torch.optim.adagrad 的源代码

```html

import torch
from torch import Tensor

from .optimizer import (Optimizer, _use_grad_for_differentiable, _get_value, _view_as_real,
                        _default_to_fused_or_foreach, _get_scalar_dtype, _differentiable_doc,
                        _foreach_doc, _maximize_doc)
from typing import List, Optional

__all__ = ["Adagrad", "adagrad"]


class Adagrad(Optimizer):
    def __init__(
        self,
        params,
        lr=1e-2,
        lr_decay=0,
        weight_decay=0,
        initial_accumulator_value=0,
        eps=1e-10,
        foreach: Optional[bool] = None,
        *,
        maximize: bool = False,
        differentiable: bool = False,
    ):
        if not 0.0 <= lr:
            raise ValueError(f"Invalid learning rate: {lr}")
        if not 0.0 <= lr_decay:
            raise ValueError(f"Invalid lr_decay value: {lr_decay}")
        if not 0.0 <= weight_decay:
            raise ValueError(f"Invalid weight_decay value: {weight_decay}")
        if not 0.0 <= initial_accumulator_value:
            raise ValueError(
                f"Invalid initial_accumulator_value value: {initial_accumulator_value}"
            )
        if not 0.0 <= eps:
            raise ValueError(f"Invalid epsilon value: {eps}")

        defaults = dict(
            lr=lr,
            lr_decay=lr_decay,
            eps=eps,
            weight_decay=weight_decay,
            initial_accumulator_value=initial_accumulator_value,
            foreach=foreach,
            maximize=maximize,
            differentiable=differentiable,
        )
        super().__init__(params, defaults)

        for group in self.param_groups:
            for p in group["params"]:
                state = self.state[p]
                state["step"] = torch.tensor(0.0, dtype=_get_scalar_dtype())
                init_value = (
                    complex(initial_accumulator_value, initial_accumulator_value)
                    if torch.is_complex(p)
                    else initial_accumulator_value
                )
                state["sum"] = torch.full_like(
                    p, init_value, memory_format=torch.preserve_format
                )

    def __setstate__(self, state):
        super().__setstate__(state)
        for group in self.param_groups:
            group.setdefault("foreach", None)
            group.setdefault("maximize", False)
            group.setdefault("differentiable", False)

        state_values = list(self.state.values())
        step_is_tensor = (len(state_values) != 0) and torch.is_tensor(
            state_values[0]["step"]
        )
        if not step_is_tensor:
            for s in state_values:
                s["step"] = torch.tensor(float(s["step"]), dtype=_get_scalar_dtype())

    def share_memory(self):
        for group in self.param_groups:
            for p in group<span class="