Takahiro Ueda tueda

Undocumented environment variables

External channels

FORM_PIPES

ParFORM

PF_LOG
PF_RBUFS

my-madminer-jupyter-env

Docker is required.
NVIDIA Container Toolkit is required to use GPU. See this.

Usage:

git clone https://gist.github.com/ac4a004de14b34e4e746e6412cf70249.git my-madminer-jupyter-env

!pip install git+https://gist.github.com/tueda/ec0181f0f4c8961d49dc659f79cbfd4a

from google.colab import drive
drive.mount('/content/drive')

import colab_cache

cache = colab_cache.Cache('hello-2.12.1')
if cache.exists():

MicrOMEGAs 5.3.35 on Homebrew/CentOS7

In short, change compilers and flags in CalcHEP_src/FlagsForSh:

CC="gcc-12"
CFLAGS="-g -O2 -fcommon -fsigned-char -std=gnu99 -fPIC"

FC="gfortran-12"
FFLAGS="-g -O2 -fno-automatic"

TensorFlow 2.4.4 build error on an old machine (CentOS6, Xeon X5690):

ERROR: An error occurred during the fetch of repository 'local_execution_config_python':
   Traceback (most recent call last):
        File "/home/tueda/build/tensorflow/third_party/py/python_configure.bzl", line 209
                _check_python_bin(<2 more arguments>)
        File "/home/tueda/build/tensorflow/third_party/py/python_configure.bzl", line 143, in _check_python_bin
                get_bash_bin(repository_ctx)
        File "/home/tueda/build/tensorflow/third_party/remote_config/common.bzl", line 75, in get_bash_bin
                which(repository_ctx, "bash")

	#!/bin/sh
	#
	# @file auto-cherry-pick.sh
	#
	# Usage:
	# auto-cherry-pick.sh ref1...ref2:
	#
	set -eu

	output=auto-cherry-pick.log

	chunkifyString[s_String, n_Integer] :=
	StringJoin @@@ Partition[Characters[s], UpTo[n]];

	writeRationalApproximation[
	stream_, x_, n_, width_:72
	] := Module[{s, lines, i},
	s = Rationalize[x, Abs[x * 10^(-n)]] // InputForm // ToString;
	lines = chunkifyString[s, width - 1];
	Do[
	s = lines[[i]];

	#!/bin/bash
	#
	# @file run-mg5_aMC.sh
	#
	# Runs MadGraph5_aMC@NLO. This script downloads and
	# sets up MadGraph5_aMC@NLO if needed.
	#
	set -eu
	set -o pipefail