Skip to content

Commit

Permalink
Added __init__.py for cc and updated readme (pyscf#120)
Browse files Browse the repository at this point in the history
* solvent for unrestricted KS

* updated dft_driver.py

* raise error if not hf or uhf

* fixed a bug in SMD gradient and Hessian

* v0.7.3

* updated readme && __init__ for cc

* fixed circular import

* support sm90
  • Loading branch information
wxj6000 authored Mar 8, 2024
1 parent b2caee7 commit 8161a89
Show file tree
Hide file tree
Showing 5 changed files with 25 additions and 5 deletions.
23 changes: 21 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,7 @@ Features
- Nonlocal functional correction (vv10) for SCF and gradient;
- ECP is supported and calculated on CPU;
- PCM models, SMD model, their analytical gradients, and semi-analytical Hessian matrix;
- Unrestricted Hartree-Fock and Unrestricted DFT, gradient, and Hessian

Limitations
--------
Expand Down Expand Up @@ -87,7 +88,25 @@ h = mf.Hessian()
h_dft = h.kernel() # compute analytical Hessian

```
Find more examples in gpu4pyscf/examples

`to_gpu` is supported since PySCF 2.5.0
```python
import pyscf
from pyscf.dft import rks

atom ='''
O 0.0000000000 -0.0000000000 0.1174000000
H -0.7570000000 -0.0000000000 -0.4696000000
H 0.7570000000 0.0000000000 -0.4696000000
'''

mol = pyscf.M(atom=atom, basis='def2-tzvpp')
mf = rks.RKS(mol, xc='LDA').density_fit().to_gpu() # move PySCF object to GPU4PySCF object
e_dft = mf.kernel() # compute total energy

```

Find more examples in [gpu4pyscf/examples](https://github.com/pyscf/gpu4pyscf/tree/master/examples)

Benchmarks
--------
Expand All @@ -108,5 +127,5 @@ Speedup with GPU4PySCF v0.6.0 on A100-80G over Q-Chem 6.1 on 32-cores CPU (Desit
| 095_Azadirachtin | 95 | 5.58 | 7.72 | 24.18 | 26.84 | 25.21 |
| 113_Taxol | 113 | 5.44 | 6.81 | 24.58 | 29.14 | nan |

Find more benchmarks in gpu4pyscf/benchmarks
Find more benchmarks in [gpu4pyscf/benchmarks](https://github.com/pyscf/gpu4pyscf/tree/master/benchmarks)

2 changes: 1 addition & 1 deletion builder/setup_libxc.py
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@
DOWNLOAD_URL = None
CLASSIFIERS = None
PLATFORMS = None
VERSION = '0.2'
VERSION = '0.3'

def get_cuda_version():
nvcc_out = subprocess.check_output(["nvcc", "--version"]).decode('utf-8')
Expand Down
2 changes: 1 addition & 1 deletion gpu4pyscf/__init__.py
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
from . import lib, grad, hessian, solvent, scf, dft

__version__ = '0.7.3'
__version__ = '0.7.4'

# monkey patch libxc reference due to a bug in nvcc
from pyscf.dft import libxc
Expand Down
1 change: 1 addition & 0 deletions gpu4pyscf/cc/__init__.py
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
from .cc import ccsd_incore
2 changes: 1 addition & 1 deletion gpu4pyscf/lib/CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ set(CMAKE_C_STANDARD "99")
if(DEFINED CUDA_ARCHITECTURES)
set(CMAKE_CUDA_ARCHITECTURES "${CUDA_ARCHITECTURES}")
else()
set(CMAKE_CUDA_ARCHITECTURES "70-real;80-real")
set(CMAKE_CUDA_ARCHITECTURES "70-real;80-real;90-real")
endif()
message("CUDA_ARCHITECTURES: ${CMAKE_CUDA_ARCHITECTURES}")

Expand Down

0 comments on commit 8161a89

Please sign in to comment.