MasterCodeSample/utils.py at main · Haskely/MasterCodeSample · GitHub

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
import torch as th

from examples import Example_Longtime2012, create_ppd, example2ppd
from loss import vertify_lossfuncs
from network import CombinedNet
from PhysicalProcessDataClass import combine, read_DataDF


def plot_diff(
    exam: Example_Longtime2012, trained_combinedNet: CombinedNet, savedir: "str"
):
    stokes_ppd, darcy_ppd = example2ppd(exam)
    pred_stokes_ppd, pred_darcy_ppd = create_ppd(
        "pred_" + exam.__name__,
        exam.NS_area,
        exam.Darcy_area,
        trained_combinedNet.NS_th,
        trained_combinedNet.Darcy_th,
    )

    SD_ppd = combine(stokes_ppd, darcy_ppd)
    SD_ppd.plot(save_dir=savedir, title="真实的数据")

    pred_SD_ppd = combine(pred_stokes_ppd, pred_darcy_ppd)
    pred_SD_ppd.plot(save_dir=savedir, title="预测的数据")

    diff_DF = SD_ppd.DataDF.copy()
    diff_DF[["U", "V", "P"]] = (SD_ppd.DataDF - pred_SD_ppd.DataDF)[["U", "V", "P"]]
    diff_ppd = read_DataDF(diff_DF, name="真解与预测解之差")
    diff_ppd.plot(save_dir=savedir)


def test_loss(exam: Example_Longtime2012, trained_combinedNet: CombinedNet):
    stokes_UVP_func = trained_combinedNet.NS_th
    darcy_UVP_func = trained_combinedNet.Darcy_th

    eqn_dict = vertify_lossfuncs(exam, stokes_UVP_func, darcy_UVP_func)
    eqnloss_dict = {k: v.abs().mean(dim=0).sum() for k, v in eqn_dict.items()}

    stokes_ppd, darcy_ppd = example2ppd(exam)
    pred_stokes_ppd, pred_darcy_ppd = create_ppd(
        "pred_" + exam.__name__,
        exam.NS_area,
        exam.Darcy_area,
        trained_combinedNet.NS_th,
        trained_combinedNet.Darcy_th,
    )

    phyloss = dict()

    for true_ppd, pred_ppd, name in zip(
        [stokes_ppd, darcy_ppd], [pred_stokes_ppd, pred_darcy_ppd], ["stokes", "darcy"]
    ):
        phy_names = ["U", "V", "P"]
        diffUVP: th.Tensor = th.from_numpy(
            pred_ppd.DataDF[phy_names].to_numpy()
        ) - th.from_numpy(true_ppd.DataDF[phy_names].to_numpy())
        _loss = diffUVP.square().mean(dim=0)
        for i, pn in enumerate(phy_names):
            phyloss[f"{name}_{pn}_loss"] = _loss[i]

    for name, loss in dict(**eqnloss_dict, **phyloss).items():
        print(f"{name}: {loss:g}")

    total_loss = sum(phyloss.values()) + sum(eqnloss_dict.values())
    print(f"\nTotalLoss:{total_loss:g}")