mlatom 3.18.2__tar.gz → 3.18.3__tar.gz

{mlatom-3.18.2/src/mlatom.egg-info → mlatom-3.18.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mlatom
-Version: 3.18.2
+Version: 3.18.3
 Summary: A Package for AI-enhanced computational chemistry
 Home-page: http://mlatom.com
 Author: Pavlo O. Dral
@@ -14,20 +14,11 @@ Classifier: Environment :: Console
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE.txt
-Requires-Dist: numpy
-Requires-Dist: scipy
-Requires-Dist: h5py
-Requires-Dist: pyh5md
-Requires-Dist: torch
-Requires-Dist: torchani
-Requires-Dist: matplotlib
-Requires-Dist: statsmodels
-Requires-Dist: tqdm
 
 # About Program
 MLatom: a Package for Atomistic Simulations with Machine Learning
 
-Version 3.18.2
+Version 3.18.3
 
 **Official website**: http://mlatom.com/    
 **Documentation**: http://mlatom.com/docs/                     
@@ -56,7 +47,7 @@ When this Software or its derivatives are used in scientific publications, it sh
 * Pavlo O. Dral, Fuchun Ge, Yi-Fan Hou, Peikun Zheng, Yuxinxin Chen, Mario Barbatti, Olexandr Isayev, Cheng Wang, Bao-Xin Xue, Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Lina Zhang, Shuang Zhang, Arif Ullah, Quanhao Zhang, Yanchi Ou. [MLatom 3: A Platform for Machine Learning-enhanced Computational Chemistry Simulations and Workflows](https://doi.org/10.1021/acs.jctc.3c01203).   *J. Chem. Theory Comput.* **2024**, *20*, 1193--1213.                                                                                               
 * Pavlo O. Dral, Fuchun Ge, Bao-Xin Xue, Yi-Fan Hou, Max Pinheiro Jr, Jianxing Huang, Mario Barbatti. [MLatom 2: An Integrative Platform for Atomistic Machine Learning](http://doi.org/10.1007/s41061-021-00339-5). *Top. Curr. Chem.* **2021**, *379*, 27.       
 * Pavlo O. Dral. [MLatom: A Program Package for Quantum Chemical Research Assisted by Machine Learning](http://dx.doi.org/10.1002/jcc.26004). *J. Comput. Chem.* **2019**, *40*, 2339--2347.                                                                                      
-* Pavlo O. Dral, Fuchun Ge, Yi-Fan Hou, Yuxinxin Chen, Peikun Zheng, Bao-Xin Xue, Mikolaj Martyka, Lina Zhang, Jakub Martinka, Quanhao Zhang, Xin-Yu Tong, Arif Ullah, Sebastian V. Pios, Vignesh B. Kumar, Yanchi Ou, Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Shuang Zhang, Matheus O. Bispo, MLatom: A Package for Atomistic Simulations with Machine Learning,  version 3.18.2, Xiamen University, Xiamen, China, 2013-2025. [MLatom.com](http://mlatom.com).
+* Pavlo O. Dral, Fuchun Ge, Yi-Fan Hou, Yuxinxin Chen, Peikun Zheng, Bao-Xin Xue, Mikolaj Martyka, Lina Zhang, Jakub Martinka, Quanhao Zhang, Xin-Yu Tong, Arif Ullah, Sebastian V. Pios, Vignesh B. Kumar, Yanchi Ou, Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Shuang Zhang, Matheus O. Bispo, MLatom: A Package for Atomistic Simulations with Machine Learning,  version 3.18.3, Xiamen University, Xiamen, China, 2013-2025. [MLatom.com](http://mlatom.com).
                                                                       
 The citations for MLatom's interfaces and features shall be eventually
 included too. See program output, header.py, ref.json, and [MLatom.com](http://mlatom.com).          

{mlatom-3.18.2 → mlatom-3.18.3}/README.md

@@ -1,7 +1,7 @@
 # About Program
 MLatom: a Package for Atomistic Simulations with Machine Learning
 
-Version 3.18.2
+Version 3.18.3
 
 **Official website**: http://mlatom.com/    
 **Documentation**: http://mlatom.com/docs/                     
@@ -30,7 +30,7 @@ When this Software or its derivatives are used in scientific publications, it sh
 * Pavlo O. Dral, Fuchun Ge, Yi-Fan Hou, Peikun Zheng, Yuxinxin Chen, Mario Barbatti, Olexandr Isayev, Cheng Wang, Bao-Xin Xue, Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Lina Zhang, Shuang Zhang, Arif Ullah, Quanhao Zhang, Yanchi Ou. [MLatom 3: A Platform for Machine Learning-enhanced Computational Chemistry Simulations and Workflows](https://doi.org/10.1021/acs.jctc.3c01203).   *J. Chem. Theory Comput.* **2024**, *20*, 1193--1213.                                                                                               
 * Pavlo O. Dral, Fuchun Ge, Bao-Xin Xue, Yi-Fan Hou, Max Pinheiro Jr, Jianxing Huang, Mario Barbatti. [MLatom 2: An Integrative Platform for Atomistic Machine Learning](http://doi.org/10.1007/s41061-021-00339-5). *Top. Curr. Chem.* **2021**, *379*, 27.       
 * Pavlo O. Dral. [MLatom: A Program Package for Quantum Chemical Research Assisted by Machine Learning](http://dx.doi.org/10.1002/jcc.26004). *J. Comput. Chem.* **2019**, *40*, 2339--2347.                                                                                      
-* Pavlo O. Dral, Fuchun Ge, Yi-Fan Hou, Yuxinxin Chen, Peikun Zheng, Bao-Xin Xue, Mikolaj Martyka, Lina Zhang, Jakub Martinka, Quanhao Zhang, Xin-Yu Tong, Arif Ullah, Sebastian V. Pios, Vignesh B. Kumar, Yanchi Ou, Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Shuang Zhang, Matheus O. Bispo, MLatom: A Package for Atomistic Simulations with Machine Learning,  version 3.18.2, Xiamen University, Xiamen, China, 2013-2025. [MLatom.com](http://mlatom.com).
+* Pavlo O. Dral, Fuchun Ge, Yi-Fan Hou, Yuxinxin Chen, Peikun Zheng, Bao-Xin Xue, Mikolaj Martyka, Lina Zhang, Jakub Martinka, Quanhao Zhang, Xin-Yu Tong, Arif Ullah, Sebastian V. Pios, Vignesh B. Kumar, Yanchi Ou, Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Shuang Zhang, Matheus O. Bispo, MLatom: A Package for Atomistic Simulations with Machine Learning,  version 3.18.3, Xiamen University, Xiamen, China, 2013-2025. [MLatom.com](http://mlatom.com).
                                                                       
 The citations for MLatom's interfaces and features shall be eventually
 included too. See program output, header.py, ref.json, and [MLatom.com](http://mlatom.com).          

{mlatom-3.18.2 → mlatom-3.18.3}/setup.py

@@ -43,7 +43,7 @@ with open(os.path.join(here, 'README.md')) as f:
 
 setup(
     name = pkg_name,
-    version = "3.18.2",
+    version = "3.18.3",
     author = "Pavlo O. Dral",
     author_email = "admin@mlatom.com",
     license = 'MIT (modified)',

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/EEEDDDD.py

@@ -18,6 +18,8 @@ from .EEEDDDD_nn import batch_jacobian, MLP, ModelEnsemble, StackModelEnsemble,
 from .EEEDDDD_descriptor import self_edge_vector_matrix, self_distance_matrix
 from typing import Any, List, Dict, Union, Iterable, Optional
 
+torch.set_default_dtype(torch.float32)
+
 # torch.autograd.set_detect_anomaly(True)
 
 @compile_mode("script")
@@ -278,6 +280,7 @@ class EEEDDDD(model_cls.ml_model, model_cls.torch_model,model_cls.downloadable_m
         self.species = model_dict['args'].pop('species')
         model = self._new_model(model_dict['args'])
         model.load_state_dict(model_dict['state'])
+        model = model.to(torch.float32)
         model.eval()
         if self.verbose:
             print(f"model loaded from {model_file}")
@@ -294,10 +297,18 @@ class EEEDDDD(model_cls.ml_model, model_cls.torch_model,model_cls.downloadable_m
                 kwargs['reset_base'] = reset_base
             self.model = ensemble(models=models, **kwargs)
     
-    def load_special(self,model_file):
-        model_name, model_path, download = self.check_model_path(model_file)
-        if download: self.download(model_name,model_path)
-        model_paths = [os.path.join(model_path,f'MDtrajNet-1.{ii}.pt') for ii in range(4)]
+    def load_special(self, model_file):
+
+        download_links = [
+            'https://zenodo.org/records/15877717/files/mdtrajnet1_model.zip?download=1',
+            'https://aitomistic.xyz/model/mdtrajnet1_model.zip']
+        model_dir = 'mdtrajnet1_model'
+        model_files = [f'MDtrajNet-1.{ii}.pt' for ii in range(4)]
+
+        mlatom_model_dir, to_download = self.check_model_path(model_dir, model_files)
+        if to_download: self.download(download_links, mlatom_model_dir)
+
+        model_paths = [os.path.join(mlatom_model_dir,f'MDtrajNet-1.{ii}.pt') for ii in range(4)]
         self.load(model_paths)
 
     def specie_to_index(self, species: np.ndarray) -> torch.Tensor:

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/MLTPA.py

@@ -1,11 +1,13 @@
 #!/usr/bin/env python
 # coding: utf-8
 
-
 '''
   !---------------------------------------------------------------------------! 
   ! mltpa: Machine learning prediction of the Two-photon absorption           ! 
-  ! Implementations by: Yuming Su, Yiheng Dai, Yangtao Chen, Fuchun Ge        ! 
+  ! Implementations by: Yuming Su, Yiheng Dai, Yangtao Chen, Fuchun Ge        !
+  !                                                                           !
+  ! Modified on 2025.07.30 by Yuming Su, to make it work,                     !
+  ! if one does NOT have conjugation structure                                !
   !---------------------------------------------------------------------------! 
 '''
 
@@ -345,6 +347,52 @@ class MLTPA(object):
         mol = Chem.MolFromSmiles(smiles)
         d_m = Chem.rdmolops.GetDistanceMatrix(mol)
         ring_list, f_a = self.find_conj(smiles)
+        
+        # 处理没有共轭结构的情况
+        if len(ring_list) == 0:
+            # 没有共轭结构时，最大共轭距离为0
+            feature_vector[0] = 0
+            
+            # 计算所有原子的PEOE电荷、logP和MR
+            AllChem.ComputeGasteigerCharges(mol, nIter=25)
+            peoe_charge = [mol.GetAtomWithIdx(i).GetDoubleProp('_GasteigerCharge') for i in range(mol.GetNumAtoms())]
+            contribs = rdMolDescriptors._CalcCrippenContribs(mol)
+            logp = [contribs[i][0] for i in range(len(contribs))]
+            mr = [contribs[i][1] for i in range(len(contribs))]
+            
+            # 使用整个分子的原子进行片段匹配
+            all_atoms = list(range(mol.GetNumAtoms()))
+            atom_props = [peoe_charge, logp, mr]
+            
+            for prop_idx in range(len(atom_props)):
+                atom_props_list = atom_props[prop_idx]
+                frag_x = []
+                
+                # 在整个分子中寻找匹配的片段
+                for i in range(len(mff_title)):
+                    patt = mff_title[i]
+                    f = Chem.MolFromSmarts(patt)
+                    atomids = mol.GetSubstructMatches(f)
+                    if len(atomids) > 0:
+                        for atomid_group in atomids:
+                            x_temp = 0
+                            for k in atomid_group:
+                                x_temp += atom_props_list[k]
+                            frag_x.append(x_temp)
+                
+                if len(frag_x) == 0:
+                    frag_x = [0]
+                
+                if prop_idx == 0 or prop_idx == 2:  # PEOE电荷和MR取最大值
+                    feature_vector[prop_idx + 1] = max(frag_x)
+                else:  # logP取最小值
+                    feature_vector[prop_idx + 1] = min(frag_x)
+            
+            feature_vector[4] = et30
+            feature_vector[5] = wavelength
+            return feature_vector
+        
+        # 原有的有共轭结构的处理逻辑
         lstmax = sorted(ring_list, key=len)[-1]
 
         # calculate max conju-distance

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/MLatom.py

@@ -4,7 +4,7 @@
   !---------------------------------------------------------------------------!
   !                                                                           !
   !     MLatom: a Package for Atomistic Simulations with Machine Learning     !
-  !                             MLatom 3.18.2                                 !
+  !                             MLatom 3.18.3                                 !
   !                                   @                                       !
   !                 Xiamen Atomistic Computing Suite (XACS)                   !
   !                                                                           !
@@ -38,7 +38,7 @@
   ! Max Pinheiro Jr, Yuming Su, Yiheng Dai, Yangtao Chen, Shuang Zhang,       !
   ! Matheus O. Bispo                                                          !
   ! MLatom: A Package for Atomistic Simulations with Machine Learning,        !
-  ! version 3.18.2, Xiamen University, Xiamen, China, 2013-2025.              !
+  ! version 3.18.3, Xiamen University, Xiamen, China, 2013-2025.              !
   !                                                                           !
   ! The citations for MLatom's interfaces and features shall be eventually    !
   ! included too. See header.py, ref.json and http://mlatom.com.              !

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/__init__.py

@@ -12,4 +12,4 @@ from .gap_md import gap_md, gap_model
 # add-ons
 from .models import uaiqm, omnip2x, vecmsani
 
-__version__ = '3.18.2'
+__version__ = '3.18.3'

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/aiqm1.py

@@ -200,25 +200,30 @@ class ani_nns_in_aiqm1(torchani_model, downloadable_model):
 
     def load_model(self):
         import torch
-        mlatomdir=os.path.dirname(__file__)
-        dirname = os.path.join(mlatomdir, 'aiqm1_model')
+        # mlatomdir = os.path.dirname(__file__)
+        # dirname = os.path.join(mlatomdir, 'aiqm1_model')
         method = 'aiqm1_' + self.level
         self.define_nn()
 
-        # check model file
-        # if 'dft' in method.lower():
-        #     if 'MODELSPATH' in os.environ:
-        #         modelspath = os.environ['MODELSPATH']
-        #     else:
-        #         raise ValueError(f'Please set environment variable MODELSPATH for using this method and put NN model files under $MODELSPATH/{method}_model/')
-        # dirname = os.path.join(os.environ['MODELSPATH'], f'{method}_model')
-        # if not os.path.exists(dirname):
-        #     raise ValueError(f'Please put model files under $MODELSPATH/{method}_model/')
-        model_name, model_path, download = self.check_model_path(self.method)
-        if download: self.download(model_name, model_path)
+        if method == 'aiqm1_cc':
+            download_links = [
+                'https://zenodo.org/records/15383390/files/aiqm1_cc_model.zip?download=1',
+                'https://aitomistic.xyz/model/uaiqm_odm2star_cc_20211202.zip']
+            model_dir = 'aiqm1_model'
+            model_files = [f'cv{ii}.pt' for ii in range(8)]
+
+        elif method == 'aiqm1_dft':
+            download_links = [
+                'https://zenodo.org/records/15383390/files/aiqm1_dft_model.zip?download=1',
+                'https://aitomistic.xyz/model/uaiqm_odm2star_dft_20211202.zip']
+            model_dir = 'aiqm1_dft_model'
+            model_files = [f'cv{ii}.pt' for ii in range(8)]
+
+        mlatom_model_dir, to_download = self.check_model_path(model_dir, model_files)
+        if to_download: self.download(download_links, mlatom_model_dir)
         
         # checkpoint = torch.load(os.path.join(dirname, f'{method}_cv{self.model_index}.pt'), map_location=self.device)
-        checkpoint = torch.load(os.path.join(model_path, f'cv{self.model_index}.pt'), map_location=self.device, weights_only=False)
+        checkpoint = torch.load(os.path.join(mlatom_model_dir, f'cv{self.model_index}.pt'), map_location=self.device, weights_only=False)
         self.nn.load_state_dict(checkpoint['nn'])
         import torchani
         self.model  = torchani.nn.Sequential(self.aev_computer, self.nn).to(self.device).double()

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/aiqm2.py

@@ -30,21 +30,23 @@ class aiqm2(method_model, downloadable_model):
     """ 
     
     supported_methods = ['AIQM2', 'AIQM2@DFT', 'AIQM2@DFT*']
-    # mlatomdir=os.path.dirname(__file__)
-    # dirname = os.path.join(mlatomdir, 'aiqm2_model')
 
     def __init__(
         self,
         method: str = 'AIQM2',
         working_directory: str = '.',
-        qm_program_kwargs: dict = {},
+        baseline_kwargs: dict = None,
+        dispersion_kwargs: dict = None,
         nthreads: int = 1
     ):
 
         self.method = method.lower()
         self.model_name = self.method.replace('@','at')
         self.working_directory = working_directory
-        self.qm_program_kwargs = qm_program_kwargs
+        if baseline_kwargs is None: self.baseline_kwargs = {}
+        else: self.baseline_kwargs = baseline_kwargs
+        if dispersion_kwargs is None: self.dispersion_kwargs = {}
+        else: self.dispersion_kwargs = dispersion_kwargs
         self.load()
         self.nthreads = nthreads
 
@@ -65,13 +67,13 @@ class aiqm2(method_model, downloadable_model):
         calculate_energy_gradients=False, 
         calculate_hessian=False,
         calculate_dipole_derivatives=False,
+        calculate_polarizability_derivatives=False,
     ):
-
         molDB = super().predict(molecular_database=molecular_database, molecule=molecule)
         for mol in molDB.molecules:
             self.predict_for_molecule(molecule=mol,
                                 calculate_energy=calculate_energy, calculate_energy_gradients=calculate_energy_gradients, calculate_hessian=calculate_hessian,
-                                calculate_dipole_derivatives=calculate_dipole_derivatives)
+                                calculate_dipole_derivatives=calculate_dipole_derivatives,calculate_polarizability_derivatives=calculate_polarizability_derivatives,)
 
     def predict_for_molecule( # no specific treatment to atomic energies currently
         self,
@@ -80,6 +82,7 @@ class aiqm2(method_model, downloadable_model):
         calculate_energy_gradients=False, 
         calculate_hessian=False,
         calculate_dipole_derivatives=False,
+        calculate_polarizability_derivatives=False,
     ):
 
         for atom in molecule.atoms:
@@ -90,7 +93,7 @@ class aiqm2(method_model, downloadable_model):
         self.aiqm2_model.predict(
             molecule=molecule,
             calculate_energy=calculate_energy, calculate_energy_gradients=calculate_energy_gradients, calculate_hessian=calculate_hessian, 
-            calculate_dipole_derivatives=calculate_dipole_derivatives,
+            calculate_dipole_derivatives=calculate_dipole_derivatives,calculate_polarizability_derivatives=calculate_polarizability_derivatives,
         )
 
         molecule.__dict__[f'{self.model_name}_nn'].standard_deviation(properties=['energy'])
@@ -98,19 +101,32 @@ class aiqm2(method_model, downloadable_model):
     def load(self):
         from .models import methods
 
-        model_name, model_path, download = self.check_model_path(self.method)
-        if download: self.download(model_name, model_path)
+        if self.model_name.lower() == 'aiqm2':
+            download_links = [
+                'https://zenodo.org/records/15383333/files/aiqm2_cc_model.zip?download=1',
+                'https://aitomistic.xyz/model/uaiqm_gfn2xtbstar_cc_20240106.zip']
+            model_dir = 'aiqm2_model'
+            model_files = [f'cv{ii}.pt' for ii in range(8)]
+        elif self.model_name.lower() in ['aiqm2atdft', 'aiqm2atdft*']:
+            download_links = [
+                'https://zenodo.org/records/15383333/files/aiqm2_dft_model.zip?download=1',
+                'https://aitomistic.xyz/model/uaiqm_gfn2xtbstar_dft_20240106.zip']
+            model_dir = 'aiqm2_dft_model'
+            model_files = [f'cv{ii}.pt' for ii in range(8)]
+
+        mlatom_model_dir, to_download = self.check_model_path(model_dir, model_files)
+        if to_download: self.download(download_links, mlatom_model_dir)
         
-        model_paths = [os.path.join(model_path, f'cv{ii}.pt') for ii in range(8)]
+        model_paths = [os.path.join(mlatom_model_dir, model_files[ii]) for ii in range(8)]
 
         baseline = model_tree_node(
             name='gfn2xtbstar',
-            model=methods(method='GFN2-xTB*', **self.qm_program_kwargs),
+            model=methods(method='GFN2-xTB*', working_directory=self.working_directory,**self.baseline_kwargs),
             operator='predict'
         )
         d4 = model_tree_node(
             name='d4wb97x',
-            model=methods(method='D4', functional='wb97x', working_directory=self.working_directory),
+            model=methods(method='D4', functional='wb97x', working_directory=self.working_directory, **self.dispersion_kwargs),
             operator='predict'
         )
         from .interfaces.torchani_interface import ani
@@ -121,6 +137,8 @@ class aiqm2(method_model, downloadable_model):
             def predict(self,**kwargs):
                 if 'calculate_dipole_derivatives' in kwargs.keys():
                     del kwargs['calculate_dipole_derivatives']
+                if 'calculate_polarizability_derivatives' in kwargs.keys():
+                    del kwargs['calculate_polarizability_derivatives']
                 super().predict(**kwargs)
         nn = model_tree_node(
             name=f'{self.model_name}_nn',

mlatom-3.18.3/src/mlatom/aiqm3.py

@@ -0,0 +1,166 @@
+import os
+from . import data, models, constants
+from .model_cls import method_model, model_tree_node, downloadable_model
+
+class aiqm3(method_model, downloadable_model):
+
+    """ 
+    GFN2-xTB based artificial intelligence quantum-mechanical method 3 (AIQM3)
+
+    Arguments:
+
+        method (str, optional): Currently supports AIQM3, AIQM3@DFT, AIQM3@DFT*
+        working_directory (str, optional): The path to save temporary calculation file. By default, current directory will be choosed.
+        baseline_kwargs (dict, optional): Keywords passed to GFN2-xTB
+        dispersion_kwargs (dict, optional): 
+
+    .. code-block:: 
+
+        # Initialize molecule
+        mol = ml.data.molecule()
+        mol.read_from_xyz_file(filename='ethanol.xyz')
+        # Run AIQM2 calculation
+        aiqm3 = ml.methods(method='aiqm3')
+        aiqm3.predict(molecule=mol, calculate_energy=True, calculate_energy_gradients=True, calculate_hessian=True)
+        # Get energy, gradient, and uncertainty of AIQM2 
+        energy = mol.energy
+        gradient = mol.get_energy_gradients()
+        hess = mol.hessian
+        std = mol.aiqm3_model_nn.energy_standard_deviation
+
+    """ 
+    
+    supported_methods = ['AIQM3', 'AIQM3@DFT', 'AIQM3@DFT*']
+
+    def __init__(
+        self,
+        method: str = 'AIQM3',
+        working_directory: str = None,
+        baseline_kwargs: dict = None,
+        dispersion_kwargs: dict = None,
+        nthreads: int = 1
+    ):
+
+        self.method = method.lower()
+        self.model_name = self.method.replace('@','at')
+        self.working_directory = working_directory
+        if baseline_kwargs is None: self.baseline_kwargs = {}
+        else: self.baseline_kwargs = baseline_kwargs
+        if dispersion_kwargs is None: self.dispersion_kwargs = {}
+        else: self.dispersion_kwargs = dispersion_kwargs
+        
+        self.load()
+        self.nthreads = nthreads
+
+    @property
+    def nthreads(self):
+        return self._nthreads
+
+    @nthreads.setter
+    def nthreads(self, value):
+        self._nthreads = value
+        self.aiqm3_model.nthreads = self._nthreads
+
+    def predict(
+        self, 
+        molecular_database=None, 
+        molecule=None,
+        calculate_energy=True, 
+        calculate_energy_gradients=False, 
+        calculate_hessian=False,
+        calculate_dipole_derivatives=False,
+    ):
+
+        molDB = super().predict(molecular_database=molecular_database, molecule=molecule)
+        for mol in molDB.molecules:
+            self.predict_for_molecule(molecule=mol,
+                                calculate_energy=calculate_energy, calculate_energy_gradients=calculate_energy_gradients, calculate_hessian=calculate_hessian,
+                                calculate_dipole_derivatives=calculate_dipole_derivatives)
+
+    def predict_for_molecule( # no specific treatment to atomic energies currently
+        self,
+        molecule=None,
+        calculate_energy=True, 
+        calculate_energy_gradients=False, 
+        calculate_hessian=False,
+        calculate_dipole_derivatives=False,
+    ):
+
+        for atom in molecule.atoms:
+            if not atom.atomic_number in [1, 6, 7, 8, 9, 16, 17 ]:
+                errmsg = ' * Warning * Molecule contains elements other than H, C, N, O, F, S, Cl, no calculations performed'
+                raise ValueError(errmsg)
+
+        self.aiqm3_model.predict(
+            molecule=molecule,
+            calculate_energy=calculate_energy, calculate_energy_gradients=calculate_energy_gradients, calculate_hessian=calculate_hessian, 
+            calculate_dipole_derivatives=calculate_dipole_derivatives,
+        )
+
+        molecule.__dict__[f'{self.model_name}_nn'].standard_deviation(properties=['energy'])
+
+    def load(self):
+        from .models import methods
+
+        if self.model_name.lower() == 'aiqm3':
+            download_links = [
+                'https://zenodo.org/records/15876694/files/aiqm3_cc_model.zip?download=1',
+                'https://aitomistic.xyz/model/uaiqm_gfn2xtbstar_cc_20250115.zip']
+            model_dir = 'aiqm3_model'
+            model_files = [f'cv{ii}.pt' for ii in range(8)]
+
+        elif self.model_name.lower() in ['aiqm3atdft', 'aiqm3atdft*']:
+            download_links = [
+                'https://zenodo.org/records/15876694/files/aiqm3_dft_model.zip?download=1',
+                'https://aitomistic.xyz/model/uaiqm_gfn2xtbstar_dft_20240619.zip']
+            model_dir = 'aiqm3_dft_model'
+            model_files = [f'cv{ii}.pt' for ii in range(8)]
+
+        mlatom_model_dir, to_download = self.check_model_path(model_dir, model_files)
+        if to_download: self.download(download_links, mlatom_model_dir)
+        
+        model_paths = [os.path.join(mlatom_model_dir, model_files[ii]) for ii in range(8)]
+
+        baseline = model_tree_node(
+            name='gfn2xtbstar',
+            model=methods(method='GFN2-xTB*', working_directory=self.working_directory, **self.baseline_kwargs),
+            operator='predict'
+        )
+        d3 = model_tree_node(
+            name='d3b973c',
+            model=methods(method='d3bj', functional='b973c', working_directory=self.working_directory, **self.dispersion_kwargs),
+            operator='predict'
+        )
+
+        from .interfaces.torchani_interface import ani
+        class ani_wrapper(ani):
+            def __init__(self,**kwargs):
+                super().__init__(**kwargs)
+
+            def predict(self,**kwargs):
+                if 'calculate_dipole_derivatives' in kwargs.keys():
+                    del kwargs['calculate_dipole_derivatives']
+                super().predict(**kwargs)
+
+        nn = model_tree_node(
+            name=f'{self.model_name}_nn',
+            children=[
+                model_tree_node(
+                    name=f'{self.model_name}_nn_{ii}',
+                    model=ani_wrapper(
+                        model_file=model_paths[ii],
+                        verbose=0),
+                    operator='predict'
+                ) for ii in range(8)
+            ],
+            operator='average'
+        )
+
+        self.aiqm3_model = model_tree_node(
+            name=self.model_name,
+            children=[baseline, nn, d3],
+            operator='sum'
+        )
+
+
+

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/gen_ts_methods.py

@@ -22,6 +22,7 @@ def gen_ts_ects(
 
     # check device and set local rank
     import torch
+    torch.set_default_dtype(torch.float32)
     if 'device' in program_kwargs: device = program_kwargs['device']
     else: device = 'cuda' if torch.cuda.is_available() else 'cpu'
     if device.lower() not in ['cuda', 'cpu']:
@@ -43,7 +44,7 @@ def gen_ts_ects(
     
     # load module
     from EcTs.comparm import GP,Loaddict2obj
-    from EcTs.utils import xyz2mol
+    from EcTs.utils import xyz2mol, xyz2AC
     from EcTs.graphs import RP_pair
     from EcTs.model import EcTs_Model
 
@@ -56,7 +57,13 @@ def gen_ts_ects(
 
     rmol = xyz2mol(ratoms,rxyz,charge=rcharge)[0]
     pmol = xyz2mol(patoms,pxyz,charge=pcharge)[0]
-    rp_pair = RP_pair(rmol=rmol,pmol=pmol,idx='ects')
+    radjs,rmol=xyz2AC(ratoms,rxyz,charge=rcharge)
+    padjs,pmol=xyz2AC(patoms,pxyz,charge=pcharge)
+    rp_pair = RP_pair(
+        rmol=rmol,pmol=pmol,
+        radjs=radjs, padjs=padjs,
+        rcoords=rxyz, pcoords=pxyz,
+        idx='ects')
 
     ects_model = EcTs_Model(modelname="EcTs_Model",local_rank=local_rank)
     

{mlatom-3.18.2 → mlatom-3.18.3}/src/mlatom/header.py

@@ -143,6 +143,8 @@ def printHeader(args):
     if args.method:
         if 'DM21' in args.method.upper():
             refItems['DM21'] = refs['DM21']
+        if 'mace-off' in args.method.lower():
+            refItems['MACE-OFF'] = refs['MACE-OFF']
     
     if args.ani1xnr:
         refItems['ANI-1xnr'] = refs['ANI-1xnr']