v1.2.0

New features of dp train:

• Polarizability and dipole fitting
• If provided with stop_lr, the decay_rate will be computed automatically
• Support non-pbc system: add an empty file named nopbc to the data system.
• Use envs TF_INTRA_OP_PARALLELISM_THREADS and TF_INTER_OP_PARALLELISM_THREADS to control the multi-threading of tf, clean up command line options.
• When the key systems is provided with a string, all possible systems will be recursively searched within the dir given by systems
• User specific atomic energy.
• Exclude types when building descriptors.
• User specific activation function and network precision.
• Transform neural network parameters.

New features of dp test:

• Recursively search all possible systems in a directory.
• Weighted average of test results

Bug fixings:

• Cuda version compatibility
• raw_to_set when nframe == 1
• Slow test when testing multiple systems
• Potentially wrong neighbor list when number of neighbors is larger than 256.
• Fix bug #163 : broken lib link of dp_ipi
• Infering number of types from data

v1.1.4

Big fixings:

• KeyError in dp test (#165 )
• Protect the std calculation

v1.1.3

Bug fixings:

• raw_to_set.sh does not work for data system that has only one frame.
• dp test ignores type_map.raw
• cannot compile lammps with deepmd-kit that does not use GPU
• compiling problems in debug mode
• GPU kernel threading bug that leads to wrong model deviation

v1.1.2

New features:

• GPU supported descriptor computation
• Support ASE evaluation of deep potential
• Trainable (true/false) descriptors

Bug fixings:

• import warnings in common.py
• relative force model deviation
• freezing bug in se_ar

v1.1.1

Bug fixing:

• pass the warning of upper/lower index bound but do not correct the index.

v1.1.0

New features:

• Atom parameter
• More robust data statistics
• Relative force loss

Code structure:

• Implement WFC and Polarizability by TensorModel
• Select the appropriate shared library suffix based on the operating system

Bug fixings:

• The minimal model devi when a proc has none local atom.
• Correctly unset USE_CUDA_TOOLKIT