Merge branch 'devel' into docs-se_atten

.github/workflows/test_python.yml

@@ -42,17 +42,18 @@ jobs:
     - name: Get durations from cache
       uses: actions/cache@v4
       with:
-        path: test_durations
+        path: .test_durations
         # the key must never match, even when restarting workflows, as that
         # will cause durations to get out of sync between groups, the
         # combined durations will be loaded if available
-        key: test-durations-split-${{ github.run_id }}-${{ github.run_number}}-${{ matrix.python }}-${{ matrix.group }}
+        key: test2-durations-split-${{ github.run_id }}-${{ github.run_number}}-${{ matrix.python }}-${{ matrix.group }}
         restore-keys: |
-          test-durations-combined-${{ matrix.python }}-${{ github.sha }}
-          test-durations-combined-${{ matrix.python }}
-    - run: pytest --cov=deepmd source/tests --durations=0 --splits 6 --group ${{ matrix.group }} --store-durations --durations-path=.test_durations_${{ matrix.group }} --splitting-algorithm least_duration
+          test2-durations-combined-${{ matrix.python }}-${{ github.sha }}
+          test2-durations-combined-${{ matrix.python }}
+    - run: pytest --cov=deepmd source/tests --durations=0 --splits 6 --group ${{ matrix.group }} --store-durations --durations-path=.test_durations --splitting-algorithm least_duration
       env:
         NUM_WORKERS: 0
+    - run: mv .test_durations .test_durations_${{ matrix.group }}
     - name: Upload partial durations
       uses: actions/upload-artifact@v4
       with:
@@ -77,15 +78,15 @@ jobs:
         # key won't match during the first run for the given commit, but
         # restore-key will if there's a previous stored durations file,
         # so cache will both be loaded and stored
-        key: test-durations-combined-${{ matrix.python }}-${{ github.sha }}
-        restore-keys: test-durations-combined-${{ matrix.python }}
+        key: test2-durations-combined-${{ matrix.python }}-${{ github.sha }}
+        restore-keys: test2-durations-combined-${{ matrix.python }}
     - name: Download artifacts
       uses: actions/download-artifact@v4
       with:
         pattern: split-${{ matrix.python }}-*
         merge-multiple: true
     - name: Combine test durations
-      run: jq '. + input' .test_durations_* > .test_durations
+      run: jq -s add .test_durations_* > .test_durations
   pass:
     name: Pass testing Python
     needs: [testpython, update_durations]

.pre-commit-config.yaml

@@ -29,7 +29,7 @@ repos:
         exclude: ^source/3rdparty
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.4.4
+    rev: v0.4.5
     hooks:
       - id: ruff
         args: ["--fix"]

deepmd/pt/utils/nlist.py

@@ -326,7 +326,7 @@ def extend_coord_with_ghosts(
         # +1: central cell
         nbuff = torch.ceil(rcut / to_face).to(torch.long)
         # 3
-        nbuff = torch.max(nbuff, dim=0, keepdim=False).values
+        nbuff = torch.amax(nbuff, dim=0)  # faster than torch.max
         nbuff_cpu = nbuff.cpu()
         xi = torch.arange(-nbuff_cpu[0], nbuff_cpu[0] + 1, 1, device="cpu")
         yi = torch.arange(-nbuff_cpu[1], nbuff_cpu[1] + 1, 1, device="cpu")

deepmd/utils/batch_size.py

@@ -61,6 +61,11 @@ def __init__(self, initial_batch_size: int = 1024, factor: float = 2.0) -> None:
             self.maximum_working_batch_size = initial_batch_size
             if self.is_gpu_available():
                 self.minimal_not_working_batch_size = 2**31
+                log.info(
+                    "If you encounter the error 'an illegal memory access was encountered', this may be due to a TensorFlow issue. "
+                    "To avoid this, set the environment variable DP_INFER_BATCH_SIZE to a smaller value than the last adjusted batch size. "
+                    "The environment variable DP_INFER_BATCH_SIZE controls the inference batch size (nframes * natoms). "
+                )
             else:
                 self.minimal_not_working_batch_size = (
                     self.maximum_working_batch_size + 1

pyproject.toml

@@ -98,7 +98,7 @@ lmp = [
     "lammps~=2023.8.2.3.0",
 ]
 ipi = [
-    "i-PI",
+    "ipi",
 ]
 gui = [
     "dpgui",

source/tests/pt/model/test_polarizability_fitting.py

@@ -269,12 +269,7 @@ def test_permu(self):
 
     def test_trans(self):
         atype = self.atype.reshape(1, 5)
-        coord_s = torch.matmul(
-            torch.remainder(
-                torch.matmul(self.coord + self.shift, torch.linalg.inv(self.cell)), 1.0
-            ),
-            self.cell,
-        )
+        coord_s = self.coord + self.shift
         for fit_diag, scale in itertools.product([True, False], [None, self.scale]):
             ft0 = PolarFittingNet(
                 self.nt,