 24.16071,64.39441 C24.23621,63.79039 24.35183,63.22886 24.50283,62.62482 C24.69159,61.90755 24.46745,61.11004 23.90117,60.6193 C21.59836,58.61849 17.10125,56.70027 12.45787,57.30428 C12.26911,57.34202 12.08036,57.33727 11.8916,57.37506 C7.701227,58.09233 4.190393,60.99919 2.453822,64.92528 C0.8682755,68.54938 0,72.52034 0,76.71071 C0,93.58545 14.27229,107.2419 31.8643,107.2419 C40.81131,107.2419 48.9278,103.6933 54.70372,97.99289 C55.04348,97.69089 54.96801,97.08916 54.51496,96.82496 C53.72219,96.37195 53.0002,95.87888 52.32068,95.35032 C51.90542,95.01055 51.72377,94.37822 52.02575,93.88746 C52.40326,93.24569 53.19603,93.13013 53.72455,93.54534 C55.00809,94.64012 59.84264,97.81364 64.25947,97.85132 C69.24262,97.88908 73.31974,93.68918 73.31974,88.55512 C73.31974,87.49809 73.12625,86.47876 72.82425,85.53503 C71.76722,83.04345 69.731,81.2409 66.74868,80.10831 C60.29324,77.65448 51.12206,79.19518 45.76143,81.91328 C40.43852,84.59361 36.96309,89.50834 36.92531,89.54608 L36.73655,89.80562 C36.39679,90.29638 35.63938,90.33423 35.2619,89.80562 L35.07314,89.54608 C30.73176,83.46815 20.19684,81.04977 14.15667,84.67383 C13.62815,84.97584 12.95098,84.8579 12.61123,84.3671 C12.23372,83.83858 12.38709,83.09064 12.95335,82.75088 C19.18228,78.93802 29.14621,80.592 34.88439,85.87715 C35.52615,86.44341 36.46522,86.41038 37.03148,85.80636 C38.69253,83.99431 41.37523,81.53571 44.73507,79.83697 C50.66199,76.85463 60.40413,75.23371 67.5391,77.91402 C69.53991,78.66905 71.19862,79.73081 72.48213,81.05208 C72.55763,80.82558 72.67792,80.55897 72.75347,80.33245 C72.2627,69.87539 68.07224,63.75265 59.72934,61.18556 C59.12532,60.99681 58.74073,60.39514 58.89173,59.82888 C59.04274,59.18711 59.68928,58.81433 60.33099,59.00308 C64.25711,60.17336 67.34564,62.09156 69.68619,64.77192 C70.1392,65.26268 70.93667,65.2674 71.31421,64.70113 C72.52224,63.04008 73.31974,60.69713 73.31974,58.13008 C73.31974,56.28027 72.90211,54.58854 72.22259,53.11626 C72.18484,53.04076 72.14238,52.95581 72.10462,52.88032 C72.10462,52.88032 72.10462,52.88267 72.10462,52.84493 C71.95362,52.58066 71.84506,52.31877 71.65633,52.05451 C71.50533,51.79025 71.35665,51.53071 71.24343,51.22871 C70.11089,49.56765 68.25637,48.16849 65.8403,47.11147 C61.34792,45.18617 55.90933,44.95968 52.28529,46.54521 C48.09492,48.39501 44.99699,52.16778 44.20419,56.39589 C43.67568,59.15172 43.94463,63.30671 48.09728,67.49708 C48.55029,67.95009 48.58805,68.74286 48.09728,69.19588 C47.87077,69.38463 47.60653,69.4554 47.34225,69.45541 C47.04024,69.45541 46.77835,69.3398 46.55184,69.1133 C42.73898,65.30043 41.15345,60.66415 42.02171,55.98299 C42.21046,55.00146 42.46999,54.0577 42.84751,53.15165 C43.22502,52.24562 43.15425,51.1862 42.58797,50.35571 C42.51247,50.24246 42.46998,50.2047 42.47,50.16696 C41.03546,47.82639 33.64098,38.87936 18.80477,40.54042 C18.12525,40.61592 17.47876,40.05439 17.55427,39.3371 C17.59202,38.77084 18.04974,38.34849 18.61602,38.31075 C32.16867,36.83845 40.84434,43.82713 44.20419,48.43276 C44.5817,48.96128 45.3792,48.99432 45.83221,48.50355 C47.38,46.80475 49.2699,45.41269 51.45948,44.4689 C55.61211,42.65684 61.76317,42.88336 66.78408,45.03516 C68.59613,45.79018 70.10388,46.76934 71.3496,47.90189 C71.46286,47.67538 71.57855,47.40877 71.69172,47.18226 C72.711,45.55896 73.31974,43.4449 73.31974,41.14208 C73.31974,38.575 72.55999,36.26985 71.31421,34.57103 C71.27646,34.53327 71.28121,34.50264 71.24343,34.46485 C70.60166,33.55882 69.45975,33.09869 68.3649,33.43849 C66.93036,33.8915 65.45803,34.08736 64.02353,34.08734 C61.90946,34.08734 59.90867,33.67443 58.36086,33.10817 C57.83235,32.91941 57.49254,32.38855 57.60584,31.82227 C57.71909,31.105 58.47416,30.72274 59.11588,30.94928 C61.98497,32.04406 66.33578,32.31544 69.54463,30.46559 C69.77113,30.31459 69.99052,30.15647 70.21707,30.0055 C70.97209,29.43923 71.58084,28.72434 72.03384,27.8938 C72.78886,26.38375 73.27252,24.53632 73.27255,22.57325 L73.27255,22.53786 C73.27255,20.95231 72.71101,19.35733 71.69172,18.14929 C68.06761,13.77016 63.57527,12.03834 58.29008,12.98211 C57.72381,13.09536 57.15754,12.82637 56.96879,12.29787 C56.70453,11.61835 57.12458,10.90343 57.84178,10.75243 C62.86268,9.846407 67.35037,11.05678 71.12545,14.3034 C71.61622,14.75641 72.41378,14.4969 72.56471,13.8551 C72.67797,13.25109 72.79128,12.41352 72.75347,11.3187 C72.60247,5.089767 70.52375,0.679967802 61.46353,0.0759477516 C60.08089,-0.0184300284 58.73346,-0.0231496684 57.42888,0.0523533116 Z M91.99474,0.0523533116 C90.69019,-0.0231489184 89.34269,-0.0184371984 87.96009,0.0759477516 C78.89983,0.679967802 76.82118,5.089767 76.67015,11.3187 C76.6324,12.41349 76.74562,13.2511 76.8589,13.8551 C77.0099,14.49688 77.80737,14.75643 78.29816,14.3034 C82.07328,11.0568 86.56094,9.846392 91.58184,10.75243 C92.2991,10.90343 92.71906,11.61837 92.45483,12.29787 C92.26608,12.82639 91.69981,13.09535 91.13354,12.98211 C85.84838,12.03833 81.35597,13.77017 77.7319,18.14929 C76.71261,19.35734 76.15107,20.9523 76.15107,22.53786 L76.15107,22.57325 C76.15109,24.53632 76.63476,26.38375 77.38978,27.8938 C77.84277,28.72434 78.45153,29.43923 79.20655,30.0055 C79.43307,30.1565 79.65252,30.31457 79.87899,30.46559 C83.08781,32.31541 87.43868,32.04406 90.30774,30.94928 C90.94949,30.72275 91.7045,31.10499 91.81778,31.82227 C91.93104,32.38855 91.59131,32.91941 91.06276,33.10817 C89.51497,33.67443 87.51412,34.08734 85.40009,34.08734 C83.96556,34.08735 82.49329,33.89152 81.05871,33.43849 C79.96393,33.09872 78.82193,33.55881 78.18019,34.46485 C78.14244,34.50261 78.14713,34.53329 78.10941,34.57103 C76.86363,36.26984 76.10388,38.57502 76.10388,41.14208 C76.10388,43.4449 76.71265,45.55896 77.7319,47.18226 C77.84513,47.40876 77.96073,47.6754 78.07402,47.90189 C79.3198,46.76934 80.82746,45.79018 82.63954,45.03516 C87.66045,42.88334 93.81151,42.65686 97.96413,44.4689 C100.1537,45.41269 102.0436,46.80475 103.5914,48.50355 C104.0444,48.9943 104.8419,48.9613 105.2194,48.43276 C108.5793,43.82713 117.255,36.83845 130.8076,38.31075 C131.3739,38.34849 131.8316,38.77084 131.8694,39.3371 C131.9449,40.05438 131.2984,40.61594 130.6188,40.54042 C115.7827,38.87936 108.3882,47.8264 106.9536,50.16696 C106.9536,50.2047 106.9111,50.24246 106.8356,50.35571 C106.2694,51.18622 106.1986,52.24561 106.5761,53.15165 C106.9536,54.05768 107.2132,55.00147 107.4019,55.98299 C108.2702,60.66414 106.6846,65.30043 102.8718,69.1133 C102.6453,69.3398 102.3834,69.45541 102.0814,69.45541 C101.8171,69.4554 101.5528,69.38463 101.3263,69.19588 C100.8356,68.74286 100.8734,67.95009 101.3263,67.49708 C105.479,63.30671 105.748,59.15172 105.2194,56.39589 C104.4267,52.16777 101.3287,48.39503 97.13833,46.54521 C93.51425,44.95967 88.07573,45.18618 83.58332,47.11147 C81.16725,48.16851 79.31273,49.56764 78.18019,51.22871 C78.06693,51.53071 77.91833,51.79025 77.76729,52.05451 C77.57852,52.31877 77.47003,52.58066 77.319,52.84493 C77.319,52.88267 77.319,52.88032 77.319,52.88032 C77.28127,52.95581 77.23874,53.04076 77.20102,53.11626 C76.52151,54.58855 76.10388,56.28026 76.10388,58.13008 C76.10388,60.69714 76.9014,63.04007 78.10941,64.70113 C78.48691,65.2674 79.28444,65.26268 79.73743,64.77192 C82.07801,62.09157 85.16648,60.17335 89.09262,59.00308 C89.73437,58.81433 90.38085,59.18711 90.53188,59.82888 C90.68292,60.39514 90.29827,60.99681 89.69428,61.18556 C81.35132,63.75263 77.16094,69.8754 76.67015,80.33245 C76.74564,80.55894 76.86601,80.8256 76.94148,81.05208 C78.22503,79.73081 79.88368,78.66905 81.88452,77.91402 C89.0195,75.23371 98.76163,76.8546 104.6886,79.83697 C108.0484,81.53574 110.7311,83.99428 112.3921,85.80636 C112.9584,86.41038 113.8975,86.44341 114.5392,85.87715 C120.2774,80.59197 130.2413,78.93805 136.4703,82.75088 C137.0365,83.09068 137.1899,83.83856 136.8124,84.3671 C136.4726,84.85786 135.7955,84.97587 135.267,84.67383 C129.2268,81.04973 118.6919,83.46819 114.3505,89.54608 L114.1617,89.80562 C113.7842,90.33417 113.0269,90.29642 112.6871,89.80562 L112.4983,89.54608 C112.4606,89.50837 108.9851,84.59358 103.6622,81.91328 C98.30153,79.19518 89.13034,77.65448 82.67493,80.10831 C79.69259,81.24087 77.65643,83.04349 76.59936,85.53503 C76.29737,86.47879 76.10388,87.49806 76.10388,88.55512 C76.10388,93.68921 80.181,97.88904 85.16415,97.85132 C89.58101,97.8136 94.4155,94.64015 95.69907,93.54534 C96.2276,93.1301 97.02036,93.24572 97.39787,93.88746 C97.69985,94.37825 97.51817,95.01053 97.10294,95.35032 C96.42342,95.87885 95.70139,96.37199 94.90865,96.82496 C94.45564,97.08922 94.38011,97.69085 94.7199,97.99289 C100.4958,103.6933 108.6123,107.2419 117.5593,107.2419 C135.1513,107.2419 149.4236,93.58545 149.4236,76.71071 C149.4236,72.52035 148.5553,68.54938 146.9698,64.92528 C145.2332,60.99917 141.7224,58.09235 137.532,57.37506 C137.3433,57.33728 137.1545,57.34201 136.9658,57.30428 C132.3224,56.70027 127.8253,58.61849 125.5224,60.6193 C124.9562,61.11006 124.732,61.90754 124.9208,62.62482 C125.0718,63.22884 125.1874,63.7904 125.2629,64.39441 C125.6404,67.60326 124.9939,70.32606 123.4461,72.0626 C123.2196,72.32685 122.9224,72.44011 122.6203,72.44011 C122.2806,72.44011 121.9738,72.32686 121.7473,72.0626 C121.3698,71.60958 121.4524,70.97018 121.8299,70.51716 C122.887,69.27138 123.2999,67.15496 122.9978,64.70113 C122.5449,61.00151 120.5747,57.49538 118.0076,55.75884 C117.4413,55.38133 117.3375,54.54138 117.866,54.01285 C118.2436,53.59758 118.8759,53.60467 119.3289,53.90668 C120.7257,54.8882 121.9809,56.24724 122.9624,57.87054 C123.3777,58.55008 124.279,58.73646 124.9208,58.28345 C125.9023,57.56618 127.0726,56.92442 128.3184,56.39589 C130.3569,55.52761 133.5705,54.58384 137.3079,55.0746 C139.1577,55.3011 141.267,55.90275 143.4188,57.1863 C144.2493,57.67706 145.3064,57.33494 145.6839,56.46667 C146.0614,55.56065 146.361,54.54138 146.4743,53.44658 C147.4936,48.7277 145.3064,39.74292 140.21,31.73969 C133.075,21.3959 125.1874,18.15403 121.4878,17.32349 C120.544,17.13473 119.6734,17.81663 119.6356,18.79814 C119.6356,19.55316 119.5177,20.30818 119.3289,21.06321 C118.6871,23.7813 117.1016,26.195 114.9875,27.70504 C114.3458,28.19579 114.0437,28.99094 114.2325,29.74596 C114.5345,30.95402 114.5793,32.12194 114.3151,33.17895 C114.2018,33.70748 113.7465,34.05195 113.2179,34.05195 C113.0669,34.05194 112.9561,34.0071 112.805,33.96937 C112.2388,33.74285 112.0075,33.14356 112.1208,32.57729 C112.4606,30.95402 111.8566,28.9178 110.422,26.87924 C108.2702,23.82141 104.9174,21.66252 101.8218,21.32274 C101.1046,21.285 100.6421,20.61257 100.8309,19.89528 C100.9819,19.36675 101.5151,19.01755 102.0814,19.09307 C105.781,19.47058 109.7826,22.00934 112.2742,25.55795 C112.5762,26.01097 113.2604,26.12187 113.7134,25.78209 C115.3367,24.61181 116.5825,22.68649 117.111,20.49694 C117.564,18.57165 117.4131,16.52834 116.7335,14.52754 C116.0163,12.45122 114.6879,10.64154 113.0646,9.206998 C108.3079,4.978877 101.1266,0.580866602 91.99474,0.0523533116 Z M74.7472,94.83124 C72.67089,98.37984 68.89343,100.7535 64.58979,100.7534 C63.07975,100.7534 61.57205,100.4467 60.21302,99.88045 C58.96723,99.38969 57.53502,99.62325 56.59128,100.5293 C50.13583,106.645 41.3422,110.1912 32.20642,110.1912 C30.13011,110.1912 28.08446,110.0001 26.08366,109.6603 C31.25556,118.3053 40.69807,124.0883 51.49488,124.0883 C59.80012,124.0883 67.05069,120.6483 72.41135,115.0989 C73.05312,114.4193 73.91425,114.0843 74.78259,114.0843 C75.65087,114.0843 76.52391,114.4193 77.16563,115.0989 C82.52629,120.6483 89.77685,124.0883 98.08211,124.0883 C108.8789,124.0883 118.3096,118.3053 123.4815,109.6603 C121.5185,110.0001 119.4847,110.1912 117.3706,110.1912 C108.1215,110.1912 99.3208,106.645 92.90313,100.5293 C91.92159,99.58551 90.52719,99.3519 89.28138,99.88045 C87.88459,100.4467 86.41471,100.7535 84.90461,100.7534 C80.60098,100.7534 76.82349,98.37989 74.7472,94.83124 Z" id="Stroke-66" stroke="#000000" stroke-width="2" stroke-linejoin="round"></path>
</g>
<g id="Group-71" transform="translate(624.697700, 144.575425)">
<path d="M63.576,0 C60.214,0 56.852,1.2167 56.852,3.6558 C56.852,25.9968 67.2,36.9633 76.959,41.5705 C76.226,44.4906 75.632,47.6602 75.195,51.116 C71.046,83.9383 63.576,85.6428 57.227,86.0164 C52.192,81.8375 44.577,79.2061 36.329,84.1885 C33.6,83.4685 30.53,83.1525 27.094,83.4296 C19.188,84.0627 12.814,87.2687 8.15199999,92.9537 C-1.05700001,104.1851 -0.524000013,122.0906 0.615999987,131.7663 C1.02399999,136.1354 14.44,134.2056 13.978,130.195 C12.485,117.5283 14.194,106.7882 18.553,101.473 C20.709,98.8433 23.628,97.3423 27.457,96.898 C25.255,120.3336 78.742,150.3456 112.126,107.1169 C155.824,50.5337 133.756,16.1369 112.126,11.5017 C104.387,9.8434 90.368,12.6604 81.758,28.7969 C76.621,25.9226 70.299,19.2189 70.299,3.6558 C70.299,1.2167 66.937,0 63.576,0 Z" id="Fill-68" fill="#95A0D4"></path>
<path d="M63.57574,0 C60.21427,0 56.85221,1.216689 56.8522,3.655721 C56.8522,25.99675 67.19983,36.9633 76.95867,41.57047 C76.22565,44.4906 75.63173,47.66018 75.19495,51.11596 C71.04547,83.93827 63.57593,85.6428 57.22633,86.01634 C52.19161,81.83744 44.57719,79.2061 36.32885,84.18848 C33.59959,83.46845 30.52998,83.15248 27.09335,83.42954 C19.18775,84.06268 12.81354,87.26869 8.15200499,92.95365 C-1.05750601,104.1851 -0.523976313,122.0906 0.616087787,131.7663 C1.02381799,136.1354 14.43977,134.2055 13.97764,130.195 C12.48457,117.5283 14.19415,106.7882 18.55264,101.473 C20.70896,98.84326 23.62747,97.3423 27.45678,96.89799 C25.25497,120.3336 78.74163,150.3456 112.1263,107.1169 C155.8239,50.53369 133.7562,16.13684 112.1263,11.50163 C104.3873,9.843339 90.36771,12.66033 81.75814,28.79682 C76.62046,25.92253 70.29925,19.21891 70.29927,3.655721 C70.29927,1.216692 66.93721,0 63.57574,0 Z" id="Stroke-70" stroke="#000000" stroke-width="2" stroke-linejoin="round"></path>
</g>
<g id="Group-75" transform="translate(371.994200, 136.191425)">
<path d="M55.3155,3.79110112e-05 C54.4245,0.00660000005 53.4905,0.1275 52.4405,0.3553 C52.0675,0.4359 51.7585,0.6967 51.6235,1.0534 C50.4905,4.0631 49.9535,7.0259 49.7305,9.7619 C49.7095,10.0277 49.5315,10.2504 49.2815,10.3416 C49.0015,10.4435 48.7293,10.5582 48.4529,10.673 C48.1118,10.8146 47.7242,10.6752 47.5655,10.3416 C46.5341,8.1759 45.2878,5.9146 44.1697,3.9878 C43.9217,3.5603 43.3693,3.4284 42.951,3.6919 C41.313,4.7241 39.8763,5.8574 38.6795,6.9221 C38.3331,7.2303 38.2087,7.7188 38.3719,8.1527 C39.2515,10.4915 39.9738,13.0595 40.4899,15.1219 C40.6451,15.7421 40.4972,16.3938 40.0876,16.8849 C38.4795,18.8125 37.1131,21.0892 36.0764,23.7002 C27.9525,44.1539 33.4268,55.8298 35.6032,59.3034 C35.9576,59.869 36.777,59.8972 37.1768,59.3625 C39.4581,56.3111 46.321,46.9979 52.9135,36.692 C58.9525,27.2512 68.2115,21.0202 71.1475,19.1921 C71.6635,18.8711 71.8025,18.1842 71.4665,17.6776 C70.2995,15.9181 68.9565,14.435 67.4795,13.2287 C67.1785,12.9834 67.0795,12.5671 67.2305,12.2111 C68.0135,10.3721 68.8515,8.5093 69.5855,6.9103 C69.7765,6.4942 69.6675,6.0045 69.3135,5.7153 C68.2595,4.8561 66.9935,3.9928 65.5035,3.2068 C65.0665,2.9766 64.5305,3.0512 64.1775,3.3961 C62.0795,5.4521 60.7235,7.4107 59.8475,9.0638 C59.7055,9.3322 59.4015,9.458 59.1015,9.4069 C58.6705,9.3331 58.2355,9.284 57.8005,9.2413 C57.4135,9.203 57.1415,8.8454 57.1965,8.4603 C57.5885,5.7467 58.0805,3.1149 58.4985,1.2308 C58.5885,0.8261 58.3335,0.4201 57.9305,0.3198 C57.0535,0.1013 56.2075,-0.00599999995 55.3155,3.79110112e-05 Z M28.1252,12.08099 C27.957,12.0804 27.7815,12.1068 27.6164,12.1756 L19.464,15.5715 C18.8029,15.8471 18.4888,16.6146 18.7659,17.2753 C20.2176,20.7334 23.782,30.9929 18.2335,37.6504 C12.4775,44.5566 8.7669,49.5086 8.4719,55.7655 C8.4291,56.672 9.5245,57.178 10.1757,56.5464 C16.9203,50.006 24.4128,45.9512 26.7645,44.7616 C27.2264,44.5272 27.5344,44.0704 27.5809,43.5547 C27.8232,40.8827 28.8329,32.2424 32.2901,24.4457 C32.7328,23.448 32.7942,22.3201 32.444,21.2864 C30.959,16.9053 29.8915,14.2493 29.3084,12.8737 C29.0991,12.3799 28.6297,12.0828 28.1252,12.08099 Z M83.5705,20.0204 C66.6675,23.2736 60.8695,32.7251 49.2935,49.3761 C40.8579,61.509 32.6507,69.3895 22.3629,74.9455 C21.3516,75.4919 20.4656,74.0419 21.4163,73.3955 C26.9125,69.6603 30.6565,66.362 32.2665,64.8645 C32.7621,64.4033 32.8381,63.6517 32.444,63.1015 C31.0555,61.164 28.0719,56.2207 27.5691,49.0093 C27.5324,48.4819 26.9616,48.176 26.5042,48.4414 C11.4737,57.1658 0,67.8321 0,88.9431 C0,121.8525 35.722,128.1863 53.3985,143.3239 C66.5105,154.5522 82.6265,154.6342 86.7535,135.9643 C91.9455,112.4778 101.9185,106.802 101.922502,87.0499 C101.9295,60.1925 79.8135,51.7796 72.5665,49.7429 C71.3095,49.3885 69.9945,50.057 69.5375,51.2811 C67.1455,57.6879 64.0295,64.1727 61.0185,69.7275 C60.9015,69.9453 60.8605,70.1913 60.8885,70.4375 C61.5395,76.1412 65.3845,89.347 87.0255,89.5938 C87.8445,89.6033 87.8985,90.8088 87.0855,90.9072 C72.6415,92.6599 63.6775,88.4898 57.6935,77.3002 C57.4785,76.8973 56.9055,76.888 56.6645,77.2765 C55.7765,78.7123 54.9495,79.9884 54.2035,81.0746 C53.4715,82.1405 53.0095,83.3736 52.8785,84.6598 C52.2645,90.6505 52.4275,95.943 53.1385,100.6569 C53.2365,101.3086 53.6785,101.8635 54.2985,102.0886 C59.9275,104.1355 65.9235,107.0096 71.5255,112.6312 C72.0715,113.1786 71.3485,114.0409 70.7215,113.5896 C62.8765,107.9415 57.4165,106.6281 54.8775,106.3482 C54.5995,106.3176 54.3885,106.5753 54.4635,106.8452 C57.0505,116.1444 61.7815,122.8372 66.1425,128.1195 C66.7925,128.9074 65.6995,129.9242 64.9585,129.2199 C60.3445,124.8304 57.0195,120.0342 54.6295,115.2934 C54.4705,114.9784 54.0335,114.9581 53.8485,115.2579 C52.9185,116.7599 51.6715,119.7031 51.3165,125.0195 C51.2545,125.9494 49.8825,125.9716 49.8015,125.0431 C49.4305,120.7745 49.9535,116.7472 52.3935,111.4953 C52.5645,111.1257 52.5745,110.7044 52.4285,110.3239 C48.5344,100.1561 48.5284,91.0894 48.6541,88.1266 C48.6685,87.7832 48.2511,87.6046 48.0152,87.8545 C46.952,88.9809 44.7838,91.2684 42.1819,93.9007 C41.8525,94.2339 41.7074,94.7061 41.7914,95.1668 C43.454,104.2668 42.0424,109.7332 40.0876,113.3056 C39.6624,114.0829 38.4633,113.5721 38.7269,112.7258 C40.5943,106.7309 39.907,102.7655 38.6204,97.9592 C38.5613,97.7385 38.2934,97.6427 38.1116,97.7817 C32.7702,101.8611 30.0958,107.186 28.7641,111.3414 C28.529,112.0752 27.4411,111.8357 27.5218,111.0693 C27.9473,107.0179 30.331,101.3763 37.5318,94.1611 C54.8405,76.8175 59.2555,66.0282 66.1185,49.4116 C72.3225,34.3898 87.7635,28.9888 90.6705,28.09 C90.8985,28.0192 90.9845,27.7449 90.8355,27.5575 C89.8515,26.313 86.7235,22.4169 84.3165,20.2215 C84.1185,20.0413 83.8335,19.9701 83.5705,20.0204 Z" id="Fill-72" fill="#D48D7B"></path>
<path d="M55.3156,-3.55271368e-15 C54.42416,0.00655689105 53.49053,0.1274897 52.44038,0.3552286 C52.06781,0.435877 51.75835,0.6966643 51.62395,1.053329 C50.49071,4.063066 49.95385,7.025856 49.7308,9.761839 C49.70921,10.02764 49.53141,10.25037 49.28118,10.34162 C49.00137,10.44347 48.72931,10.5582 48.45292,10.67292 C48.11177,10.81453 47.72419,10.67519 47.5655,10.34162 C46.53411,8.17585 45.28775,5.914568 44.16966,3.987719 C43.92166,3.560244 43.36929,3.428399 42.95094,3.691913 C41.31298,4.724061 39.87631,5.85738 38.67951,6.922108 C38.33306,7.230313 38.20864,7.718744 38.37187,8.152658 C39.25143,10.49146 39.97379,13.05943 40.48984,15.12183 C40.64508,15.74203 40.49722,16.39375 40.08754,16.88483 C38.4795,18.81244 37.1131,21.08915 36.07642,23.70019 C27.95252,44.15383 33.42676,55.82975 35.60313,59.30332 C35.95753,59.869 36.77693,59.89712 37.17682,59.36248 C39.45807,56.3111 46.32098,46.99791 52.91366,36.69196 C58.95283,27.25118 68.21114,21.02015 71.14711,19.19212 C71.66318,18.87104 71.80241,18.1842 71.46658,17.67759 C70.29926,15.91809 68.95611,14.435 67.47912,13.22868 C67.17849,12.98333 67.07916,12.56703 67.23065,12.21111 C68.01328,10.3721 68.85108,8.509232 69.58526,6.910276 C69.77608,6.49416 69.6679,6.004506 69.31311,5.715222 C68.25901,4.856108 66.99345,3.992817 65.50314,3.206792 C65.06658,2.976585 64.53045,3.051178 64.17793,3.396108 C62.07918,5.452074 60.72309,7.410626 59.84734,9.063739 C59.70536,9.332188 59.40139,9.457991 59.10191,9.406873 C58.67065,9.33304 58.23544,9.283998 57.80037,9.241222 C57.41303,9.20298 57.14133,8.84537 57.19693,8.460296 C57.58805,5.746649 58.08084,3.11488 58.49847,1.230813 C58.58821,0.8260561 58.33302,0.4200665 57.93052,0.319732 C57.05362,0.1012619 56.20704,-0.00603864295 55.3156,-3.55271368e-15 Z M28.12517,12.0809397 C27.957,12.08034 27.78146,12.1068 27.61639,12.17561 L19.46399,15.57146 C18.8029,15.8471 18.48873,16.61459 18.76589,17.2753 C20.21756,20.73333 23.78193,30.99292 18.23344,37.65037 C12.4775,44.5566 8.766838,49.50862 8.471866,55.76549 C8.42908,56.67194 9.524476,57.17799 10.1757,56.54641 C16.92026,50.00595 24.41279,45.9512 26.76447,44.76153 C27.2264,44.52716 27.53433,44.07036 27.58089,43.55465 C27.82322,40.88265 28.83287,32.24239 32.29011,24.44562 C32.73273,23.44793 32.79417,22.32007 32.44393,21.28642 C30.95894,16.90528 29.89151,14.24928 29.3084,12.87371 C29.09911,12.37988 28.62969,12.0828 28.12517,12.0809397 Z M83.57093,20.02037 C66.66771,23.27357 60.86924,32.72504 49.29301,49.37609 C40.85787,61.50894 32.65065,69.38949 22.36289,74.94551 C21.35156,75.49188 20.46556,74.04183 21.41631,73.39549 C26.91251,69.6603 30.65651,66.36195 32.26645,64.86446 C32.76208,64.40329 32.83804,63.65163 32.44393,63.10146 C31.05549,61.16401 28.0719,56.22069 27.56906,49.00929 C27.53233,48.48187 26.96158,48.17596 26.50416,48.44135 C11.47365,57.16576 0,67.83207 0,88.94302 C0,121.8525 35.72201,128.1862 53.39879,143.3239 C66.51077,154.5522 82.62676,154.6341 86.7538,135.9642 C91.94559,112.4778 101.9181,106.802 101.922701,87.04986 C101.9291,60.19244 79.81361,51.77955 72.56697,49.74289 C71.30954,49.3885 69.9949,50.05694 69.53793,51.28108 C67.14498,57.68789 64.02917,64.17262 61.01873,69.7275 C60.90098,69.94522 60.8602,70.19132 60.88858,70.43744 C61.53907,76.14113 65.38488,89.34697 87.02594,89.59379 C87.84454,89.60326 87.89804,90.80877 87.0851,90.90717 C72.64147,92.65985 63.6778,88.48977 57.69388,77.30012 C57.47844,76.89726 56.90498,76.88801 56.66448,77.27646 C55.77659,78.71223 54.94931,79.98832 54.20338,81.0746 C53.47186,82.14044 53.00953,83.37355 52.87817,84.65976 C52.26479,90.65046 52.42748,95.94294 53.13848,100.6569 C53.23692,101.3085 53.67863,101.8634 54.29803,102.0886 C59.9279,104.1355 65.92314,107.0096 71.52574,112.6311 C72.07134,113.1786 71.34853,114.0409 70.72115,113.5895 C62.8763,107.9415 57.41692,106.628 54.87781,106.3482 C54.59952,106.3175 54.3887,106.5752 54.46369,106.8452 C57.05011,116.1443 61.78138,122.8372 66.14208,128.1195 C66.79257,128.9074 65.69903,129.9241 64.95886,129.2199 C60.34411,124.8304 57.01921,120.0342 54.62934,115.2934 C54.47069,114.9783 54.0339,114.9581 53.84841,115.2579 C52.91887,116.7599 51.6719,119.7031 51.31632,125.0194 C51.25422,125.9494 49.88241,125.9715 49.80179,125.0431 C49.43074,120.7744 49.95393,116.7471 52.39305,111.4952 C52.56457,111.1257 52.57436,110.7044 52.42854,110.3238 C48.53432,100.1561 48.52836,91.08933 48.65407,88.1266 C48.66845,87.78318 48.25105,87.60455 48.01513,87.85446 C46.95193,88.98088 44.7838,91.26841 42.18185,93.90072 C41.85244,94.23391 41.70735,94.70602 41.79138,95.16676 C43.45395,104.2668 42.04239,109.7332 40.08754,113.3056 C39.66234,114.0829 38.46329,113.5721 38.72684,112.7258 C40.59425,106.7309 39.90698,102.7655 38.62035,97.95917 C38.56128,97.73843 38.29333,97.64267 38.11156,97.78169 C32.77022,101.8611 30.09576,107.186 28.76411,111.3414 C28.52898,112.0752 27.44109,111.8357 27.52173,111.0693 C27.94731,107.0179 30.33099,101.3763 37.53179,94.16103 C54.84052,76.81746 59.25499,66.02818 66.11842,49.41159 C72.32266,34.38979 87.7635,28.98881 90.67026,28.08994 C90.89896,28.01914 90.98477,27.7449 90.83591,27.55749 C89.8511,26.31293 86.72333,22.41682 84.31636,20.22152 C84.11872,20.04129 83.8337,19.97002 83.57093,20.02037 Z" id="Stroke-74" stroke="#000000" stroke-width="2" stroke-linejoin="round"></path>
</g>
<g id="Group-79" transform="translate(510.539294, 155.173225)">
<path d="M47.7614055,0.00230000028 C36.8694055,-0.0806999997 24.8474055,2.0753 14.6594055,9.7059 C-0.831594487,21.3502 -0.0925944875,50.3551 0.0484055125,78.866 C0.154405513,98.2379 1.14040551,113.0601 8.37340551,113.0601 C24.9934055,113.0601 26.1974055,84.4014 41.4764055,78.9321 C61.3064055,71.8397 78.1694055,62.0325 97.8234055,46.9654 C97.8234055,46.9654 142.675406,16.9329 136.747406,7.5116 C131.984406,-0.0748999997 102.017406,-1.7642 84.1274055,1.7997 C77.9884055,2.9994 71.6784055,3.0303 65.5034055,1.8658 C60.4224055,0.8999 54.2964055,0.0521000003 47.7614055,0.00230000028 Z" id="Fill-76" fill="#D4C393"></path>
<path d="M47.7618155,0.00228989028 C36.8700955,-0.0807576097 24.8481255,2.075328 14.6593655,9.705873 C-0.831092687,21.35016 -0.0922903175,50.35507 0.0488527225,78.86596 C0.154709313,98.23786 1.14050551,113.0601 8.37408751,113.0601 C24.9936755,113.0601 26.1978055,84.40141 41.4765455,78.93212 C61.3071355,71.83969 78.1693255,62.03245 97.8234955,46.96543 C97.8234955,46.96543 142.676106,16.93284 136.748106,7.51154 C131.984506,-0.0748959497 102.018106,-1.764203 84.1282055,1.799658 C77.9884455,2.999369 71.6789455,3.030249 65.5039455,1.865819 C60.4227855,0.8998703 54.2968555,0.0521183203 47.7618155,0.00228989028 Z" id="Stroke-78" stroke="#000000" stroke-width="2" stroke-linejoin="round"></path>
</g>
</g>
</g>
</svg>
\ No newline at end of file
docs/source/nnDetectionFunctional.svg 0 → 100644
View file @ a960f549
<?xml version="1.0" encoding="UTF-8"?>
<svg width="1186px" height="522px" viewBox="0 0 1186 522" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<title>Group 32</title>
<defs>
<rect id="path-1" x="0" y="0" width="190" height="439"></rect>
<mask id="mask-2" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="190" height="439" fill="white">
<use xlink:href="#path-1"></use>
</mask>
<rect id="path-3" x="0" y="0" width="257" height="183" rx="11"></rect>
<mask id="mask-4" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="257" height="183" fill="white">
<use xlink:href="#path-3"></use>
</mask>
<rect id="path-5" x="0" y="0" width="257" height="183" rx="11"></rect>
<mask id="mask-6" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="257" height="183" fill="white">
<use xlink:href="#path-5"></use>
</mask>
<polygon id="path-7" points="0 0 565 0 565 439 278.49536 439 0 439"></polygon>
<mask id="mask-8" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="565" height="439" fill="white">
<use xlink:href="#path-7"></use>
</mask>
<rect id="path-9" x="0" y="0" width="193" height="440"></rect>
<mask id="mask-10" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="193" height="440" fill="white">
<use xlink:href="#path-9"></use>
</mask>
<rect id="path-11" x="0" y="0" width="202" height="439"></rect>
<mask id="mask-12" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="202" height="439" fill="white">
<use xlink:href="#path-11"></use>
</mask>
</defs>
<g id="Page-1" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
<g id="Group-28">
<g id="Group-3" transform="translate(0.000000, 83.000000)">
<g id="Group-6" transform="translate(11.000000, 12.000000)">
<rect id="Rectangle" stroke="#000000" fill="#D0B8EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Distribution-of-Spac" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="28.9760742" y="27">Distribution of</tspan>
<tspan x="46.9760742" y="49">Spacings</tspan>
</text>
</g>
<g id="Group-7" transform="translate(11.000000, 100.000000)">
<rect id="Rectangle" stroke="#000000" fill="#D0B8EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Median-Shape" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="25.9570312" y="37">Median Shape</tspan>
</text>
</g>
<g id="Group" transform="translate(11.000000, 188.000000)">
<rect id="Rectangle" stroke="#000000" fill="#D0B8EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Intensity-Distributi" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="50.4829102" y="27">Intensity</tspan>
<tspan x="38.9824219" y="49">Distribution</tspan>
</text>
</g>
<g id="Group-8" transform="translate(11.000000, 276.000000)">
<rect id="Rectangle" stroke="#000000" fill="#D0B8EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Image-Modality" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="58.9863281" y="27">Image</tspan>
<tspan x="50.4873047" y="49">Modality</tspan>
</text>
</g>
<g id="Group-9" transform="translate(11.000000, 364.000000)">
<rect id="Rectangle" stroke="#000000" fill="#D0B8EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Object-Sizes" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="33.4804688" y="36">Object Sizes</tspan>
</text>
</g>
<use id="Rectangle" stroke="#A261F4" mask="url(#mask-2)" stroke-width="6" stroke-dasharray="3" xlink:href="#path-1"></use>
</g>
<g id="Group-18" transform="translate(202.000000, 83.000000)">
<g id="Group-16" transform="translate(368.000000, 14.000000)">
<g id="Group-4">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Resampling-Strategy" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="36.4775391" y="27">Resampling</tspan>
<tspan x="50.4829102" y="49">Strategy</tspan>
</text>
</g>
<g id="Group-4" transform="translate(0.000000, 74.000000)">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Intensity-Normalizat" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="50.4829102" y="27">Intensity</tspan>
<tspan x="28.9804688" y="49">Normalization</tspan>
</text>
</g>
<g id="Group-4" transform="translate(0.000000, 148.000000)">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Image-Target-Spacing" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="31.6303711" y="27">Image Target</tspan>
<tspan x="51.4760742" y="49">Spacing</tspan>
</text>
</g>
</g>
<g id="Group-17" transform="translate(17.000000, 19.000000)">
<g id="Group-13" transform="translate(0.000000, 112.000000)">
<rect id="Rectangle" stroke="#000000" fill="#B8EEDC" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Low-Resolution-Model" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="22.4633789" y="27">Low Resolution</tspan>
<tspan x="59.4873047" y="49">Model</tspan>
</text>
</g>
<g id="Group-4">
<g id="Rectangle-3" fill="#B8EEDC" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Full-Resolution-Mode" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="24.4716797" y="27">Full Resolution</tspan>
<tspan x="59.4873047" y="49">Model</tspan>
</text>
</g>
</g>
<g id="Group-12" transform="translate(301.000000, 245.000000)">
<use id="Rectangle" stroke="#000000" mask="url(#mask-4)" stroke-width="6" stroke-dasharray="5" xlink:href="#path-3"></use>
<g id="Group-5" transform="translate(10.000000, 9.000000)">
<g id="Group-4" transform="translate(34.000000, 0.000000)">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Anchor-Sizes" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="30.9755859" y="38">Anchor Sizes</tspan>
</text>
</g>
<g id="Group-4" transform="translate(33.000000, 103.000000)">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="IoU-Maximisation" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="14.9794922" y="38">IoU Maximisation</tspan>
</text>
</g>
<path id="Line" d="M202.454243,39.0865269 L203.183858,39.7703849 C225.379833,60.5743626 236.5,76.4121768 236.5,87.5 C236.5,97.058331 228.248793,110.118623 211.785975,126.852127 L211.252069,127.393192 L205.397,133.079 L215.017314,130.536861 L215.984101,130.281279 L216.495266,132.214853 L215.528479,132.470436 L202.217588,135.989341 C201.499591,136.179152 200.846999,135.551516 200.977527,134.845354 L201.003069,134.738933 L204.907999,121.53613 L205.191619,120.577193 L207.109492,121.144433 L206.825872,122.10337 L204.003,131.645 L209.844568,125.972433 C226.299827,109.345709 234.5,96.4672512 234.5,87.5 C234.5,77.2651261 223.836676,61.9756191 202.466985,41.8412308 L201.816142,41.2296151 L201.086527,40.545757 L202.454243,39.0865269 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line" d="M0.45424297,32.0865269 L1.18385805,32.7703849 C23.3798326,53.5743626 34.5,69.4121768 34.5,80.5 C34.5,90.058331 26.2487926,103.118623 9.78597538,119.852127 L9.25206852,120.393192 L3.397,126.079 L13.0173136,123.536861 L13.9841009,123.281279 L14.4952661,125.214853 L13.5284789,125.470436 L0.217588223,128.989341 L-1.50114461,129.44371 L-0.996931108,127.738933 L2.90799856,114.53613 L3.19161865,113.577193 L5.1094921,114.144433 L4.825872,115.10337 L2.003,124.645 L7.84456847,118.972433 C24.2998272,102.345709 32.5,89.4672512 32.5,80.5 C32.5,70.2651261 21.8366761,54.9756191 0.4669855,34.8412308 L-0.183858051,34.2296151 L-0.913473132,33.545757 L0.45424297,32.0865269 Z" fill="#000000" fill-rule="nonzero"></path>
</g>
</g>
<g id="Group-11" transform="translate(9.000000, 245.000000)">
<use id="Rectangle" stroke="#000000" mask="url(#mask-6)" stroke-width="6" stroke-dasharray="5" xlink:href="#path-5"></use>
<g id="Group-10" transform="translate(11.500000, 9.000000)">
<g id="Group-4" transform="translate(33.500000, 0.000000)">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Patch-Size" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="40.9775391" y="38">Patch Size</tspan>
</text>
</g>
<g id="Group-4" transform="translate(32.500000, 103.000000)">
<g id="Rectangle-3" fill="#BEEEB8" stroke="#000000">
<rect id="Rectangle" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
</g>
<text id="Network-Topology-&amp;-F" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="12.6240234" y="27">Network Topology</tspan>
<tspan x="28.9804688" y="49">&amp; FPN Levels</tspan>
</text>
</g>
<path id="Line" d="M201.954243,39.0865269 L202.683858,39.7703849 C224.879833,60.5743626 236,76.4121768 236,87.5 C236,97.058331 227.748793,110.118623 211.285975,126.852127 L210.752069,127.393192 L204.897,133.079 L214.517314,130.536861 L215.484101,130.281279 L215.995266,132.214853 L215.028479,132.470436 L201.717588,135.989341 C200.999591,136.179152 200.346999,135.551516 200.477527,134.845354 L200.503069,134.738933 L204.407999,121.53613 L204.691619,120.577193 L206.609492,121.144433 L206.325872,122.10337 L203.503,131.645 L209.344568,125.972433 C225.799827,109.345709 234,96.4672512 234,87.5 C234,77.2651261 223.336676,61.9756191 201.966985,41.8412308 L201.316142,41.2296151 L200.586527,40.545757 L201.954243,39.0865269 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line" d="M-0.0457570301,32.0865269 L0.683858051,32.7703849 C22.8798326,53.5743626 34,69.4121768 34,80.5 C34,90.058331 25.7487926,103.118623 9.28597538,119.852127 L8.75206852,120.393192 L2.897,126.079 L12.5173136,123.536861 L13.4841009,123.281279 L13.9952661,125.214853 L13.0284789,125.470436 L-0.282411777,128.989341 L-2.00114461,129.44371 L-1.49693111,127.738933 L2.40799856,114.53613 L2.69161865,113.577193 L4.6094921,114.144433 L4.325872,115.10337 L1.503,124.645 L7.34456847,118.972433 C23.7998272,102.345709 32,89.4672512 32,80.5 C32,70.2651261 21.3366761,54.9756191 -0.0330145001,34.8412308 L-0.683858051,34.2296151 L-1.41347313,33.545757 L-0.0457570301,32.0865269 Z" fill="#000000" fill-rule="nonzero"></path>
</g>
</g>
<use id="Rectangle" stroke="#64BF58" mask="url(#mask-8)" stroke-width="8" stroke-dasharray="2" xlink:href="#path-7"></use>
</g>
<g id="Group-21" transform="translate(993.000000, 82.000000)">
<g id="Group-14" transform="translate(13.000000, 16.000000)">
<rect id="Rectangle" stroke="#000000" fill-opacity="0.449260578" fill="#EE7400" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Model-NMS-IoU-Thresh" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="36.9873047" y="29">Model NMS</tspan>
<tspan x="27.6313477" y="51">IoU Threshold</tspan>
</text>
</g>
<g id="Group-14" transform="translate(13.000000, 188.000000)">
<rect id="Rectangle" stroke="#000000" fill-opacity="0.449260578" fill="#EE7400" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Model-Min-Probabilit" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="42.4848633" y="29">Model Min</tspan>
<tspan x="41.9794922" y="51">Probability</tspan>
</text>
</g>
<g id="Group-14" transform="translate(13.000000, 102.000000)">
<rect id="Rectangle" stroke="#000000" fill-opacity="0.449260578" fill="#EE7400" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Ensemble-WBC-IoU-Thr" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="20.4814453" y="29">Ensemble WBC</tspan>
<tspan x="27.6313477" y="51">IoU Threshold</tspan>
</text>
</g>
<g id="Group-14" transform="translate(12.000000, 274.000000)">
<rect id="Rectangle" stroke="#000000" fill-opacity="0.449260578" fill="#EE7400" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Min-Object-Size" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="20.9780273" y="38">Min Object Size</tspan>
</text>
</g>
<g id="Group-14" transform="translate(13.000000, 360.000000)">
<rect id="Rectangle" stroke="#000000" fill-opacity="0.45" fill="#EE7400" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Model-Selection" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="59.4873047" y="29">Model</tspan>
<tspan x="46.9760742" y="51">Selection</tspan>
</text>
</g>
<use id="Rectangle" stroke="#EE7400" mask="url(#mask-10)" stroke-width="6" stroke-dasharray="3" xlink:href="#path-9"></use>
</g>
<g id="Group-20" transform="translate(779.000000, 83.000000)">
<use id="Rectangle" stroke="#537EF9" mask="url(#mask-12)" stroke-width="8" stroke-dasharray="2" xlink:href="#path-11"></use>
<g id="Group-4" transform="translate(17.000000, 276.000000)">
<rect id="Rectangle" stroke="#000000" fill="#B8C6EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Optimizer-&amp;-Learning" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="36.9916992" y="27">Optimizer &amp;</tspan>
<tspan x="27.4599609" y="49">Learning Rate</tspan>
</text>
</g>
<g id="Group-4" transform="translate(17.000000, 364.000000)">
<rect id="Rectangle" stroke="#000000" fill="#B8C6EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Data-Augmentation" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="7.44287109" y="38">Data Augmentation</tspan>
</text>
</g>
<g id="Group-4" transform="translate(17.000000, 100.000000)">
<rect id="Rectangle" stroke="#000000" fill="#B8C6EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Anchor-Matching" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="16.4648438" y="38">Anchor Matching</tspan>
</text>
</g>
<g id="Group-4" transform="translate(17.000000, 188.000000)">
<rect id="Rectangle" stroke="#000000" fill="#B8C6EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Loss-Functions" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="23.4697266" y="38">Loss Functions</tspan>
</text>
</g>
<g id="Group-4" transform="translate(17.000000, 12.000000)">
<rect id="Rectangle" stroke="#000000" fill="#B8C6EE" x="0.5" y="0.5" width="167" height="61" rx="7"></rect>
<text id="Network-Blueprint" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="50.9926758" y="27">Network</tspan>
<tspan x="48.4790039" y="49">Blueprint</tspan>
</text>
</g>
</g>
<g id="Group-9" transform="translate(11.000000, 0.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#D0B8EE" x="1" y="1" width="166" height="60" rx="7"></rect>
<text id="Data-Fingerpint" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="64.9892578" y="27">Data</tspan>
<tspan x="43.9790039" y="49">Fingerpint</tspan>
</text>
</g>
<g id="Group-4" transform="translate(406.000000, 1.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#BEEEB8" x="1" y="1" width="166" height="60" rx="7"></rect>
<text id="Rule-based-Parameter" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="38.4682617" y="31">Rule based</tspan>
<tspan x="37.4838867" y="53">Parameters</tspan>
</text>
</g>
<g id="Group-4" transform="translate(796.000000, 2.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="166" height="60" rx="7"></rect>
<text id="Fixed-Parameters" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="61.9921875" y="27">Fixed</tspan>
<tspan x="37.4838867" y="49">Parameters</tspan>
</text>
</g>
<g id="Group-4" transform="translate(1007.000000, 1.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill-opacity="0.45" fill="#EE7400" x="1" y="1" width="166" height="60" rx="7"></rect>
<text id="Empricial-Optimiztio" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="46.9936523" y="27">Empricial</tspan>
<tspan x="38.9868164" y="49">Optimiztion</tspan>
</text>
</g>
<text id="Triggers" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="311.322266" y="194">Triggers</tspan>
</text>
<path id="Line-8" d="M490.349078,129.629546 L490.859766,130.489312 L562.71,251.452 L567.868729,248.388459 L569,264 L555.832004,255.538092 L560.99,252.474 L489.140234,131.510688 L488.629546,130.650922 L490.349078,129.629546 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-6" d="M489.340536,129.748958 L491.251042,130.340536 L490.955253,131.295789 L434.095,314.921 L439.82782,316.696978 L429,328 L426.454274,312.555931 L432.185,314.33 L489.044747,130.704211 L489.340536,129.748958 Z" fill="#000000" fill-rule="nonzero"></path>
<line x1="388" y1="132" x2="485" y2="132" id="Line-2" stroke="#000000" stroke-width="2" stroke-linecap="square"></line>
<line x1="388" y1="242" x2="428.5" y2="241.5" id="Line-2" stroke="#000000" stroke-width="2" stroke-linecap="square"></line>
<path id="Line-3" d="M430.940057,240.587057 L432.412943,241.940057 L431.736443,242.6765 L362.207,318.366 L366.626097,322.425299 L352,328 L356.315897,312.954302 L360.734,317.013 L430.263557,241.3235 L430.940057,240.587057 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-4" d="M430.806036,242.599151 L431.603443,243.202594 L531.439,318.754 L535.060409,313.969958 L542,328 L526.612212,325.133647 L530.232,320.349 L430.396557,244.797406 L429.599151,244.193964 L430.806036,242.599151 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-5" d="M429.779606,240.283023 L430.748291,240.531315 L555.686,272.554 L557.176442,266.743125 L569,277 L553.700365,280.304721 L555.19,274.492 L430.251709,242.468685 L429.283023,242.220394 L429.779606,240.283023 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-7" d="M490.595962,129.717491 L490.939235,130.656726 L558.132,314.507 L563.768814,312.447788 L562,328 L550.619519,317.253622 L556.255,315.193 L489.060765,131.343274 L488.717491,130.404038 L490.595962,129.717491 Z" fill="#000000" fill-rule="nonzero"></path>
<circle id="Oval" stroke="#979797" fill="#000000" cx="490" cy="131" r="4.5"></circle>
<circle id="Oval" stroke="#979797" fill="#000000" cx="430" cy="242" r="4.5"></circle>
<path id="Line-9" d="M302.00995,163.990149 L304.009851,164.00995 L303.999951,165.009901 L303.638,201.51 L309.638264,201.56999 L302.5,215.5 L295.63895,201.431383 L301.638,201.49 L302.000049,164.990099 L302.00995,163.990149 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-10" d="M1091.47754,331.47803 L1091.49975,332.477783 L1091.75048,343.7606 L1091.931,351.911 L1096.61614,343.132037 L1097.08692,342.249785 L1098.85142,343.191338 L1098.38065,344.07359 L1091.89891,356.220592 L1091.06198,357.78904 L1090.15619,356.259319 L1083.14126,344.412244 L1082.63176,343.551775 L1084.35269,342.532768 L1084.8622,343.393236 L1089.932,351.955 L1089.75098,343.805034 L1089.50025,332.522217 L1089.47803,331.522464 L1091.47754,331.47803 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-10" d="M1091.47754,331.47803 L1091.49975,332.477783 L1091.75048,343.7606 L1091.931,351.911 L1096.61614,343.132037 L1097.08692,342.249785 L1098.85142,343.191338 L1098.38065,344.07359 L1091.89891,356.220592 L1091.06198,357.78904 L1090.15619,356.259319 L1083.14126,344.412244 L1082.63176,343.551775 L1084.35269,342.532768 L1084.8622,343.393236 L1089.932,351.955 L1089.75098,343.805034 L1089.50025,332.522217 L1089.47803,331.522464 L1091.47754,331.47803 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-10" d="M1091.47754,245.47803 L1091.49975,246.477783 L1091.75048,257.7606 L1091.931,265.911 L1096.61614,257.132037 L1097.08692,256.249785 L1098.85142,257.191338 L1098.38065,258.07359 L1091.89891,270.220592 L1091.06198,271.78904 L1090.15619,270.259319 L1083.14126,258.412244 L1082.63176,257.551775 L1084.35269,256.532768 L1084.8622,257.393236 L1089.932,265.955 L1089.75098,257.805034 L1089.50025,246.522217 L1089.47803,245.522464 L1091.47754,245.47803 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-10" d="M1091.47754,159.47803 L1091.49975,160.477783 L1091.75048,171.7606 L1091.931,179.911 L1096.61614,171.132037 L1097.08692,170.249785 L1098.85142,171.191338 L1098.38065,172.07359 L1091.89891,184.220592 L1091.06198,185.78904 L1090.15619,184.259319 L1083.14126,172.412244 L1082.63176,171.551775 L1084.35269,170.532768 L1084.8622,171.393236 L1089.932,179.955 L1089.75098,171.805034 L1089.50025,160.522217 L1089.47803,159.522464 L1091.47754,159.47803 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-11" d="M387.928029,21.4572448 L389.235392,22.1926367 L405.235392,31.1926367 L407.559593,32.5 L405.235392,33.8073633 L389.235392,42.8073633 L387.928029,43.5427552 L386.457245,40.9280286 L387.764608,40.1926367 L398.772,34 L397,34 L397,31 L398.775,31 L387.764608,24.8073633 L386.457245,24.0719714 L387.928029,21.4572448 Z M190,31 L190,34 L177,34 L177,31 L190,31 Z M210,31 L210,34 L197,34 L197,31 L210,31 Z M230,31 L230,34 L217,34 L217,31 L230,31 Z M250,31 L250,34 L237,34 L237,31 L250,31 Z M270,31 L270,34 L257,34 L257,31 L270,31 Z M290,31 L290,34 L277,34 L277,31 L290,31 Z M310,31 L310,34 L297,34 L297,31 L310,31 Z M330,31 L330,34 L317,34 L317,31 L330,31 Z M350,31 L350,34 L337,34 L337,31 L350,31 Z M370,31 L370,34 L357,34 L357,31 L370,31 Z M391,31 L391,34 L377,34 L377,31 L391,31 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-11" d="M777.978614,21.3862558 L779.282606,22.1276096 L795.241343,31.2005744 L797.559551,32.5185368 L795.229404,33.8152738 L779.188475,42.7421213 L777.877768,43.4715359 L776.418939,40.8501207 L777.729646,40.1207061 L791.44,32.49 L777.799898,24.7355931 L776.495906,23.9942393 L777.978614,21.3862558 Z M773.00478,30.9063989 L774.504764,30.9132481 L778.515861,30.9315636 L784.50466,30.9589098 L786.004644,30.965759 L785.990946,33.9657277 L784.490962,33.9588785 L778.502163,33.9315324 L774.491066,33.9132169 L772.991081,33.9063676 L773.00478,30.9063989 Z M753.004988,30.8150756 L754.504973,30.8219249 L764.504868,30.8675865 L766.004853,30.8744357 L765.991154,33.8744045 L764.49117,33.8675552 L754.491274,33.8218936 L752.99129,33.8150444 L753.004988,30.8150756 Z M733.005197,30.7237524 L734.505181,30.7306016 L744.505077,30.7762633 L746.005061,30.7831125 L745.991363,33.7830812 L744.491379,33.776232 L734.491483,33.7305704 L732.991498,33.7237211 L733.005197,30.7237524 Z M713.005405,30.6324291 L714.50539,30.6392784 L724.505285,30.68494 L726.00527,30.6917892 L725.991571,33.691758 L724.491587,33.6849087 L714.491691,33.6392471 L712.991707,33.6323979 L713.005405,30.6324291 Z M693.005614,30.5411059 L694.505598,30.5479551 L704.505494,30.5936168 L706.005478,30.600466 L705.99178,33.6004347 L704.491796,33.5935855 L694.4919,33.5479239 L692.991915,33.5410746 L693.005614,30.5411059 Z M673.005822,30.4497826 L674.505807,30.4566319 L684.505702,30.5022935 L686.005687,30.5091427 L685.991988,33.5091115 L684.492004,33.5022622 L674.492108,33.4566006 L672.992124,33.4497514 L673.005822,30.4497826 Z M653.006031,30.3584594 L654.506015,30.3653086 L664.505911,30.4109703 L666.005895,30.4178195 L665.992197,33.4177882 L664.492213,33.410939 L654.492317,33.3652774 L652.992332,33.3584281 L653.006031,30.3584594 Z M633.006239,30.2671361 L634.506224,30.2739854 L644.506119,30.319647 L646.006104,30.3264963 L645.992405,33.326465 L644.492421,33.3196157 L634.492525,33.2739541 L632.992541,33.2671049 L633.006239,30.2671361 Z M613.006448,30.1758129 L614.506432,30.1826621 L624.506328,30.2283238 L626.006312,30.235173 L625.992614,33.2351417 L624.49263,33.2282925 L614.492734,33.1826309 L612.992749,33.1757816 L613.006448,30.1758129 Z M593.006656,30.0844896 L594.506641,30.0913389 L604.506536,30.1370005 L606.006521,30.1438498 L605.992822,33.1438185 L604.492838,33.1369692 L594.492942,33.0913076 L592.992958,33.0844584 L593.006656,30.0844896 Z M573.006865,29.9931664 L574.506849,30.0000156 L584.506745,30.0456773 L586.006729,30.0525265 L585.993031,33.0524952 L584.493047,33.045646 L574.493151,32.9999844 L572.993166,32.9931351 L573.006865,29.9931664 Z" fill="#000000" fill-rule="nonzero"></path>
<path id="Line-11" d="M988.928029,21.4572448 L990.235392,22.1926367 L1006.23539,31.1926367 L1008.55959,32.5 L1006.23539,33.8073633 L990.235392,42.8073633 L988.928029,43.5427552 L987.457245,40.9280286 L988.764608,40.1926367 L1002.44,32.5 L988.764608,24.8073633 L987.457245,24.0719714 L988.928029,21.4572448 Z M976,31 L976,34 L963,34 L963,31 L976,31 Z M996,31 L996,34 L983,34 L983,31 L996,31 Z" fill="#000000" fill-rule="nonzero"></path>
</g>
</g>
</svg>
\ No newline at end of file
docs/source/nnDetectionFunctionalDetails.svg 0 → 100644
View file @ a960f549
<?xml version="1.0" encoding="UTF-8"?>
<svg width="1198px" height="948px" viewBox="0 0 1198 948" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<title>Group 33</title>
<defs>
<path d="M0,0 L1198,0 L1198,47 L0,47 L0,0 Z" id="path-1"></path>
<mask id="mask-2" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="1198" height="47" fill="white">
<use xlink:href="#path-1"></use>
</mask>
</defs>
<g id="Page-1" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
<g id="Group-27">
<g id="Group-19">
<g id="Group-15" transform="translate(0.000000, 65.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="102"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#BEEEB8" x="1.5" y="1.5" width="184" height="102"></rect>
<text id="Resampling-Strategy" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="44.4775391" y="18">Resampling</tspan>
<tspan x="58.4829102" y="40">Strategy</tspan>
</text>
<text id="Image:-We-use-the-sa" font-family="Helvetica-Bold, Helvetica" font-size="18" font-weight="bold" fill="#000000">
<tspan x="196" y="27">Image</tspan>
<tspan x="248.022461" y="27" font-family="Helvetica" font-weight="normal">: We use the same image resampling </tspan>
<tspan x="196" y="49" font-family="Helvetica" font-weight="normal">procedure as nnU-Net</tspan>
<tspan x="196" y="71">Annotation</tspan>
<tspan x="290.974609" y="71" font-family="Helvetica" font-weight="normal">: Annotations are resampled with </tspan>
<tspan x="196" y="93" font-family="Helvetica" font-weight="normal">nearest neighbor</tspan>
</text>
</g>
<g id="Group-15">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="65"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#FFFFFF" x="1.5" y="1.5" width="184" height="65"></rect>
<text id="Parameter" font-family="Helvetica-Bold, Helvetica" font-size="18" font-weight="bold" fill="#000000">
<tspan x="49.9711914" y="18">Parameter</tspan>
</text>
<text id="Description" font-family="Helvetica-Bold, Helvetica" font-size="18" font-weight="bold" fill="#000000">
<tspan x="340.491211" y="18">Description</tspan>
</text>
</g>
<g id="Group-15" transform="translate(0.000000, 167.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="214"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#BEEEB8" x="1.5" y="1.5" width="184" height="214"></rect>
<text id="Network-Topology-&amp;-F" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="12.1206055" y="18">Network Topology &amp;</tspan>
<tspan x="36.9804688" y="40">FPN Levels &amp;</tspan>
<tspan x="48.9775391" y="62">Patch Size</tspan>
</text>
<text id="The-anisotric-axis-o" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="27">The anisotric axis of the patch size is initialized </tspan>
<tspan x="195" y="49">with the median shape of the anisotropic axis of </tspan>
<tspan x="195" y="71">the dataset. The isotropic axes are initialized </tspan>
<tspan x="195" y="93">with the minimum size of the isotropic axes of </tspan>
<tspan x="195" y="115">the dataset. </tspan>
<tspan x="195" y="137">The patch size is decreased while adapting the </tspan>
<tspan x="195" y="159">network architecture and feature pyramid </tspan>
<tspan x="195" y="181">network levels until the memory constrains are </tspan>
<tspan x="195" y="203">fulfilled. The batch size is fixed to four.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(0.000000, 381.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="121"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#BEEEB8" x="1.5" y="1.5" width="184" height="121"></rect>
<text id="Anchor-Optimization" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="63.4838867" y="18">Anchor</tspan>
<tspan x="41.9814453" y="40">Optimization</tspan>
</text>
<text id="The-anchor-sizes-are" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="25">The anchor sizes are determined by maximising </tspan>
<tspan x="195" y="47">the IoU of the best fitting anchor on the given </tspan>
<tspan x="195" y="69">object sizes extracted from the training set. </tspan>
<tspan x="195" y="91">Optimization of three anchor sizes per axis is </tspan>
<tspan x="195" y="113">performed via differential evolution.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(0.000000, 502.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="168"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#B8EEDC" x="1.5" y="1.5" width="184" height="168"></rect>
<text id="Low-Resolution-Model" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="30.4633789" y="18">Low Resolution</tspan>
<tspan x="67.4873047" y="40">Model</tspan>
</text>
<text id="The-low-resolution-c" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="24">The low resolution configuration will be triggered </tspan>
<tspan x="195" y="46">if the 99.5 percentile of object sizes along any</tspan>
<tspan x="195" y="68">axes exceeds the patch size of the full </tspan>
<tspan x="195" y="90">resolution model. If the low resolution </tspan>
<tspan x="195" y="112">configuration is triggered, the target spacing </tspan>
<tspan x="195" y="134">along each axes will be increased by two to </tspan>
<tspan x="195" y="156">incorporate more contextual information.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(605.000000, 263.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="275"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#B8C6EE" x="1.5" y="1.5" width="184" height="275"></rect>
<text id="Optimizer-&amp;-Learning" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="44.9916992" y="18">Optimizer &amp;</tspan>
<tspan x="35.4599609" y="40">Learning Rate</tspan>
</text>
<text id="All-configurations-a" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="24">All configurations are trained for 60 epochs with </tspan>
<tspan x="195" y="46">2500 mini batches per epoch and half of the </tspan>
<tspan x="195" y="68">batch is forced to contain at least one object. </tspan>
<tspan x="195" y="90">SGD with Nesterov momentum 0.9 is used.</tspan>
<tspan x="195" y="112">At the beginning of the training the learning rate </tspan>
<tspan x="195" y="134">is linearly ramped up from 1e-6 to 1e-2 over the </tspan>
<tspan x="195" y="156">first 4000 iterations. Poly learning rate schedule </tspan>
<tspan x="195" y="178">is used until epoch 50. The last 10 epochs are </tspan>
<tspan x="195" y="200">trained with a cyclic learning rate fluctuating </tspan>
<tspan x="195" y="222">between 1e-3 and 1e-6 during every epoch.</tspan>
<tspan x="195" y="244">We snapshot the model weights after each </tspan>
<tspan x="195" y="266">epoch for Stochastic Weight Averaging.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(0.000000, 671.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="121"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#B8C6EE" x="1.5" y="1.5" width="184" height="121"></rect>
<text id="Architecture-Templat" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="43.980957" y="18">Architecture</tspan>
<tspan x="55.4814453" y="40">Template</tspan>
</text>
<text id="Retina-U-Net-with-an" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="25">Retina U-Net with an encoder which consists of </tspan>
<tspan x="195" y="47">plain convolutions, ReLU and instance </tspan>
<tspan x="195" y="69">normalization blocks. The detection heads used </tspan>
<tspan x="195" y="91">for anchor classification and regression consist </tspan>
<tspan x="195" y="113">of three convolutions with group norm.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(605.000000, 0.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="263"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#B8C6EE" x="1.5" y="1.5" width="184" height="263"></rect>
<text id="Loss-Functions" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="31.4697266" y="18">Loss Functions</tspan>
</text>
<text id="Detection-Branch:-To" font-family="Helvetica-Bold, Helvetica" font-size="18" font-weight="bold" fill="#000000">
<tspan x="195" y="30">Detection Branch</tspan>
<tspan x="344.027344" y="30" font-family="Helvetica" font-weight="normal">: To balance positive and </tspan>
<tspan x="195" y="52" font-family="Helvetica" font-weight="normal">negative anchors, hard negative mining is used </tspan>
<tspan x="195" y="74" font-family="Helvetica" font-weight="normal">while selecting 1/3 positive and 2/3 negative </tspan>
<tspan x="195" y="96" font-family="Helvetica" font-weight="normal">anchors. The classification branch is trained with </tspan>
<tspan x="195" y="118" font-family="Helvetica" font-weight="normal">the Binary Cross-Entropy loss and the </tspan>
<tspan x="195" y="140" font-family="Helvetica" font-weight="normal">Generalized IoU Loss is used for anchor </tspan>
<tspan x="195" y="162" font-family="Helvetica" font-weight="normal">regression.</tspan>
<tspan x="195" y="184">Segmentation Branch</tspan>
<tspan x="381.029297" y="184" font-family="Helvetica" font-weight="normal">: The segmentation </tspan>
<tspan x="195" y="206" font-family="Helvetica" font-weight="normal">branch is trained with the Dice and Cross-</tspan>
<tspan x="195" y="228" font-family="Helvetica" font-weight="normal">Entropy loss to distinguish foreground and </tspan>
<tspan x="195" y="250" font-family="Helvetica" font-weight="normal">background pixels.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(605.000000, 538.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="72"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#B8C6EE" x="1.5" y="1.5" width="184" height="72"></rect>
<text id="Data-Augmentation" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="15.4428711" y="18">Data Augmentation</tspan>
</text>
<text id="We-use-the-same-augm" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="22">We use the same augmentation strategy as </tspan>
<tspan x="195" y="44">nnU-Net without simulating low resolution </tspan>
<tspan x="195" y="66">samples.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(0.000000, 792.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="99"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill="#B8C6EE" x="1.5" y="1.5" width="184" height="99"></rect>
<text id="Anchor-Matching" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="24.4648438" y="18">Anchor Matching</tspan>
</text>
<text id="Adaptive-Training-Sa" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="24">Adaptive Training Sample Selection (ATSS) is </tspan>
<tspan x="195" y="46">used to match anchors and ground truth boxes. </tspan>
<tspan x="195" y="68">The center of the anchor boxes do not need to </tspan>
<tspan x="195" y="90">lie within the ground truth box.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(605.000000, 610.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="187"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill-opacity="0.45" fill="#EE7400" x="1.5" y="1.5" width="184" height="187"></rect>
<text id="Empirical-Parameter" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="10.4770508" y="18">Empirical Parameter</tspan>
<tspan x="41.9814453" y="40">Optimization</tspan>
</text>
<text id="Parameters-which-are" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="23">Parameters which are only required during the </tspan>
<tspan x="195" y="45">inference procedure are empirically optimized </tspan>
<tspan x="195" y="67">by evaluating the performance on the validation </tspan>
<tspan x="195" y="89">set. This includes: the IoU threshold required for </tspan>
<tspan x="195" y="111">the NMS of the model, the IoU threshold </tspan>
<tspan x="195" y="133">required to perform WBC, a minimum probability </tspan>
<tspan x="195" y="155">for predictions of the model, a minimum object </tspan>
<tspan x="195" y="177">size.</tspan>
</text>
</g>
<g id="Group-15" transform="translate(605.000000, 797.000000)">
<rect id="Rectangle" stroke="#979797" stroke-width="3" x="185.5" y="1.5" width="406" height="94"></rect>
<rect id="Rectangle" stroke="#979797" stroke-width="3" fill-opacity="0.45" fill="#EE7400" x="1.5" y="1.5" width="184" height="94"></rect>
<text id="Model-Selection" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="67.4873047" y="18">Model</tspan>
<tspan x="54.9760742" y="40">Selection</tspan>
</text>
<text id="If-the-low-resolutio" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="195" y="30">If the low resolution model was triggered, only </tspan>
<tspan x="195" y="52">the best model as determined by the five fold </tspan>
<tspan x="195" y="74">cross-validation will be used for the test set.</tspan>
</text>
</g>
</g>
<g id="Group-25" transform="translate(0.000000, 901.000000)">
<use id="Rectangle-2" stroke="#585757" mask="url(#mask-2)" stroke-width="8" stroke-dasharray="3" xlink:href="#path-1"></use>
<g id="Group-23" transform="translate(110.000000, 13.000000)" fill="#000000">
<path id="Line-8" d="M57,4 L71,11 L57,18 L57,12 L-1,12 L-1,10 L57,10 L57,4 Z" fill-rule="nonzero"></path>
<text id="Symbolizes-a-depende" font-family="Helvetica" font-size="18" font-weight="normal">
<tspan x="82" y="18">Symbolizes a dependency</tspan>
</text>
</g>
<g id="Group-22" transform="translate(720.000000, 13.000000)" fill="#000000">
<path id="Line-10" d="M59.3686857,3.88816322 L60.2402612,4.37842446 L72.2402612,11.1284245 L73.7897289,12 L72.2402612,12.8715755 L60.2402612,19.6215755 L59.3686857,20.1118368 L58.3881632,18.3686857 L59.2597388,17.8784245 L69.71,12 L59.2597388,6.12157554 L58.3881632,5.6313143 L59.3686857,3.88816322 Z M8,11 L8,13 L-1,13 L-1,11 L8,11 Z M22,11 L22,13 L13,13 L13,11 L22,11 Z M36,11 L36,13 L27,13 L27,11 L36,11 Z M50,11 L50,13 L41,13 L41,11 L50,11 Z M64,11 L64,13 L55,13 L55,11 L64,11 Z" fill-rule="nonzero"></path>
<text id="Denotes-sequential-p" font-family="Helvetica" font-size="18" font-weight="normal">
<tspan x="85" y="18">Denotes sequential procedures</tspan>
</text>
</g>
</g>
</g>
</g>
</svg>
\ No newline at end of file
docs/source/nnDetectionModule.svg 0 → 100644
View file @ a960f549
<?xml version="1.0" encoding="UTF-8"?>
<svg width="1186px" height="312px" viewBox="0 0 1186 312" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<title>Group 31</title>
<defs>
<polygon id="path-1" points="0 0 526 0 526 312 259.271786 312 0 312"></polygon>
<mask id="mask-2" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="526" height="312" fill="white">
<use xlink:href="#path-1"></use>
</mask>
<rect id="path-3" x="25" y="13" width="238" height="284"></rect>
<mask id="mask-4" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="238" height="284" fill="white">
<use xlink:href="#path-3"></use>
</mask>
<polygon id="path-5" points="0 1.08246745e-15 442 1.08246745e-15 648 0 648 147 442 147 442 312 0 312"></polygon>
<mask id="mask-6" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="648" height="312" fill="white">
<use xlink:href="#path-5"></use>
</mask>
<rect id="path-7" x="455" y="156" width="193" height="156"></rect>
<mask id="mask-8" maskContentUnits="userSpaceOnUse" maskUnits="objectBoundingBox" x="0" y="0" width="193" height="156" fill="white">
<use xlink:href="#path-7"></use>
</mask>
</defs>
<g id="Page-1" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
<g id="Group-31">
<g id="Group-29">
<g id="Group-9" transform="translate(38.000000, 25.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#D0B8EE" x="1" y="1" width="209" height="60" rx="7"></rect>
<text id="planning.properties." font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="25.9467773" y="27">planning.properties.</tspan>
<tspan x="71.4785156" y="49">instance</tspan>
</text>
</g>
<g id="Group-9" transform="translate(38.000000, 125.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#D0B8EE" x="1" y="1" width="209" height="60" rx="7"></rect>
<text id="planning.properties." font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="25.9467773" y="27">planning.properties.</tspan>
<tspan x="71.9838867" y="49">intensity</tspan>
</text>
</g>
<g id="Group-9" transform="translate(36.000000, 225.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#D0B8EE" x="1" y="1" width="209" height="60" rx="7"></rect>
<text id="planning.properties." font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="25.9467773" y="27">planning.properties.</tspan>
<tspan x="73.987793" y="49">medical</tspan>
</text>
</g>
<g id="Group-9" transform="translate(282.000000, 13.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#BEEEB8" x="1" y="1" width="221" height="60" rx="7"></rect>
<text id="planning.-plan_exper" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="74.9682617" y="27">planning.</tspan>
<tspan x="44.9580078" y="49">plan_experiment</tspan>
</text>
</g>
<g id="Group-9" transform="translate(282.000000, 87.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#BEEEB8" x="1" y="1" width="221" height="60" rx="7"></rect>
<text id="planning.-plan_archi" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="74.9682617" y="27">planning.</tspan>
<tspan x="42.4575195" y="49">plan_architecture</tspan>
</text>
</g>
<g id="Group-9" transform="translate(282.000000, 161.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#BEEEB8" x="1" y="1" width="221" height="60" rx="7"></rect>
<text id="planning.-estimator" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="74.9682617" y="27">planning.</tspan>
<tspan x="74.4892578" y="49">estimator</tspan>
</text>
</g>
<g id="Group-9" transform="translate(282.000000, 235.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#BEEEB8" x="1" y="1" width="221" height="60" rx="7"></rect>
<text id="preprocessing" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="54.96875" y="38">preprocessing</tspan>
</text>
</g>
<use id="Rectangle" stroke="#64BF58" mask="url(#mask-2)" stroke-width="8" stroke-dasharray="2" xlink:href="#path-1"></use>
<use id="Rectangle" stroke="#A261F4" mask="url(#mask-4)" stroke-width="6" stroke-dasharray="3" xlink:href="#path-3"></use>
</g>
<g id="Group-30" transform="translate(538.000000, 0.000000)">
<g id="Group-9" transform="translate(226.000000, 161.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="io.datamodule.-bg_lo" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="30.4594727" y="27">io.datamodule.</tspan>
<tspan x="49.4658203" y="49">bg_loader</tspan>
</text>
</g>
<g id="Group-9" transform="translate(226.000000, 235.000000)" fill="#E0E0E0" stroke="#000000" stroke-width="2">
<rect id="Rectangle" x="1" y="1" width="177" height="60" rx="7"></rect>
</g>
<g id="Group-9" transform="translate(454.000000, 8.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#E0E0E0" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="evaluator" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="52.4760742" y="38">evaluator</tspan>
</text>
</g>
<g id="Group-9" transform="translate(454.000000, 75.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#E0E0E0" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="utils" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="73.4951172" y="38">utils</tspan>
</text>
</g>
<g id="Group-9" transform="translate(12.000000, 13.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#E0E0E0" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="ptmodule" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="52.4760742" y="38">ptmodule</tspan>
</text>
</g>
<g id="Group-9" transform="translate(226.000000, 87.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="io.-augmentation" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="79.9946289" y="27">io.</tspan>
<tspan x="34.9594727" y="49">augmentation</tspan>
</text>
</g>
<g id="Group-9" transform="translate(226.000000, 13.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="training" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="59.9819336" y="38">training</tspan>
</text>
</g>
<g id="Group-9" transform="translate(12.000000, 161.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="models" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="60.4873047" y="38">models</tspan>
</text>
</g>
<g id="Group-9" transform="translate(462.000000, 169.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#F7C08C" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="inferene" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="56.9760742" y="38">inferene</tspan>
</text>
</g>
<g id="Group-9" transform="translate(462.000000, 241.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#F7C08C" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="inferene.-sweeper" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="54.4755859" y="27">inferene.</tspan>
<tspan x="55.4819336" y="49">sweeper</tspan>
</text>
</g>
<g id="Group-9" transform="translate(12.000000, 87.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="detection" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="52.9726562" y="38">detection</tspan>
</text>
</g>
<g id="Group-9" transform="translate(12.000000, 235.000000)">
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<rect id="Rectangle" stroke="#000000" stroke-width="2" fill="#B8C6EE" x="1" y="1" width="177" height="60" rx="7"></rect>
<text id="losses" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="63.9897461" y="38">losses</tspan>
</text>
</g>
<use id="Rectangle" stroke="#537EF9" mask="url(#mask-6)" stroke-width="8" stroke-dasharray="2" xlink:href="#path-5"></use>
<use id="Rectangle" stroke="#EE7400" mask="url(#mask-8)" stroke-width="6" stroke-dasharray="3" xlink:href="#path-7"></use>
<text id="io.datamodule.-bg_mo" font-family="Helvetica" font-size="18" font-weight="normal" fill="#000000">
<tspan x="256.459473" y="262">io.datamodule.</tspan>
<tspan x="270.96582" y="284">bg_module</tspan>
</text>
</g>
</g>
</g>
</svg>
\ No newline at end of file
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment