Qihong Lu qihongl

Installing emergent

The current process for setting up the new emergent is distributed across several webpages and can be a bit confusing or even contradictory. This is an attempt to put it all in one place, potentially into simpler terms, and with troubleshooting tips informed by my own experience getting everything running.

Download Go

The first step to setting up emergent is to download Go. Pick your favored binary release here, download it, and run it. The MSI installers for Windows and Mac do all the work for you.

Test Your Go Installation

You might want to make sure you installed Go successfully. To do this:

Head to your Go workspace directory. The MacOS package installer places the Go distribution to /usr/local/go while the Windows once places it at c:\Go.

	import numpy as np
	import matplotlib.pyplot as plt
	import seaborn as sns

	# Parameters
	input_size = 100 # Number of input neurons
	output_size = 100 # Number of output neurons
	num_samples = 1000 # Number of input samples
	sparsity_levels = np.linspace(0.1, 1.0, 10) # Connectivity levels (from sparse to dense)

	import numpy as np
	import matplotlib.pyplot as plt
	import seaborn as sns


	sns.set(style='white', palette='colorblind', context='talk')
	cpal = sns.color_palette('colorblind')


	import numpy as np

	import numpy as np
	from scipy.linalg import qr
	import matplotlib.pyplot as plt
	import seaborn as sns
	sns.set(style='white', palette='colorblind', context='talk')
	cpal = sns.color_palette()

	n_pts = 3
	n_dim = 2 # number of hidden units in the simulation

	import numpy as np
	import matplotlib.pyplot as plt
	import seaborn as sns
	sns.set(style='white', palette='colorblind', context='talk')

	np.random.seed(1)

	# Define the function (error surface)
	def f(x, y):
	return x2 + y2 # Example of a simple quadratic error surface

	squeue -u qlu \| awk '
	BEGIN {
	abbrev["R"]="(Running)"
	abbrev["PD"]="(Pending)"
	abbrev["CG"]="(Completing)"
	abbrev["F"]="(Failed)"
	}
	NR>1 {a[$5]++}
	END {
	for (i in a) {

	#!/bin/bash

	exp_name=0916-widesim-prandom
	n_epoch=1000
	n_def_tps=0
	enc_size=16
	similarity_max=.9
	similarity_min=0

	p_rm_ob_enc=0.3

	#!/bin/bash
	#SBATCH -t 13:55:00
	#SBATCH -c 1
	#SBATCH --mem-per-cpu 4G

	#SBATCH --job-name=lcarnn
	#SBATCH --mail-type=FAIL
	#SBATCH [email protected]
	#SBATCH --output slurm_log/lcarnn-%j.log

	# general shortcuts
	alias ls='ls -GFh'
	alias l='ls -ltr'
	alias f='open -a Finder ./'
	alias cppath='pwd\|pbcopy'
	alias cdgit='cd /Users/Qihong/Dropbox/github'
	alias cdnorman='cd /Volumes/norman/qlu'

	# customize the terminal
	export PS1="\[\033[36m\]\u\[\033[m\]@:\[\033[33;1m\]\W\[\033[m\]\$ "

	import numpy as np
	import pandas as pd
	import dabest


	cond_means = {'control': 0, 'test': 2}
	scale = 1
	n_data = 64
	data_dict = {
	cond: np.random.normal(loc=cond_means[cond], scale=scale, size=(n_data,))