Skip to content

Instantly share code, notes, and snippets.

@mrocklin
Created April 10, 2017 22:29
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 2 You must be signed in to fork a gist
  • Save mrocklin/4fc08482f33d60cc90cc3f8723146de5 to your computer and use it in GitHub Desktop.
Save mrocklin/4fc08482f33d60cc90cc3f8723146de5 to your computer and use it in GitHub Desktop.
Display the source blob
Display the rendered blob
Raw
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Imports and initial setup"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import dask.array as da\n",
"import pandas as pd\n",
"import numpy as np\n",
"import time\n",
"import random\n",
"\n",
"from dask import delayed, persist\n",
"from dask.distributed import Client, as_completed, wait\n",
"from dask_glm.algorithms import local_update\n",
"from toolz import partition_all, partial"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<Client: scheduler='tcp://localhost:8786' processes=7 cores=56>"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"client = Client('localhost:8786')\n",
"client.restart()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Data Setup"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We will create a base coefficient vector that actually has many 0's to compare against:"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [],
"source": [
"## create inputs with a bunch of independent normals\n",
"beta = np.random.random(100) # random beta coefficients, no intercept\n",
"zero_idx = np.random.choice(len(beta), size=10)\n",
"beta[zero_idx] = 0 # set some to 0\n",
"X = da.random.normal(0, 1, size=(50000000, 100), chunks=(50000, 100))\n",
"y = X.dot(beta) + da.random.normal(0, 2, size=50000000, chunks=(50000,)) # increase noise a little\n",
"\n",
"## make sure all chunks are ~equally sized\n",
"X, y = persist(X, y)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"metrics_v1 = []\n",
"metrics_v2 = []\n",
"metrics_base = []\n",
"\n",
"def update_metrics(hist):\n",
" '''Will look in current namespace for all the variables it needs;\n",
" written purely for convenience and DRY'''\n",
" \n",
" ## update metrics\n",
" hist.append({\n",
" 'time_elapsed': time.time() - start, ## this isn't quite appropriate; time passed is what we want\n",
" 'primal_res': np.linalg.norm(betas - z),\n",
" 'dual_res': np.linalg.norm(rho * (zold - z)),\n",
" 'eps_primal': np.sqrt(p * nchunks) * abstol \\\n",
" + nchunks * reltol * np.maximum(np.linalg.norm(betas), np.linalg.norm(z)),\n",
" 'eps_dual': np.sqrt(p * nchunks) * abstol \\\n",
" + nchunks * reltol * np.linalg.norm(rho * u),\n",
" 'num_zeros': (z == 0).sum()})"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Algorithmic Setup"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The only two parameters worth calling out here are `rho` and `over_relax`; `rho` can be interpreted as the step-size for the dual problem updates, and `over_relax` is a parameter in $(0, 2)$ which weights the current local update with the current global parameter estimate (`z`)."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"def local_f(beta, X, y, z, u, rho):\n",
" return ((y - X.dot(beta)) **2).sum() + (rho / 2) * np.dot(beta - z + u, \n",
" beta - z + u)\n",
"\n",
"def local_grad(beta, X, y, z, u, rho):\n",
" return 2 * X.T.dot(X.dot(beta) - y) + rho * (beta - z + u)\n",
"\n",
"\n",
"def shrinkage(beta, t):\n",
" return np.maximum(0, beta - t) - np.maximum(0, -beta - t)\n",
"\n",
"local_update2 = partial(local_update, f=local_f, fprime=local_grad)\n",
"\n",
"## set some algorithm parameters\n",
"MAX_TIME = 20 # in seconds\n",
"lamduh = 7.2\n",
"rho = 1.2\n",
"over_relax = 1.7\n",
"\n",
"abstol = 1e-4\n",
"reltol = 1e-2\n",
"\n",
"(n, p) = X.shape\n",
"nchunks = X.npartitions\n",
"ncores = sum(client.ncores().values())\n",
"XD = X.to_delayed().flatten().tolist() # imagine a list of numpy arrays, one for each chunk\n",
"yD = y.to_delayed().flatten().tolist()\n",
"XD = client.compute(XD)\n",
"yD = client.compute(yD)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"del X, y"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"wait([XD, yD]);"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Standard ADMM"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# the initial consensus estimate\n",
"z = np.zeros(p)\n",
"\n",
"# an array of the individual \"dual variables\" and parameter estimates,\n",
"# one for each chunk of data\n",
"u = np.array([np.zeros(p) for i in range(nchunks)])\n",
"betas = np.array([np.zeros(p) for i in range(nchunks)])"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"start = time.time()\n",
"time_elapsed = 0\n",
"\n",
"while time.time() - start < MAX_TIME:\n",
" \n",
" # process each chunk in parallel, using the black-box 'local_update' magic\n",
" betas = [delayed(local_update2)(xx, yy, bb, z, uu, rho) \n",
" for xx, yy, bb, uu in zip(XD, yD, betas, u)]\n",
" betas = np.array(da.compute(*betas))\n",
"\n",
" # create consensus estimate\n",
" zold = z.copy()\n",
" ztilde = np.mean(betas + np.array(u), axis=0)\n",
" z = shrinkage(ztilde, lamduh / (rho * nchunks))\n",
" \n",
" # update dual variables\n",
" u += betas - z\n",
" \n",
" update_metrics(metrics_base)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Async version 1: single updates"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# the initial consensus estimate\n",
"z = np.zeros(p)\n",
"\n",
"# an array of the individual \"dual variables\" and parameter estimates,\n",
"# one for each chunk of data\n",
"\n",
"count = 0\n",
"u = np.array([np.zeros(p) for i in range(nchunks)])\n",
"betas = np.array([np.zeros(p) for i in range(nchunks)])\n",
"\n",
"starting_indices = np.random.choice(nchunks, size=ncores*2, replace=True)\n",
"\n",
"futures = [client.submit(local_update, XD[i], yD[i], betas[i], z, u[i],\n",
" rho, f=local_f, fprime=local_grad)\n",
" for i in starting_indices]\n",
"index = dict(zip(futures, starting_indices))\n",
"pool = as_completed(futures, with_results=True)\n",
"\n",
"\n",
"start = time.time()\n",
"\n",
"while time.time() - start < MAX_TIME:\n",
" future, local_beta = next(pool)\n",
" i = index.pop(future)\n",
" betas[i] = local_beta\n",
" count += 1\n",
"\n",
" ztilde = np.mean(betas + np.array(u), axis=0)\n",
"\n",
" if count < nchunks: # artificially inflate beta in the beginning\n",
" ztilde *= nchunks / (count + 1)\n",
" \n",
" zold = z.copy()\n",
" z = shrinkage(ztilde, lamduh / (rho * nchunks))\n",
"\n",
" update_metrics(metrics_v1)\n",
"\n",
" i = random.randint(0, nchunks - 1)\n",
" u[i] += betas[i] - z\n",
"\n",
" new_future = client.submit(local_update2, XD[i], yD[i], betas[i], z, u[i], rho)\n",
" index[new_future] = i\n",
" pool.add(new_future)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Async version 2: batched updates"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# the initial consensus estimate\n",
"z = np.zeros(p)\n",
"\n",
"# an array of the individual \"dual variables\" and parameter estimates,\n",
"# one for each chunk of data\n",
"\n",
"u = np.array([np.zeros(p) for i in range(nchunks)])\n",
"betas = np.array([np.zeros(p) for i in range(nchunks)])\n",
"\n",
"starting_idx = np.random.choice(nchunks, size=ncores*2, replace=True)\n",
"new_betas = [client.submit(local_update, XD[i], yD[i], betas[i], z, u[i], \n",
" rho, f=local_f, fprime=local_grad) for\n",
" i in starting_idx]\n",
"index = dict(zip(new_betas, starting_idx))\n",
"pool = as_completed(new_betas, with_results=True)\n",
"batches = pool.batches()\n",
"\n",
"start = time.time()\n",
"time_elapsed = 0\n",
"count = 0\n",
"\n",
"while time.time() - start < MAX_TIME:\n",
" batch = next(batches)\n",
" for future, result in batch:\n",
" i = index.pop(future)\n",
" # betas[i] = over_relax * local_beta + (1 - over_relax) * z\n",
" betas[i] = result\n",
" count += 1\n",
"\n",
" ztilde = np.mean(betas + np.array(u), axis=0)\n",
"\n",
" if count < nchunks:\n",
" ztilde *= nchunks / (count + 1)\n",
"\n",
" z = shrinkage(ztilde, lamduh / (rho * nchunks))\n",
"\n",
" update_metrics(metrics_v2)\n",
"\n",
" for _ in batch: # submit new futures\n",
" i = random.randint(0, nchunks - 1)\n",
" u[i] += betas[i] - z\n",
" new_fut = client.submit(local_update2, XD[i], yD[i], betas[i], z, u[i],\n",
" rho)\n",
" index[new_fut] = i\n",
" pool.add(new_fut)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Plots"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Below we will plot the \"primal residual\" (`primal_res`) as well as the \"dual tolerance\" (`eps_dual`); the former attempts to measure how much consensus there is, and the latter tracks the size of the dual variables (`u`). Both are used in the ultimate convergence criterion."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"pd_metrics_base = pd.DataFrame(metrics_base)\n",
"pd_metrics_v1 = pd.DataFrame(metrics_v1)\n",
"pd_metrics_v2 = pd.DataFrame(metrics_v2)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x7f96ba64c940>"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA9YAAAHVCAYAAAAU4mM1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f96ba768e80>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"%matplotlib inline\n",
"import matplotlib.pyplot as plt\n",
"fig, ax = plt.subplots(1, 2, sharex=True, figsize=(12, 5))\n",
"\n",
"x, y = 'time_elapsed', 'primal_res'\n",
"\n",
"pd_metrics_base.plot(ax=ax[0], x=x, y=y, label='Synchronous ADMM', title='Primal residual')\n",
"pd_metrics_v1.plot(ax=ax[0], x=x, y=y, label='Asynchronouns ADMM')\n",
"pd_metrics_v2.plot(ax=ax[0], x=x, y=y, label='Batched Async ADMM')\n",
"\n",
"y = 'eps_dual'\n",
"\n",
"pd_metrics_base.plot(ax=ax[1], x=x, y=y, label='Synchronous ADMM', title='Eps Dual')\n",
"pd_metrics_v1.plot(ax=ax[1], x=x, y=y, label='Asynchronouns ADMM')\n",
"pd_metrics_v2.plot(ax=ax[1], x=x, y=y, label='Batched Async ADMM')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It is a bit strange to see this upward trend in the dual tolerance plot; we expect $u \\rightarrow u^*$ to converge to the optimal dual variables which would imply asymptotic stability in the above.\n",
"\n",
"Also, it isn't clear to me why the dual variables appear to be on different scales."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Actual Convergence Time"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The above plots were all skewed in that `matplotlib` interpolates values for us; this makes synchronous ADMM look better than asynchronous ADMM. In fact, if we track the actual `time_elapsed` to convergence, we find async converges faster.\n",
"\n",
"This is because synchronous is blocked from an update until all workers have reported back, whereas asynchronous is constantly updating it's estimates, allowing it to converge in the interim times between synchronous updates.\n",
"\n",
"The standard convergence criterion for ADMM can be informally stated as follows: \n",
"- the size of the difference between the global consensus estimate and the individual worker estimates should be \"small\", where \"small\" is measured by the current size of all the estimates\n",
"- the size of the difference between the current global consensus estimate and the old global consensus estimate should be \"small\", where \"small\" is measured by the current size of the dual variables (this difference is a proxy for dual variable feasibility)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"pd_metrics_base['primal_conv'] = pd_metrics_base.primal_res - pd_metrics_base.eps_primal\n",
"pd_metrics_v1['primal_conv'] = pd_metrics_v1.primal_res - pd_metrics_v1.eps_primal\n",
"pd_metrics_v2['primal_conv'] = pd_metrics_v2.primal_res - pd_metrics_v2.eps_primal"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"pd_metrics_base['dual_conv'] = pd_metrics_base.dual_res - pd_metrics_base.eps_dual\n",
"pd_metrics_v1['dual_conv'] = pd_metrics_v1.dual_res - pd_metrics_v1.eps_dual\n",
"pd_metrics_v2['dual_conv'] = pd_metrics_v2.dual_res - pd_metrics_v2.eps_dual"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"standard_primal_conv = pd_metrics_base.loc[np.where(pd_metrics_base['primal_conv'] < 0)[0][0], 'time_elapsed']\n",
"async_primal_conv = pd_metrics_v1.loc[np.where(pd_metrics_v1['primal_conv'] < 0)[0][0], 'time_elapsed']\n",
"batched_primal_conv = pd_metrics_v2.loc[np.where(pd_metrics_v2['primal_conv'] < 0)[0][0], 'time_elapsed']\n",
"\n",
"standard_dual_conv = pd_metrics_base.loc[np.where(pd_metrics_base['dual_conv'] < 0)[0][0], 'time_elapsed']\n",
"async_dual_conv = pd_metrics_v1.loc[np.where(pd_metrics_v1['dual_conv'] < 0)[0][0], 'time_elapsed']\n",
"batched_dual_conv = pd_metrics_v2.loc[np.where(pd_metrics_v2['dual_conv'] < 0)[0][0], 'time_elapsed']\n",
"\n",
"standard_converged = max(standard_primal_conv, standard_dual_conv)\n",
"async_converged = max(async_primal_conv, async_dual_conv)\n",
"batched_converged = max(batched_primal_conv, batched_dual_conv)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Standard converged in 6.113s\n",
"Async converged in 0.031s\n",
"Batched converged in 0.030s\n"
]
}
],
"source": [
"print('Standard converged in {:.3f}s'.format(standard_converged))\n",
"print('Async converged in {:.3f}s'.format(async_converged))\n",
"print('Batched converged in {:.3f}s'.format(batched_converged))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"## Convergence Plots"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now that we interpolated between update times with constants, we can better see the path to convergence of both synchronous and asynchronous ADMM; the algorithms are converged whenever both the primal and dual lines are below 0."
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA+oAAAHVCAYAAABmLQGDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJElqyP8BFPCJMUgSmkQAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f96ba1ba208>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pyplot as plt\n",
"fig, ax = plt.subplots(1, 2, sharex=True, figsize=(12, 5))\n",
"\n",
"for name, df in [('Synchronous ADMM', pd_metrics_base),\n",
" ('Asynchronous ADMM', pd_metrics_v1),\n",
" ('Batched Async ADMM', pd_metrics_v2)]:\n",
"\n",
" df.plot(ax=ax[0], x=x, y='primal_conv', label=name, title='Primal Convergence')\n",
" \n",
" df.plot(ax=ax[1], x=x, y='dual_conv', label=name, title='Dual Convergence')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As before, I'm really not sure why the dual variables are scaled so differently between synchronous and asynchronous."
]
}
],
"metadata": {
"anaconda-cloud": {},
"kernelspec": {
"display_name": "Python [default]",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.5.2"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
@minhthangbk
Copy link

Hello,

Could you please specify the version of Python, DASK, Numpy, and Pandas. I got some errors while using Google Colab. It may be from the old version of DASK, which you used.

Good day

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment