Skip to content

Instantly share code, notes, and snippets.

View MartinNowak's full-sized avatar
🎯
Focusing

dawg MartinNowak

🎯
Focusing
View GitHub Profile
@MartinNowak
MartinNowak / dnf-automatic-restart
Last active August 28, 2023 21:31
dnf-automatic restart hook
#!/usr/bin/env bash
set -ueEo pipefail
if ! dnf needs-restarting --reboothint; then
systemctl reboot --message='Rebooting after dnf-automatic updates'
fi
mapfile -t services < <(dnf needs-restarting --services)
if [[ "${#services[*]}" -ne 0 ]]; then
@MartinNowak
MartinNowak / convert_saved_model.py
Created November 28, 2019 16:17
Convert TF saved model from protobuf binary to text format
import os, sys
import google.protobuf
from tensorflow.core.protobuf import saved_model_pb2
import tensorflow as tf
def convert_saved_model_to_pbtxt(path):
saved_model = saved_model_pb2.SavedModel()
with open(os.path.join(path, 'saved_model.pb'), 'rb') as f:
@MartinNowak
MartinNowak / BvSshServer-Settings.wst
Last active April 14, 2023 03:52
Preparing a Win7x64 base box.
@MartinNowak
MartinNowak / release.sh
Last active October 17, 2022 12:29
Orchestration for dlang release building
#!/usr/bin/env bash
set -ueo pipefail
set -x
ROOT="$PWD"
LATEST_D_VER=v$(curl -fsS http://downloads.dlang.org/releases/LATEST)
BUILD_LDC_VER=v1.30.0
D_VER=v2.101.0-beta.1
@MartinNowak
MartinNowak / OSX-10.8.ova
Last active October 11, 2022 21:16
curl follow redirects
@MartinNowak
MartinNowak / fir.py
Created April 28, 2022 10:31
FIR low-pass filter
#!python
from numpy import cos, sin, pi, absolute, arange
from scipy.signal import windows, kaiserord, lfilter, firwin, freqz
from pylab import figure, clf, plot, xlabel, ylabel, xlim, ylim, title, grid, axes, show
N = window_size_in_days = 20
#------------------------------------------------
@MartinNowak
MartinNowak / Scikit-HTS.py
Created March 11, 2022 15:46
hts for m5 dataset
#!/usr/bin/env python
# coding: utf-8
# # scikit-hts showcase: [M5 kaggle competition](https://www.kaggle.com/c/m5-forecasting-accuracy)
#
# In this notebook we will use the data from the Kaggle competiton to perform some hierarchical forecasting. The problem is particularly well suited for the library, as there is a clear hierarchical relationship between each of the series: we have states, stores, categories, departments, and items; sum of sales of items resolve to departments, which summed resolved to categories and so forth.
#
#
# We will however limit the scope of the forecasting task to producing forecasts at the
# department level, rather than going to the full extent and forecasting for single items.

send aws credentials

gpg –keyring $PWD/d-keyring.gpg –no-default-keyring –list-keys –with-fingerprint

Bump ldc version with major dlang releases, keep it for minor dlang release

Bump D version (to -beta.1)

Bump Dub version (mostly only with major release, compare e.g. dub v1.26.0…stable to check for bugfixes and minor releases)

We do -beta.1 for all releases and an additional -rc.1 for major releases

(status quo: 2 weeks beta+rc phase for major, 1 week beta for minor releases)

@MartinNowak
MartinNowak / terraform_update.sh
Last active November 29, 2021 19:03
script to update terraform binary to (latest) version
#!/usr/bin/env bash
set -ueo pipefail
function cleanup {
rm -rf "$TMP"
}
TMP=$(mktemp -d)
trap cleanup EXIT
cd "$TMP"
import std.stdio, std.variant;
//==============================================================================
// registry for query field names
template registerQuery(T, string file, size_t line)
{
__gshared string[] fields;
}