# Copyright (c) OpenMMLab. All rights reserved.
from typing import List

import torch

from mmengine.registry import PARAM_SCHEDULERS
from .param_scheduler import (INF, ConstantParamScheduler,
                              CosineAnnealingParamScheduler,
                              ExponentialParamScheduler, LinearParamScheduler,
                              MultiStepParamScheduler, PolyParamScheduler,
                              StepParamScheduler)


@PARAM_SCHEDULERS.register_module()
class ConstantLR(ConstantParamScheduler):
    """Decays the learning rate value of each parameter group by a small
    constant factor until the number of epoch reaches a pre-defined milestone:
    ``end``. Notice that such decay can happen simultaneously with other
    changes to the learning rate value from outside this scheduler.

    Args:
        optimizer (Optimizer): Wrapped optimizer.
        factor (float): The number we multiply learning rate until the
            milestone. Defaults to 1./3.
        begin (int): Step at which to start updating the learning rate.
            Defaults to 0.
        end (int): Step at which to stop updating the learning rate.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without state
            dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled learning rate is updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the learning rate for each update.
            Defaults to False.
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 factor: float = 1.0 / 3,
                 begin: int = 0,
                 end: int = INF,
                 last_step: int = -1,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            factor=factor,
            begin=begin,
            end=end,
            last_step=last_step,
            by_epoch=by_epoch,
            verbose=verbose)


@PARAM_SCHEDULERS.register_module()
class CosineAnnealingLR(CosineAnnealingParamScheduler):
    r"""Set the learning rate of each parameter group using a cosine annealing
    schedule, where :math:`\eta_{max}` is set to the initial value and
    :math:`T_{cur}` is the number of epochs since the last restart in SGDR:

    .. math::
        \begin{aligned}
            \eta_t & = \eta_{min} + \frac{1}{2}(\eta_{max} - \eta_{min})\left(1
            + \cos\left(\frac{T_{cur}}{T_{max}}\pi\right)\right),
            & T_{cur} \neq (2k+1)T_{max}; \\
            \eta_{t+1} & = \eta_{t} + \frac{1}{2}(\eta_{max} - \eta_{min})
            \left(1 - \cos\left(\frac{1}{T_{max}}\pi\right)\right),
            & T_{cur} = (2k+1)T_{max}.
        \end{aligned}

    Notice that because the schedule
    is defined recursively, the learning rate can be simultaneously modified
    outside this scheduler by other operators. If the learning rate is set
    solely by this scheduler, the learning rate at each step becomes:

    .. math::
        \eta_t = \eta_{min} + \frac{1}{2}(\eta_{max} - \eta_{min})\left(1 +
        \cos\left(\frac{T_{cur}}{T_{max}}\pi\right)\right)

    It has been proposed in
    `SGDR: Stochastic Gradient Descent with Warm Restarts`_. Note that this
    only implements the cosine annealing part of SGDR, and not the restarts.

    Args:
        optimizer (Optimizer): Wrapped optimizer.
        T_max (int): Maximum number of iterations.
        eta_min (float): Minimum learning rate. Defaults to 0.
        begin (int): Step at which to start updating the learning rate.
            Defaults to 0.
        end (int): Step at which to stop updating the learning rate.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without
            state dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled learning rate is updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the learning rate for each update.
            Defaults to False.

    .. _SGDR\: Stochastic Gradient Descent with Warm Restarts:
        https://arxiv.org/abs/1608.03983
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 T_max: int,
                 eta_min: int = 0,
                 begin: int = 0,
                 end: int = INF,
                 last_step: int = -1,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            T_max=T_max,
            eta_min=eta_min,
            begin=begin,
            end=end,
            last_step=last_step,
            by_epoch=by_epoch,
            verbose=verbose)


@PARAM_SCHEDULERS.register_module()
class ExponentialLR(ExponentialParamScheduler):
    """Decays the learning rate of each parameter group by gamma every epoch.

    Args:
        optimizer (Optimizer): Wrapped optimizer.
        gamma (float): Multiplicative factor of learning rate decay.
        begin (int): Step at which to start updating the learning rate.
            Defaults to 0.
        end (int): Step at which to stop updating the learning rate.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without
            state dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled learning rate is updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the learning rate for each update.
            Defaults to False.
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 gamma: float,
                 begin: int = 0,
                 end: int = INF,
                 last_step: int = -1,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            gamma=gamma,
            begin=begin,
            end=end,
            last_step=last_step,
            by_epoch=by_epoch,
            verbose=verbose)


@PARAM_SCHEDULERS.register_module()
class LinearLR(LinearParamScheduler):
    """Decays the learning rate of each parameter group by linearly changing
    small multiplicative factor until the number of epoch reaches a pre-defined
    milestone: ``end``.

    Notice that such decay can happen simultaneously with other changes to the
    learning rate from outside this scheduler.
    Args:
        optimizer (Optimizer): Wrapped optimizer.
        start_factor (float): The number we multiply learning rate in the
            first epoch. The multiplication factor changes towards end_factor
            in the following epochs. Defaults to 1./3.
        end_factor (float): The number we multiply learning rate at the end
            of linear changing process. Defaults to 1.0.
        begin (int): Step at which to start updating the learning rate.
            Defaults to 0.
        end (int): Step at which to stop updating the learning rate.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without
            state dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled learning rate is updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the learning rate for each update.
            Defaults to False.
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 start_factor: float = 1.0 / 3,
                 end_factor: float = 1.0,
                 begin: int = 0,
                 end: int = INF,
                 last_step: int = -1,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            start_factor=start_factor,
            end_factor=end_factor,
            begin=begin,
            end=end,
            last_step=last_step,
            by_epoch=by_epoch,
            verbose=verbose)


@PARAM_SCHEDULERS.register_module()
class MultiStepLR(MultiStepParamScheduler):
    """Decays the specified learning rate in each parameter group by gamma once
    the number of epoch reaches one of the milestones. Notice that such decay
    can happen simultaneously with other changes to the learning rate from
    outside this scheduler.

    Args:
        optimizer (Optimizer): Wrapped optimizer.
        milestones (list): List of epoch indices. Must be increasing.
        gamma (float): Multiplicative factor of learning rate decay.
            Defaults to 0.1.
        begin (int): Step at which to start updating the learning rate.
            Defaults to 0.
        end (int): Step at which to stop updating the learning rate.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without
            state dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled learning rate is updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the learning rate for each update.
            Defaults to False.
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 milestones: List[int],
                 gamma: float = 0.1,
                 last_step: int = -1,
                 begin: int = 0,
                 end: int = INF,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            milestones=milestones,
            gamma=gamma,
            last_step=last_step,
            begin=begin,
            end=end,
            by_epoch=by_epoch,
            verbose=verbose)


@PARAM_SCHEDULERS.register_module()
class StepLR(StepParamScheduler):
    """Decays the learning rate of each parameter group by gamma every
    step_size epochs. Notice that such decay can happen simultaneously with
    other changes to the learning rate from outside this scheduler.

    Args:
        optimizer (Optimizer): Wrapped optimizer.
        step_size (int): Period of learning rate decay.
        gamma (float): Multiplicative factor of learning rate decay.
            Defaults to 0.1.
        begin (int): Step at which to start updating the learning rate.
            Defaults to 0.
        end (int): Step at which to stop updating the learning rate.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without
            state dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled learning rate is updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the learning rate for each update.
            Defaults to False.
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 step_size: int,
                 gamma: float = 0.1,
                 begin: int = 0,
                 end: int = INF,
                 last_step: int = -1,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            step_size=step_size,
            gamma=gamma,
            begin=begin,
            end=end,
            last_step=last_step,
            by_epoch=by_epoch,
            verbose=verbose)


@PARAM_SCHEDULERS.register_module()
class PolyLR(PolyParamScheduler):
    """Decays the learning rate of each parameter group in a polynomial decay
    scheme.

    Notice that such decay can happen simultaneously with other changes to the
    parameter value from outside this scheduler.

    Args:
        optimizer (Optimizer): Wrapped optimizer.
        eta_min (float): Minimum learning rate at the end of scheduling.
            Defaults to 0.
        power (float): The power of the polynomial. Defaults to 1.0.
        begin (int): Step at which to start updating the parameters.
            Defaults to 0.
        end (int): Step at which to stop updating the parameters.
            Defaults to INF.
        last_step (int): The index of last step. Used for resume without
            state dict. Defaults to -1.
        by_epoch (bool): Whether the scheduled parameters are updated by
            epochs. Defaults to True.
        verbose (bool): Whether to print the value for each update.
            Defaults to False.
    """

    def __init__(self,
                 optimizer: torch.optim.Optimizer,
                 eta_min: float = 0,
                 power: float = 1,
                 begin: int = 0,
                 end: int = INF,
                 last_step: int = -1,
                 by_epoch: bool = True,
                 verbose: bool = False):
        super().__init__(
            optimizer,
            param_name='lr',
            eta_min=eta_min,
            power=power,
            begin=begin,
            end=end,
            last_step=last_step,
            by_epoch=by_epoch,
            verbose=verbose)