feat: add use_loc_mapping

deepmd/dpmodel/descriptor/dpa3.py

@@ -271,6 +271,8 @@ class DescrptDPA3(NativeOP, BaseDescriptor):
         Whether to use electronic configuration type embedding.
     use_tebd_bias : bool, Optional
         Whether to use bias in the type embedding layer.
+    use_loc_mapping : bool, Optional
+        Whether to use local atom index mapping in non-parallel inference.
     type_map : list[str], Optional
         A list of strings. Give the name to each type of atoms.
     """
@@ -290,6 +292,7 @@ def __init__(
         seed: Optional[Union[int, list[int]]] = None,
         use_econf_tebd: bool = False,
         use_tebd_bias: bool = False,
+        use_loc_mapping: bool = True,
         type_map: Optional[list[str]] = None,
     ) -> None:
         super().__init__()
@@ -335,6 +338,7 @@ def init_subclass_params(sub_data, sub_class):
             use_exp_switch=self.repflow_args.use_exp_switch,
             use_dynamic_sel=self.repflow_args.use_dynamic_sel,
             sel_reduce_factor=self.repflow_args.sel_reduce_factor,
+            use_loc_mapping=use_loc_mapping,
             exclude_types=exclude_types,
             env_protection=env_protection,
             precision=precision,
@@ -343,6 +347,7 @@ def init_subclass_params(sub_data, sub_class):
 
         self.use_econf_tebd = use_econf_tebd
         self.use_tebd_bias = use_tebd_bias
+        self.use_loc_mapping = use_loc_mapping
         self.type_map = type_map
         self.tebd_dim = self.repflow_args.n_dim
         self.type_embedding = TypeEmbedNet(
@@ -541,10 +546,16 @@ def call(
         nall = xp.reshape(coord_ext, (nframes, -1)).shape[1] // 3
 
         type_embedding = self.type_embedding.call()
-        node_ebd_ext = xp.reshape(
-            xp.take(type_embedding, xp.reshape(atype_ext, [-1]), axis=0),
-            (nframes, nall, self.tebd_dim),
-        )
+        if self.use_loc_mapping:
+            node_ebd_ext = xp.reshape(
+                xp.take(type_embedding, xp.reshape(atype_ext[:, :nloc], [-1]), axis=0),
+                (nframes, nloc, self.tebd_dim),
+            )
+        else:
+            node_ebd_ext = xp.reshape(
+                xp.take(type_embedding, xp.reshape(atype_ext, [-1]), axis=0),
+                (nframes, nall, self.tebd_dim),
+            )
         node_ebd_inp = node_ebd_ext[:, :nloc, :]
         # repflows
         node_ebd, edge_ebd, h2, rot_mat, sw = self.repflows(

deepmd/dpmodel/descriptor/repflows.py

@@ -145,6 +145,8 @@ class DescrptBlockRepflows(NativeOP, DescriptorBlock):
         In the dynamic selection case, neighbor-scale normalization will use `e_sel / sel_reduce_factor`
         or `a_sel / sel_reduce_factor` instead of the raw `e_sel` or `a_sel` values,
         accommodating larger selection numbers.
+    use_loc_mapping : bool, optional
+        Whether to use local atom index mapping in non-parallel inference.
     ntypes : int
         Number of element types
     activation_function : str, optional
@@ -196,6 +198,7 @@ def __init__(
         use_exp_switch: bool = False,
         use_dynamic_sel: bool = False,
         sel_reduce_factor: float = 10.0,
+        use_loc_mapping: bool = True,
         seed: Optional[Union[int, list[int]]] = None,
     ) -> None:
         super().__init__()
@@ -229,6 +232,7 @@ def __init__(
         self.smooth_edge_update = smooth_edge_update
         self.use_exp_switch = use_exp_switch
         self.use_dynamic_sel = use_dynamic_sel
+        self.use_loc_mapping = use_loc_mapping
         self.sel_reduce_factor = sel_reduce_factor
         if self.use_dynamic_sel and not self.smooth_edge_update:
             raise NotImplementedError(
@@ -527,10 +531,18 @@ def call(
             cosine_ij, (nframes, nloc, self.a_sel, self.a_sel, 1)
         ) / (xp.pi**0.5)
 
+        if self.use_loc_mapping:
+            assert mapping is not None
+            flat_map = xp.reshape(mapping, (nframes, -1))
+            nlist = xp.reshape(
+                xp_take_along_axis(flat_map, xp.reshape(nlist, (nframes, -1)), axis=1),
+                nlist.shape,
+            )
+
         if self.use_dynamic_sel:
             # get graph index
             edge_index, angle_index = get_graph_index(
-                nlist, nlist_mask, a_nlist_mask, nall
+                nlist, nlist_mask, a_nlist_mask, nall, use_loc_mapping=self.use_loc_mapping
             )
             # flat all the tensors
             # n_edge x 1
@@ -561,7 +573,11 @@ def call(
         for idx, ll in enumerate(self.layers):
             # node_ebd:     nb x nloc x n_dim
             # node_ebd_ext: nb x nall x n_dim
-            node_ebd_ext = xp_take_along_axis(node_ebd, mapping, axis=1)
+            node_ebd_ext = (
+                node_ebd
+                if self.use_loc_mapping
+                else xp_take_along_axis(node_ebd, mapping, axis=1)
+            )
             node_ebd, edge_ebd, angle_ebd = ll.call(
                 node_ebd_ext,
                 edge_ebd,

deepmd/dpmodel/utils/network.py

@@ -1006,6 +1006,7 @@ def get_graph_index(
     nlist_mask: np.ndarray,
     a_nlist_mask: np.ndarray,
     nall: int,
+    use_loc_mapping: bool = True,
 ):
     """
     Get the index mapping for edge graph and angle graph, ready in `aggregate` or `index_select`.
@@ -1020,6 +1021,8 @@ def get_graph_index(
         Masks of the neighbor list for angle. real nei 1 otherwise 0
     nall
         The number of extended atoms.
+    use_loc_mapping
+        Whether to use local atom index mapping in non-parallel inference.
 
     Returns
     -------
@@ -1060,7 +1063,7 @@ def get_graph_index(
     n2e_index = n2e_index[xp.astype(nlist_mask, xp.bool)]
 
     # node_ext(j) to edge(ij) index_select
-    frame_shift = xp.arange(nf, dtype=nlist.dtype) * nall
+    frame_shift = xp.arange(nf, dtype=nlist.dtype) * (nall if not use_loc_mapping else nloc)
     shifted_nlist = nlist + frame_shift[:, xp.newaxis, xp.newaxis]
     # n_edge
     n_ext2e_index = shifted_nlist[xp.astype(nlist_mask, xp.bool)]

deepmd/pt/model/descriptor/dpa3.py

@@ -89,6 +89,8 @@ class DescrptDPA3(BaseDescriptor, torch.nn.Module):
         Whether to use electronic configuration type embedding.
     use_tebd_bias : bool, Optional
         Whether to use bias in the type embedding layer.
+    use_loc_mapping : bool, Optional
+        Whether to use local atom index mapping in non-parallel inference.
     type_map : list[str], Optional
         A list of strings. Give the name to each type of atoms.
     """
@@ -108,6 +110,7 @@ def __init__(
         seed: Optional[Union[int, list[int]]] = None,
         use_econf_tebd: bool = False,
         use_tebd_bias: bool = False,
+        use_loc_mapping: bool = True,
         type_map: Optional[list[str]] = None,
     ) -> None:
         super().__init__()
@@ -153,13 +156,15 @@ def init_subclass_params(sub_data, sub_class):
             use_exp_switch=self.repflow_args.use_exp_switch,
             use_dynamic_sel=self.repflow_args.use_dynamic_sel,
             sel_reduce_factor=self.repflow_args.sel_reduce_factor,
+            use_loc_mapping=use_loc_mapping,
             exclude_types=exclude_types,
             env_protection=env_protection,
             precision=precision,
             seed=child_seed(seed, 1),
         )
 
         self.use_econf_tebd = use_econf_tebd
+        self.use_loc_mapping = use_loc_mapping
         self.use_tebd_bias = use_tebd_bias
         self.type_map = type_map
         self.tebd_dim = self.repflow_args.n_dim
@@ -469,12 +474,16 @@ def forward(
             The smooth switch function. shape: nf x nloc x nnei
 
         """
+        parallel_mode = comm_dict is not None
         # cast the input to internal precsion
         extended_coord = extended_coord.to(dtype=self.prec)
         nframes, nloc, nnei = nlist.shape
         nall = extended_coord.view(nframes, -1).shape[1] // 3
 
-        node_ebd_ext = self.type_embedding(extended_atype)
+        if not parallel_mode and self.use_loc_mapping:
+            node_ebd_ext = self.type_embedding(extended_atype[:, :nloc])
+        else:
+            node_ebd_ext = self.type_embedding(extended_atype)
         node_ebd_inp = node_ebd_ext[:, :nloc, :]
         # repflows
         node_ebd, edge_ebd, h2, rot_mat, sw = self.repflows(

deepmd/pt/model/descriptor/repflow_layer.py

@@ -684,7 +684,7 @@ def optim_edge_update_dynamic(
 
     def forward(
         self,
-        node_ebd_ext: torch.Tensor,  # nf x nall x n_dim
+        node_ebd_ext: torch.Tensor,  # nf x nall x n_dim [OR] nf x nloc x n_dim when not parallel_mode
         edge_ebd: torch.Tensor,  # nf x nloc x nnei x e_dim
         h2: torch.Tensor,  # nf x nloc x nnei x 3
         angle_ebd: torch.Tensor,  # nf x nloc x a_nnei x a_nnei x a_dim

deepmd/pt/model/descriptor/repflows.py

@@ -209,6 +209,7 @@ def __init__(
         use_exp_switch: bool = False,
         use_dynamic_sel: bool = False,
         sel_reduce_factor: float = 10.0,
+        use_loc_mapping: bool = True,
         optim_update: bool = True,
         seed: Optional[Union[int, list[int]]] = None,
     ) -> None:
@@ -239,6 +240,7 @@ def __init__(
         self.fix_stat_std = fix_stat_std
         self.set_stddev_constant = fix_stat_std != 0.0
         self.a_compress_use_split = a_compress_use_split
+        self.use_loc_mapping = use_loc_mapping
         self.optim_update = optim_update
         self.smooth_edge_update = smooth_edge_update
         self.use_exp_switch = use_exp_switch
@@ -416,9 +418,9 @@ def forward(
         mapping: Optional[torch.Tensor] = None,
         comm_dict: Optional[dict[str, torch.Tensor]] = None,
     ):
-        if comm_dict is None:
+        parallel_mode = comm_dict is not None
+        if not parallel_mode:
             assert mapping is not None
-            assert extended_atype_embd is not None
         nframes, nloc, nnei = nlist.shape
         nall = extended_coord.view(nframes, -1).shape[1] // 3
         atype = extended_atype[:, :nloc]
@@ -470,12 +472,9 @@ def forward(
 
         # get node embedding
         # [nframes, nloc, tebd_dim]
-        if comm_dict is None:
-            assert isinstance(extended_atype_embd, torch.Tensor)  # for jit
-            atype_embd = extended_atype_embd[:, :nloc, :]
-            assert list(atype_embd.shape) == [nframes, nloc, self.n_dim]
-        else:
-            atype_embd = extended_atype_embd
+        assert extended_atype_embd is not None
+        atype_embd = extended_atype_embd[:, :nloc, :]
+        assert list(atype_embd.shape) == [nframes, nloc, self.n_dim]
         assert isinstance(atype_embd, torch.Tensor)  # for jit
         node_ebd = self.act(atype_embd)
         n_dim = node_ebd.shape[-1]
@@ -494,10 +493,22 @@ def forward(
         cosine_ij = torch.matmul(normalized_diff_i, normalized_diff_j) * (1 - 1e-6)
         angle_input = cosine_ij.unsqueeze(-1) / (torch.pi**0.5)
 
+        if not parallel_mode and self.use_loc_mapping:
+            assert mapping is not None
+            # convert nlist from nall to nloc index
+            nlist = torch.gather(
+                mapping,
+                1,
+                index=nlist.reshape(nframes, -1),
+            ).reshape(nlist.shape)
         if self.use_dynamic_sel:
             # get graph index
             edge_index, angle_index = get_graph_index(
-                nlist, nlist_mask, a_nlist_mask, nall
+                nlist,
+                nlist_mask,
+                a_nlist_mask,
+                nall,
+                use_loc_mapping=self.use_loc_mapping,
             )
             # flat all the tensors
             # n_edge x 1
@@ -524,18 +535,23 @@ def forward(
         angle_ebd = self.angle_embd(angle_input)
 
         # nb x nall x n_dim
-        if comm_dict is None:
+        if not parallel_mode:
             assert mapping is not None
             mapping = (
                 mapping.view(nframes, nall).unsqueeze(-1).expand(-1, -1, self.n_dim)
             )
         for idx, ll in enumerate(self.layers):
             # node_ebd:     nb x nloc x n_dim
-            # node_ebd_ext: nb x nall x n_dim
-            if comm_dict is None:
+            # node_ebd_ext: nb x nall x n_dim [OR] nb x nloc x n_dim when not parallel_mode
+            if not parallel_mode:
                 assert mapping is not None
-                node_ebd_ext = torch.gather(node_ebd, 1, mapping)
+                node_ebd_ext = (
+                    torch.gather(node_ebd, 1, mapping)
+                    if not self.use_loc_mapping
+                    else node_ebd
+                )
             else:
+                assert comm_dict is not None
                 has_spin = "has_spin" in comm_dict
                 if not has_spin:
                     n_padding = nall - nloc

deepmd/pt/model/network/utils.py

@@ -51,6 +51,7 @@ def get_graph_index(
     nlist_mask: torch.Tensor,
     a_nlist_mask: torch.Tensor,
     nall: int,
+    use_loc_mapping: bool = True,
 ):
     """
     Get the index mapping for edge graph and angle graph, ready in `aggregate` or `index_select`.
@@ -100,7 +101,9 @@ def get_graph_index(
     n2e_index = n2e_index[nlist_mask]  # graph node index, atom_graph[:, 0]
 
     # node_ext(j) to edge(ij) index_select
-    frame_shift = torch.arange(0, nf, dtype=nlist.dtype, device=nlist.device) * nall
+    frame_shift = torch.arange(0, nf, dtype=nlist.dtype, device=nlist.device) * (
+        nall if not use_loc_mapping else nloc
+    )
     shifted_nlist = nlist + frame_shift[:, None, None]
     # n_edge
     n_ext2e_index = shifted_nlist[nlist_mask]  # graph neighbor index, atom_graph[:, 1]

deepmd/utils/argcheck.py

@@ -1421,6 +1421,14 @@ def descrpt_dpa3_args():
             default=False,
             doc=doc_use_tebd_bias,
         ),
+        Argument(
+            "use_loc_mapping",
+            bool,
+            optional=True,
+            default=True,
+            doc="Whether to use local atom index mapping in non-parallel inference. "
+            "When True, local indexing and mapping are applied to neighbor lists and embeddings during descriptor computation.",
+        ),
     ]