Skip to content

IPNDMScheduler

IPNDMScheduler is a fourth-order Improved Pseudo Linear Multistep scheduler. The original implementation can be found at crowsonkb/v-diffusion-pytorch.

mindone.diffusers.IPNDMScheduler

Bases: SchedulerMixin, ConfigMixin

A fourth-order Improved Pseudo Linear Multistep scheduler.

This model inherits from [SchedulerMixin] and [ConfigMixin]. Check the superclass documentation for the generic methods the library implements for all schedulers such as loading and saving.

PARAMETER DESCRIPTION
num_train_timesteps

The number of diffusion steps to train the model.

TYPE: `int`, defaults to 1000 DEFAULT: 1000

trained_betas

Pass an array of betas directly to the constructor to bypass beta_start and beta_end.

TYPE: `np.ndarray`, *optional* DEFAULT: None

Source code in mindone/diffusers/schedulers/scheduling_ipndm.py 
 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
class IPNDMScheduler(SchedulerMixin, ConfigMixin):
    """
    A fourth-order Improved Pseudo Linear Multistep scheduler.

    This model inherits from [`SchedulerMixin`] and [`ConfigMixin`]. Check the superclass documentation for the generic
    methods the library implements for all schedulers such as loading and saving.

    Args:
        num_train_timesteps (`int`, defaults to 1000):
            The number of diffusion steps to train the model.
        trained_betas (`np.ndarray`, *optional*):
            Pass an array of betas directly to the constructor to bypass `beta_start` and `beta_end`.
    """

    order = 1

    @register_to_config
    def __init__(self, num_train_timesteps: int = 1000, trained_betas: Optional[Union[np.ndarray, List[float]]] = None):
        # set `betas`, `alphas`, `timesteps`
        self.set_timesteps(num_train_timesteps)

        # standard deviation of the initial noise distribution
        self.init_noise_sigma = 1.0

        # For now we only support F-PNDM, i.e. the runge-kutta method
        # For more information on the algorithm please take a look at the paper: https://arxiv.org/pdf/2202.09778.pdf
        # mainly at formula (9), (12), (13) and the Algorithm 2.
        self.pndm_order = 4

        # running values
        self.ets = []
        self._step_index = None
        self._begin_index = None

    @property
    def step_index(self):
        """
        The index counter for current timestep. It will increase 1 after each scheduler step.
        """
        return self._step_index

    @property
    def begin_index(self):
        """
        The index for the first timestep. It should be set from pipeline with `set_begin_index` method.
        """
        return self._begin_index

    # Copied from diffusers.schedulers.scheduling_dpmsolver_multistep.DPMSolverMultistepScheduler.set_begin_index
    def set_begin_index(self, begin_index: int = 0):
        """
        Sets the begin index for the scheduler. This function should be run from pipeline before the inference.

        Args:
            begin_index (`int`):
                The begin index for the scheduler.
        """
        self._begin_index = begin_index

    def set_timesteps(self, num_inference_steps: int):
        """
        Sets the discrete timesteps used for the diffusion chain (to be run before inference).

        Args:
            num_inference_steps (`int`):
                The number of diffusion steps used when generating samples with a pre-trained model.
        """
        self.num_inference_steps = num_inference_steps
        steps = ms.tensor(np.linspace(1, 0, num_inference_steps + 1), dtype=ms.float32)[:-1]
        steps = ops.cat([steps, ms.tensor([0.0])])

        if self.config.trained_betas is not None:
            self.betas = ms.tensor(self.config.trained_betas, dtype=ms.float32)
        else:
            self.betas = ops.sin(steps * math.pi / 2) ** 2

        self.alphas = (1.0 - self.betas**2) ** 0.5

        timesteps = (ops.atan2(self.betas, self.alphas) / math.pi * 2)[:-1]
        self.timesteps = timesteps

        self.ets = []
        self._step_index = None
        self._begin_index = None

    # Copied from diffusers.schedulers.scheduling_euler_discrete.EulerDiscreteScheduler.index_for_timestep
    def index_for_timestep(self, timestep, schedule_timesteps=None):
        if schedule_timesteps is None:
            schedule_timesteps = self.timesteps

        if (schedule_timesteps == timestep).sum() > 1:
            pos = 1
        else:
            pos = 0

        # The sigma index that is taken for the **very** first `step`
        # is always the second index (or the last index if there is only 1)
        # This way we can ensure we don't accidentally skip a sigma in
        # case we start in the middle of the denoising schedule (e.g. for image-to-image)
        indices = (schedule_timesteps == timestep).nonzero()

        return int(indices[pos])

    # Copied from diffusers.schedulers.scheduling_euler_discrete.EulerDiscreteScheduler._init_step_index
    def _init_step_index(self, timestep):
        if self.begin_index is None:
            self._step_index = self.index_for_timestep(timestep)
        else:
            self._step_index = self._begin_index

    def step(
        self,
        model_output: ms.Tensor,
        timestep: Union[int, ms.Tensor],
        sample: ms.Tensor,
        return_dict: bool = False,
    ) -> Union[SchedulerOutput, Tuple]:
        """
        Predict the sample from the previous timestep by reversing the SDE. This function propagates the sample with
        the linear multistep method. It performs one forward pass multiple times to approximate the solution.

        Args:
            model_output (`ms.Tensor`):
                The direct output from learned diffusion model.
            timestep (`int`):
                The current discrete timestep in the diffusion chain.
            sample (`ms.Tensor`):
                A current instance of a sample created by the diffusion process.
            return_dict (`bool`):
                Whether or not to return a [`~schedulers.scheduling_utils.SchedulerOutput`] or tuple.

        Returns:
            [`~schedulers.scheduling_utils.SchedulerOutput`] or `tuple`:
                If return_dict is `True`, [`~schedulers.scheduling_utils.SchedulerOutput`] is returned, otherwise a
                tuple is returned where the first element is the sample tensor.
        """
        if self.num_inference_steps is None:
            raise ValueError(
                "Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler"
            )
        if self.step_index is None:
            self._init_step_index(timestep)

        timestep_index = self.step_index
        prev_timestep_index = self.step_index + 1

        ets = (sample * self.betas[timestep_index] + model_output * self.alphas[timestep_index]).to(sample.dtype)
        self.ets.append(ets)

        if len(self.ets) == 1:
            ets = self.ets[-1]
        elif len(self.ets) == 2:
            ets = (3 * self.ets[-1] - self.ets[-2]) / 2
        elif len(self.ets) == 3:
            ets = (23 * self.ets[-1] - 16 * self.ets[-2] + 5 * self.ets[-3]) / 12
        else:
            ets = (1 / 24) * (55 * self.ets[-1] - 59 * self.ets[-2] + 37 * self.ets[-3] - 9 * self.ets[-4])

        prev_sample = self._get_prev_sample(sample, timestep_index, prev_timestep_index, ets)

        # upon completion increase step index by one
        self._step_index += 1

        if not return_dict:
            return (prev_sample,)

        return SchedulerOutput(prev_sample=prev_sample)

    def scale_model_input(self, sample: ms.Tensor, *args, **kwargs) -> ms.Tensor:
        """
        Ensures interchangeability with schedulers that need to scale the denoising model input depending on the
        current timestep.

        Args:
            sample (`ms.Tensor`):
                The input sample.

        Returns:
            `ms.Tensor`:
                A scaled input sample.
        """
        return sample

    def _get_prev_sample(self, sample, timestep_index, prev_timestep_index, ets):
        alpha = self.alphas[timestep_index]
        sigma = self.betas[timestep_index]

        next_alpha = self.alphas[prev_timestep_index]
        next_sigma = self.betas[prev_timestep_index]

        pred = (sample - sigma * ets) / max(alpha, 1e-8)
        prev_sample = (next_alpha * pred + ets * next_sigma).to(sample.dtype)

        return prev_sample

    def __len__(self):
        return self.config.num_train_timesteps

mindone.diffusers.IPNDMScheduler.begin_index property

The index for the first timestep. It should be set from pipeline with set_begin_index method.

mindone.diffusers.IPNDMScheduler.step_index property

The index counter for current timestep. It will increase 1 after each scheduler step.

mindone.diffusers.IPNDMScheduler.scale_model_input(sample, *args, **kwargs)

Ensures interchangeability with schedulers that need to scale the denoising model input depending on the current timestep.

PARAMETER DESCRIPTION
sample

The input sample.

TYPE: `ms.Tensor`

RETURNS DESCRIPTION
Tensor

ms.Tensor: A scaled input sample.
Source code in mindone/diffusers/schedulers/scheduling_ipndm.py 
195
196
197
198
199
200
201
202
203
204
205
206
207
208
def scale_model_input(self, sample: ms.Tensor, *args, **kwargs) -> ms.Tensor:
    """
    Ensures interchangeability with schedulers that need to scale the denoising model input depending on the
    current timestep.

    Args:
        sample (`ms.Tensor`):
            The input sample.

    Returns:
        `ms.Tensor`:
            A scaled input sample.
    """
    return sample

mindone.diffusers.IPNDMScheduler.set_begin_index(begin_index=0)

Sets the begin index for the scheduler. This function should be run from pipeline before the inference.

PARAMETER DESCRIPTION
begin_index

The begin index for the scheduler.

TYPE: `int` DEFAULT: 0

Source code in mindone/diffusers/schedulers/scheduling_ipndm.py 
76
77
78
79
80
81
82
83
84
def set_begin_index(self, begin_index: int = 0):
    """
    Sets the begin index for the scheduler. This function should be run from pipeline before the inference.

    Args:
        begin_index (`int`):
            The begin index for the scheduler.
    """
    self._begin_index = begin_index

mindone.diffusers.IPNDMScheduler.set_timesteps(num_inference_steps)

Sets the discrete timesteps used for the diffusion chain (to be run before inference).

PARAMETER DESCRIPTION
num_inference_steps

The number of diffusion steps used when generating samples with a pre-trained model.

TYPE: `int`

Source code in mindone/diffusers/schedulers/scheduling_ipndm.py 
 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
def set_timesteps(self, num_inference_steps: int):
    """
    Sets the discrete timesteps used for the diffusion chain (to be run before inference).

    Args:
        num_inference_steps (`int`):
            The number of diffusion steps used when generating samples with a pre-trained model.
    """
    self.num_inference_steps = num_inference_steps
    steps = ms.tensor(np.linspace(1, 0, num_inference_steps + 1), dtype=ms.float32)[:-1]
    steps = ops.cat([steps, ms.tensor([0.0])])

    if self.config.trained_betas is not None:
        self.betas = ms.tensor(self.config.trained_betas, dtype=ms.float32)
    else:
        self.betas = ops.sin(steps * math.pi / 2) ** 2

    self.alphas = (1.0 - self.betas**2) ** 0.5

    timesteps = (ops.atan2(self.betas, self.alphas) / math.pi * 2)[:-1]
    self.timesteps = timesteps

    self.ets = []
    self._step_index = None
    self._begin_index = None

mindone.diffusers.IPNDMScheduler.step(model_output, timestep, sample, return_dict=False)

Predict the sample from the previous timestep by reversing the SDE. This function propagates the sample with the linear multistep method. It performs one forward pass multiple times to approximate the solution.

PARAMETER DESCRIPTION
model_output

The direct output from learned diffusion model.

TYPE: `ms.Tensor`

timestep

The current discrete timestep in the diffusion chain.

TYPE: `int`

sample

A current instance of a sample created by the diffusion process.

TYPE: `ms.Tensor`

return_dict

Whether or not to return a [~schedulers.scheduling_utils.SchedulerOutput] or tuple.

TYPE: `bool` DEFAULT: False

RETURNS DESCRIPTION
Union[SchedulerOutput, Tuple]

[~schedulers.scheduling_utils.SchedulerOutput] or tuple: If return_dict is True, [~schedulers.scheduling_utils.SchedulerOutput] is returned, otherwise a tuple is returned where the first element is the sample tensor.
Source code in mindone/diffusers/schedulers/scheduling_ipndm.py 
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
def step(
    self,
    model_output: ms.Tensor,
    timestep: Union[int, ms.Tensor],
    sample: ms.Tensor,
    return_dict: bool = False,
) -> Union[SchedulerOutput, Tuple]:
    """
    Predict the sample from the previous timestep by reversing the SDE. This function propagates the sample with
    the linear multistep method. It performs one forward pass multiple times to approximate the solution.

    Args:
        model_output (`ms.Tensor`):
            The direct output from learned diffusion model.
        timestep (`int`):
            The current discrete timestep in the diffusion chain.
        sample (`ms.Tensor`):
            A current instance of a sample created by the diffusion process.
        return_dict (`bool`):
            Whether or not to return a [`~schedulers.scheduling_utils.SchedulerOutput`] or tuple.

    Returns:
        [`~schedulers.scheduling_utils.SchedulerOutput`] or `tuple`:
            If return_dict is `True`, [`~schedulers.scheduling_utils.SchedulerOutput`] is returned, otherwise a
            tuple is returned where the first element is the sample tensor.
    """
    if self.num_inference_steps is None:
        raise ValueError(
            "Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler"
        )
    if self.step_index is None:
        self._init_step_index(timestep)

    timestep_index = self.step_index
    prev_timestep_index = self.step_index + 1

    ets = (sample * self.betas[timestep_index] + model_output * self.alphas[timestep_index]).to(sample.dtype)
    self.ets.append(ets)

    if len(self.ets) == 1:
        ets = self.ets[-1]
    elif len(self.ets) == 2:
        ets = (3 * self.ets[-1] - self.ets[-2]) / 2
    elif len(self.ets) == 3:
        ets = (23 * self.ets[-1] - 16 * self.ets[-2] + 5 * self.ets[-3]) / 12
    else:
        ets = (1 / 24) * (55 * self.ets[-1] - 59 * self.ets[-2] + 37 * self.ets[-3] - 9 * self.ets[-4])

    prev_sample = self._get_prev_sample(sample, timestep_index, prev_timestep_index, ets)

    # upon completion increase step index by one
    self._step_index += 1

    if not return_dict:
        return (prev_sample,)

    return SchedulerOutput(prev_sample=prev_sample)