freifrauvonbleifrei

<script src="https://unpkg.com/leaflet@1.6.0/dist/leaflet.js"></script>
<link href="https://unpkg.com/leaflet@1.6.0/dist/leaflet.css" rel="stylesheet"/>


<script>
function cloneMarker(markerToClone) {
  var marker = L.marker(markerToClone.getLatLng()).bindPopup(markerToClone.getPopup());
  return marker;
}  

freifrauvonbleifrei

Checkliste Bauvorhaben in der Lokalpolitik

Oh oh eine neue Vorlage der Stadt zu einem Bauvorhaben. Wie prüfe ich das kritisch?

Beschreibung

• was soll gemacht werden?
• was ist da bisher?
• mit welchen zwingenden Gründen wird die Maßnahme begründet? Ist jeder einzelne stichhaltig?
  • Im Zweifelsfall bei dienstälteren Mitgliedern nachfragen, wie sie das bewerten.
• wurden alternative Pläne ernsthaft geprüft?

freifrauvonbleifrei

# Based on NERSC Cori config
# These package should be upgraded whenever there is a CDT or PE upgrade
packages:
    all:
        # default compilers defined by the system
        # these reflect the current installed PE
        compiler: [gcc@8.3.0, cce@9.0.2, intel/19.0.1.144, pgi/19.7.0]
        providers:
            mpi: [mpich]
            mkl: [intel-mkl]

freifrauvonbleifrei

compilers:
- compiler:
    environment: {
      unset:
        F77
    }
    target: x86_64
    spec: intel@19.0.5.281
    paths:
      cc: /lrz/sys/intel/studio2019_u5/compilers_and_libraries_2019.5.281/linux/bin/intel64/icc

freifrauvonbleifrei

Getting github access on a firewalled system

This approach, based on [0] and [1] lets me reverse-tunnel through the local machine, to get github (and other) access on protected machines, such as HPC compute systems.

I put these lines in my local ~/.ssh/config file

Host somehpcsocks
 ProxyCommand ssh -D 2020 localhost nc -q 1 localhost 22

freifrauvonbleifrei

from __future__ import division
import os
import numpy as np
import pandas as pd
import math

import sys

# In[5]:

freifrauvonbleifrei

import numpy as np

def bound_combinations(lmin, lmax):
    """
    example: bound_combinations([1,1], [2,3]) == [[1,1], [1,2], [1,3], [2,1], [2,2], [2,3]]
    """
    assert len(lmax) == len(lmin)
    d = len(lmax)
    # get all the ranges as slices
    ranges = tuple(slice(lmin[i],lmax[i]+1, 1) for i in range(llen))

freifrauvonbleifrei

#from tensorflow internals
def _safe_mean(losses, num_present):
    """Computes a safe mean of the losses.
    Args:
    losses: `Tensor` whose elements contain individual loss measurements.
    num_present: The number of measurable elements in `losses`.
    Returns:
    A scalar representing the mean of `losses`. If `num_present` is zero,
      then zero is returned.

freifrauvonbleifrei

# custom combined loss cf https://towardsdatascience.com/custom-tensorflow-loss-functions-for-advanced-machine-learning-f13cdd1d188a
loss_1 = tf.losses.huber_loss

def loss_2(labels, predictions):
    k = tf.clip_by_value(predictions, -np.inf, 0)
#         k = tf.cond(predictions < 0, lambda: 1 * -predictions, lambda: 0)
    loss = tf.reduce_sum(-k)
    return loss

negativity_loss_rate = 0.1