karpathy

"""
The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
This file is the complete algorithm.
Everything else is just efficiency.

@karpathy
"""

import os       # os.path.exists
import math     # math.log, math.exp

fschiettecatte

High Traffic Server Settings on RHEL / AlmaLinux / Rocky / CentOS 8-10

I recently did some work to optimize the network configuration of an AlmaLinux 9 based web server that receives a lot of traffic.

Of course these settings also apply to 8, and I think these should also work on 10 as well but I have not yet tested them.

There is a lot of information on the web for this and it distills down to a minimum recommended configuration, and a recommended configuration.

The minimum recommended configuration should be sufficient for servers with less than 10Gb, and the recommended configuration should be sufficient for servers with 10Gb or more.

zhuizhuhaomeng

package main

import (
    "fmt"
    "golang.org/x/net/http2/hpack"
)

func main() {
    fmt.Println("nginx", "→", Encode("MYWS"))
    fmt.Println("nginx", "→", Encode("nginx"))

quasilyte

package benchmark

import (
	"context"
	"runtime"
	"sync"
	"sync/atomic"
	"testing"
	"time"
)

arnabdas

#!/bin/bash

add-apt-repository -y ppa:hvr/ghc
apt-get install -y cabal-install-2.4

apt install ghc-8.6

curl -sSL https://get.haskellstack.org/ | sh

# to install intero, either install

yudai

Problem

Now tmux, mosh and iTerm2 support the OSC 52 sequence that enables clipboard sharing. However, there is a trap that prevents them from working together.

Mosh accepts OSC 52 sequences with the c; option. However, tmux doesn't send that option when it emits OSC 52 sequences, which means you cannot use tmux and mosh together with the default configuration.

Workaround

You can override the OSC 52 sequence generated by tmux by adding the following line to your tmux.conf.

tombigel

How to Change Open Files Limit on OS X and macOS

This text is the section about OS X Yosemite (which also works for macOS Sierra) from https://docs.basho.com/riak/kv/2.1.4/using/performance/open-files-limit/#mac-os-x

The last time i visited this link it was dead (403), so I cloned it here from the latest snapshot in Archive.org's Wayback Machine https://web.archive.org/web/20170523131633/https://docs.basho.com/riak/kv/2.1.4/using/performance/open-files-limit/

Mac OS X

To check the current limits on your Mac OS X system, run:

itamarhaber

SHOBJ_CFLAGS ?= -fno-common -g -ggdb
SHOBJ_LDFLAGS ?= -shared -Bsymbolic
CFLAGS = -Wall -g -fPIC -lc -lm -Og -std=gnu99  
CC=gcc

all: example.so

example.so: example.o
	$(LD) -o $@ example.o $(SHOBJ_LDFLAGS) $(LIBS) -lc 

karpathy

""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym

# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward

lilydjwg

#!/bin/bash -e

NETNS_FILE=/var/run/netns/mynet
MNTNS_FILE=/var/run/ns/mynet_mnt

if [[ ! -f $NETNS_FILE ]]; then
  ip netns add mynet

  ip link add mynet0 type veth peer name mynet1
  ip link set mynet0 up