Skip to content

Commit

Permalink
【PPSCI Doc No.75, 77-85】 (PaddlePaddle#703)
Browse files Browse the repository at this point in the history 
* update docs

* update func

* Update ppsci/loss/func.py

Co-authored-by: HydrogenSulfate <490868991@qq.com>

---------

Co-authored-by: HydrogenSulfate <490868991@qq.com>
  • Loading branch information
@NKNaN @HydrogenSulfate
NKNaN and HydrogenSulfate committed Dec 21, 2023
1 parent fb1a380 commit 9a50799
Show file tree
Hide file tree
Showing 6 changed files with 249 additions and 33 deletions.
44 changes: 27 additions & 17 deletions ppsci/loss/func.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -19,44 +19,54 @@
from typing import Optional
from typing import Union

from typing_extensions import Literal

from ppsci.loss import base


class FunctionalLoss(base.Loss):
r"""Functional loss class, which allows to use custom loss computing function from given loss_expr for complex computation cases.
$$
L = f(\mathbf{x}, \mathbf{y})
$$
$$
\mathbf{x}, \mathbf{y} \in \mathcal{R}^{N}
$$
Args:
loss_expr (Callable): expression of loss calculation.
reduction (Literal["mean", "sum"], optional): Reduction method. Defaults to "mean".
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> import paddle
>>> from ppsci.loss import FunctionalLoss
>>> import paddle.nn.functional as F
>>> def loss_expr(output_dict, *args):
>>> def mse_sum_loss(output_dict, label_dict, weight_dict=None):
... losses = 0
... for key in output_dict:
... length = int(len(output_dict[key])/2)
... out_dict = {key: output_dict[key][:length]}
... label_dict = {key: output_dict[key][length:]}
... losses += F.mse_loss(out_dict, label_dict, "sum")
... for key in output_dict.keys():
... loss = F.mse_loss(output_dict[key], label_dict[key], "sum")
... if weight_dict:
... loss *= weight_dict[key]
... losses += loss
... return losses
>>> loss = ppsci.loss.FunctionalLoss(loss_expr)
>>> loss = FunctionalLoss(mse_sum_loss)
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight_dict = {'u': 0.8, 'v': 0.2}
>>> result = loss(output_dict, label_dict, weight_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
17.89600182)
"""

def __init__(
self,
loss_expr: Callable,
reduction: Literal["mean", "sum"] = "mean",
weight: Optional[Union[float, Dict[str, float]]] = None,
):
if reduction not in ["mean", "sum"]:
raise ValueError(
f"reduction should be 'mean' or 'sum', but got {reduction}"
)
super().__init__(reduction, weight)
super().__init__(None, weight)
self.loss_expr = loss_expr

def forward(self, output_dict, label_dict=None, weight_dict=None):
Expand Down
22 changes: 20 additions & 2 deletions ppsci/loss/integral.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -44,8 +44,26 @@ class IntegralLoss(base.Loss):
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.IntegralLoss("mean")
>>> import paddle
>>> from ppsci.loss import IntegralLoss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 2.2, 0.9], [1.1, 0.8, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 2.2, 0.9], [1.1, 0.8, -1.3]]),
... 'area': paddle.to_tensor([[0.01, 0.02, 0.03], [0.01, 0.02, 0.03]])}
>>> label_dict = {'u': paddle.to_tensor([-1.8, 0.0]),
... 'v': paddle.to_tensor([0.1, 0.1])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = IntegralLoss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
1.40911996)
>>> loss = IntegralLoss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
2.81823993)
"""

def __init__(
Expand Down
33 changes: 31 additions & 2 deletions ppsci/loss/l1.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -119,13 +119,42 @@ class PeriodicL1Loss(base.Loss):
$\mathbf{x_l} \in \mathcal{R}^{N}$ is the first half of batch output,
$\mathbf{x_r} \in \mathcal{R}^{N}$ is the second half of batch output.
when `reduction` is set to "mean"
$$
L = MEAN \left( \Vert \mathbf{x_l}-\mathbf{x_r} \Vert_1 \right)
$$
when `reduction` is set to "sum"
$$
L = SUM \left( \Vert \mathbf{x_l}-\mathbf{x_r} \Vert_1 \right)
$$
Args:
reduction (Literal["mean", "sum"], optional): Reduction method. Defaults to "mean".
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.PeriodicL1Loss("mean")
>>> import paddle
>>> from ppsci.loss import PeriodicL1Loss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 2.2, 0.9], [1.1, 0.8, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 2.2, 0.9], [1.1, 0.8, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 0.0, 1.0], [-0.2, 0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1, 0.1], [0.1, 0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = PeriodicL1Loss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
4.19999981)
>>> loss = PeriodicL1Loss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
4.19999981)
"""

def __init__(
Expand Down
97 changes: 91 additions & 6 deletions ppsci/loss/l2.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -36,13 +36,40 @@ class L2Loss(base.Loss):
\mathbf{x}, \mathbf{y} \in \mathcal{R}^{N}
$$
when `reduction` is set to "mean"
$$
L = MEAN \left( \Vert \mathbf{x} - \mathbf{y} \Vert_2 \right)
$$
when `reduction` is set to "sum"
$$
L = SUM \left( \Vert \mathbf{x} - \mathbf{y} \Vert_2 \right)
$$
Args:
reduction (Literal["mean", "sum"], optional): Reduction method. Defaults to "mean".
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.L2Loss()
>>> import paddle
>>> from ppsci.loss import L2Loss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = L2Loss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
2.78884506)
>>> loss = L2Loss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
5.57769012)
"""

def __init__(
Expand Down Expand Up @@ -92,13 +119,42 @@ class PeriodicL2Loss(base.Loss):
$\mathbf{x_l} \in \mathcal{R}^{N}$ is the first half of batch output,
$\mathbf{x_r} \in \mathcal{R}^{N}$ is the second half of batch output.
when `reduction` is set to "mean"
$$
L = MEAN \left( \Vert \mathbf{x_l}-\mathbf{x_r} \Vert_2 \right)
$$
when `reduction` is set to "sum"
$$
L = SUM \left( \Vert \mathbf{x_l}-\mathbf{x_r} \Vert_2 \right)
$$
Args:
reduction (Literal["mean", "sum"], optional): Reduction method. Defaults to "mean".
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.PeriodicL2Loss()
>>> import paddle
>>> from ppsci.loss import PeriodicL2Loss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 2.2, 0.9], [1.1, 0.8, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 2.2, 0.9], [1.1, 0.8, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 0.0, 1.0], [-0.2, 0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1, 0.1], [0.1, 0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = PeriodicL2Loss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
2.67581749)
>>> loss = PeriodicL2Loss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
2.67581749)
"""

def __init__(
Expand Down Expand Up @@ -158,13 +214,42 @@ class L2RelLoss(base.Loss):
\mathbf{x}, \mathbf{y} \in \mathcal{R}^{N}
$$
when `reduction` is set to "mean"
$$
L = MEAN \left( \dfrac{\Vert \mathbf{x} - \mathbf{y} \Vert_2}{\Vert \mathbf{y} \Vert_2} \right)
$$
when `reduction` is set to "sum"
$$
L = SUM \left( \dfrac{\Vert \mathbf{x} - \mathbf{y} \Vert_2}{\Vert \mathbf{y} \Vert_2} \right)
$$
Args:
reduction (Literal["mean", "sum"], optional): Specifies the reduction to apply to the output: 'mean' | 'sum'. Defaults to "mean".
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.L2RelLoss()
>>> import paddle
>>> from ppsci.loss import L2RelLoss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = L2RelLoss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
2.93676996)
>>> loss = L2RelLoss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
5.87353992)
"""

def __init__(
Expand Down
21 changes: 19 additions & 2 deletions ppsci/loss/mae.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -44,8 +44,25 @@ class MAELoss(base.Loss):
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.MAELoss("mean")
>>> import paddle
>>> from ppsci.loss import MAELoss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = MAELoss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
1.67999995)
>>> loss = MAELoss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
6.71999979)
"""

def __init__(
Expand Down
65 changes: 61 additions & 4 deletions ppsci/loss/mse.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -44,8 +44,25 @@ class MSELoss(base.Loss):
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.MSELoss("mean")
>>> import paddle
>>> from ppsci.loss import MSELoss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = MSELoss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
4.47400045)
>>> loss = MSELoss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
17.89600182)
"""

def __init__(
Expand Down Expand Up @@ -108,8 +125,27 @@ class MSELossWithL2Decay(MSELoss):
ValueError: reduction should be 'mean' or 'sum'.
Examples:
>>> import ppsci
>>> loss = ppsci.loss.MSELossWithL2Decay("mean", {"k_matrix": 2.0})
>>> import paddle
>>> from ppsci.loss import MSELossWithL2Decay
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> regularization_dict = {'u': 2.0}
>>> loss = MSELossWithL2Decay(regularization_dict=regularization_dict, weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
12.39400005)
>>> regularization_dict = {'v': 1.0}
>>> loss = MSELossWithL2Decay(reduction="sum", regularization_dict=regularization_dict, weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
21.85600090)
"""

def __init__(
Expand Down Expand Up @@ -152,6 +188,27 @@ class PeriodicMSELoss(base.Loss):
Args:
reduction (Literal["mean", "sum"], optional): Reduction method. Defaults to "mean".
weight (Optional[Union[float, Dict[str, float]]]): Weight for loss. Defaults to None.
Examples:
>>> import paddle
>>> from ppsci.loss import PeriodicMSELoss
>>> output_dict = {'u': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]]),
... 'v': paddle.to_tensor([[0.5, 0.9], [1.1, -1.3]])}
>>> label_dict = {'u': paddle.to_tensor([[-1.8, 1.0], [-0.2, 2.5]]),
... 'v': paddle.to_tensor([[0.1, 0.1], [0.1, 0.1]])}
>>> weight = {'u': 0.8, 'v': 0.2}
>>> loss = PeriodicMSELoss(weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
2.59999967)
>>> loss = PeriodicMSELoss(reduction="sum", weight=weight)
>>> result = loss(output_dict, label_dict)
>>> print(result)
Tensor(shape=[], dtype=float32, place=Place(gpu:0), stop_gradient=True,
5.19999933)
"""

def __init__(
Expand Down

0 comments on commit 9a50799

Please sign in to comment.