{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Getting started with csky\n", "\n", ".. contents:: :local:\n", "\n", "In this tutorial, we'll demonstrate how to get up and running with csky. The entire tutorial runs in about 15 minutes on my laptop (see timing summary at the bottom).\n", "\n", "If not working on a `cobalt`, it's easiest to get started if you set up a couple of environment variables. Here are the settings on my laptop:\n", "\n", " export CSKY_DATA_ANALYSES_DIR=\"/home/mike/work/i3/data/analyses\"\n", " export CSKY_REMOTE_USER=\"mrichman\"\n", " \n", "You can confirm your own settings like so:" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "None\n", "None\n" ] } ], "source": [ "import os\n", "print(os.getenv('CSKY_DATA_ANALYSES_DIR'))\n", "print(os.getenv('CSKY_REMOTE_USER'))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "If you are working on a `cobalt`, you **do not** need to set these variables because you can read directly from the official data repositories." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Basic Setup" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll import a few tools:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "import histlite as hl\n", "import csky as cy\n", "\n", "%matplotlib inline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "I like to set up my plotting like so:" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/mnt/lfs7/user/ssclafani/software/external/csky/csky/plotting.py:92: MatplotlibDeprecationWarning: Support for setting the 'text.latex.preamble' or 'pgf.preamble' rcParam to a list of strings is deprecated since 3.3 and will be removed two minor releases later; set it to a single string instead.\n", " r'\\SetSymbolFont{operators} {sans}{OT1}{cmss} {m}{n}'\n" ] } ], "source": [ "cy.plotting.mrichman_mpl()\n", "\n", "# for serif fonts:\n", "# cy.plotting.mrichman_mpl(sans=False)\n", "\n", "# for default rendering instead of LaTeX for everything\n", "# cy.plotting.mrichman_mpl(tex=False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll use a `Timer` to keep track of compute times:" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "timer = cy.timing.Timer()\n", "time = timer.time" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Initial Setup" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For our first example, we'll fire up an Analysis for the MESC-7yr dataset. csky works just as well for tracks, but cascades are nice for a tutorial because you can generate a reasonably accurate all-sky scan in just a minute or so. Below is what happens the first time you load the data on a non-cobalt machine. csky automatically fetches the required dataset files and stores them under `$CSKY_DATA_ANALYSES_DIR`." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Setting up Analysis for:\n", "MESC_2010_2016\n", "Setting up MESC_2010_2016...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2013_MC.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC79_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2011_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2012_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2013_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2014_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2015_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/IC86_2016_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC79_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC86_2011_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC86_2012_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC86_2013_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC86_2014_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC86_2015_exp.npy ...\n", "Reading /data/ana/analyses/mese_cascades/version-001-p02/GRL/IC86_2016_exp.npy ...\n", "Energy PDF Ratio Model...\n", " * gamma = 4.0000 ...\n", "Signal Acceptance Model...\n", " * gamma = 4.0000 ...\n", "Done.\n", "\n", "0:00:08.400903 elapsed.\n" ] } ], "source": [ "with time('ana setup (from scratch)'):\n", " ana = cy.get_analysis(cy.selections.Repository(), 'version-001-p02' , cy.selections.MESEDataSpecs.mesc_7yr)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "When you use a `cy.selections.Repository` to mirror files like this from within a notebook, you can check the stdout stream in the terminal where jupyter is running to monitor the download progress." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "You are probably interested in tracks; here are some commonly used selections:\n", "\n", "* `cy.selections.PSDataSpecs.ps_7yr` -- Stefan's 7yr dataset\n", "* `cy.selections.GFUDataSpecs.gfu_3yr` -- GFU subset used for TXS analysis\n", "* `cy.selections.NTDataSpecs.nt_8yr` -- Northern tracks\n", "* `cy.selections.PSDataSpecs.ps_10yr` -- Tessa's 10yr dataset" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We have used `cy.selections.mrichman_repo` here only because adding the cascade dataset to `/data/ana/analyses` is still TODO; for datasets already organized in the official repository, you can instead use `cy.selections.repo` to read from there." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Once you have an `Analysis` instance, you can cache it to disk:" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "ana_dir = cy.utils.ensure_dir('/home/mike/work/i3/csky/ana')" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "-> /home/mike/work/i3/csky/ana/MESC_2010_2016_DNN.subanalysis.npy \n" ] } ], "source": [ "ana.save(ana_dir)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "From now on, you can accelerate this setup step by reading from that directory (note that `Repository` objects also keep track of loaded files so they do not need to be re-loaded if used multiple times within the same session):" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Setting up Analysis for:\n", "MESC_2010_2016_DNN\n", "Setting up MESC_2010_2016_DNN...\n", "<- /home/mike/work/i3/csky/ana/MESC_2010_2016_DNN.subanalysis.npy \n", "Done.\n", "\n", "0:00:00.376400 elapsed.\n" ] } ], "source": [ "with time('ana setup (from cache-to-disk)'):\n", " ana = cy.get_analysis(cy.selections.Repository(), 'version-001-p02', cy.selections.MESEDataSpecs.mesc_7yr, dir=ana_dir)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "csky's \"global\" default configuration is stored as `cy.CONF`. By default it is just:" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'mp_cpus': 1}\n" ] } ], "source": [ "print(cy.CONF)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For the rest of the tutorial, we will be using this `ana`. We will also use 3 cores for multiprocessing-enabled steps (more can be appropriate, but this is a laptop)." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "cy.CONF['ana'] = ana\n", "cy.CONF['mp_cpus'] = 3" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Point Source Analysis" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The hottest spot in my analysis was found at $(\\alpha,\\delta)=(271.23^\\circ,7.78^\\circ)$. Let's get a trial runner for that position in the sky:" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "src = cy.sources(271.23, 7.78, deg=True)\n", "tr = cy.get_trial_runner(src=src)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "`cy.sources(ra,dec,**kw)` is a convenience function for constructing `cy.utils.Sources(**kw)` instances. If `ra` and `dec` are arrays, then stacking analysis will be performed. If nonzero `extension` value(s) are given, Gaussian source extension is assumed." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "`cy.get_trial_runner(...)` is a powerful function for obtaining `cy.trial.TrialRunner` instances. A *trial runner* is an object which is capable of working with \"trials\" (basically, event ensembles plus counts of \"excluded\" events pre-emptively counted as purely background, $S/B \\to 0$). Trial runners not only produce trials, but also invoke likelihood fitting machinery for them; run them in (parallelizable) batches; and perform batches specifically useful for estimating sensitivities and discovery potentials. They also provide an interface for converting between event rates and fluxes/fluences." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Background estimation (PS)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The first step in most point source analyses is to characterize the background by running scrambled trials. In csky, this is done as follows:" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performing 10000 background trials using 3 cores:\n", " 10000/10000 trials complete. \n", "\n", "0:00:34.311516 elapsed.\n" ] } ], "source": [ "with time('ps bg trials'):\n", " n_trials = 10000\n", " bg = cy.dists.Chi2TSD(tr.get_many_fits(n_trials))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "`Chi2TSD` stores the trials as well as a $\\chi^2$ fit to the nonzero part of the TS distribution. We can look at a barebones summary:" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Chi2TSD(10000 trials, eta=0.589, ndof=1.140, median=0.058 (from fit 0.059))\n" ] } ], "source": [ "print(bg)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "or a more detailed one:" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Chi2TSD from 10000 trials:\n", " eta = 0.589\n", " ndof = 1.140\n", " loc = 0.000\n", " scale = 0.972\n", "Thresholds from trials:\n", " median = 0.058\n", " 1 sigma = 1.39\n", " 2 sigma = 4.50\n", " 3 sigma = 9.48\n", " 4 sigma = 16.37\n", " 5 sigma = 25.18\n", "Thresholds from fit:\n", " median = 0.059\n", " 1 sigma = 1.39\n", " 2 sigma = 4.50\n", " 3 sigma = 9.48\n", " 4 sigma = 16.37\n", " 5 sigma = 25.18\n" ] } ], "source": [ "print(bg.description)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, we can plot the distribution as follows:" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots()\n", "\n", "# csky uses histlite all over the place for PDF management\n", "# the background distribution fit integrates with histlite as well\n", "h = bg.get_hist(bins=50)\n", "hl.plot1d(ax, h, crosses=True,\n", " label='{} bg trials'.format(bg.n_total))\n", "\n", "# compare with the chi2 fit:\n", "x = h.centers[0]\n", "norm = h.integrate().values\n", "ax.semilogy(x, norm * bg.pdf(x), lw=1, ls='--',\n", " label=r'$\\chi^2[{:.2f}\\text{{dof}},\\ \\eta={:.3f}]$'.format(bg.ndof, bg.eta))\n", "\n", "# always label your plots, folks\n", "ax.set_xlabel(r'TS')\n", "ax.set_ylabel(r'number of trials')\n", "ax.legend()\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Sensitivity and discovery potential (PS)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can estimate the sensitivity and discovery potential as follows:" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Start time: 2019-09-26 10:41:14.457200\n", "Using 3 cores.\n", "* Starting initial scan for 90% of 50 trials with TS >= 0.058...\n", " n_sig = 5.000 ... frac = 0.84000\n", " n_sig = 10.000 ... frac = 0.96000\n", "* Generating batches of 500 trials...\n", "n_trials | n_inj 0.00 4.00 8.00 12.00 16.00 20.00 | n_sig(relative error)\n", "500 | 57.0% 86.8% 97.8% 99.0% 100.0% 100.0% | 4.818 (+/- 6.8%) [spline]\n", "1000 | 53.5% 84.6% 96.8% 99.2% 100.0% 100.0% | 5.317 (+/- 3.7%) [spline]\n", "End time: 2019-09-26 10:41:47.328998\n", "Elapsed time: 0:00:32.871798\n", "\n", "0:00:32.873298 elapsed.\n" ] } ], "source": [ "with time('ps sensitivity'):\n", " sens = tr.find_n_sig(\n", " # ts, threshold\n", " bg.median(),\n", " # beta, fraction of trials which should exceed the threshold\n", " 0.9,\n", " # n_inj step size for initial scan\n", " n_sig_step=5,\n", " # this many trials at a time\n", " batch_size=500,\n", " # tolerance, as estimated relative error\n", " tol=.05\n", " )" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Start time: 2019-09-26 10:41:47.342592\n", "Using 3 cores.\n", "* Starting initial scan for 50% of 50 trials with TS >= 25.182...\n", " n_sig = 5.000 ... frac = 0.00000\n", " n_sig = 10.000 ... frac = 0.06000\n", " n_sig = 15.000 ... frac = 0.30000\n", " n_sig = 20.000 ... frac = 0.52000\n", "* Generating batches of 500 trials...\n", "n_trials | n_inj 0.00 8.00 16.00 24.00 32.00 40.00 | n_sig(relative error)\n", "500 | 0.0% 3.2% 31.8% 70.0% 96.2% 100.0% | 19.787 (+/- 1.5%) [spline]\n", "End time: 2019-09-26 10:42:06.300172\n", "Elapsed time: 0:00:18.957580\n", "\n", "0:00:18.963805 elapsed.\n" ] } ], "source": [ "with time('ps discovery potential'):\n", " disc = tr.find_n_sig(bg.isf_nsigma(5), 0.5, n_sig_step=5, batch_size=500, tol=.05)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now `sens` and `disc` are `dict` objects including various information about the calculation, as well as the result under the key `n_sig`:" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "5.316699361560448 19.786576686695103\n" ] } ], "source": [ "print(sens['n_sig'], disc['n_sig'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So far, we've just found the Poisson event rates that correspond to the sensitivity and discovery potential fluxes. These can be converted to fluxes as follows:" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "5.259e-12 TeV/cm2/s @ 100 TeV\n", "1.957e-11 TeV/cm2/s @ 100 TeV\n" ] } ], "source": [ "fmt = '{:.3e} TeV/cm2/s @ 100 TeV'\n", "# either the number of events or the whole dict will work\n", "print(fmt.format(tr.to_E2dNdE(sens, E0=100, unit=1e3)))\n", "print(fmt.format(tr.to_E2dNdE(disc['n_sig'], E0=100, unit=1e3)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For more counts <--> flux conversions, see [below](#Counts-<-->-flux-conversions)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Test for fit bias (PS)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "One powerful test of analysis readiness is to demonstrate that we can properly fit the parameters of an injected signal. For this we start by getting some batches of signal trials:" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "0:00:11.244517 elapsed.\n" ] } ], "source": [ "n_sigs = np.r_[:101:10]\n", "with time('ps fit bias trials'):\n", " trials = [tr.get_many_fits(100, n_sig=n_sig, logging=False, seed=n_sig) for n_sig in n_sigs]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We add the true number of events injected for bookkeeping convenience:" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "for (n_sig, t) in zip(n_sigs, trials):\n", " t['ntrue'] = np.repeat(n_sig, len(t))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Concatenate the trial batches:" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Arrays(1100 items | columns: gamma, ns, ntrue, ts)" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "allt = cy.utils.Arrays.concatenate(trials)\n", "allt" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, we make the plots:" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axs = plt.subplots(1, 2, figsize=(6,3))\n", "\n", "dns = np.mean(np.diff(n_sigs))\n", "ns_bins = np.r_[n_sigs - 0.5*dns, n_sigs[-1] + 0.5*dns]\n", "expect_kw = dict(color='C0', ls='--', lw=1, zorder=-10)\n", "expect_gamma = tr.sig_injs[0].flux[0].gamma\n", "\n", "ax = axs[0]\n", "h = hl.hist((allt.ntrue, allt.ns), bins=(ns_bins, 100))\n", "hl.plot1d(ax, h.contain_project(1),errorbands=True, drawstyle='default')\n", "\n", "lim = ns_bins[[0, -1]]\n", "ax.set_xlim(ax.set_ylim(lim))\n", "ax.plot(lim, lim, **expect_kw)\n", "ax.set_aspect('equal')\n", "\n", "\n", "ax = axs[1]\n", "h = hl.hist((allt.ntrue, allt.gamma), bins=(ns_bins, 100))\n", "hl.plot1d(ax, h.contain_project(1),errorbands=True, drawstyle='default')\n", "ax.axhline(expect_gamma, **expect_kw)\n", "ax.set_xlim(axs[0].get_xlim())\n", "\n", "for ax in axs:\n", " ax.set_xlabel(r'$n_\\text{inj}$')\n", " ax.grid()\n", "axs[0].set_ylabel(r'$n_s$')\n", "axs[1].set_ylabel(r'$\\gamma$')\n", "\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Unblinding: confirming hottest spot" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Since this analysis is already unblinded, and paper submitted, in this tutorial we do not simulate the analysis review period and instead skip straight to unblinding!" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[9.143492673487607, 32.29657745559301, 2.9952804375887756]\n" ] } ], "source": [ "result = ts, ns, gamma = tr.get_one_fit(TRUTH=True)\n", "print(result)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can view the result in a few different ways. For a nice table," ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "TS 9.143492673487607\n", "ns 32.29657745559301\n", "gamma 2.9952804375887756\n" ] } ], "source": [ "print(tr.format_result(tr.get_one_fit(TRUTH=True)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For a more raw view, with named fit parameters rather than a flat array," ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(9.143492673487607, {'gamma': 2.9952804375887756, 'ns': 32.29657745559301})" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.get_one_fit(TRUTH=True, flat=False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, for full details including the complete minimizer configuration and convergence report," ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(9.143492673487607,\n", " {'gamma': 2.9952804375887756, 'ns': 32.29657745559301},\n", " {},\n", " {'_full_output': True,\n", " '_ns_min': 0,\n", " '_ts_min': 0,\n", " '_ns_tol': 0.0001,\n", " '_taylor_tol': 0.001,\n", " '_fit_null': False,\n", " '_fmin_method': 'l_bfgs_b',\n", " '_fmin_factr': 1000000000.0,\n", " '_fmin_epsilon': 1e-08,\n", " '_seed_with_prior': True,\n", " '_seed_with_null': False,\n", " '_log_params': [],\n", " '_masks': None},\n", " {'grad': array([-1.24344979e-06]),\n", " 'task': b'CONVERGENCE: NORM_OF_PROJECTED_GRADIENT_<=_PGTOL',\n", " 'funcalls': 10,\n", " 'nit': 3,\n", " 'warnflag': 0})" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.get_one_fit(TRUTH=True, _full_output=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can also perform the fit with non-default parameters. For example, to constrain the spectral fit to $\\gamma\\in[1.5,2.5]$, seeding with $\\gamma=2$, we can say:" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(6.519182178972011, {'gamma': 2.5, 'ns': 16.262526192118198})" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.get_one_fit(TRUTH=True, flat=False, gamma=[1.5, 2, 2.5])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We won't do a thorough tour of the underlying objects performing these calculations, but let's dig in only a little bit. A :class:`LLHEvaluator` instance is responsible for the actual fitting:" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(9.143492673487607, {'gamma': 2.9952804375887756, 'ns': 32.29657745559301})" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "L = tr.get_one_llh(TRUTH=True)\n", "L.fit(**tr.fitter_args)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can also take a look at the trial data itself:" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Trial(evss=[[Events(1875 items | columns: dec, idx, inj, log10energy, ra, sigma, sindec)]], n_exs=[104])\n" ] }, { "data": { "text/plain": [ "(9.143492673487607, {'gamma': 2.9952804375887756, 'ns': 32.29657745559301})" ] }, "execution_count": 30, "metadata": {}, "output_type": "execute_result" } ], "source": [ "trial = tr.get_one_trial(TRUTH=True) # event ensemble and count of masked-out events\n", "print(trial)\n", "L = tr.get_one_llh_from_trial(trial)\n", "L.fit(**tr.fitter_args)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Ok, now finally, we use the pre-computed `bg` to calculate the p-value:" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "p = 1.63e-03 (2.94 sigma)\n" ] } ], "source": [ "print('p = {:.2e} ({:.2f} sigma)'.format(bg.sf(ts), bg.sf_nsigma(ts)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In my [paper draft](https://wiki.icecube.wisc.edu/index.php/Cascade_7yr_PS_GP_Paper), I claim 2.8 sigma based on a whole lot more trials." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## All-sky scan" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In many analyses, we aren't just interested in specific sky coordinates, but instead a scan over the whole sky. In csky we can do this using a `cy.trial.SkyScanTrialRunner`, obtained like so:" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [], "source": [ "sstr = cy.get_sky_scan_trial_runner(mp_scan_cpus=3, nside=32)" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Scanning 12288 locations using 3 cores:\n", " 12288/12288 coordinates complete. \n", "\n", "0:00:33.151653 elapsed.\n" ] } ], "source": [ "with time('ps all-sky scan'):\n", " scan = sstr.get_one_scan(TRUTH=True, logging=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "What do we have here?" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(4, 12288)" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "scan.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The second axis is npix for the selected healpy nside. The first axis has items $(-\\log_{10}(p),\\text{TS},n_s,\\gamma)$. In this example we haven't bothered to characterize the background in detail, so the `scan[0]` is simply a copy of `scan[1]`. To obtain proper p-values, pass an argument `get_sky_scan_trial_runner(..., ts_to_p=[function])`, where `[function]` should accept `dec[radians],ts` and return `p`, typically by using `cy.dists.ts_to_p`." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can plot the TS map like so:" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots (subplot_kw=dict (projection='aitoff'))\n", "sp = cy.plotting.SkyPlotter(pc_kw=dict(cmap=cy.plotting.skymap_cmap, vmin=0, vmax=10))\n", "mesh, cb = sp.plot_map(ax, scan[1], n_ticks=2)\n", "kw = dict(color='.5', alpha=.5)\n", "sp.plot_gp(ax, lw=.5, **kw)\n", "sp.plot_gc(ax, **kw)\n", "ax.grid(**kw)\n", "cb.set_label(r'TS')\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Galactic Plane" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "csky provides two types of template analysis: space-only templates, and per-pixel energy-binned spectra. In the former, more common type, we provide a space-only healpix map describing the signal distribution across the sky. My `mrichman_repo` provides the *Fermi*-LAT $\\pi^0$-decay template:" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Reading /home/mike/work/i3/data/analyses/templates/Fermi-LAT_pi0_map.npy ...\n" ] } ], "source": [ "pi0_map = cy.selections.mrichman_repo.get_template ('Fermi-LAT_pi0_map')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The desired template analysis can be expressed as a configuration dict, which will ultimately be overlaied on top of `cy.CONF`:" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [], "source": [ "pi0_conf = {\n", " # desired template\n", " 'template': pi0_map,\n", " # desired baseline spectrum\n", " 'flux': cy.hyp.PowerLawFlux(2.5),\n", " # desired fixed spectrum in likelihood\n", " 'fitter_args': dict(gamma=2.5),\n", " # use signal subtracting likelihood\n", " 'sigsub': True,\n", " # fastest construction:\n", " # weight from acceptance parameterization rather\n", " # than pixel-by-pixel directly from MC\n", " 'fast_weight': True,\n", " # cache angular-resolution-smeared maps to disk\n", " 'dir': cy.utils.ensure_dir('{}/templates/pi0'.format(ana_dir))\n", "}" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "MESC_2010_2016_DNN | Acceptance weighting complete. \n", "MESC_2010_2016_DNN | Smearing complete. \n", "-> /home/mike/work/i3/csky/ana/templates/pi0/MESC_2010_2016_DNN.template.fastweight.npy \n", "\n", "0:00:14.135515 elapsed.\n" ] } ], "source": [ "with time('pi0 template construction'):\n", " pi0_tr = cy.get_trial_runner(pi0_conf)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "`mrichman_repo` also provides the KRAγ templates:" ] }, { "cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Reading /home/mike/work/i3/data/analyses/templates/KRA-gamma_5PeV_maps_energies.tuple.npy ...\n" ] } ], "source": [ "krag5_map, krag5_energy_bins = cy.selections.mrichman_repo.get_template(\n", " 'KRA-gamma_5PeV_maps_energies', per_pixel_flux=True)" ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [], "source": [ "krag5_conf = {\n", " 'template': krag5_map,\n", " 'bins_energy': krag5_energy_bins,\n", " 'fitter_args': dict(gamma=2.5),\n", " 'update_bg' : True,\n", " 'sigsub': True,\n", " 'dir': cy.utils.ensure_dir('{}/templates/kra'.format(ana_dir))}" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "MESC_2010_2016_DNN | Acceptance weighting complete. \n", "MESC_2010_2016_DNN | Smearing complete. \n", "-> /home/mike/work/i3/csky/ana/templates/kra/MESC_2010_2016_DNN.template.npy \n", "\n", "0:00:26.555365 elapsed.\n" ] } ], "source": [ "with time('KRAγ5 template construction'):\n", " krag5_tr = cy.get_trial_runner(krag5_conf)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As we will see, all the usual trial runner functionality is available." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Background characterization (GP)" ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performing 10000 background trials using 3 cores:\n", " 10000/10000 trials complete. \n", "\n", "0:00:28.301643 elapsed.\n" ] } ], "source": [ "with time('pi0 bg trials'):\n", " n_trials = 10000\n", " pi0_bg = cy.dists.Chi2TSD(pi0_tr.get_many_fits(n_trials))" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performing 10000 background trials using 3 cores:\n", " 10000/10000 trials complete. \n", "\n", "0:00:11.343195 elapsed.\n" ] } ], "source": [ "with time('KRAγ5 bg trials'):\n", " n_trials = 10000\n", " krag5_bg = cy.dists.Chi2TSD(krag5_tr.get_many_fits(n_trials))" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "bgs = pi0_bg, krag5_bg\n", "titles = '$\\pi^0$-decay', 'KRA$_\\gamma^5$'\n", "for (b, title) in zip(bgs, titles):\n", " fig, ax = plt.subplots()\n", " h = b.get_hist(bins=30)\n", " hl.plot1d(ax, h, crosses=True,\n", " label='{} bg trials'.format(bg.n_total))\n", " x = h.centers[0]\n", " norm = h.integrate().values\n", " ax.semilogy(x, norm * b.pdf(x), lw=1, ls='--',\n", " label=r'$\\chi^2[{:.2f}\\text{{dof}},\\ \\eta={:.3f}]$'.format(b.ndof, b.eta))\n", " ax.set_xlabel(r'TS')\n", " ax.set_ylabel(r'number of trials')\n", " ax.set_title(title)\n", " ax.legend()\n", " plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Sensitivity and discovery potential (GP)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll find the sensitivity and discovery potential the same way as we did above for a point source:" ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Start time: 2019-09-26 10:44:26.457045\n", "Using 3 cores.\n", "* Starting initial scan for 90% of 50 trials with TS >= 0.000...\n", " n_sig = 10.000 ... frac = 0.56000\n", " n_sig = 20.000 ... frac = 0.78000\n", " n_sig = 30.000 ... frac = 0.86000\n", " n_sig = 40.000 ... frac = 0.98000\n", "* Generating batches of 500 trials...\n", "n_trials | n_inj 0.00 16.00 32.00 48.00 64.00 80.00 | n_sig(relative error)\n", "500 | 44.6% 78.2% 90.2% 97.2% 99.4% 99.8% | 31.639 (+/- 6.6%) [spline]\n", "1000 | 46.6% 75.2% 88.5% 96.8% 99.5% 99.7% | 34.347 (+/- 3.5%) [spline]\n", "End time: 2019-09-26 10:47:24.457213\n", "Elapsed time: 0:02:58.000168\n", "\n", "0:02:58.001318 elapsed.\n" ] } ], "source": [ "with time('pi0 sensitivity'):\n", " pi0_sens = pi0_tr.find_n_sig(pi0_bg.median(), 0.9, n_sig_step=10, batch_size=500, tol=.05)" ] }, { "cell_type": "code", "execution_count": 47, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Start time: 2019-09-26 10:47:24.467355\n", "Using 3 cores.\n", "* Starting initial scan for 50% of 50 trials with TS >= 24.101...\n", " n_sig = 20.000 ... frac = 0.00000\n", " n_sig = 40.000 ... frac = 0.00000\n", " n_sig = 60.000 ... frac = 0.00000\n", " n_sig = 80.000 ... frac = 0.00000\n", " n_sig = 100.000 ... frac = 0.08000\n", " n_sig = 120.000 ... frac = 0.22000\n", " n_sig = 140.000 ... frac = 0.40000\n", " n_sig = 160.000 ... frac = 0.58000\n", "* Generating batches of 500 trials...\n", "n_trials | n_inj 0.00 64.00 128.00 192.00 256.00 320.00 | n_sig(relative error)\n", "500 | 0.0% 0.8% 32.4% 84.4% 94.6% 96.6% | 148.888 (+/- 1.1%) [spline]\n", "End time: 2019-09-26 10:49:08.611245\n", "Elapsed time: 0:01:44.143890\n", "\n", "0:01:44.146194 elapsed.\n" ] } ], "source": [ "with time('pi0 discovery potential'):\n", " pi0_disc = pi0_tr.find_n_sig(pi0_bg.isf_nsigma(5), 0.5, n_sig_step=20, batch_size=500, tol=.05)" ] }, { "cell_type": "code", "execution_count": 48, "metadata": { "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Start time: 2019-09-26 10:49:08.618539\n", "Using 3 cores.\n", "* Starting initial scan for 90% of 50 trials with TS >= 0.000...\n", " n_sig = 10.000 ... frac = 0.70000\n", " n_sig = 20.000 ... frac = 0.88000\n", " n_sig = 30.000 ... frac = 0.98000\n", "* Generating batches of 500 trials...\n", "n_trials | n_inj 0.00 12.00 24.00 36.00 48.00 60.00 | n_sig(relative error)\n", "500 | 47.8% 76.0% 91.8% 98.6% 99.8% 100.0% | 22.002 (+/- 6.2%) [spline]\n", "1000 | 48.1% 78.1% 92.1% 97.9% 99.8% 100.0% | 21.373 (+/- 4.1%) [spline]\n", "End time: 2019-09-26 10:53:05.232254\n", "Elapsed time: 0:03:56.613715\n", "\n", "0:03:56.615738 elapsed.\n" ] } ], "source": [ "with time('KRAγ5 sensitivity'):\n", " krag5_sens = krag5_tr.find_n_sig(krag5_bg.median(), 0.9, n_sig_step=10, batch_size=500, tol=.05)" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Start time: 2019-09-26 10:53:05.244537\n", "Using 3 cores.\n", "* Starting initial scan for 50% of 50 trials with TS >= 23.837...\n", " n_sig = 20.000 ... frac = 0.00000\n", " n_sig = 40.000 ... frac = 0.00000\n", " n_sig = 60.000 ... frac = 0.18000\n", " n_sig = 80.000 ... frac = 0.34000\n", " n_sig = 100.000 ... frac = 0.74000\n", "* Generating batches of 500 trials...\n", "n_trials | n_inj 0.00 40.00 80.00 120.00 160.00 200.00 | n_sig(relative error)\n", "500 | 0.0% 0.8% 39.8% 91.6% 99.6% 100.0% | 87.707 (+/- 1.6%) [spline]\n", "End time: 2019-09-26 10:55:17.222720\n", "Elapsed time: 0:02:11.978183\n", "\n", "0:02:11.981155 elapsed.\n" ] } ], "source": [ "with time('KRAγ5 discovery potential'):\n", " krag5_disc = krag5_tr.find_n_sig(krag5_bg.isf_nsigma(5), 0.5, n_sig_step=20, batch_size=500, tol=.05)" ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "pi0-decay sens/disc:\n", "2.340e-11 TeV/cm2/s @ 100 TeV\n", "1.014e-10 TeV/cm2/s @ 100 TeV\n" ] } ], "source": [ "print('pi0-decay sens/disc:')\n", "print(fmt.format(pi0_tr.to_E2dNdE(pi0_sens, E0=100, unit=1e3)))\n", "print(fmt.format(pi0_tr.to_E2dNdE(pi0_disc, E0=100, unit=1e3)))" ] }, { "cell_type": "code", "execution_count": 51, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "KRAγ5 sens/disc:\n", "0.581590224006752 x KRAγ5\n", "2.386635444613829 x KRAγ5\n" ] } ], "source": [ "print('KRAγ5 sens/disc:')\n", "print('{} x KRAγ5'.format(krag5_tr.to_model_norm(krag5_sens, E0=100, unit=1e3)))\n", "print('{} x KRAγ5'.format(krag5_tr.to_model_norm(krag5_disc, E0=100, unit=1e3)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Test for fit bias (π⁰)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Once again, we'll follow the pattern from the PS analysis described above." ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "0:00:37.967059 elapsed.\n" ] } ], "source": [ "with time('pi0 fit bias trials'):\n", " pi0_trials = [pi0_tr.get_many_fits(100, n_sig=n_sig, logging=False, seed=n_sig) for n_sig in n_sigs]" ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [], "source": [ "for (n_sig, t) in zip(n_sigs, pi0_trials):\n", " t['ntrue'] = np.repeat(n_sig, len(t))" ] }, { "cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Arrays(1100 items | columns: ns, ntrue, ts)" ] }, "execution_count": 54, "metadata": {}, "output_type": "execute_result" } ], "source": [ "allt = cy.utils.Arrays.concatenate(pi0_trials)\n", "allt" ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(1, figsize=(3,3))\n", "\n", "dns = np.mean(np.diff(n_sigs))\n", "ns_bins = np.r_[n_sigs - 0.5*dns, n_sigs[-1] + 0.5*dns]\n", "expect_kw = dict(color='C0', ls='--', lw=1, zorder=-10)\n", "expect_gamma = tr.sig_injs[0].flux[0].gamma\n", "\n", "h = hl.hist((allt.ntrue, allt.ns), bins=(ns_bins, 100))\n", "hl.plot1d(ax, h.contain_project(1),errorbands=True, drawstyle='default')\n", "\n", "lim = 1.05 * ns_bins[[0, -1]]\n", "ax.set_xlim(ax.set_ylim(lim))\n", "ax.plot(lim, lim, **expect_kw)\n", "ax.set_aspect('equal')\n", "\n", "ax.set_xlabel(r'$n_\\text{inj}$')\n", "ax.grid()\n", "ax.set_ylabel(r'$n_s$')\n", "\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Unblinding (GP)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, we obtain the unblinded result:" ] }, { "cell_type": "code", "execution_count": 56, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "TS 3.0484314194823994\n", "ns 44.68690065047474\n" ] } ], "source": [ "pi0_result = pi0_tr.get_one_fit(TRUTH=True)\n", "print(pi0_tr.format_result(pi0_result))" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "p = 3.70e-02 (1.79 sigma)\n" ] } ], "source": [ "pi0_ts = pi0_result[0]\n", "print('p = {:.2e} ({:.2f} sigma)'.format(pi0_bg.sf(pi0_ts), pi0_bg.sf_nsigma(pi0_ts)))" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "TS 4.055486117489724\n", "ns 31.091825690526317\n" ] } ], "source": [ "krag5_result = krag5_tr.get_one_fit(TRUTH=True)\n", "print(krag5_tr.format_result(krag5_result))" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "p = 1.92e-02 (2.07 sigma)\n" ] } ], "source": [ "krag5_ts = krag5_result[0]\n", "print('p = {:.2e} ({:.2f} sigma)'.format(krag5_bg.sf(krag5_ts), krag5_bg.sf_nsigma(krag5_ts)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In the [paper draft](https://wiki.icecube.wisc.edu/index.php/Cascade_7yr_PS_GP_Paper), I claim p=0.040 and p=0.021 respectively, again based on a whole lot more trials." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Counts <--> flux conversions" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Counts <--> flux conversions can be pretty error prone; csky tries to make them easy. Here are a few examples, assuming for now that `ns` from the all-sky hottest spot actually corresponds to an $E^{-2}$ spectrum:" ] }, { "cell_type": "code", "execution_count": 60, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "3.1945011757861e-08" ] }, "execution_count": 60, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.to_dNdE(ns) # 1/GeV/cm2/s @ 1 GeV" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "3.1945011757861e-18" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.to_dNdE(ns, E0=1e5) # 1/GeV/cm2/s @ 10^5 GeV" ] }, { "cell_type": "code", "execution_count": 62, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "3.1945011757861e-08" ] }, "execution_count": 62, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.to_E2dNdE(ns, E0=1e5) # GeV/cm2/s @ 10^5 GeV" ] }, { "cell_type": "code", "execution_count": 63, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "3.1945011757861e-11" ] }, "execution_count": 63, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.to_E2dNdE(ns, E0=100, unit=1e3) # TeV/cm2/s @ 100 TeV" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "But it was actually a different spectrum right?" ] }, { "cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "all-sky hottest spot TS, ns, gamma = 9.143492673487607, 32.29657745559301, 2.9952804375887756\n" ] } ], "source": [ "print('all-sky hottest spot TS, ns, gamma = {}, {}, {}'.format(ts, ns, gamma))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can convert to a flux for that spectrum too:" ] }, { "cell_type": "code", "execution_count": 65, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "6.4612895507279714e-12" ] }, "execution_count": 65, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.to_E2dNdE(ns, gamma=gamma, E0=100, unit=1e3) # TeV/cm2/s @ 100 TeV, for gamma ~ 3.002" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Of course, any of these flux values can be converted back to event rates:" ] }, { "cell_type": "code", "execution_count": 66, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "32.29657745559335" ] }, "execution_count": 66, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tr.to_ns(6.46128955072804e-12, gamma=gamma, E0=100, unit=1e3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Timing summary" ] }, { "cell_type": "code", "execution_count": 67, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "ana setup (from scratch) | 0:00:19.693964\n", "ana setup (from cache-to-disk) | 0:00:00.376400\n", "ps bg trials | 0:00:34.311516\n", "ps sensitivity | 0:00:32.873298\n", "ps discovery potential | 0:00:18.963805\n", "ps fit bias trials | 0:00:11.244517\n", "ps all-sky scan | 0:00:33.151653\n", "pi0 template construction | 0:00:14.135515\n", "KRAγ5 template construction | 0:00:26.555365\n", "pi0 bg trials | 0:00:28.301643\n", "KRAγ5 bg trials | 0:00:11.343195\n", "pi0 sensitivity | 0:02:58.001318\n", "pi0 discovery potential | 0:01:44.146194\n", "KRAγ5 sensitivity | 0:03:56.615738\n", "KRAγ5 discovery potential | 0:02:11.981155\n", "pi0 fit bias trials | 0:00:37.967059\n", "-----------------------------------------------\n", "total | 0:15:19.662335\n" ] } ], "source": [ "print(timer)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Closing remarks" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In this tutorial, we've just scratched the surface of what you can do with csky. I hope I've convinced you that it's easy, fast, and maybe even fun.\n", "\n", "What now? Try it yourself, ask questions, and consult the other tutorials and documentation." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.7.5" } }, "nbformat": 4, "nbformat_minor": 4 }