Merge pull request #36 from mesh-adaptation/21_rename

Rename from WarpMesh to UM2N

.github/workflows/test_warpmesh.yml → .github/workflows/test_suite.yml

@@ -1,4 +1,4 @@
-name: Install and Test WarpMesh
+name: Install and test UM2N
 
 on:
   push:
@@ -9,7 +9,7 @@ on:
 
 jobs:
   test-warpmesh:
-    name: Test WarpMesh
+    name: Test UM2N
     runs-on: ubuntu-latest
     container:
       image: firedrakeproject/firedrake:latest
@@ -52,12 +52,12 @@ jobs:
           . /home/firedrake/firedrake/bin/activate
           python3 -m pip install 'git+https://github.com/facebookresearch/pytorch3d.git'
 
-      - name: Install WarpMesh
+      - name: Install UM2N
         run: |
           . /home/firedrake/firedrake/bin/activate
           python3 -m pip install -e .
 
-      - name: Run WarpMesh test suite
+      - name: Run UM2N test suite
         run: |
           . /home/firedrake/firedrake/bin/activate
           python3 -m pytest tests/test* -v

README.md

@@ -17,7 +17,7 @@ Solving complex Partial Differential Equations (PDEs) accurately and efficiently
 
 The latest test status:
 
-[![UM2N](https://github.com/mesh-adaptation/UM2N/actions/workflows/test_warpmesh.yml/badge.svg)](https://github.com/mesh-adaptation/UM2N/actions/workflows/test_warpmesh.yml)
+[![UM2N](https://github.com/mesh-adaptation/UM2N/actions/workflows/test_suite.yml/badge.svg)](https://github.com/mesh-adaptation/UM2N/actions/workflows/test_suite.yml)
 
 
 ## 🛠️ Installation
@@ -31,7 +31,7 @@ Just navigate to **project root** folder, open terminal and execute the
 ```
 This will install [Firedrake](https://www.firedrakeproject.org/download.html)
 and [Movement](https://github.com/mesh-adaptation/movement) under the `install`
-folder, as well as the `WarpMesh` package. Note that the pytorch installed is a cpu version.
+folder, as well as the `UM2N` package. Note that the pytorch installed is a cpu version.
 
 - GPU (cuda) support
 
@@ -114,7 +114,7 @@ The number of samples in the dataset can be changed by modifying the variable
 
 ## 🚀 Train the model (This is outdated)
 
-A training notebook is provided: `script/train_warpmesh.ipynb`. Further training
+A training notebook is provided: `script/train_um2n.ipynb`. Further training
 details can be found in the notebook.
 
 Here is also a link to pre-trained models:
@@ -136,7 +136,7 @@ The documentation is generated by Sphinx. To build the documentation, under the
 ## 🧩 Project Layout
 
 ```
-├── warpmesh (Implementation of the project)
+├── UM2N (Implementation of the project)
 │   ├── __init__.py
 │   ├── generator (Dataset generator)
 │   ├── processor (Data processor)

warpmesh/generator/burgers_solver.py → UM2N/generator/burgers_solver.py

@@ -9,8 +9,6 @@
 import movement as mv
 import numpy as np  # noqa
 
-import warpmesh as wm  # noqa
-
 __all__ = ["BurgersSolver"]
 
 

warpmesh/generator/mesh_generator.py → UM2N/generator/mesh_generator.py

@@ -3,7 +3,7 @@
 import firedrake as fd
 import movement as mv
 
-from warpmesh.generator.equation_solver import EquationSolver
+from UM2N.generator.equation_solver import EquationSolver
 
 os.environ["OMP_NUM_THREADS"] = "1"
 __all__ = ["MeshGenerator"]

warpmesh/generator/swirl_solver.py → UM2N/generator/swirl_solver.py

@@ -8,7 +8,7 @@
 import firedrake as fd  # noqa
 import torch
 import movement as mv  # noqa
-import warpmesh as wm  # noqa
+import UM2N
 import numpy as np
 import pandas as pd
 
@@ -20,7 +20,7 @@
 import matplotlib.pyplot as plt  # noqa
 
 from tqdm import tqdm  # noqa
-from warpmesh.model.train_util import model_forward
+from UM2N.model.train_util import model_forward
 
 
 def get_log_og(log_path, idx):
@@ -826,7 +826,7 @@ def eval_problem(
             # self.u_prev_adapt.project(self.u_cur)
 
             # check mesh integrity - Only perform evaluation on non-tangling mesh  # noqa
-            num_tangle = wm.get_sample_tangle(out, sample.x[:, :2], sample.face)  # noqa
+            num_tangle = UM2N.get_sample_tangle(out, sample.x[:, :2], sample.face)  # noqa
             if isinstance(num_tangle, torch.Tensor):
                 num_tangle = num_tangle.item()
             if num_tangle > 0:  # has tangled elems:
@@ -859,7 +859,7 @@ def eval_problem(
             self.mesh_new.coordinates.dat.data[:] = y
 
             if ((step + 1) % self.save_interval == 0) or (step == 0):
-                fig, plot_data_dict = wm.plot_compare(
+                fig, plot_data_dict = UM2N.plot_compare(
                     self.mesh_fine,
                     self.mesh,
                     self.mesh_new,
@@ -943,16 +943,16 @@ def make_all_dirs(self, log_path, plot_path, plot_more_path, plot_data_path):
         self.make_plot_data_dir(plot_data_path)
 
     def make_log_dir(self, log_path):
-        wm.mkdir_if_not_exist(log_path)
+        UM2N.mkdir_if_not_exist(log_path)
 
     def make_plot_dir(self, plot_path):
-        wm.mkdir_if_not_exist(plot_path)
+        UM2N.mkdir_if_not_exist(plot_path)
 
     def make_plot_more_dir(self, plot_more_path):
-        wm.mkdir_if_not_exist(plot_more_path)
+        UM2N.mkdir_if_not_exist(plot_more_path)
 
     def make_plot_data_dir(self, plot_data_path):
-        wm.mkdir_if_not_exist(plot_data_path)
+        UM2N.mkdir_if_not_exist(plot_data_path)
 
 
 if __name__ == "__main__":

warpmesh/generator/swirl_solver_step.py → UM2N/generator/swirl_solver_step.py

@@ -7,7 +7,6 @@
 import firedrake as fd  # noqa
 
 import movement as mv  # noqa
-import warpmesh as wm  # noqa
 import numpy as np
 
 import firedrake.function as ffunc

warpmesh/test/bench_burgers.py → UM2N/test/bench_burgers.py

@@ -13,8 +13,8 @@
 import torch  # noqa
 from torch_geometric.loader import DataLoader
 
-import warpmesh as wm  # noqa
-from warpmesh.model.train_util import generate_samples
+import UM2N
+from UM2N.model.train_util import generate_samples
 
 
 def get_log_og(log_path, idx):
@@ -302,7 +302,7 @@ def eval_problem(self):
                     dur_ms = (end - start) * 1000
 
                 # check mesh integrity - Only perform evaluation on non-tangling mesh  # noqa
-                num_tangle = wm.get_sample_tangle(out, sample.x[:, :2], sample.face)  # noqa
+                num_tangle = UM2N.get_sample_tangle(out, sample.x[:, :2], sample.face)  # noqa
                 if isinstance(num_tangle, torch.Tensor):
                     num_tangle = num_tangle.item()
                 if num_tangle > 0:  # has tangled elems:
@@ -368,7 +368,7 @@ def eval_problem(self):
                 df.to_csv(os.path.join(self.log_path, f"log{self.idx}_{cur_step}.csv"))  # noqa
 
                 # plot compare mesh
-                compare_plot = wm.plot_mesh_compare_benchmark(
+                compare_plot = UM2N.plot_mesh_compare_benchmark(
                     out.detach().cpu().numpy(),
                     sample.y.detach().cpu().numpy(),
                     sample.face.detach().cpu().numpy(),
@@ -485,13 +485,13 @@ def get_error(self):
         return error_og, error_adapt
 
     def make_log_dir(self):
-        wm.mkdir_if_not_exist(self.log_path)
+        UM2N.mkdir_if_not_exist(self.log_path)
 
     def make_plot_dir(self):
-        wm.mkdir_if_not_exist(self.plot_path)
+        UM2N.mkdir_if_not_exist(self.plot_path)
 
     def make_plot_more_dir(self):
-        wm.mkdir_if_not_exist(self.plot_more_path)
+        UM2N.mkdir_if_not_exist(self.plot_more_path)
 
     def make_plot_data_dir(self):
-        wm.mkdir_if_not_exist(self.plot_data_path)
+        UM2N.mkdir_if_not_exist(self.plot_data_path)

warpmesh/test/bench_swirl.py → UM2N/test/bench_swirl.py

@@ -13,8 +13,8 @@
 import torch
 from torch_geometric.loader import DataLoader
 
-import warpmesh as wm
-from warpmesh.model.train_util import model_forward
+import UM2N
+from UM2N.model.train_util import model_forward
 
 
 def get_log_og(log_path, idx):
@@ -126,10 +126,10 @@ def __init__(
         self.y_0 = kwargs.pop("y_0", 0.25)
 
         # initital condition of u on coarse / fine mesh
-        u_init_exp = wm.get_u_0(
+        u_init_exp = UM2N.get_u_0(
             self.x, self.y, self.r_0, self.x_0, self.y_0, self.sigma
         )  # noqa
-        u_init_exp_fine = wm.get_u_0(
+        u_init_exp_fine = UM2N.get_u_0(
             self.x_fine, self.y_fine, self.r_0, self.x_0, self.y_0, self.sigma
         )  # noqa
         self.u_init = fd.Function(self.scalar_space).interpolate(u_init_exp)
@@ -265,7 +265,7 @@ def solve_u(self, t):
         Solve the PDE problem using RK (SSPRK) scheme on the coarse mesh
         store the solution field to a varaible: self.u_cur
         """
-        c_exp = wm.get_c(self.x, self.y, t, alpha=self.alpha)
+        c_exp = UM2N.get_c(self.x, self.y, t, alpha=self.alpha)
         c_temp = fd.Function(self.vector_space).interpolate(c_exp)
         self.c.project(c_temp)
 
@@ -283,7 +283,7 @@ def solve_u_fine(self, t):
         Solve the PDE problem using RK (SSPRK) scheme on the fine mesh
         store the solution field to a varaible: self.u_cur_fine
         """
-        c_exp = wm.get_c(self.x_fine, self.y_fine, t, alpha=self.alpha)
+        c_exp = UM2N.get_c(self.x_fine, self.y_fine, t, alpha=self.alpha)
         c_temp = fd.Function(self.vector_space_fine).interpolate(c_exp)
         self.c_fine.project(c_temp)
 
@@ -303,16 +303,16 @@ def project_u_(self):
         return
 
     def make_log_dir(self):
-        wm.mkdir_if_not_exist(self.log_path)
+        UM2N.mkdir_if_not_exist(self.log_path)
 
     def make_plot_dir(self):
-        wm.mkdir_if_not_exist(self.plot_path)
+        UM2N.mkdir_if_not_exist(self.plot_path)
 
     def make_plot_more_dir(self):
-        wm.mkdir_if_not_exist(self.plot_more_path)
+        UM2N.mkdir_if_not_exist(self.plot_more_path)
 
     def make_plot_data_dir(self):
-        wm.mkdir_if_not_exist(self.plot_data_path)
+        UM2N.mkdir_if_not_exist(self.plot_data_path)
 
     def eval_problem(self, model_name="model"):
         print("In eval problem")
@@ -484,7 +484,7 @@ def eval_problem(self, model_name="model"):
                 self.uh_model = fd.Function(function_space_model).project(self.u_cur)  # noqa
 
                 # check mesh integrity - Only perform evaluation on non-tangling mesh  # noqa
-                num_tangle = wm.get_sample_tangle(out, sample.x[:, :2], sample.face)  # noqa
+                num_tangle = UM2N.get_sample_tangle(out, sample.x[:, :2], sample.face)  # noqa
                 if isinstance(num_tangle, torch.Tensor):
                     num_tangle = num_tangle.item()
                 if num_tangle > 0:  # has tangled elems:
@@ -500,7 +500,7 @@ def eval_problem(self, model_name="model"):
                 # self.solve_u_fine(self.t)
                 u_fine = fd.Function(function_space_fine).project(self.u_cur_fine)  # noqa
 
-                fig, plot_data_dict = wm.plot_compare(
+                fig, plot_data_dict = UM2N.plot_compare(
                     self.mesh_fine,
                     self.mesh_coarse,
                     self.mesh_new,
@@ -574,7 +574,7 @@ def eval_problem(self, model_name="model"):
                 df = pd.DataFrame(res, index=[0])
                 df.to_csv(os.path.join(self.log_path, f"log_{idx:04d}.csv"))
                 # plot compare mesh
-                plot_fig = wm.plot_mesh_compare_benchmark(
+                plot_fig = UM2N.plot_mesh_compare_benchmark(
                     out.detach().cpu().numpy(),
                     sample.y.detach().cpu().numpy(),
                     sample.face.detach().cpu().numpy(),

warpmesh/test/compare_hlmhltz.py → UM2N/test/compare_hlmhltz.py

@@ -4,7 +4,7 @@
 import firedrake as fd
 import matplotlib.pyplot as plt  # noqa
 
-import warpmesh as wm
+import UM2N
 
 __all__ = ["compare_error"]
 
@@ -66,21 +66,21 @@ def func(x, y):
     # construct the helmholtz equation
     eq = None
     if problem_type == "helmholtz":
-        eq = wm.HelmholtzEqGenerator(
+        eq = UM2N.HelmholtzEqGenerator(
             params={
                 "u_exact_func": u_exact,
             }
         )
     elif problem_type == "poisson":
-        eq = wm.PoissonEqGenerator(
+        eq = UM2N.PoissonEqGenerator(
             params={
                 "u_exact_func": u_exact,
             }
         )
 
     # solution on og mesh
     og_res = eq.discretise(mesh)
-    og_solver = wm.EquationSolver(
+    og_solver = UM2N.EquationSolver(
         params={
             "function_space": og_res["function_space"],
             "LHS": og_res["LHS"],
@@ -93,7 +93,7 @@ def func(x, y):
     # solution on MA mesh
     mesh_MA.coordinates.dat.data[:] = data_in.y.detach().cpu().numpy()
     ma_res = eq.discretise(mesh_MA)
-    ma_solver = wm.EquationSolver(
+    ma_solver = UM2N.EquationSolver(
         params={
             "function_space": ma_res["function_space"],
             "LHS": ma_res["LHS"],
@@ -107,7 +107,7 @@ def func(x, y):
     uh_model = None
     if num_tangle == 0:
         model_res = eq.discretise(mesh_model)
-        model_solver = wm.EquationSolver(
+        model_solver = UM2N.EquationSolver(
             params={
                 "function_space": model_res["function_space"],
                 "LHS": model_res["LHS"],
@@ -125,7 +125,7 @@ def func(x, y):
     res_high_res = eq.discretise(high_res_mesh)
     uh_exact = fd.interpolate(res_high_res["u_exact"], high_res_function_space)
 
-    fig, plot_data_dict = wm.plot_compare(
+    fig, plot_data_dict = UM2N.plot_compare(
         mesh_fine,
         mesh,
         mesh_MA,