Skip to content

Instantly share code, notes, and snippets.

View yueyericardo's full-sized avatar

Jinze Xue yueyericardo

View GitHub Profile

300k atoms 8 GPUs

gpu/ware off

Loop time of 30.3401 on 8 procs for 2000 steps with 300003 atoms

Performance: 2.848 ns/day, 8.428 hours/ns, 65.919 timesteps/s
77.5% CPU use with 8 MPI tasks x 1 OpenMP threads
====================================================================== erf ======================================================================
/opt/conda/lib/python3.8/site-packages/torch/profiler/profiler.py:395: UserWarning: use_cuda is deprecated, use activities argument instead
warn("use_cuda is deprecated, use activities argument instead")
------------------------------------------------------- ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------
Name Self CPU % Self CPU CPU total % CPU total CPU time avg Self CUDA Self CUDA % CUDA total CUDA time avg # of Calls
------------------------------------------------------- ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------
void at::native::vectorized_elementwise_kernel<4, at... 0.00% 0.000us
import time
import argparse
from functorch import vmap, jacrev, jacfwd
import torch
import torch.nn as nn
torch.backends.cuda.matmul.allow_tf32 = False
_ = torch.manual_seed(0)
Compiler Toolset Versions Currently Tested
XCode 13.2.1 & 12.5.1
GCC 111 & 10[^2]
Clang 12[^2] & 11[^2]
Visual Studio with MSVC VS2022[^3] & VS2019[^4]
Visual Studio with LLVM VS2022[^3] & VS2019[^4]

Footnotes

  1. Precise version may be found in the latest CI results.

#!/usr/bin/env python
total = 10
def info(text):
print('\033[32m{}\33[0m'.format(text)) # green
def cal(points):
eq = points.replace(' ', ' + ')
eq = f'{total} - ({eq})'
from openmm.app import *
from openmm import *
from openmm.unit import *
from simtk import openmm, unit
from openmmtools import testsystems
water_box = testsystems.WaterBox(box_edge=20.0*unit.nanometer)
positions = water_box.positions
atoms = list(water_box.topology.atoms())
elements = [a.element.symbol for a in atoms]
from torchani.datasets import AniH5Dataset
import os
import cudf
try:
path = os.path.dirname(os.path.realpath(__file__))
except NameError:
path = os.getcwd()
h5file = os.path.join(path, '../dataset/ani-1x/sample.h5')

快捷键

Spark

  • Cmd + M: Mark as Read/Unread
  • Cmd + Shift + P: Pin

Magnet

  • U I J K: 四分之一
  • D F G: 三分之一
  • E T: 三分之二
# Configuration file for jupyter-notebook.
#------------------------------------------------------------------------------
# Application(SingletonConfigurable) configuration
#------------------------------------------------------------------------------
## This is an application.
## The date format used by logging formatters for %(asctime)s
#c.Application.log_datefmt = '%Y-%m-%d %H:%M:%S'