newtontech · newtontech · Feb 27, 2026 · Feb 28, 2026
diff --git a/dpdata/__init__.py b/dpdata/__init__.py
@@ -1,7 +1,7 @@
 from __future__ import annotations
 
-from . import lammps, md, vasp
 from .bond_order_system import BondOrderSystem
+from .formats import lammps, md, vasp
 from .system import LabeledSystem, MultiSystems, System
 
 try:

diff --git a/dpdata/bond_order_system.py b/dpdata/bond_order_system.py
@@ -6,11 +6,11 @@
 
 import numpy as np
 
-import dpdata.rdkit.utils
-from dpdata.rdkit.sanitize import Sanitizer
+import dpdata.formats.rdkit.utils
+from dpdata.formats.rdkit.sanitize import Sanitizer
 from dpdata.system import Axis, DataType, System
 
-# import dpdata.rdkit.mol2
+# import dpdata.formats.rdkit.mol2
 
 
 class BondOrderSystem(System):
@@ -79,7 +79,7 @@ def __init__(
         self.sanitizer = Sanitizer(sanitize_level, raise_errors, verbose)
 
         if data:
-            mol = dpdata.rdkit.utils.system_data_to_mol(data)
+            mol = dpdata.formats.rdkit.utils.system_data_to_mol(data)
             self.from_rdkit_mol(mol)
         if file_name:
             self.from_fmt(
@@ -161,7 +161,7 @@ def __add__(self, other):
     #         magic method "+" operation
     #     '''
     #     if isinstance(other, BondOrderSystem):
-    #         if dpdata.rdkit.utils.check_same_molecule(self.rdkit_mol, other.rdkit_mol):
+    #         if dpdata.formats.rdkit.utils.check_same_molecule(self.rdkit_mol, other.rdkit_mol):
     #             self.__class__(self, data=other.data)
     #         else:
     #             raise RuntimeError("The two systems are not of the same topology.")
@@ -171,7 +171,7 @@ def __add__(self, other):
     def from_rdkit_mol(self, rdkit_mol):
         """Initialize from a rdkit.Chem.rdchem.Mol object."""
         rdkit_mol = self.sanitizer.sanitize(rdkit_mol)
-        self.data = dpdata.rdkit.utils.mol_to_system_data(rdkit_mol)
+        self.data = dpdata.formats.rdkit.utils.mol_to_system_data(rdkit_mol)
         self.data["bond_dict"] = dict(
             [(f"{int(bond[0])}-{int(bond[1])}", bond[2]) for bond in self.data["bonds"]]
         )

diff --git a/dpdata/formats/__init__.py b/dpdata/formats/__init__.py
@@ -0,0 +1 @@
+# Format modules for dpdata
diff --git a/dpdata/abacus/__init__.py → dpdata/formats/abacus/__init__.py b/dpdata/abacus/__init__.py → dpdata/formats/abacus/__init__.py
diff --git a/dpdata/abacus/md.py → dpdata/formats/abacus/md.py b/dpdata/abacus/md.py → dpdata/formats/abacus/md.py
diff --git a/dpdata/abacus/relax.py → dpdata/formats/abacus/relax.py b/dpdata/abacus/relax.py → dpdata/formats/abacus/relax.py
diff --git a/dpdata/abacus/scf.py → dpdata/formats/abacus/scf.py b/dpdata/abacus/scf.py → dpdata/formats/abacus/scf.py
@@ -8,7 +8,7 @@
 
 from dpdata.utils import open_file
 
-from ..unit import LengthConversion, PressureConversion
+from ...unit import LengthConversion, PressureConversion
 from .stru import get_frame_from_stru
 
 bohr2ang = LengthConversion("bohr", "angstrom").value()

diff --git a/dpdata/abacus/stru.py → dpdata/formats/abacus/stru.py b/dpdata/abacus/stru.py → dpdata/formats/abacus/stru.py
@@ -6,7 +6,7 @@
 
 import numpy as np
 
-from ..unit import LengthConversion
+from ...unit import LengthConversion
 
 bohr2ang = LengthConversion("bohr", "angstrom").value()
 

diff --git a/dpdata/amber/__init__.py → dpdata/formats/amber/__init__.py b/dpdata/amber/__init__.py → dpdata/formats/amber/__init__.py
diff --git a/dpdata/amber/mask.py → dpdata/formats/amber/mask.py b/dpdata/amber/mask.py → dpdata/formats/amber/mask.py
diff --git a/dpdata/amber/md.py → dpdata/formats/amber/md.py b/dpdata/amber/md.py → dpdata/formats/amber/md.py
@@ -5,11 +5,11 @@
 
 import numpy as np
 
-from dpdata.amber.mask import pick_by_amber_mask
+from dpdata.formats.amber.mask import pick_by_amber_mask
 from dpdata.unit import EnergyConversion
 from dpdata.utils import open_file
 
-from ..periodic_table import ELEMENTS
+from ...periodic_table import ELEMENTS
 
 kcalmol2eV = EnergyConversion("kcal_mol", "eV").value()
 symbols = ["X"] + ELEMENTS

diff --git a/dpdata/amber/sqm.py → dpdata/formats/amber/sqm.py b/dpdata/amber/sqm.py → dpdata/formats/amber/sqm.py
diff --git a/dpdata/cp2k/__init__.py → dpdata/formats/cp2k/__init__.py b/dpdata/cp2k/__init__.py → dpdata/formats/cp2k/__init__.py
diff --git a/dpdata/cp2k/cell.py → dpdata/formats/cp2k/cell.py b/dpdata/cp2k/cell.py → dpdata/formats/cp2k/cell.py
diff --git a/dpdata/cp2k/output.py → dpdata/formats/cp2k/output.py b/dpdata/cp2k/output.py → dpdata/formats/cp2k/output.py
@@ -7,7 +7,7 @@
 
 import numpy as np
 
-from ..unit import (
+from ...unit import (
     EnergyConversion,
     ForceConversion,
     LengthConversion,

diff --git a/dpdata/deepmd/__init__.py → dpdata/formats/deepmd/__init__.py b/dpdata/deepmd/__init__.py → dpdata/formats/deepmd/__init__.py
diff --git a/dpdata/deepmd/comp.py → dpdata/formats/deepmd/comp.py b/dpdata/deepmd/comp.py → dpdata/formats/deepmd/comp.py
diff --git a/dpdata/deepmd/hdf5.py → dpdata/formats/deepmd/hdf5.py b/dpdata/deepmd/hdf5.py → dpdata/formats/deepmd/hdf5.py
diff --git a/dpdata/deepmd/mixed.py → dpdata/formats/deepmd/mixed.py b/dpdata/deepmd/mixed.py → dpdata/formats/deepmd/mixed.py
diff --git a/dpdata/deepmd/raw.py → dpdata/formats/deepmd/raw.py b/dpdata/deepmd/raw.py → dpdata/formats/deepmd/raw.py
diff --git a/dpdata/dftbplus/__init__.py → dpdata/formats/dftbplus/__init__.py b/dpdata/dftbplus/__init__.py → dpdata/formats/dftbplus/__init__.py
diff --git a/dpdata/dftbplus/output.py → dpdata/formats/dftbplus/output.py b/dpdata/dftbplus/output.py → dpdata/formats/dftbplus/output.py
diff --git a/dpdata/fhi_aims/__init__.py → dpdata/formats/fhi_aims/__init__.py b/dpdata/fhi_aims/__init__.py → dpdata/formats/fhi_aims/__init__.py
diff --git a/dpdata/fhi_aims/output.py → dpdata/formats/fhi_aims/output.py b/dpdata/fhi_aims/output.py → dpdata/formats/fhi_aims/output.py
diff --git a/dpdata/gaussian/__init__.py → dpdata/formats/gaussian/__init__.py b/dpdata/gaussian/__init__.py → dpdata/formats/gaussian/__init__.py
diff --git a/dpdata/gaussian/fchk.py → dpdata/formats/gaussian/fchk.py b/dpdata/gaussian/fchk.py → dpdata/formats/gaussian/fchk.py
@@ -9,8 +9,8 @@
 if TYPE_CHECKING:
     from dpdata.utils import FileType
 
-from ..periodic_table import ELEMENTS
-from ..unit import (
+from ...periodic_table import ELEMENTS
+from ...unit import (
     EnergyConversion,
     ForceConversion,
     HessianConversion,

diff --git a/dpdata/gaussian/gjf.py → dpdata/formats/gaussian/gjf.py b/dpdata/gaussian/gjf.py → dpdata/formats/gaussian/gjf.py
diff --git a/dpdata/gaussian/log.py → dpdata/formats/gaussian/log.py b/dpdata/gaussian/log.py → dpdata/formats/gaussian/log.py
@@ -9,8 +9,8 @@
 if TYPE_CHECKING:
     from dpdata.utils import FileType
 
-from ..periodic_table import ELEMENTS
-from ..unit import EnergyConversion, ForceConversion, LengthConversion
+from ...periodic_table import ELEMENTS
+from ...unit import EnergyConversion, ForceConversion, LengthConversion
 
 length_convert = LengthConversion("bohr", "angstrom").value()
 energy_convert = EnergyConversion("hartree", "eV").value()

diff --git a/dpdata/gromacs/__init__.py → dpdata/formats/gromacs/__init__.py b/dpdata/gromacs/__init__.py → dpdata/formats/gromacs/__init__.py
diff --git a/dpdata/gromacs/gro.py → dpdata/formats/gromacs/gro.py b/dpdata/gromacs/gro.py → dpdata/formats/gromacs/gro.py
@@ -11,7 +11,7 @@
 if TYPE_CHECKING:
     from dpdata.utils import FileType
 
-from ..unit import LengthConversion
+from ...unit import LengthConversion
 
 nm2ang = LengthConversion("nm", "angstrom").value()
 ang2nm = LengthConversion("angstrom", "nm").value()

diff --git a/dpdata/lammps/__init__.py → dpdata/formats/lammps/__init__.py b/dpdata/lammps/__init__.py → dpdata/formats/lammps/__init__.py
diff --git a/dpdata/lammps/dump.py → dpdata/formats/lammps/dump.py b/dpdata/lammps/dump.py → dpdata/formats/lammps/dump.py
diff --git a/dpdata/lammps/lmp.py → dpdata/formats/lammps/lmp.py b/dpdata/lammps/lmp.py → dpdata/formats/lammps/lmp.py
diff --git a/dpdata/lmdb/__init__.py → dpdata/formats/lmdb/__init__.py b/dpdata/lmdb/__init__.py → dpdata/formats/lmdb/__init__.py
diff --git a/dpdata/lmdb/format.py → dpdata/formats/lmdb/format.py b/dpdata/lmdb/format.py → dpdata/formats/lmdb/format.py
diff --git a/dpdata/md/__init__.py → dpdata/formats/md/__init__.py b/dpdata/md/__init__.py → dpdata/formats/md/__init__.py
diff --git a/dpdata/md/msd.py → dpdata/formats/md/msd.py b/dpdata/md/msd.py → dpdata/formats/md/msd.py
diff --git a/dpdata/md/pbc.py → dpdata/formats/md/pbc.py b/dpdata/md/pbc.py → dpdata/formats/md/pbc.py
diff --git a/dpdata/md/rdf.py → dpdata/formats/md/rdf.py b/dpdata/md/rdf.py → dpdata/formats/md/rdf.py
diff --git a/dpdata/md/water.py → dpdata/formats/md/water.py b/dpdata/md/water.py → dpdata/formats/md/water.py
diff --git a/dpdata/openmx/__init__.py → dpdata/formats/openmx/__init__.py b/dpdata/openmx/__init__.py → dpdata/formats/openmx/__init__.py
diff --git a/dpdata/openmx/omx.py → dpdata/formats/openmx/omx.py b/dpdata/openmx/omx.py → dpdata/formats/openmx/omx.py
@@ -10,7 +10,7 @@
 if TYPE_CHECKING:
     from dpdata.utils import FileType
 
-from ..unit import (
+from ...unit import (
     EnergyConversion,
     ForceConversion,
     LengthConversion,

diff --git a/dpdata/orca/__init__.py → dpdata/formats/orca/__init__.py b/dpdata/orca/__init__.py → dpdata/formats/orca/__init__.py
diff --git a/dpdata/orca/output.py → dpdata/formats/orca/output.py b/dpdata/orca/output.py → dpdata/formats/orca/output.py
diff --git a/dpdata/psi4/__init__.py → dpdata/formats/psi4/__init__.py b/dpdata/psi4/__init__.py → dpdata/formats/psi4/__init__.py
diff --git a/dpdata/psi4/input.py → dpdata/formats/psi4/input.py b/dpdata/psi4/input.py → dpdata/formats/psi4/input.py
diff --git a/dpdata/psi4/output.py → dpdata/formats/psi4/output.py b/dpdata/psi4/output.py → dpdata/formats/psi4/output.py
diff --git a/dpdata/pwmat/__init__.py → dpdata/formats/pwmat/__init__.py b/dpdata/pwmat/__init__.py → dpdata/formats/pwmat/__init__.py
diff --git a/dpdata/pwmat/atomconfig.py → dpdata/formats/pwmat/atomconfig.py b/dpdata/pwmat/atomconfig.py → dpdata/formats/pwmat/atomconfig.py
@@ -3,7 +3,7 @@
 
 import numpy as np
 
-from ..periodic_table import ELEMENTS
+from ...periodic_table import ELEMENTS
 
 
 def _to_system_data_lower(lines):

diff --git a/dpdata/pwmat/movement.py → dpdata/formats/pwmat/movement.py b/dpdata/pwmat/movement.py → dpdata/formats/pwmat/movement.py
@@ -4,7 +4,7 @@
 
 import numpy as np
 
-from ..periodic_table import ELEMENTS
+from ...periodic_table import ELEMENTS
 
 
 def system_info(lines, type_idx_zero=False):

diff --git a/dpdata/pymatgen/__init__.py → dpdata/formats/pymatgen/__init__.py b/dpdata/pymatgen/__init__.py → dpdata/formats/pymatgen/__init__.py
diff --git a/dpdata/pymatgen/molecule.py → dpdata/formats/pymatgen/molecule.py b/dpdata/pymatgen/molecule.py → dpdata/formats/pymatgen/molecule.py
diff --git a/dpdata/pymatgen/structure.py → dpdata/formats/pymatgen/structure.py b/dpdata/pymatgen/structure.py → dpdata/formats/pymatgen/structure.py
diff --git a/dpdata/qe/__init__.py → dpdata/formats/qe/__init__.py b/dpdata/qe/__init__.py → dpdata/formats/qe/__init__.py
diff --git a/dpdata/qe/scf.py → dpdata/formats/qe/scf.py b/dpdata/qe/scf.py → dpdata/formats/qe/scf.py
diff --git a/dpdata/qe/traj.py → dpdata/formats/qe/traj.py b/dpdata/qe/traj.py → dpdata/formats/qe/traj.py
@@ -13,7 +13,7 @@
 
 import os
 
-from ..unit import (
+from ...unit import (
     EnergyConversion,
     ForceConversion,
     LengthConversion,

diff --git a/dpdata/rdkit/__init__.py → dpdata/formats/rdkit/__init__.py b/dpdata/rdkit/__init__.py → dpdata/formats/rdkit/__init__.py
diff --git a/dpdata/rdkit/sanitize.py → dpdata/formats/rdkit/sanitize.py b/dpdata/rdkit/sanitize.py → dpdata/formats/rdkit/sanitize.py
diff --git a/dpdata/rdkit/utils.py → dpdata/formats/rdkit/utils.py b/dpdata/rdkit/utils.py → dpdata/formats/rdkit/utils.py
diff --git a/dpdata/siesta/__init__.py → dpdata/formats/siesta/__init__.py b/dpdata/siesta/__init__.py → dpdata/formats/siesta/__init__.py
diff --git a/dpdata/siesta/aiMD_output.py → dpdata/formats/siesta/aiMD_output.py b/dpdata/siesta/aiMD_output.py → dpdata/formats/siesta/aiMD_output.py
diff --git a/dpdata/siesta/output.py → dpdata/formats/siesta/output.py b/dpdata/siesta/output.py → dpdata/formats/siesta/output.py
diff --git a/dpdata/vasp/__init__.py → dpdata/formats/vasp/__init__.py b/dpdata/vasp/__init__.py → dpdata/formats/vasp/__init__.py
diff --git a/dpdata/vasp/outcar.py → dpdata/formats/vasp/outcar.py b/dpdata/vasp/outcar.py → dpdata/formats/vasp/outcar.py
diff --git a/dpdata/vasp/poscar.py → dpdata/formats/vasp/poscar.py b/dpdata/vasp/poscar.py → dpdata/formats/vasp/poscar.py
diff --git a/dpdata/vasp/xml.py → dpdata/formats/vasp/xml.py b/dpdata/vasp/xml.py → dpdata/formats/vasp/xml.py
diff --git a/dpdata/xyz/__init__.py → dpdata/formats/xyz/__init__.py b/dpdata/xyz/__init__.py → dpdata/formats/xyz/__init__.py
diff --git a/dpdata/xyz/quip_gap_xyz.py → dpdata/formats/xyz/quip_gap_xyz.py b/dpdata/xyz/quip_gap_xyz.py → dpdata/formats/xyz/quip_gap_xyz.py
diff --git a/dpdata/xyz/xyz.py → dpdata/formats/xyz/xyz.py b/dpdata/xyz/xyz.py → dpdata/formats/xyz/xyz.py
diff --git a/dpdata/plugins/3dmol.py b/dpdata/plugins/3dmol.py
@@ -3,7 +3,7 @@
 import numpy as np
 
 from dpdata.format import Format
-from dpdata.xyz.xyz import coord_to_xyz
+from dpdata.formats.xyz.xyz import coord_to_xyz
 
 
 @Format.register("3dmol")

diff --git a/dpdata/plugins/abacus.py b/dpdata/plugins/abacus.py
@@ -5,12 +5,12 @@
 
 import numpy as np
 
-import dpdata.abacus.md
-import dpdata.abacus.relax
-import dpdata.abacus.scf
-from dpdata.abacus.stru import get_frame_from_stru, make_unlabeled_stru
+import dpdata.formats.abacus.md
+import dpdata.formats.abacus.relax
+import dpdata.formats.abacus.scf
 from dpdata.data_type import Axis, DataType
 from dpdata.format import Format
+from dpdata.formats.abacus.stru import get_frame_from_stru, make_unlabeled_stru
 from dpdata.utils import open_file
 
 if TYPE_CHECKING:
@@ -90,7 +90,7 @@ def register_move_data(data):
 class AbacusSCFFormat(Format):
     # @Format.post("rot_lower_triangular")
     def from_labeled_system(self, file_name, **kwargs):
-        data = dpdata.abacus.scf.get_frame(file_name)
+        data = dpdata.formats.abacus.scf.get_frame(file_name)
         register_mag_data(data)
         register_move_data(data)
         return data
@@ -102,7 +102,7 @@ def from_labeled_system(self, file_name, **kwargs):
 class AbacusMDFormat(Format):
     # @Format.post("rot_lower_triangular")
     def from_labeled_system(self, file_name, **kwargs):
-        data = dpdata.abacus.md.get_frame(file_name)
+        data = dpdata.formats.abacus.md.get_frame(file_name)
         register_mag_data(data)
         register_move_data(data)
         return data
@@ -114,7 +114,7 @@ def from_labeled_system(self, file_name, **kwargs):
 class AbacusRelaxFormat(Format):
     # @Format.post("rot_lower_triangular")
     def from_labeled_system(self, file_name, **kwargs):
-        data = dpdata.abacus.relax.get_frame(file_name)
+        data = dpdata.formats.abacus.relax.get_frame(file_name)
         register_mag_data(data)
         register_move_data(data)
         return data
diff --git a/dpdata/plugins/amber.py b/dpdata/plugins/amber.py
@@ -4,8 +4,8 @@
 import subprocess as sp
 import tempfile
 
-import dpdata.amber.md
-import dpdata.amber.sqm
+import dpdata.formats.amber.md
+import dpdata.formats.amber.sqm
 from dpdata.driver import Driver, Minimizer
 from dpdata.format import Format
 from dpdata.utils import open_file
@@ -26,7 +26,7 @@ def from_system(
             parm7_file = file_name + ".parm7"
         if nc_file is None:
             nc_file = file_name + ".nc"
-        return dpdata.amber.md.read_amber_traj(
+        return dpdata.formats.amber.md.read_amber_traj(
             parm7_file=parm7_file,
             nc_file=nc_file,
             use_element_symbols=use_element_symbols,
@@ -55,7 +55,7 @@ def from_labeled_system(
             mden_file = file_name + ".mden"
         if mdout_file is None:
             mdout_file = file_name + ".mdout"
-        return dpdata.amber.md.read_amber_traj(
+        return dpdata.formats.amber.md.read_amber_traj(
             parm7_file, nc_file, mdfrc_file, mden_file, mdout_file, use_element_symbols
         )
 
@@ -64,11 +64,11 @@ def from_labeled_system(
 class SQMOutFormat(Format):
     def from_system(self, fname, **kwargs):
         """Read from ambertools sqm.out."""
-        return dpdata.amber.sqm.parse_sqm_out(fname)
+        return dpdata.formats.amber.sqm.parse_sqm_out(fname)
 
     def from_labeled_system(self, fname, **kwargs):
         """Read from ambertools sqm.out."""
-        data = dpdata.amber.sqm.parse_sqm_out(fname)
+        data = dpdata.formats.amber.sqm.parse_sqm_out(fname)
         assert "forces" in list(data.keys()), f"No forces in {fname}"
         return data
 
@@ -104,7 +104,7 @@ def to_system(self, data, fname=None, frame_idx=0, **kwargs):
                 mult : int, default=1
                     multiplicity. Only 1 is allowed.
         """
-        return dpdata.amber.sqm.make_sqm_in(data, fname, frame_idx, **kwargs)
+        return dpdata.formats.amber.sqm.make_sqm_in(data, fname, frame_idx, **kwargs)
 
 
 @Driver.register("sqm")

diff --git a/dpdata/plugins/cp2k.py b/dpdata/plugins/cp2k.py
@@ -2,9 +2,9 @@
 
 import glob
 
-import dpdata.cp2k.output
-from dpdata.cp2k.output import Cp2kSystems
+import dpdata.formats.cp2k.output
 from dpdata.format import Format
+from dpdata.formats.cp2k.output import Cp2kSystems
 
 string_warning = """
 Hi, you got an error from dpdata,
@@ -42,7 +42,7 @@ def from_labeled_system(self, file_name, restart=False, **kwargs):
                 data["energies"],
                 data["forces"],
                 tmp_virial,
-            ) = dpdata.cp2k.output.get_frames(file_name)
+            ) = dpdata.formats.cp2k.output.get_frames(file_name)
             if tmp_virial is not None:
                 data["virials"] = tmp_virial
             return data