Skip to content

Latest commit

 

History

History
1300 lines (894 loc) · 41.1 KB

README.md

File metadata and controls

1300 lines (894 loc) · 41.1 KB

Quantifying the redundancy between prosody and text

Paper EMNLP 2023

Prosody Experiments Super Quick Start:

  1. Create an environment, e.g with anaconda.
conda create --name prosody python=3.10
conda activate prosody
conda install --file conda_requirements.txt
  1. Download the already prepared prominence dataset:
cd data
git clone https://github.com/Helsinki-NLP/prosody.git
  1. Run your first experiment
cd .. # go back to project root
python src/train.py experiment=duration_regression_abs_gpt2

For details on the full functionality of this repository please refer to the full README below!

Powered by Pytorch-Lightning and Hydra

python pytorch lightning hydra black isort
tests code-quality codecov
license PRs contributors

Main Technologies

PyTorch Lightning - a lightweight PyTorch wrapper for high-performance AI research. Think of it as a framework for organizing your PyTorch code.

Hydra - a framework for elegantly configuring complex applications. The key feature is the ability to dynamically create a hierarchical configuration by composition and override it through config files and the command line.


Main Ideas

  • Rapid Experimentation: thanks to hydra command line superpowers
  • Minimal Boilerplate: thanks to automating pipelines with config instantiation
  • Main Configs: allow you to specify default training configuration
  • Experiment Configs: allow you to override chosen hyperparameters and version control experiments
  • Workflow: comes down to 4 simple steps
  • Experiment Tracking: Tensorboard, W&B, Neptune, Comet, MLFlow and CSVLogger
  • Logs: all logs (checkpoints, configs, etc.) are stored in a dynamically generated folder structure
  • Hyperparameter Search: simple search is effortless with Hydra plugins like Optuna Sweeper
  • Tests: generic, easy-to-adapt smoke tests for speeding up the development
  • Continuous Integration: automatically test and lint your repo with Github Actions
  • Best Practices: a couple of recommended tools, practices and standards

Project Structure

The directory structure of new project looks like this:

├── .github                   <- Github Actions workflows
│
├── configs                   <- Hydra configs
│   ├── callbacks                <- Callbacks configs
│   ├── data                     <- Data configs
│   ├── debug                    <- Debugging configs
│   ├── experiment               <- Experiment configs
│   ├── extras                   <- Extra utilities configs
│   ├── hparams_search           <- Hyperparameter search configs
│   ├── hydra                    <- Hydra configs
│   ├── local                    <- Local configs
│   ├── logger                   <- Logger configs
│   ├── model                    <- Model configs
│   ├── paths                    <- Project paths configs
│   ├── trainer                  <- Trainer configs
│   │
│   ├── eval.yaml             <- Main config for evaluation
│   └── train.yaml            <- Main config for training
│
├── data                   <- Project data
│
├── logs                   <- Logs generated by hydra and lightning loggers
│
├── notebooks              <- Jupyter notebooks. Naming convention is a number (for ordering),
│                             the creator's initials, and a short `-` delimited description,
│                             e.g. `1.0-jqp-initial-data-exploration.ipynb`.
│
├── scripts                <- Shell scripts
│
├── src                    <- Source code
│   ├── data                     <- Data scripts
│   ├── models                   <- Model scripts
│   ├── utils                    <- Utility scripts
│   │
│   ├── eval.py                  <- Run evaluation
│   └── train.py                 <- Run training
│
├── tests                  <- Tests of any kind
│
├── .env.example              <- Example of file for storing private environment variables
├── .gitignore                <- List of files ignored by git
├── .pre-commit-config.yaml   <- Configuration of pre-commit hooks for code formatting
├── .project-root             <- File for inferring the position of project root directory
├── environment.yaml          <- File for installing conda environment
├── Makefile                  <- Makefile with commands like `make train` or `make test`
├── pyproject.toml            <- Configuration options for testing and linting
├── requirements.txt          <- File for installing python dependencies
├── setup.py                  <- File for installing project as a package
└── README.md

⚡  Your Superpowers

Override any config parameter from command line
python train.py trainer.max_epochs=20 model.optimizer.lr=1e-4

Note: You can also add new parameters with + sign.

python train.py +model.new_param="owo"
Train on CPU, GPU, multi-GPU and TPU
# train on CPU
python train.py trainer=cpu

# train on 1 GPU
python train.py trainer=gpu

# train on TPU
python train.py +trainer.tpu_cores=8

# train with DDP (Distributed Data Parallel) (4 GPUs)
python train.py trainer=ddp trainer.devices=4

# train with DDP (Distributed Data Parallel) (8 GPUs, 2 nodes)
python train.py trainer=ddp trainer.devices=4 trainer.num_nodes=2

# simulate DDP on CPU processes
python train.py trainer=ddp_sim trainer.devices=2

# accelerate training on mac
python train.py trainer=mps

Warning: Currently there are problems with DDP mode, read this issue to learn more.

Train with mixed precision
# train with pytorch native automatic mixed precision (AMP)
python train.py trainer=gpu +trainer.precision=16
Train model with any logger available in PyTorch Lightning, like W&B or Tensorboard
# set project and entity names in `configs/logger/wandb`
wandb:
  project: "your_project_name"
  entity: "your_wandb_team_name"
# train model with Weights&Biases (link to wandb dashboard should appear in the terminal)
python train.py logger=wandb

Note: Lightning provides convenient integrations with most popular logging frameworks. Learn more here.

Note: Using wandb requires you to setup account first. After that just complete the config as below.

Note: Click here to see example wandb dashboard generated with this template.

Train model with chosen experiment config
python train.py experiment=example

Note: Experiment configs are placed in configs/experiment/.

Attach some callbacks to run
python train.py callbacks=default

Note: Callbacks can be used for things such as as model checkpointing, early stopping and many more.

Note: Callbacks configs are placed in configs/callbacks/.

Use different tricks available in Pytorch Lightning
# gradient clipping may be enabled to avoid exploding gradients
python train.py +trainer.gradient_clip_val=0.5

# run validation loop 4 times during a training epoch
python train.py +trainer.val_check_interval=0.25

# accumulate gradients
python train.py +trainer.accumulate_grad_batches=10

# terminate training after 12 hours
python train.py +trainer.max_time="00:12:00:00"

Note: PyTorch Lightning provides about 40+ useful trainer flags.

Easily debug
# runs 1 epoch in default debugging mode
# changes logging directory to `logs/debugs/...`
# sets level of all command line loggers to 'DEBUG'
# enforces debug-friendly configuration
python train.py debug=default

# run 1 train, val and test loop, using only 1 batch
python train.py debug=fdr

# print execution time profiling
python train.py debug=profiler

# try overfitting to 1 batch
python train.py debug=overfit

# raise exception if there are any numerical anomalies in tensors, like NaN or +/-inf
python train.py +trainer.detect_anomaly=true

# use only 20% of the data
python train.py +trainer.limit_train_batches=0.2 \
+trainer.limit_val_batches=0.2 +trainer.limit_test_batches=0.2

Note: Visit configs/debug/ for different debugging configs.

Resume training from checkpoint
python train.py ckpt_path="/path/to/ckpt/name.ckpt"

Note: Checkpoint can be either path or URL.

Note: Currently loading ckpt doesn't resume logger experiment, but it will be supported in future Lightning release.

Evaluate checkpoint on test dataset
python eval.py ckpt_path="/path/to/ckpt/name.ckpt"

Note: Checkpoint can be either path or URL.

Create a sweep over hyperparameters
# this will run 6 experiments one after the other,
# each with different combination of batch_size and learning rate
python train.py -m data.batch_size=32,64,128 model.lr=0.001,0.0005

Note: Hydra composes configs lazily at job launch time. If you change code or configs after launching a job/sweep, the final composed configs might be impacted.

Create a sweep over hyperparameters with Optuna
# this will run hyperparameter search defined in `configs/hparams_search/mnist_optuna.yaml`
# over chosen experiment config
python train.py -m hparams_search=mnist_optuna experiment=example

Note: Using Optuna Sweeper doesn't require you to add any boilerplate to your code, everything is defined in a single config file.

Warning: Optuna sweeps are not failure-resistant (if one job crashes then the whole sweep crashes).

Execute all experiments from folder
python train.py -m 'experiment=glob(*)'

Note: Hydra provides special syntax for controlling behavior of multiruns. Learn more here. The command above executes all experiments from configs/experiment/.

Execute run for multiple different seeds
python train.py -m seed=1,2,3,4,5 trainer.deterministic=True logger=csv tags=["benchmark"]

Note: trainer.deterministic=True makes pytorch more deterministic but impacts the performance.

Execute sweep on a remote AWS cluster

Note: This should be achievable with simple config using Ray AWS launcher for Hydra. Example is not implemented in this template.

Use Hydra tab completion

Note: Hydra allows you to autocomplete config argument overrides in shell as you write them, by pressing tab key. Read the docs.

Apply pre-commit hooks
pre-commit run -a

Note: Apply pre-commit hooks to do things like auto-formatting code and configs, performing code analysis or removing output from jupyter notebooks. See # Best Practices for more.

Update pre-commit hook versions in .pre-commit-config.yaml with:

pre-commit autoupdate
Run tests
# run all tests
pytest

# run tests from specific file
pytest tests/test_train.py

# run all tests except the ones marked as slow
pytest -k "not slow"
Use tags

Each experiment should be tagged in order to easily filter them across files or in logger UI:

python train.py tags=["mnist","experiment_X"]

Note: You might need to escape the bracket characters in your shell with python train.py tags=\["mnist","experiment_X"\].

If no tags are provided, you will be asked to input them from command line:

>>> python train.py tags=[]
[2022-07-11 15:40:09,358][src.utils.utils][INFO] - Enforcing tags! <cfg.extras.enforce_tags=True>
[2022-07-11 15:40:09,359][src.utils.rich_utils][WARNING] - No tags provided in config. Prompting user to input tags...
Enter a list of comma separated tags (dev):

If no tags are provided for multirun, an error will be raised:

>>> python train.py -m +x=1,2,3 tags=[]
ValueError: Specify tags before launching a multirun!

Note: Appending lists from command line is currently not supported in hydra :(

How It Works

All PyTorch Lightning modules are dynamically instantiated from module paths specified in config. Example model config:

_target_: src.models.mnist_model.MNISTLitModule
lr: 0.001
net:
  _target_: src.models.components.simple_dense_net.SimpleDenseNet
  input_size: 784
  lin1_size: 256
  lin2_size: 256
  lin3_size: 256
  output_size: 10

Using this config we can instantiate the object with the following line:

model = hydra.utils.instantiate(config.model)

This allows you to easily iterate over new models! Every time you create a new one, just specify its module path and parameters in appropriate config file.

Switch between models and datamodules with command line arguments:

python train.py model=mnist

Example pipeline managing the instantiation logic: src/train.py.


Main Config

Location: configs/train.yaml
Main project config contains default training configuration.
It determines how config is composed when simply executing command python train.py.

Show main project config
# order of defaults determines the order in which configs override each other
defaults:
  - _self_
  - data: mnist.yaml
  - model: mnist.yaml
  - callbacks: default.yaml
  - logger: null # set logger here or use command line (e.g. `python train.py logger=csv`)
  - trainer: default.yaml
  - paths: default.yaml
  - extras: default.yaml
  - hydra: default.yaml

  # experiment configs allow for version control of specific hyperparameters
  # e.g. best hyperparameters for given model and datamodule
  - experiment: null

  # config for hyperparameter optimization
  - hparams_search: null

  # optional local config for machine/user specific settings
  # it's optional since it doesn't need to exist and is excluded from version control
  - optional local: default.yaml

  # debugging config (enable through command line, e.g. `python train.py debug=default)
  - debug: null

# task name, determines output directory path
task_name: "train"

# tags to help you identify your experiments
# you can overwrite this in experiment configs
# overwrite from command line with `python train.py tags="[first_tag, second_tag]"`
# appending lists from command line is currently not supported :(
# https://github.com/facebookresearch/hydra/issues/1547
tags: ["dev"]

# set False to skip model training
train: True

# evaluate on test set, using best model weights achieved during training
# lightning chooses best weights based on the metric specified in checkpoint callback
test: True

# simply provide checkpoint path to resume training
ckpt_path: null

# seed for random number generators in pytorch, numpy and python.random
seed: null

Experiment Config

Location: configs/experiment
Experiment configs allow you to overwrite parameters from main config.
For example, you can use them to version control best hyperparameters for each combination of model and dataset.

Show example experiment config
# @package _global_

# to execute this experiment run:
# python train.py experiment=example

defaults:
  - override /data: mnist.yaml
  - override /model: mnist.yaml
  - override /callbacks: default.yaml
  - override /trainer: default.yaml

# all parameters below will be merged with parameters from default configurations set above
# this allows you to overwrite only specified parameters

tags: ["mnist", "simple_dense_net"]

seed: 12345

trainer:
  min_epochs: 10
  max_epochs: 10
  gradient_clip_val: 0.5

model:
  optimizer:
    lr: 0.002
  net:
    lin1_size: 128
    lin2_size: 256
    lin3_size: 64

data:
  batch_size: 64

logger:
  wandb:
    tags: ${tags}
    group: "mnist"

Workflow

Basic workflow

  1. Write your PyTorch Lightning module (see models/mnist_module.py for example)
  2. Write your PyTorch Lightning datamodule (see data/mnist_datamodule.py for example)
  3. Write your experiment config, containing paths to model and datamodule
  4. Run training with chosen experiment config:
    python src/train.py experiment=experiment_name.yaml

Experiment design

Say you want to execute many runs to plot how accuracy changes in respect to batch size.

  1. Execute the runs with some config parameter that allows you to identify them easily, like tags:

    python train.py -m logger=csv data.batch_size=16,32,64,128 tags=["batch_size_exp"]
  2. Write a script or notebook that searches over the logs/ folder and retrieves csv logs from runs containing given tags in config. Plot the results.


Logs

Hydra creates new output directory for every executed run.

Default logging structure:

├── logs
│   ├── task_name
│   │   ├── runs                        # Logs generated by single runs
│   │   │   ├── YYYY-MM-DD_HH-MM-SS       # Datetime of the run
│   │   │   │   ├── .hydra                  # Hydra logs
│   │   │   │   ├── csv                     # Csv logs
│   │   │   │   ├── wandb                   # Weights&Biases logs
│   │   │   │   ├── checkpoints             # Training checkpoints
│   │   │   │   └── ...                     # Any other thing saved during training
│   │   │   └── ...
│   │   │
│   │   └── multiruns                   # Logs generated by multiruns
│   │       ├── YYYY-MM-DD_HH-MM-SS       # Datetime of the multirun
│   │       │   ├──1                        # Multirun job number
│   │       │   ├──2
│   │       │   └── ...
│   │       └── ...
│   │
│   └── debugs                          # Logs generated when debugging config is attached
│       └── ...

You can change this structure by modifying paths in hydra configuration.


Experiment Tracking

PyTorch Lightning supports many popular logging frameworks: Weights&Biases, Neptune, Comet, MLFlow, Tensorboard.

These tools help you keep track of hyperparameters and output metrics and allow you to compare and visualize results. To use one of them simply complete its configuration in configs/logger and run:

python train.py logger=logger_name

You can use many of them at once (see configs/logger/many_loggers.yaml for example).

You can also write your own logger.

Lightning provides convenient method for logging custom metrics from inside LightningModule. Read the docs or take a look at MNIST example.


Tests

Template comes with generic tests implemented with pytest.

# run all tests
pytest

# run tests from specific file
pytest tests/test_train.py

# run all tests except the ones marked as slow
pytest -k "not slow"

Most of the implemented tests don't check for any specific output - they exist to simply verify that executing some commands doesn't end up in throwing exceptions. You can execute them once in a while to speed up the development.

Currently, the tests cover cases like:

  • running 1 train, val and test step
  • running 1 epoch on 1% of data, saving ckpt and resuming for the second epoch
  • running 2 epochs on 1% of data, with DDP simulated on CPU

And many others. You should be able to modify them easily for your use case.

There is also @RunIf decorator implemented, that allows you to run tests only if certain conditions are met, e.g. GPU is available or system is not windows. See the examples.


Hyperparameter Search

You can define hyperparameter search by adding new config file to configs/hparams_search.

Show example hyperparameter search config
# @package _global_

defaults:
  - override /hydra/sweeper: optuna

# choose metric which will be optimized by Optuna
# make sure this is the correct name of some metric logged in lightning module!
optimized_metric: "val/acc_best"

# here we define Optuna hyperparameter search
# it optimizes for value returned from function with @hydra.main decorator
hydra:
  sweeper:
    _target_: hydra_plugins.hydra_optuna_sweeper.optuna_sweeper.OptunaSweeper

    # 'minimize' or 'maximize' the objective
    direction: maximize

    # total number of runs that will be executed
    n_trials: 20

    # choose Optuna hyperparameter sampler
    # docs: https://optuna.readthedocs.io/en/stable/reference/samplers.html
    sampler:
      _target_: optuna.samplers.TPESampler
      seed: 1234
      n_startup_trials: 10 # number of random sampling runs before optimization starts

    # define hyperparameter search space
    params:
      model.optimizer.lr: interval(0.0001, 0.1)
      data.batch_size: choice(32, 64, 128, 256)
      model.net.lin1_size: choice(64, 128, 256)
      model.net.lin2_size: choice(64, 128, 256)
      model.net.lin3_size: choice(32, 64, 128, 256)

Next, execute it with: python train.py -m hparams_search=mnist_optuna

Using this approach doesn't require adding any boilerplate to code, everything is defined in a single config file. The only necessary thing is to return the optimized metric value from the launch file.

You can use different optimization frameworks integrated with Hydra, like Optuna, Ax or Nevergrad.

The optimization_results.yaml will be available under logs/task_name/multirun folder.

This approach doesn't support resuming interrupted search and advanced techniques like prunning - for more sophisticated search and workflows, you should probably write a dedicated optimization task (without multirun feature).


Continuous Integration

Template comes with CI workflows implemented in Github Actions:

  • .github/workflows/test.yaml: running all tests with pytest
  • .github/workflows/code-quality-main.yaml: running pre-commits on main branch for all files
  • .github/workflows/code-quality-pr.yaml: running pre-commits on pull requests for modified files only

Distributed Training

Lightning supports multiple ways of doing distributed training. The most common one is DDP, which spawns separate process for each GPU and averages gradients between them. To learn about other approaches read the lightning docs.

You can run DDP on mnist example with 4 GPUs like this:

python train.py trainer=ddp

Note: When using DDP you have to be careful how you write your models - read the docs.


Accessing Datamodule Attributes In Model

The simplest way is to pass datamodule attribute directly to model on initialization:

# ./src/train.py
datamodule = hydra.utils.instantiate(config.data)
model = hydra.utils.instantiate(config.model, some_param=datamodule.some_param)

Note: Not a very robust solution, since it assumes all your datamodules have some_param attribute available.

Similarly, you can pass a whole datamodule config as an init parameter:

# ./src/train.py
model = hydra.utils.instantiate(config.model, dm_conf=config.data, _recursive_=False)

You can also pass a datamodule config parameter to your model through variable interpolation:

# ./configs/model/my_model.yaml
_target_: src.models.my_module.MyLitModule
lr: 0.01
some_param: ${data.some_param}

Another approach is to access datamodule in LightningModule directly through Trainer:

# ./src/models/mnist_module.py
def on_train_start(self):
  self.some_param = self.trainer.datamodule.some_param

Note: This only works after the training starts since otherwise trainer won't be yet available in LightningModule.


Best Practices

Use Miniconda

It's usually unnecessary to install full anaconda environment, miniconda should be enough (weights around 80MB).

Big advantage of conda is that it allows for installing packages without requiring certain compilers or libraries to be available in the system (since it installs precompiled binaries), so it often makes it easier to install some dependencies e.g. cudatoolkit for GPU support.

It also allows you to access your environments globally which might be more convenient than creating new local environment for every project.

Example installation:

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh

Update conda:

conda update -n base -c defaults conda

Create new conda environment:

conda create -n myenv python=3.10
conda activate myenv
Use automatic code formatting

Use pre-commit hooks to standardize code formatting of your project and save mental energy.
Simply install pre-commit package with:

pip install pre-commit

Next, install hooks from .pre-commit-config.yaml:

pre-commit install

After that your code will be automatically reformatted on every new commit.

To reformat all files in the project use command:

pre-commit run -a

To update hook versions in .pre-commit-config.yaml use:

pre-commit autoupdate
Set private environment variables in .env file

System specific variables (e.g. absolute paths to datasets) should not be under version control or it will result in conflict between different users. Your private keys also shouldn't be versioned since you don't want them to be leaked.

Template contains .env.example file, which serves as an example. Create a new file called .env (this name is excluded from version control in .gitignore). You should use it for storing environment variables like this:

MY_VAR=/home/user/my_system_path

All variables from .env are loaded in train.py automatically.

Hydra allows you to reference any env variable in .yaml configs like this:

path_to_data: ${oc.env:MY_VAR}
Name metrics using '/' character

Depending on which logger you're using, it's often useful to define metric name with / character:

self.log("train/loss", loss)

This way loggers will treat your metrics as belonging to different sections, which helps to get them organised in UI.

Use torchmetrics

Use official torchmetrics library to ensure proper calculation of metrics. This is especially important for multi-GPU training!

For example, instead of calculating accuracy by yourself, you should use the provided Accuracy class like this:

from torchmetrics.classification.accuracy import Accuracy


class LitModel(LightningModule):
    def __init__(self)
        self.train_acc = Accuracy()
        self.val_acc = Accuracy()

    def training_step(self, batch, batch_idx):
        ...
        acc = self.train_acc(predictions, targets)
        self.log("train/acc", acc)
        ...

    def validation_step(self, batch, batch_idx):
        ...
        acc = self.val_acc(predictions, targets)
        self.log("val/acc", acc)
        ...

Make sure to use different metric instance for each step to ensure proper value reduction over all GPU processes.

Torchmetrics provides metrics for most use cases, like F1 score or confusion matrix. Read documentation for more.

Follow PyTorch Lightning style guide

The style guide is available here.

  1. Be explicit in your init. Try to define all the relevant defaults so that the user doesn’t have to guess. Provide type hints. This way your module is reusable across projects!

    class LitModel(LightningModule):
        def __init__(self, layer_size: int = 256, lr: float = 0.001):
  2. Preserve the recommended method order.

    class LitModel(LightningModule):
    
        def __init__():
            ...
    
        def forward():
            ...
    
        def training_step():
            ...
    
        def training_step_end():
            ...
    
        def on_train_epoch_end():
            ...
    
        def validation_step():
            ...
    
        def validation_step_end():
            ...
    
        def on_validation_epoch_end():
            ...
    
        def test_step():
            ...
    
        def test_step_end():
            ...
    
        def on_test_epoch_end():
            ...
    
        def configure_optimizers():
            ...
    
        def any_extra_hook():
            ...
Version control your data and models with DVC

Use DVC to version control big files, like your data or trained ML models.
To initialize the dvc repository:

dvc init

To start tracking a file or directory, use dvc add:

dvc add data/MNIST

DVC stores information about the added file (or a directory) in a special .dvc file named data/MNIST.dvc, a small text file with a human-readable format. This file can be easily versioned like source code with Git, as a placeholder for the original data:

git add data/MNIST.dvc data/.gitignore
git commit -m "Add raw data"
Support installing project as a package

It allows other people to easily use your modules in their own projects. Change name of the src folder to your project name and complete the setup.py file.

Now your project can be installed from local files:

pip install -e .

Or directly from git repository:

pip install git+git://github.com/YourGithubName/your-repo-name.git --upgrade

So any file can be easily imported into any other file like so:

from project_name.models.mnist_module import MNISTLitModule
from project_name.data.mnist_datamodule import MNISTDataModule
Keep local configs out of code versioning

Some configurations are user/machine/installation specific (e.g. configuration of local cluster, or harddrive paths on a specific machine). For such scenarios, a file configs/local/default.yaml can be created which is automatically loaded but not tracked by Git.

For example, you can use it for a SLURM cluster config:

# @package _global_

defaults:
  - override /hydra/launcher@_here_: submitit_slurm

data_dir: /mnt/scratch/data/

hydra:
  launcher:
    timeout_min: 1440
    gpus_per_task: 1
    gres: gpu:1
  job:
    env_set:
      MY_VAR: /home/user/my/system/path
      MY_KEY: asdgjhawi8y23ihsghsueity23ihwd

Resources

This template was inspired by:

Other useful repositories:


License

Lightning-Hydra-Template is licensed under the MIT License.

MIT License

Copyright (c) 2021 ashleve

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.