From e25b1cd42e4a81d32a20c2273133338cbbff96e5 Mon Sep 17 00:00:00 2001
From: zhiminzhang0830 <452516515@qq.com>
Date: Mon, 15 May 2023 06:49:57 +0000
Subject: [PATCH] update euler beam example

---
 examples/euler_beam/euler_beam.py | 53 ++++++++++++++++++++++---------
 1 file changed, 38 insertions(+), 15 deletions(-)

diff --git a/examples/euler_beam/euler_beam.py b/examples/euler_beam/euler_beam.py
index 772c08cf2..73d35952c 100644
--- a/examples/euler_beam/euler_beam.py
+++ b/examples/euler_beam/euler_beam.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import paddle
+import numpy as np
 
 import ppsci
 from ppsci.autodiff import hessian
@@ -22,8 +22,6 @@
 
 if __name__ == "__main__":
     args = config.parse_args()
-    # enable computation for fourth-order differentiation of matmul
-    paddle.fluid.core.set_prim_eager_enabled(True)
     # set random seed for reproducibility
     ppsci.utils.misc.set_random_seed(42)
     # set training hyper-parameters
@@ -58,24 +56,49 @@
         random="Hammersley",
         name="EQ",
     )
-    bc = ppsci.constraint.BoundaryConstraint(
-        {
-            "u0": lambda d: d["u"][0:1],
-            "u__x": lambda d: jacobian(d["u"], d["x"])[1:2],
-            "u__x__x": lambda d: hessian(d["u"], d["x"])[2:3],
-            "u__x__x__x": lambda d: jacobian(hessian(d["u"], d["x"]), d["x"])[3:4],
-        },
-        {"u0": 0, "u__x": 0, "u__x__x": 0, "u__x__x__x": 0},
+    bc1 = ppsci.constraint.BoundaryConstraint(
+        {"u0": lambda d: d["u"]},
+        {"u0": 0},
         geom["interval"],
-        {**dataloader_cfg, "batch_size": 4},
+        {**dataloader_cfg, "batch_size": 1},
         ppsci.loss.MSELoss("sum"),
-        evenly=True,
-        name="BC",
+        criteria=lambda x: np.isclose(x, 0.0),
+        name="BC1",
+    )
+    bc2 = ppsci.constraint.BoundaryConstraint(
+        {"u__x": lambda d: jacobian(d["u"], d["x"])},
+        {"u__x": 0},
+        geom["interval"],
+        {**dataloader_cfg, "batch_size": 1},
+        ppsci.loss.MSELoss("sum"),
+        criteria=lambda x: np.isclose(x, 0.0),
+        name="BC2",
+    )
+    bc3 = ppsci.constraint.BoundaryConstraint(
+        {"u__x__x": lambda d: hessian(d["u"], d["x"])},
+        {"u__x__x": 0},
+        geom["interval"],
+        {**dataloader_cfg, "batch_size": 1},
+        ppsci.loss.MSELoss("sum"),
+        criteria=lambda x: np.isclose(x, 1.0),
+        name="BC3",
+    )
+    bc4 = ppsci.constraint.BoundaryConstraint(
+        {"u__x__x__x": lambda d: jacobian(hessian(d["u"], d["x"]), d["x"])},
+        {"u__x__x__x": 0},
+        geom["interval"],
+        {**dataloader_cfg, "batch_size": 1},
+        ppsci.loss.MSELoss("sum"),
+        criteria=lambda x: np.isclose(x, 1.0),
+        name="BC4",
     )
     # wrap constraints together
     constraint = {
         pde_constraint.name: pde_constraint,
-        bc.name: bc,
+        bc1.name: bc1,
+        bc2.name: bc2,
+        bc3.name: bc3,
+        bc4.name: bc4,
     }
 
     # set optimizer