File size: 9,872 Bytes
2ba4412
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
import math
import torch
from torch.optim import Optimizer
from torch.optim.lr_scheduler import LambdaLR

__all__ = ['Adafactor']

class Adafactor(Optimizer):
    """
    AdaFactor pytorch implementation can be used as a drop in replacement for Adam original fairseq code:
    https://github.com/pytorch/fairseq/blob/master/fairseq/optim/adafactor.py
    Paper: *Adafactor: Adaptive Learning Rates with Sublinear Memory Cost* https://arxiv.org/abs/1804.04235 Note that
    this optimizer internally adjusts the learning rate depending on the `scale_parameter`, `relative_step` and
    `warmup_init` options. To use a manual (external) learning rate schedule you should set `scale_parameter=False` and
    `relative_step=False`.
    Arguments:
        params (`Iterable[nn.parameter.Parameter]`):
            Iterable of parameters to optimize or dictionaries defining parameter groups.
        lr (`float`, *optional*):
            The external learning rate.
        eps (`Tuple[float, float]`, *optional*, defaults to (1e-30, 1e-3)):
            Regularization constants for square gradient and parameter scale respectively
        clip_threshold (`float`, *optional*, defaults 1.0):
            Threshold of root mean square of final gradient update
        decay_rate (`float`, *optional*, defaults to -0.8):
            Coefficient used to compute running averages of square
        beta1 (`float`, *optional*):
            Coefficient used for computing running averages of gradient
        weight_decay (`float`, *optional*, defaults to 0):
            Weight decay (L2 penalty)
        scale_parameter (`bool`, *optional*, defaults to `True`):
            If True, learning rate is scaled by root mean square
        relative_step (`bool`, *optional*, defaults to `True`):
            If True, time-dependent learning rate is computed instead of external learning rate
        warmup_init (`bool`, *optional*, defaults to `False`):
            Time-dependent learning rate computation depends on whether warm-up initialization is being used
    This implementation handles low-precision (FP16, bfloat) values, but we have not thoroughly tested.
    Recommended T5 finetuning settings (https://discuss.huggingface.co/t/t5-finetuning-tips/684/3):
        - Training without LR warmup or clip_threshold is not recommended.
           - use scheduled LR warm-up to fixed LR
           - use clip_threshold=1.0 (https://arxiv.org/abs/1804.04235)
        - Disable relative updates
        - Use scale_parameter=False
        - Additional optimizer operations like gradient clipping should not be used alongside Adafactor
    Example:
    ```python
    Adafactor(model.parameters(), scale_parameter=False, relative_step=False, warmup_init=False, lr=1e-3)
    ```
    Others reported the following combination to work well:
    ```python
    Adafactor(model.parameters(), scale_parameter=True, relative_step=True, warmup_init=True, lr=None)
    ```
    When using `lr=None` with [`Trainer`] you will most likely need to use [`~optimization.AdafactorSchedule`]
    scheduler as following:
    ```python
    from transformers.optimization import Adafactor, AdafactorSchedule
    optimizer = Adafactor(model.parameters(), scale_parameter=True, relative_step=True, warmup_init=True, lr=None)
    lr_scheduler = AdafactorSchedule(optimizer)
    trainer = Trainer(..., optimizers=(optimizer, lr_scheduler))
    ```
    Usage:
    ```python
    # replace AdamW with Adafactor
    optimizer = Adafactor(
        model.parameters(),
        lr=1e-3,
        eps=(1e-30, 1e-3),
        clip_threshold=1.0,
        decay_rate=-0.8,
        beta1=None,
        weight_decay=0.0,
        relative_step=False,
        scale_parameter=False,
        warmup_init=False,
    )
    ```"""

    def __init__(
        self,
        params,
        lr=None,
        eps=(1e-30, 1e-3),
        clip_threshold=1.0,
        decay_rate=-0.8,
        beta1=None,
        weight_decay=0.0,
        scale_parameter=True,
        relative_step=True,
        warmup_init=False,
    ):
        r"""require_version("torch>=1.5.0")  # add_ with alpha
        """
        if lr is not None and relative_step:
            raise ValueError("Cannot combine manual `lr` and `relative_step=True` options")
        if warmup_init and not relative_step:
            raise ValueError("`warmup_init=True` requires `relative_step=True`")

        defaults = dict(
            lr=lr,
            eps=eps,
            clip_threshold=clip_threshold,
            decay_rate=decay_rate,
            beta1=beta1,
            weight_decay=weight_decay,
            scale_parameter=scale_parameter,
            relative_step=relative_step,
            warmup_init=warmup_init,
        )
        super().__init__(params, defaults)

    @staticmethod
    def _get_lr(param_group, param_state):
        rel_step_sz = param_group["lr"]
        if param_group["relative_step"]:
            min_step = 1e-6 * param_state["step"] if param_group["warmup_init"] else 1e-2
            rel_step_sz = min(min_step, 1.0 / math.sqrt(param_state["step"]))
        param_scale = 1.0
        if param_group["scale_parameter"]:
            param_scale = max(param_group["eps"][1], param_state["RMS"])
        return param_scale * rel_step_sz

    @staticmethod
    def _get_options(param_group, param_shape):
        factored = len(param_shape) >= 2
        use_first_moment = param_group["beta1"] is not None
        return factored, use_first_moment

    @staticmethod
    def _rms(tensor):
        return tensor.norm(2) / (tensor.numel() ** 0.5)

    @staticmethod
    def _approx_sq_grad(exp_avg_sq_row, exp_avg_sq_col):
        # copy from fairseq's adafactor implementation:
        # https://github.com/huggingface/transformers/blob/8395f14de6068012787d83989c3627c3df6a252b/src/transformers/optimization.py#L505
        r_factor = (exp_avg_sq_row / exp_avg_sq_row.mean(dim=-1, keepdim=True)).rsqrt_().unsqueeze(-1)
        c_factor = exp_avg_sq_col.unsqueeze(-2).rsqrt()
        return torch.mul(r_factor, c_factor)

    def step(self, closure=None):
        """
        Performs a single optimization step
        Arguments:
            closure (callable, optional): A closure that reevaluates the model
                and returns the loss.
        """
        loss = None
        if closure is not None:
            loss = closure()

        for group in self.param_groups:
            for p in group["params"]:
                if p.grad is None:
                    continue
                grad = p.grad.data
                if grad.dtype in {torch.float16, torch.bfloat16}:
                    grad = grad.float()
                if grad.is_sparse:
                    raise RuntimeError("Adafactor does not support sparse gradients.")

                state = self.state[p]
                grad_shape = grad.shape

                factored, use_first_moment = self._get_options(group, grad_shape)
                # State Initialization
                if len(state) == 0:
                    state["step"] = 0

                    if use_first_moment:
                        # Exponential moving average of gradient values
                        state["exp_avg"] = torch.zeros_like(grad)
                    if factored:
                        state["exp_avg_sq_row"] = torch.zeros(grad_shape[:-1]).to(grad)
                        state["exp_avg_sq_col"] = torch.zeros(grad_shape[:-2] + grad_shape[-1:]).to(grad)
                    else:
                        state["exp_avg_sq"] = torch.zeros_like(grad)

                    state["RMS"] = 0
                else:
                    if use_first_moment:
                        state["exp_avg"] = state["exp_avg"].to(grad)
                    if factored:
                        state["exp_avg_sq_row"] = state["exp_avg_sq_row"].to(grad)
                        state["exp_avg_sq_col"] = state["exp_avg_sq_col"].to(grad)
                    else:
                        state["exp_avg_sq"] = state["exp_avg_sq"].to(grad)

                p_data_fp32 = p.data
                if p.data.dtype in {torch.float16, torch.bfloat16}:
                    p_data_fp32 = p_data_fp32.float()

                state["step"] += 1
                state["RMS"] = self._rms(p_data_fp32)
                lr = self._get_lr(group, state)

                beta2t = 1.0 - math.pow(state["step"], group["decay_rate"])
                update = (grad**2) + group["eps"][0]
                if factored:
                    exp_avg_sq_row = state["exp_avg_sq_row"]
                    exp_avg_sq_col = state["exp_avg_sq_col"]

                    exp_avg_sq_row.mul_(beta2t).add_(update.mean(dim=-1), alpha=(1.0 - beta2t))
                    exp_avg_sq_col.mul_(beta2t).add_(update.mean(dim=-2), alpha=(1.0 - beta2t))

                    # Approximation of exponential moving average of square of gradient
                    update = self._approx_sq_grad(exp_avg_sq_row, exp_avg_sq_col)
                    update.mul_(grad)
                else:
                    exp_avg_sq = state["exp_avg_sq"]

                    exp_avg_sq.mul_(beta2t).add_(update, alpha=(1.0 - beta2t))
                    update = exp_avg_sq.rsqrt().mul_(grad)

                update.div_((self._rms(update) / group["clip_threshold"]).clamp_(min=1.0))
                update.mul_(lr)

                if use_first_moment:
                    exp_avg = state["exp_avg"]
                    exp_avg.mul_(group["beta1"]).add_(update, alpha=(1 - group["beta1"]))
                    update = exp_avg

                if group["weight_decay"] != 0:
                    p_data_fp32.add_(p_data_fp32, alpha=(-group["weight_decay"] * lr))

                p_data_fp32.add_(-update)

                if p.data.dtype in {torch.float16, torch.bfloat16}:
                    p.data.copy_(p_data_fp32)

        return loss