diff --git a/notebooks/ALeRCE_ZTF_Stamp_Classifier.ipynb b/notebooks/ALeRCE_ZTF_Stamp_Classifier.ipynb
new file mode 100644
index 0000000..dbffc35
--- /dev/null
+++ b/notebooks/ALeRCE_ZTF_Stamp_Classifier.ipynb
@@ -0,0 +1,1668 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# ALeRCE Stamp Classifier notebook\n",
+ "\n",
+ "```Author: Rodrigo Carrasco-Davis, Last updated: 20200804```"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Table of contents:\n",
+ "* [Introduction](#intro)\n",
+ " * [Training data](#training_data) \n",
+ " * [Convolutional neural network architecture](#CNN)\n",
+ " * [Classification results](#Results)\n",
+ "* [ALeRCE frontend and Supernova Hunter](#Front)\n",
+ "* [Requirements](#req)\n",
+ "* [ALeRCE client](#client)\n",
+ " * [Plotting light curves](#plot_lc)\n",
+ " * [Retrieve classified alerts](#get_alerts)\n",
+ " * [Plot some examples](#examples)\n",
+ " * [Get a sequence of stamps](#stamp_series)\n",
+ " * [Get spatial distribution](#spatial_distribution)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Introduction "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "In this notebook, we will explore the results of the ALeRCE stamp classifier, explained in **Carrasco-Davis et al 2020 \"Alert Classification for the ALeRCE Broker System: The Real-time Stamp Classifier\"**, which is based on a convolutional neural network (CNN) that uses the images and metadata of the first alert only as input. The stamp classifier uses the first alert to quickly discriminate between active galactic nuclei (AGN), supernovae(SNe), variable stars (VS), asteroids, and bogus alerts. The predictions of the classifier are used to sort the alerts by their probability of being a supernova so an astronomer can see them in the [Supernova Hunter](https://snhunter.alerce.online/) web interface, to then report the candidates to TNS.\n",
+ "\n",
+ "### Training Data: \n",
+ "Using the labeled set from P. Sánchez-Sáez et al 2020 (The ALeRCE light curve classifier paper), we gathered 52,244 first alerts to build our training set. The number of samples of AGN, SN, VS, asteroid, and bogus are 14,966(29%), 1620 (3%), 14,996 (29%), 9899 (19%), and 10,763(20%) respectively. The images in the alert are the science, template and difference image. We noticed each of these classes can be classified fast since they present specific properties that can be found in the information of the first alert:\n",
+ "\n",
+ "+ AGN: Being stochastically variable objects, an alert generated by an AGN should have flux from the source in both the reference and science stamps. Considering this feature alone, it is difficul. t to discriminate AGNs from other variable sources. Nevertheless, AGNs should lie at the centers of their host galaxies, or appear as (quasi-)stellar objects, in relatively lower stellar density fields. Thus, a change in flux will appear as a variable source, which may lie at the center of a galaxy, or even when the galaxy is not visible they tend to be in lower stellar density fields. In these cases, the alert is likely to be triggered by an AGN. In addition, AGNs are commonly found outside the Galactic plane.\n",
+ " \n",
+ "+ Supernovae (SNe): An alert generated by a SN should appear as a change in flux where no unresolved sources were present. These transients tend to appear near their host galaxies, and their location should be consistent with the underlying host stellar population distribution (e.g., a SN will have a higher probability of arising from a location aligned with the disk than perpendicular to it). As such, most SN detections exhibit a visible host galaxy in both the science and reference stamps, with the flux from the SN arising only in the science and difference images. SN candidates tend to appear outside the Galactic plane due to occlusion.\n",
+ " \n",
+ "+ Variable Stars (VS): The flux coming from variable stars usually appears in both the reference and science stamps. With ZTF's sensitivity, variable stars can be detected within the Milky Way or the Local Group, and thus the alert will typically not be associated with a visible host galaxy in the stamp, but rather with other point-like sources. In addition, such alerts will have a higher probability of residing at lower Galactic latitudes and in crowded fields with multiple point sources within the stamps, given the high concentration of stars in the disk and bulge of our Galaxy.\n",
+ "\n",
+ "+ Asteroids: Alerts from moving Solar-system objects will appear only one time at a given position, and thus will show flux only in the science and difference images. Depending on their distance and speed, they may appear elongated in the direction of motion. In addition, such alerts should have a higher probability of residing near the ecliptic.\n",
+ " \n",
+ "+ Bogus alerts: Camera and telescope optics effects, such as saturated pixels at the centers of bright sources, bad columns, hot pixels, astrometric misalignment in the subtraction to compute the difference image, unbaffled internal reflections, etc., can produce bogus alerts with no interesting real source. Bogus alerts are characterized by the presence of NaN pixels due to saturation, single or multiple bright pixels with little or no spatial extension (i.e., smaller than the telescope point spread function (PSF) and nightly seeing), or lines with high or low pixel values that extend over a large portion of the stamp (hot or cold columns/rows).\n",
+ "\n",
+ "\n",
+ "\n",
+ "Here we show examples of the five classes that are to be discriminated by using only the first detection. For each class, the triplet of images in each row are science, reference and difference images from left to right. Each row corresponds to a different candidate. The images are cropped at the center resulting in 21x21 images, and normalized. Further details about the pre-processing of the images is explained in Carrasco-Davis et al 2020.\n",
+ "\n",
+ "### Convolutional neural network architecture: \n",
+ "\n",
+ "
\n",
+ "\n",
+ "The Figure on the left represent the stamp classifier. The box Convolutional Layers refers to those described in the Table at the right side, from the first convolutional layer to the last pooling layer. For each sample, the science, reference and difference images are concatenated in the channel dimension, obtaining an image input of dimension $21 \\times 21 \\times 3$. For each sample within the sampled batch, rotated versions are generated and fed to the CNN. After the first dense layer, the dense vectors of each the rotated version for the same objects are averaged. The metadata features pass through a batch normalization layer, the output of which is concatenated with the cyclic pooling output. Then, the concatenation goes through 2 fully connected layers, and finally a softmax function is applied to estimate the output probabilities. For the convolutional layers, the parameters shown in the table at the right are the filter dimensions and number of output channels. All the convolutional layers and fully connected layers have a Recified Linear Unit (ReLU) activation function (except for the last fully connected one that has a softmax output).\n",
+ "\n",
+ "When the CNN model is trained using cross-entropy as the loss function to be minimized, the classification confidence of the model is very high, resulting in a distribution of output probabilities with saturated values of 0s and 1s without populating the values in between, even for wrong classifications. In this case there is no insight of certainty (relative probabilities between classes) of the prediction because most estimated probabilities for each class were either 0 or 1. In order to provide more granularity to the astronomers who revise SN candidates reported by the model to later request follow-up observing time, we added the entropy of the predicted probabilities of the models as a regularization term, to be maximized during training. By maximizing the entropy of the output probabilities, we penalize predictions with high confidence, in order to get better insight in cases where the stamps seem equally likely to belong to more than one class. The loss function $\\mathcal{L}$ per sample is as follows:\n",
+ "\n",
+ "$$ \\mathcal{L} = \\underbrace{-\\sum_{c=1}^{N}y_{c}\\log{(\\hat{y}_{c})}}_{\\text{cross-entropy}} + \\underbrace{\\beta \\sum_{c=1}^{N}\\hat{y}_{c}\\log{(\\hat{y}_{c})}}_{\\text{entropy regularization}},$$\n",
+ "where $N$ is the number of classes, $y_{c}$ is the one-hot encoding label (a value of 1 in the corresponding index of class, and 0 for the rest) indexed by $c$, $\\hat{y}_{c}$ is the model prediction for class $c$, and $\\beta$ controls the regularization term in the loss function.\n",
+ "\n",
+ "### Clasification results: \n",
+ "\n",
+ "
\n",
+ "\n",
+ "In the left image we show the confusion matrix in the test set for the proposed model. We use accuracy to compare models since the validation and test sets are balanced; achieving $0.95 \\pm 0.005$ in the validation set and $0.941 \\pm 0.004$ in the test set. To compute the confusion matrix we run five realizations of the mentioned model. With our five class model, we recover $87 \\pm 1\\%$ of the SNe, with only $5 \\pm 2\\%$ of false positives. By inspecting the predictions made by our model for each SN sample in the test set, we found that the results are in agreement with our initial expectations regarding the class discrimination described in the Training Data section. It is worth highlighting that the results of our model are achieved by using **the first alert only**.\n",
+ "\n",
+ "In the right image, we show the probability distribution for each of the classes in the training set, for different values of the regularization constant $\\beta=\\{ 0, 0.5, 1.0 \\}$. For the model without regularization ($\\beta = 0$ shown on the top plot), the probability distribution saturates to 1 or 0. Increasing $\\beta$ to 0.5 or 1.0 decreases the saturation and spreads the distribution of predictions made by the model (mid and bottom plots). In the case of $\\beta=0$, the predictions are mostly saturated around 0 or 1 for the SN, VS, Asteroids and Bogus alert classes, creating difficulties to identify stamps that seem equally likely to belong to more than one class, because every sample is mapped to similar levels of high certainty. As the value of $\\beta$ increases, the saturation of predicted values decreases, spreading the predicted probability distributions and emphasizing the different levels of certainty between predictions of different samples. The use of regularization to find noticeable differences in the predicted probabilities are helpful to experts for evaluating the output of the classifier, gaining better insight into how reliable the classifications are.\n",
+ "\n",
+ "# ALeRCE frontend and Supernova Hunter \n",
+ "\n",
+ "The [Supernova Hunter](https://snhunter.alerce.online/) is a visualization tool that allows the user to inspect SN candidates classified by the model in real time, in order to select good targets for follow-up observations.\n",
+ "\n",
+ "\n",
+ "\n",
+ "On the bottom left side, the location of each candidate in sky coordinates with respect to the Galactic plane and the ecliptic are depicted. On the bottom right side, a selection of the top candidates is listed, initially ordered by SN probability score from the stamp classifier. The list of candidates can be sorted by other parameters, and updated/refreshed to include newly ingested alerts. On the top left side, the SN candidate ID is shown as a clickable link to the [ALeRCE frontend](https://alerce.online/object/ZTF20abpvolc), with relevant metadata such as radec coordinates, magnitude, date, etc. At the bottom there are links to other sources of information, including ALeRCE, NED, TNS, and the Simbad Astronomical Database. In the middle of the figure there is a colored image from Aladin. On the top right side, the stamps of the first detection are shown, along with buttons for reporting the candidate as eventual bogus or as a possible SN.\n",
+ "\n",
+ "\n",
+ "\n",
+ "When the candidate is inspected using the [ALeRCE frontend](https://alerce.online/object/ZTF20abpvolc), the full prediction of classes by the stamp classifier is shown at the bottom in addition to the full information of the light curve available, metadata, cross matches, etc. When the light curve of the object has six or more points in one of the bands, it is classified by the light curve classifier (Sánchez-Sáez et al. 2020), with the prediction also available in the ALeRCE frontend with a more complex taxonomy."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Requirements "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from IPython.display import display\n",
+ "import sys\n",
+ "import pandas as pd\n",
+ "import numpy as np\n",
+ "import matplotlib.pyplot as plt\n",
+ "from astropy.time import Time\n",
+ "import ephem\n",
+ "from coords_utils import ec2gal\n",
+ "from tqdm import tqdm\n",
+ "import copy\n",
+ "import pickle"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# ALeRCE client "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Install and import the alerce API client https://alerce.readthedocs.io/en/latest/"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#!python3 -m pip install --user alerce\n",
+ "from alerce.api import AlerceAPI\n",
+ "client = AlerceAPI()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Plotting light curves "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We will create a simple function that plots the light curve given an object id (oid), a dataframe with detections and a dataframe with non detections. This function will be used to plot light curves of some objects predicted by the stamp classifier."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def plotLC(oid, SN_det, SN_nondet):\n",
+ " \n",
+ " fig, ax = plt.subplots(figsize = (14, 7))\n",
+ " labels = {1: 'g', 2: 'r'}\n",
+ " colors = {1: '#56E03A', 2: '#D42F4B'} # color blind friendly green and red \n",
+ " markers = {1: 'o', 2: 's'}\n",
+ " sizes = {1: 30, 2: 60}\n",
+ " \n",
+ " # loop the passbands\n",
+ " for fid in [1, 2]:\n",
+ " \n",
+ " # plot detections if available\n",
+ " mask = SN_det.fid == fid\n",
+ " if np.sum(mask) > 0:\n",
+ " # note that the detections index is candid and that we are plotting the psf corrected magnitudes\n",
+ " ax.errorbar(SN_det[mask].mjd, SN_det[mask].magpsf, \n",
+ " yerr = SN_det[mask].sigmapsf, c=colors[fid], label=labels[fid], marker=markers[fid])\n",
+ " \n",
+ " # plot non detections if available\n",
+ " mask = (SN_nondet.fid == fid) & (SN_nondet.diffmaglim > -900)\n",
+ " if np.sum(mask) > 0: \n",
+ " # non detections index is mjd\n",
+ " ax.scatter(SN_nondet[mask].index, SN_nondet[mask].diffmaglim, c=colors[fid], alpha = 0.5,\n",
+ " marker='v', label=\"lim.mag. %s\" % labels[fid], s=sizes[fid])\n",
+ " \n",
+ " ax.set_title(oid)\n",
+ " ax.set_xlabel(\"MJD\")\n",
+ " ax.set_ylabel(\"Apparent magnitude\")\n",
+ " ax.legend()\n",
+ " ax.set_ylim(ax.get_ylim()[::-1])\n",
+ "\n",
+ "def getSNdata(oid, doLC=False, dostamp=False):\n",
+ "\n",
+ " results = {\"oid\": oid}\n",
+ " \n",
+ " # query detections\n",
+ " SN_det = client.get_detections(oid, format='pandas')\n",
+ " results[\"lc_det\"] = SN_det\n",
+ " \n",
+ " # query non detections\n",
+ " SN_nondet = client.get_non_detections(oid, format='pandas')\n",
+ " results[\"lc_nondet\"] = SN_nondet\n",
+ " \n",
+ " # plot the LC\n",
+ " if doLC:\n",
+ " plotLC(oid, SN_det, SN_nondet)\n",
+ " \n",
+ " # show the first image stamp\n",
+ " if dostamp:\n",
+ " client.plot_stamp(oid) \n",
+ " \n",
+ " # return data\n",
+ " return results"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Classified Alerts "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Get objects per class\n",
+ "Here, we build a function that uses the ALeRCE client to query objects according to the predictions of the stamp classifier. In order to select the classifier, we define the query in terms of the **pcclassearly**, the classes are the following strings:\n",
+ "+ \"agn\": Active Galactic Nuclei\n",
+ "+ \"sn\": Supernova\n",
+ "+ \"vs\": Variable star\n",
+ "+ \"asteroid\"\n",
+ "+ \"bogus\"\n",
+ "\n",
+ "Using the ALeRCE client we can select the min and max number of observation per object, also sort them by firstmjd, number of observations (nobs), probability of the selected class (classearly), etc. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def get_objects_per_class(classearly=\"sn\", pclassearly=0.5, min_nobs=10, max_nobs=50, n_objects=100):\n",
+ "\n",
+ " total = n_objects\n",
+ " records_per_page = n_objects\n",
+ " page = 1\n",
+ " sortBy = \"pclassearly\"\n",
+ " min_firstmjd = Time(\"2017-11-01T00:00:00\", format=\"isot\", scale=\"utc\").mjd + 70\n",
+ "\n",
+ " params = {\n",
+ " \"records_per_pages\": records_per_page, # Number of queried objects\n",
+ " \"page\": page,\n",
+ " \"sortBy\": sortBy,\n",
+ " \"query_parameters\": {\n",
+ " \"filters\": {\n",
+ " \"pclassearly\": pclassearly, # Probability of objects for classearly\n",
+ " \"classearly\": classearly, # string or identifier of the queried class\n",
+ " \"nobs\": {\n",
+ " \"min\": min_nobs, # Min number of observations\n",
+ " \"max\": max_nobs # Max number of observations\n",
+ " },\n",
+ " \"dates\": {\n",
+ " \"firstmjd\": {\n",
+ " \"min\": min_firstmjd # Min mjd of the first detection\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " \n",
+ " objects = client.query(params, format='pandas')\n",
+ " objects.sort_values(by=\"nobs\", ascending=False, inplace=True)\n",
+ " print(objects.shape)\n",
+ " \n",
+ " return objects"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The previous function returns a pandas dataframe with a list of objects by their ZTF oid as index, and relevant information in their columns as we will see next.\n",
+ "Now, lets query some objects for each of the class available from the stamp classifier. For now, the ALeRCE client works with class ids numbers as shown in the dictionary defined below as early_classes. Here, we are querying a few thousands of objects from the ALeRCE database. We will also sort the classes by ascending number of observations in the cases of bogus and asteroids in order to get a clean examples. AGNs, VS and SNe are sorted in descending order of number of observations to appreciate better the light curves for each of these classes.\n",
+ "\n",
+ "**This query will take a few seconds**"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(10000, 43)\n",
+ "Result of a query using the ALeRCE client\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "
"
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "for cl, df in objects.items():\n",
+ " print(\"######### \"+ cl + \" class #########\")\n",
+ " if cl in [\"asteroid\", \"bogus\"]: # Just to have well classified samples to show\n",
+ " df.sort_values(by=\"nobs\", inplace=True)\n",
+ " plot_some_LC(df, n_examples=2)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The distribution of number of detection per class can give us insights on how well the classifier is working. For instance, for AGNs, SNe, and VS, we expect to see more than one detection for these classes. The objects belonging to these classes and have one detection could be new objects or the ones visited only once, short transients, or errors in the classification. For asteroids and bogus class, we expect to have objects with one detection only."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "
"
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "f, ax = plt.subplots(2, 3, figsize=(20, 8))\n",
+ "bins = np.arange(0, 40)\n",
+ "ax = ax.flatten()\n",
+ "fontsize = 16\n",
+ "\n",
+ "# Concatenated dataframe with all the classes\n",
+ "h, _ = np.histogram(pd.concat(list(objects.values()), axis=0).nobs, bins=bins, density=True)\n",
+ "ax[0].bar(bins[:-1], h, label=\"all_data\")\n",
+ "ax[0].set_title(\"all data\", fontsize=fontsize)\n",
+ "ax[0].tick_params(axis='both', which='major', labelsize=fontsize)\n",
+ "ax[0].set_ylim([0, 1])\n",
+ "\n",
+ "for i, (cl, df) in enumerate(objects.items()):\n",
+ " # Histogram per class\n",
+ " h, bins = np.histogram(df.nobs, bins=bins, density=True)\n",
+ " ax[i+1].bar(bins[:-1], h, label=cl)\n",
+ " ax[i+1].set_title(cl, fontsize=fontsize)\n",
+ " ax[i+1].tick_params(axis='both', which='major', labelsize=fontsize)\n",
+ " ax[i+1].set_ylim([0, 1])\n",
+ " \n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "In our small sample we can see that most of the asteroids and bogus has only one detection, while AGNs, SNe and VS have a long tail of objects with more than one detection."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Get a sequence of stamps \n",
+ "\n",
+ "In this section, we will implement a function to get a series of stamps from the alerts availables for a given object. The stamp classifier uses the first alert only, but in ALeRCE we are working to extend the stamp classifier to use a multi-stamp approach considering band g and r simultaneusly. One example of this kind of classifier is the one presented in [Carrasco-Davis et al. 2019](https://iopscience.iop.org/article/10.1088/1538-3873/aaef12), where the stamps are processed by a convolutional neural network, to then use a recurrent neural network that can store previous inputs of the sequence of stamps. Then, in case you want to experiment with a sequence of stamps, here is an example:\n",
+ "+ We first get the detections of an object using get_detection of the ALeRCE client\n",
+ "+ Then we sort the values by mjd\n",
+ "+ The resulting dataframe uses the candidate id as index, then we use the get_stamp function for each candidate to obtain their respective stamp\n",
+ "+ We finally plot the stamps and return a list of stamps"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [],
+ "source": [
+ "def get_stamp_series(oid, n_stamps=5, show_stamps=False, show_LC=False):\n",
+ " band_labels = {1: 'g', 2: 'r'}\n",
+ " band_cmap = {\"g\": \"viridis\", \"r\": \"inferno\"}\n",
+ " object_det = client.get_detections(oid, format='pandas') # Get object detections\n",
+ " object_det.sort_values(by=\"mjd\", inplace=True) # Sort values by MJD\n",
+ " object_det = object_det.iloc[:n_stamps] # Use n stamps only\n",
+ " candids = object_det.index.values # Obtain the candidate id to use it in the get_stamp function\n",
+ " stamps = []\n",
+ " for c in candids:\n",
+ " stamps.append(client.get_stamps(oid, candid=c)) # Save the stamps in a list to return\n",
+ " \n",
+ " if show_LC:\n",
+ " getSNdata(oid, doLC=True, dostamp=False) # Plot the light curve of the object\n",
+ " \n",
+ " if show_stamps:\n",
+ " f, ax = plt.subplots(3, n_stamps, figsize=(n_stamps*3, 10))\n",
+ " for i, s in enumerate(stamps):\n",
+ " corresponding_band = band_labels[object_det.iloc[i].fid] # Check band\n",
+ " cmap = band_cmap[corresponding_band] # use a different cmap for each band\n",
+ " # The get_stamp function returns a list with three images \n",
+ " #(science, reference and difference) in hdu format. Using .data we get the numpy array.\n",
+ " science, ref, difference = s[0].data, s[1].data, s[2].data \n",
+ " ax[0, i].imshow(np.log(science), cmap=cmap) # Log scale for visualization\n",
+ " ax[1, i].imshow(np.log(ref), cmap=cmap)\n",
+ " ax[2, i].imshow(difference, cmap=cmap)\n",
+ " title = \"MjD \" + str(np.round(object_det.iloc[i].mjd, 2)) + \", band: \"+corresponding_band\n",
+ " ax[0, i].set_title(title)\n",
+ " ax[0, 0].set_ylabel(\"Science\")\n",
+ " ax[1, 0].set_ylabel(\"Reference\")\n",
+ " ax[2, 0].set_ylabel(\"Difference\")\n",
+ " plt.show()\n",
+ " return object_det, stamps"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Let's use [ZTF19acftude](https://alerce.online/object/ZTF19acftude) since it has a very nice looking SN light curve and host galaxy."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "WARNING: VerifyWarning: Keyword name 'STAMP_TYPE' is greater than 8 characters or contains characters not allowed by the FITS standard; a HIERARCH card will be created. [astropy.io.fits.card]\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "
"
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "detections, stamps = get_stamp_series(oid=\"ZTF19acftude\", show_stamps=True, show_LC=True)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The detections dataframe contains important information regarding the specific properties measured at a given time for the same object. For instance, the measuring time in mjd, mean and std of the position in radec, the magnitude, etc."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "
\n",
+ "\n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
\n",
+ "
candid_str
\n",
+ "
dec
\n",
+ "
diffmaglim
\n",
+ "
distpsnr1
\n",
+ "
fid
\n",
+ "
field
\n",
+ "
has_stamps
\n",
+ "
isdiffpos
\n",
+ "
magap
\n",
+ "
magap_corr
\n",
+ "
...
\n",
+ "
rb
\n",
+ "
rcid
\n",
+ "
sgscore1
\n",
+ "
sigmadec
\n",
+ "
sigmagap
\n",
+ "
sigmagap_corr
\n",
+ "
sigmagnr
\n",
+ "
sigmapsf
\n",
+ "
sigmapsf_corr
\n",
+ "
sigmara
\n",
+ "
\n",
+ "
\n",
+ "
candid
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ "
\n",
+ " \n",
+ " \n",
+ "
\n",
+ "
1022193432915015007
\n",
+ "
1022193432915015007
\n",
+ "
25.629601
\n",
+ "
20.080343
\n",
+ "
5.715140
\n",
+ "
2
\n",
+ "
645
\n",
+ "
None
\n",
+ "
1
\n",
+ "
19.083401
\n",
+ "
19.083401
\n",
+ "
...
\n",
+ "
0.755714
\n",
+ "
29
\n",
+ "
0.047375
\n",
+ "
0.09
\n",
+ "
0.1378
\n",
+ "
0.1378
\n",
+ "
0.090
\n",
+ "
0.160765
\n",
+ "
0.160765
\n",
+ "
0.09
\n",
+ "
\n",
+ "
\n",
+ "
1025183682915015015
\n",
+ "
1025183682915015015
\n",
+ "
25.629646
\n",
+ "
20.761301
\n",
+ "
6.058198
\n",
+ "
1
\n",
+ "
645
\n",
+ "
None
\n",
+ "
1
\n",
+ "
18.732500
\n",
+ "
18.732500
\n",
+ "
...
\n",
+ "
0.885714
\n",
+ "
29
\n",
+ "
0.047375
\n",
+ "
0.09
\n",
+ "
0.0712
\n",
+ "
0.0712
\n",
+ "
0.085
\n",
+ "
0.095342
\n",
+ "
0.095342
\n",
+ "
0.09
\n",
+ "
\n",
+ "
\n",
+ "
1025217712915015002
\n",
+ "
1025217712915015002
\n",
+ "
25.629655
\n",
+ "
20.306961
\n",
+ "
6.009156
\n",
+ "
2
\n",
+ "
645
\n",
+ "
None
\n",
+ "
1
\n",
+ "
19.118999
\n",
+ "
19.118999
\n",
+ "
...
\n",
+ "
0.942857
\n",
+ "
29
\n",
+ "
0.047375
\n",
+ "
0.09
\n",
+ "
0.1493
\n",
+ "
0.1493
\n",
+ "
0.090
\n",
+ "
0.094740
\n",
+ "
0.094740
\n",
+ "
0.09
\n",
+ "
\n",
+ "
\n",
+ "
1028185342915015006
\n",
+ "
1028185342915015006
\n",
+ "
25.629672
\n",
+ "
20.332592
\n",
+ "
5.922150
\n",
+ "
2
\n",
+ "
645
\n",
+ "
None
\n",
+ "
1
\n",
+ "
18.897900
\n",
+ "
18.897900
\n",
+ "
...
\n",
+ "
0.947143
\n",
+ "
29
\n",
+ "
0.047375
\n",
+ "
0.09
\n",
+ "
0.1201
\n",
+ "
0.1201
\n",
+ "
0.090
\n",
+ "
0.102625
\n",
+ "
0.102625
\n",
+ "
0.09
\n",
+ "
\n",
+ "
\n",
+ "
1028226352915015006
\n",
+ "
1028226352915015006
\n",
+ "
25.629628
\n",
+ "
20.651066
\n",
+ "
6.055933
\n",
+ "
1
\n",
+ "
645
\n",
+ "
None
\n",
+ "
1
\n",
+ "
18.590799
\n",
+ "
18.590799
\n",
+ "
...
\n",
+ "
0.837143
\n",
+ "
29
\n",
+ "
0.047375
\n",
+ "
0.09
\n",
+ "
0.0649
\n",
+ "
0.0649
\n",
+ "
0.085
\n",
+ "
0.111296
\n",
+ "
0.111296
\n",
+ "
0.09
\n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
5 rows × 27 columns
\n",
+ "
"
+ ],
+ "text/plain": [
+ " candid_str dec diffmaglim distpsnr1 \\\n",
+ "candid \n",
+ "1022193432915015007 1022193432915015007 25.629601 20.080343 5.715140 \n",
+ "1025183682915015015 1025183682915015015 25.629646 20.761301 6.058198 \n",
+ "1025217712915015002 1025217712915015002 25.629655 20.306961 6.009156 \n",
+ "1028185342915015006 1028185342915015006 25.629672 20.332592 5.922150 \n",
+ "1028226352915015006 1028226352915015006 25.629628 20.651066 6.055933 \n",
+ "\n",
+ " fid field has_stamps isdiffpos magap magap_corr \\\n",
+ "candid \n",
+ "1022193432915015007 2 645 None 1 19.083401 19.083401 \n",
+ "1025183682915015015 1 645 None 1 18.732500 18.732500 \n",
+ "1025217712915015002 2 645 None 1 19.118999 19.118999 \n",
+ "1028185342915015006 2 645 None 1 18.897900 18.897900 \n",
+ "1028226352915015006 1 645 None 1 18.590799 18.590799 \n",
+ "\n",
+ " ... rb rcid sgscore1 sigmadec sigmagap \\\n",
+ "candid ... \n",
+ "1022193432915015007 ... 0.755714 29 0.047375 0.09 0.1378 \n",
+ "1025183682915015015 ... 0.885714 29 0.047375 0.09 0.0712 \n",
+ "1025217712915015002 ... 0.942857 29 0.047375 0.09 0.1493 \n",
+ "1028185342915015006 ... 0.947143 29 0.047375 0.09 0.1201 \n",
+ "1028226352915015006 ... 0.837143 29 0.047375 0.09 0.0649 \n",
+ "\n",
+ " sigmagap_corr sigmagnr sigmapsf sigmapsf_corr sigmara \n",
+ "candid \n",
+ "1022193432915015007 0.1378 0.090 0.160765 0.160765 0.09 \n",
+ "1025183682915015015 0.0712 0.085 0.095342 0.095342 0.09 \n",
+ "1025217712915015002 0.1493 0.090 0.094740 0.094740 0.09 \n",
+ "1028185342915015006 0.1201 0.090 0.102625 0.102625 0.09 \n",
+ "1028226352915015006 0.0649 0.085 0.111296 0.111296 0.09 \n",
+ "\n",
+ "[5 rows x 27 columns]"
+ ]
+ },
+ "execution_count": 22,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "detections"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "Index(['candid_str', 'dec', 'diffmaglim', 'distpsnr1', 'fid', 'field',\n",
+ " 'has_stamps', 'isdiffpos', 'magap', 'magap_corr', 'magnr', 'magpsf',\n",
+ " 'magpsf_corr', 'mjd', 'oid', 'parent_candid', 'ra', 'rb', 'rcid',\n",
+ " 'sgscore1', 'sigmadec', 'sigmagap', 'sigmagap_corr', 'sigmagnr',\n",
+ " 'sigmapsf', 'sigmapsf_corr', 'sigmara'],\n",
+ " dtype='object')"
+ ]
+ },
+ "execution_count": 23,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "detections.columns"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Get spatial distribution of predictions "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "One issue when training models in astronomy is the possible biases found in the training set. One of the problems we had at the begining was the abundance of extragalactic objects predicted near the galactic plane. As mentioned in the Introduction, SNe and AGNs are less likely to be found near the galactic plane due to occlusion. We fixed this problem by adding some features to the CNN at the fully connected layers, one of the most important are the ecliptic and galactic coordinates, so let's see how they look like per class."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Compute galactic and ecliptic coordinates\n",
+ "We computed the coordinates from the meanra and mean dec using ephem python package. Here (https://rhodesmill.org/pyephem/coordinates.html) there is a more detailed tutorial of how to use this transformation between coordinates. Now, we will take our dataframes and add ecliptic and galactic latitude and longitude as columns."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def ecliptic_coordinates(df):\n",
+ " # Define the lambda function to be applied to each row of the dataframe\n",
+ " ecl = df.apply(lambda row: ephem.Ecliptic(ephem.Equatorial('%s' % (row.meanra / 15.),\n",
+ " '%s' % row.meandec, epoch=ephem.J2000)), axis=1)\n",
+ " \n",
+ " # Apply the function to each row, the result is append as a new column\n",
+ " df[\"ecl_lat\"] = ecl.apply(lambda row: np.rad2deg(row.lat))\n",
+ " df[\"ecl_long\"] = ecl.apply(lambda row: np.rad2deg(row.long))\n",
+ " \n",
+ " # Return the updated dataframe\n",
+ " return df\n",
+ "\n",
+ "\n",
+ "def galactic_coordinates(df):\n",
+ " # Define the lambda function to be applied to each row of the dataframe\n",
+ " gal = df.apply(lambda row: ephem.Galactic(ephem.Equatorial('%s' % (row.meanra / 15.),\n",
+ " '%s' % row.meandec, epoch=ephem.J2000)), axis=1)\n",
+ " # Apply the function to each row, the result is append as a new column\n",
+ " df[\"gal_lat\"] = gal.apply(lambda row: np.rad2deg(row.lat))\n",
+ " df[\"gal_long\"] = gal.apply(lambda row: np.rad2deg(row.long))\n",
+ " \n",
+ " # Return the updated dataframe\n",
+ " return df\n",
+ "\n",
+ "# For each of our dataframe corresponding to one of the classes of the stamp classifier\n",
+ "# we add the ecliptic and galactic coordinates\n",
+ "for cl, df in objects.items():\n",
+ " ecl_df = ecliptic_coordinates(df)\n",
+ " gal_df = galactic_coordinates(ecl_df)\n",
+ " objects[cl] = gal_df"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def plot_spatial_distribution(df, ax, index=0, cmin_val=0.1, vmax_val=5, titles=\"\", fontsize=18):\n",
+ " dims = ['gal_long', 'gal_lat'] # Let's use our new column to plot the objects\n",
+ " _, _, _, im = ax.hist2d(df[dims].values[:, 0], df[dims].values[:, 1],\n",
+ " (300, 300), cmap=\"viridis\", vmax=vmax_val, cmin=cmin_val)\n",
+ " ax.tick_params(axis='both', labelsize=fontsize-2)\n",
+ " ax.set_xlim([0, 360])\n",
+ " ax.set_ylim([-80, 80])\n",
+ " cbar = plt.colorbar(im,ax=ax, pad=0)\n",
+ " cbar.ax.tick_params(labelsize=fontsize-4)\n",
+ " \n",
+ " ecliptic_lat = np.linspace(0, 0, num=500)\n",
+ " ecliptic_longi = np.linspace(0, 360, num=500)\n",
+ " # This auxiliary function computes the ecliptic plane by converting\n",
+ " # the ecliptic latitude 0 to galactic coordinates\n",
+ " galact_long, galact_lat = ec2gal(ecliptic_longi, ecliptic_lat)\n",
+ " \n",
+ " ax.set_title(titles, fontsize=fontsize)\n",
+ " ax.plot(galact_long, galact_lat, \"-k\", lw=6)\n",
+ " ax.plot(galact_long, galact_lat, \"-y\", lw=4)\n",
+ " \n",
+ " return ax"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "metadata": {
+ "scrolled": false
+ },
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "f, ax = plt.subplots(2, 3, figsize=(3*10, 2*6))\n",
+ "plt.subplots_adjust(wspace=0.0, hspace=0.15)\n",
+ "ax = ax.flatten()\n",
+ "titles = [\"all objects\", ] + list(objects.keys())\n",
+ "max_count_per_class = dict(zip(titles, [20, 5, 5, 10, 5, 5]))\n",
+ "\n",
+ "for i in range(6):\n",
+ " if i == 0:\n",
+ " df = pd.concat(list(objects.values()), axis=0)\n",
+ " else:\n",
+ " df = list(objects.values())[i-1]\n",
+ " \n",
+ " ax[i] = plot_spatial_distribution(df, ax[i], titles=titles[i], vmax_val=max_count_per_class[titles[i]])\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The probabilities in this case are from a previous model without features, this is why you see an abundance of extragalactic objects near the galactic plane (AGNs and SNe). This problem was fixed as mentioned earlier by adding some features from the alert metadata. **This section will be updated with the probabilities of the new database that includes the predictions of the model with features**."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "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": "3.7.7"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/notebooks/coords_utils.py b/notebooks/coords_utils.py
new file mode 100644
index 0000000..4cef257
--- /dev/null
+++ b/notebooks/coords_utils.py
@@ -0,0 +1,1208 @@
+"""
+ Taken from Erin Sheldon's esutil package at:
+ https://github.com/esheldon/esutil/blob/master/esutil/coords.py
+ 2020
+ NAME
+ coords
+ PURPOSE
+ A set of astronomical utilities for dealing with coordinates and
+ coordinate transformations.
+ COORDINATE TRANSFORMATIONS
+ euler:
+ A generic routine for transforming between Galactic, Celestial,
+ and ecliptic coords. The following wrapper routines are also
+ supplied for convenience:
+ l,b = eq2gal(ra, dec, b1950=False, dtype='f8')
+ Convert equatorial to glactic coordinates.
+ # The following use the same interface:
+ gal2eq
+ Convert galactic to equatorial coordinates.
+ eq2ec
+ Convert equatorial to ecliptic coordinates.
+ ec2eq
+ Convert ecliptic to equatorial coordinates.
+ ec2gal
+ Convert ecliptic to galactic coordinates.
+ gal2ec
+ Convert galactic to ecliptic coordinates.
+ # These SDSS specific functions do not use euler
+ eq2sdss
+ Convert between equatorial and corrected SDSS survey coords.
+ sdss2eq
+ Convert between corrected SDSS survey and equatorial coords.
+ eq2xyz: Convert equatorial to x,y,z on the sphere according to
+ the following transform:
+ x = sin(pi/2-dec)*cos(ra)
+ y = sin(pi/2-dec)*sin(ra)
+ z = cos(pi/2-dec)
+ xyz2eq:
+ inverse of eq2xyz
+ sphdist:
+ Calculate the arc length between two sets of points on the sphere.
+ Currently only takes ra,dec.
+ shiftlon:
+ shift the input longitude. By default wrap the coordinate to
+ -180,180. If a shift is entered, return the new value
+ lon-shift such that the range is still [0,360)
+ shiftra:
+ shift right ascension. This just calls shiftlon
+ radec2aitoff:
+ Convert ra,dec to aitoff coordinates.
+ dec_parse(decstring)
+ parse a colon separated string representing declination ito
+ degrees.
+ ra_parse(decstring)
+ parse a colon separated string representing right ascension ito
+ degrees.
+ randsphere(numrand, system='eq', ra_range=[0,360], dec_range=[-90,90]):
+ Generate random points on the sphere. By default ra,dec are
+ returned. If system='xyz' then x,y,z are returned.
+ randcap(nrand,ra,dec,rad,get_radius=False):
+ Create random points in a cap, or disc, centered at the
+ input ra,dec location and with radius rad.
+ rect_area(lon_min, lon_max, lat_min, lat_max)
+ Calculate the area of a rectangle on the sphere.
+"""
+import math
+import numpy as np
+
+try:
+ import numpy
+ from numpy import where, sin, cos, arccos, arcsin, arctan2, sqrt, rad2deg, deg2rad, zeros
+ have_numpy = True
+except:
+ have_numpy = False
+
+
+_license = """
+ Copyright (C) 2009 Erin Sheldon
+ This program is free software; you can redistribute it and/or modify it
+ under the terms of version 2 of the GNU General Public License as
+ published by the Free Software Foundation.
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software
+ Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+"""
+
+PI = math.pi
+HALFPI = PI / 2.0
+D2R = PI / 180.0
+R2D = 1.0 / D2R
+
+_centra = 185.
+_centdec = 32.5
+_sdsspar = {'center_ra': _centra,
+ 'center_dec': _centdec,
+ 'node': (_centra - 90.0) * D2R,
+ 'etapole': _centdec * D2R,
+ 'etaoffset': 91.25,
+ 'doc': """
+ A set of transformation functions for use with SDSS coordinate systems.
+ eq2sdss(): Convert between equatorial and corrected SDSS survey coords.
+ sdss2eq(): Convert between corrected SDSS survey and equatorial coords.
+ Don't use these unless you have to, as these are the old coordinates
+ eq2survey(): Convert between equatorial and uncorrected survey coords.
+ survey2eq(): Convert between uncorrected survey and equatorial coords.
+ Adapted from astrotools
+ Erin Sheldon, NYU, 2006-03-11
+ Force data type and allow selection of dtype through keyword.
+ Erin Sheldon, NYU, 2007-05-23
+"""
+ }
+
+def euler(ai_in, bi_in, select, b1950=False, dtype='f8'):
+ """
+ NAME:
+ euler
+ PURPOSE:
+ Transform between Galactic, celestial, and ecliptic coordinates.
+ CALLING SEQUENCE:
+ long_out, lat_out =
+ euler(long_in, lat_in, type, b1950=False, dtype='f8')
+ INPUTS:
+ long_in - Input Longitude in DEGREES, scalar or vector.
+ lat_in - Input Latitude in DEGREES
+ select - Integer (1-6) specifying type of coordinate transformation.
+ select From To | select From To
+ 1 RA-Dec (2000) Galactic | 4 Ecliptic RA-Dec
+ 2 Galactic RA-DEC | 5 Ecliptic Galactic
+ 3 RA-Dec Ecliptic | 6 Galactic Ecliptic
+ Celestial coordinates (RA, Dec) should be given in equinox J2000
+ unless the b1950=True keyword is set.
+ OUTPUTS:
+ long_out - Output Longitude in DEGREES
+ lat_out - Output Latitude in DEGREES
+ INPUT KEYWORD:
+ b1950 - If this keyword is true then input and output
+ celestial and ecliptic coordinates should be given in equinox
+ B1950.
+ REVISION HISTORY:
+ Written W. Landsman, February 1987
+ Adapted from Fortran by Daryl Yentis NRL
+ Converted to IDL V5.0 W. Landsman September 1997
+ Made J2000 the default, added /FK4 keyword W. Landsman December 1998
+ Add option to specify SELECT as a keyword W. Landsman March 2003
+ Converted from IDL to numerical Python: Erin Sheldon, NYU, 2008-07-02
+ """
+
+ # Make a copy as an array. ndmin=1 to avoid messed up scalar arrays
+ ai = numpy.array(ai_in, ndmin=1, copy=True, dtype=dtype)
+ bi = numpy.array(bi_in, ndmin=1, copy=True, dtype=dtype)
+
+ twopi = 2.0 * PI
+ fourpi = 4.0 * PI
+
+ # J2000 coordinate conversions are based on the following constants
+ # (see the Hipparcos explanatory supplement).
+ # eps = 23.4392911111d Obliquity of the ecliptic
+ # alphaG = 192.85948d Right Ascension of Galactic North Pole
+ # deltaG = 27.12825d Declination of Galactic North Pole
+ # lomega = 32.93192d Galactic longitude of celestial equator
+ # alphaE = 180.02322d Ecliptic longitude of Galactic North Pole
+ # deltaE = 29.811438523d Ecliptic latitude of Galactic North Pole
+ # Eomega = 6.3839743d Galactic longitude of ecliptic equator
+ # Parameters for all the different conversions
+ if b1950:
+ # equinox = '(B1950)'
+ psi = numpy.array([0.57595865315, 4.9261918136,
+ 0.00000000000, 0.0000000000,
+ 0.11129056012, 4.7005372834], dtype=dtype)
+ stheta = numpy.array([0.88781538514, -0.88781538514,
+ 0.39788119938, -0.39788119938,
+ 0.86766174755, -0.86766174755], dtype=dtype)
+ ctheta = numpy.array([0.46019978478, 0.46019978478,
+ 0.91743694670, 0.91743694670,
+ 0.49715499774, 0.49715499774], dtype=dtype)
+ phi = numpy.array([4.9261918136, 0.57595865315,
+ 0.0000000000, 0.00000000000,
+ 4.7005372834, 0.11129056012], dtype=dtype)
+ else:
+ # equinox = '(J2000)'
+
+ psi = numpy.array([0.57477043300, 4.9368292465,
+ 0.00000000000, 0.0000000000,
+ 0.11142137093, 4.71279419371], dtype=dtype)
+ stheta = numpy.array([0.88998808748, -0.88998808748,
+ 0.39777715593, -0.39777715593,
+ 0.86766622025, -0.86766622025], dtype=dtype)
+ ctheta = numpy.array([0.45598377618, 0.45598377618,
+ 0.91748206207, 0.91748206207,
+ 0.49714719172, 0.49714719172], dtype=dtype)
+ phi = numpy.array([4.9368292465, 0.57477043300,
+ 0.0000000000, 0.00000000000,
+ 4.71279419371, 0.11142137093], dtype=dtype)
+
+ # zero offset
+ i = select - 1
+ a = (ai * D2R) - phi[i]
+
+ b = bi * D2R
+ sb = sin(b)
+ cb = cos(b)
+ cbsa = cb * sin(a)
+ b = -stheta[i] * cbsa + ctheta[i] * sb
+ w, = numpy.where(b > 1.0)
+ if w.size > 0:
+ b[w] = 1.0
+ bo = arcsin(b) * R2D
+
+ a = arctan2(ctheta[i] * cbsa + stheta[i] * sb, cb * cos(a))
+
+ ao = ((a + psi[i] + fourpi) % twopi) * R2D
+
+ return ao, bo
+
+#
+# Some clearer shortcut functions which call Euler
+#
+def eq2gal(ra, dec, b1950=False, dtype='f8'):
+ """
+ NAME
+ eq2gal
+ PURPOSE
+ Convert from equatorial to galactic coordinates in units of degrees.
+ CALLING SEQUENCE
+ l,b = eq2gal(ra, dec, b1950=False, dtype='f8')
+ INPUTS
+ ra, dec: Equatorial coordinates. May be Numpy arrays, sequences, or
+ scalars as long as they are all the same length. They must be
+ convertible to a Numpy array with the specified datatype.
+ KEYWORDS
+ b1950: If True, use b1950 coordiates. By default j2000 are used.
+ dtype: The datatype of the output arrays. Default is f8
+ OUTPUTS
+ l, b: Galactic longitude and latitude. The returned value is always
+ a Numpy array with the specified dtype
+ REVISION HISTORY
+ Created Erin Sheldon, NYU, 2008-07-02
+ """
+ return euler(ra, dec, 1, b1950=b1950, dtype=dtype)
+
+
+def gal2eq(l, b, b1950=False, dtype='f8'):
+ """
+ NAME
+ gal2eq
+ PURPOSE
+ Convert from galactice to equatorial coordinates in units of degrees.
+ CALLING SEQUENCE
+ ra,dec = gal2eq(l, b, b1950=False, dtype='f8')
+ INPUTS
+ l, b: Galactic coordinates. May be Numpy arrays, sequences, or
+ scalars as long as they are all the same length. They must be
+ convertible to a Numpy array with the specified datatype.
+ KEYWORDS
+ b1950: If True, use b1950 coordiates. By default j2000 are used.
+ dtype: The datatype of the output arrays. Default is f8
+ OUTPUTS
+ ra, dec: Equatorial longitude and latitude. The returned value is
+ always a Numpy array with the specified dtype
+ REVISION HISTORY
+ Created Erin Sheldon, NYU, 2008-07-02
+ """
+
+ return euler(l, b, 2, b1950=b1950, dtype=dtype)
+
+def eq2ec(ra, dec, b1950=False, dtype='f8'):
+ """
+ NAME
+ eq2ec
+ PURPOSE
+ Convert from equatorial to ecliptic coordinates in units of degrees.
+ CALLING SEQUENCE
+ lam,beta = eq2ec(ra, dec, b1950=False, dtype='f8')
+ INPUTS
+ ra, dec: Equatorial coordinates. May be Numpy arrays, sequences, or
+ scalars as long as they are all the same length. They must be
+ convertible to a Numpy array with the specified datatype.
+ KEYWORDS
+ b1950: If True, use b1950 coordiates. By default j2000 are used.
+ dtype: The datatype of the output arrays. Default is f8
+ OUTPUTS
+ lam, beta: Ecliptic longitude and latitude. The returned value is
+ always a Numpy array with the specified dtype
+ REVISION HISTORY
+ Created Erin Sheldon, NYU, 2008-07-02
+ """
+
+ return euler(ra, dec, 3, b1950=b1950, dtype=dtype)
+
+def ec2eq(lam, beta, b1950=False, dtype='f8'):
+ """
+ NAME
+ ec2eq
+ PURPOSE
+ Convert from ecliptic to equatorial coordinates in units of degrees.
+ CALLING SEQUENCE
+ ra,dec = eq2gal(lam, beta, b1950=False, dtype='f8')
+ INPUTS
+ lam,beta: Ecliptic coordinates. May be Numpy arrays, sequences, or
+ scalars as long as they are all the same length. They must be
+ convertible to a Numpy array with the specified datatype.
+ KEYWORDS
+ b1950: If True, use b1950 coordiates. By default j2000 are used.
+ dtype: The datatype of the output arrays. Default is f8
+ OUTPUTS
+ ra,dec: Equatorial longitude and latitude. The returned value is
+ always a Numpy array with the specified dtype
+ REVISION HISTORY
+ Created Erin Sheldon, NYU, 2008-07-02
+ """
+
+ return euler(lam, beta, 4, b1950=b1950, dtype=dtype)
+
+def ec2gal(lam, beta, b1950=False, dtype='f8'):
+ """
+ NAME
+ ec2gal
+ PURPOSE
+ Convert from ecliptic to galactic coordinates in units of degrees.
+ CALLING SEQUENCE
+ l,b = eq2gal(lam, beta, b1950=False, dtype='f8')
+ INPUTS
+ lam, beta: Ecliptic coordinates. May be Numpy arrays, sequences, or
+ scalars as long as they are all the same length. They must be
+ convertible to a Numpy array with the specified datatype.
+ KEYWORDS
+ b1950: If True, use b1950 coordiates. By default j2000 are used.
+ dtype: The datatype of the output arrays. Default is f8
+ OUTPUTS
+ l, b: Galactic longitude and latitude. The returned value is always
+ a Numpy array with the specified dtype
+ REVISION HISTORY
+ Created Erin Sheldon, NYU, 2008-07-02
+ """
+
+ galact_long, galact_lat = euler(lam, beta, 5, b1950=b1950, dtype=dtype)
+ sort_idxs = np.argsort(galact_long)
+ return galact_long[sort_idxs], galact_lat[sort_idxs]
+
+def gal2ec(l, b, b1950=False, dtype='f8'):
+ """
+ NAME
+ gal2ec
+ PURPOSE
+ Convert from Galactic to Ecliptic coordinates in units of degrees.
+ CALLING SEQUENCE
+ lam,beta = eq2gal(l, b, b1950=False, dtype='f8')
+ INPUTS
+ l, b: Galactic coordinates. May be Numpy arrays, sequences, or
+ scalars as long as they are all the same length. They must be
+ convertible to a Numpy array with the specified datatype.
+ KEYWORDS
+ b1950: If True, use b1950 coordiates. By default j2000 are used.
+ dtype: The datatype of the output arrays. Default is f8
+ OUTPUTS
+ lam,beta: Ecliptic longitude and latitude. The returned value is
+ always a Numpy array with the specified dtype
+ REVISION HISTORY
+ Created Erin Sheldon, NYU, 2008-07-02
+ """
+
+ return euler(l, b, 6, b1950=b1950, dtype=dtype)
+
+
+def _thetaphi2xyz(theta, phi):
+ """
+ theta and phi in radians relative to the SDSS node at ra=95 degrees
+ """
+ x = cos(theta) * cos(phi)
+ y = sin(theta) * cos(phi)
+ z = sin(phi)
+
+ return x, y, z
+
+def _xyz2thetaphi(x, y, z):
+ """
+ returns theta, phi in radians relative to the SDSS node at ra=95 degrees
+ """
+ phi = arcsin(z)
+ theta = arctan2(y, x)
+
+ return theta, phi
+
+def eq2xyz(ra, dec, dtype='f8', units='deg'):
+ """
+ Convert equatorial coordinates RA and DEC to x,y,z on the unit sphere
+ parameters
+ ----------
+ ra: scalar or array
+ Right ascension. Can be an array
+ dec: scalar or array
+ Declination. Can be an array
+ units: string, optional
+ 'deg' if the input is degrees, 'rad' if input
+ is in radians. Default is degrees.
+ Notes:
+ This follows the same convention as the STOMP package.
+ """
+
+ theta = numpy.array(ra, ndmin=1, copy=True, dtype=dtype)
+ phi = numpy.array(dec, ndmin=1, copy=True, dtype=dtype)
+
+ # in place is more efficient
+ if units == 'deg':
+ numpy.deg2rad(theta, theta)
+ numpy.deg2rad(phi, phi)
+
+ theta -= _sdsspar['node']
+
+ return _thetaphi2xyz(theta, phi)
+
+def xyz2eq(xin, yin, zin, units='deg'):
+ """
+ Convert x,y,z on the unit sphere to RA DEC.
+ parameters
+ ----------
+ x,y,z:
+ scalars or arrays as given by eq2xyz
+ units: string, optional
+ 'deg' if the output is to be degrees, 'rad' if it is to be radians.
+ Default is degrees.
+ Notes:
+ This follows the same convention as the STOMP package.
+ """
+
+ x = numpy.array(xin, ndmin=1, copy=False)
+ y = numpy.array(yin, ndmin=1, copy=False)
+ z = numpy.array(zin, ndmin=1, copy=False)
+
+ theta, phi = _xyz2thetaphi(x, y, z)
+ theta += _sdsspar['node']
+
+ if units == 'deg':
+ numpy.rad2deg(theta, theta)
+ numpy.rad2deg(phi, phi)
+
+ atbound(theta, 0.0, 360.0)
+
+ # theta->ra, phi->dec
+ return theta, phi
+
+def sphdist(ra1, dec1, ra2, dec2, units=['deg', 'deg']):
+ """
+ Get the arc length between two points on the unit sphere
+ parameters
+ ----------
+ ra1,dec1,ra2,dec2: scalar or array
+ Coordinates of two points or sets of points.
+ Must be the same length.
+ units: sequence
+ A sequence containing the units of the input and output. Default
+ ['deg',deg'], which means inputs and outputs are in degrees. Units
+ can be 'deg' or 'rad'
+ """
+
+ units_in, units_out = units
+
+ # note x,y,z from eq2xyz always returns 8-byte float
+ x1, y1, z1 = eq2xyz(ra1, dec1, units=units_in)
+ x2, y2, z2 = eq2xyz(ra2, dec2, units=units_in)
+
+ costheta = x1 * x2 + y1 * y2 + z1 * z2
+ costheta.clip(-1.0, 1.0, out=costheta)
+
+ theta = arccos(costheta)
+
+ if units_out == 'deg':
+ numpy.rad2deg(theta, theta)
+ return theta
+
+
+def gcirc(ra1deg, dec1deg, ra2deg, dec2deg, getangle=False):
+ """
+ This is currently very inflexible: degrees in, radians out
+ """
+ ra1 = numpy.array(ra1deg, dtype='f8', ndmin=1)
+ dec1 = numpy.array(dec1deg, dtype='f8', ndmin=1)
+ ra2 = numpy.array(ra2deg, dtype='f8', ndmin=1)
+ dec2 = numpy.array(dec2deg, dtype='f8', ndmin=1)
+
+ deg2rad(ra1, ra1)
+ deg2rad(dec1, dec1)
+ deg2rad(ra2, ra2)
+ deg2rad(dec2, dec2)
+
+ sindec1 = sin(dec1)
+ cosdec1 = cos(dec1)
+
+ sindec2 = sin(dec2)
+ cosdec2 = cos(dec2)
+
+ radiff = (ra2 - ra1)
+ cosradiff = cos(radiff)
+ cosdis = sindec1 * sindec2 + cosdec1 * cosdec2 * cosradiff
+
+ cosdis.clip(-1.0, 1.0, out=cosdis)
+ dis = arccos(cosdis)
+
+ if getangle:
+ theta = arctan2(sin(radiff),
+ (sindec1 * cosradiff - cosdec1 * sindec2 / cosdec2)) - HALFPI
+ return dis, theta
+
+ return dis
+
+# utility functions
+def atbound(longitude, minval, maxval):
+ """
+ Wraps the longitude until it is within the bounds. The bounds must define an entire circle (e.g
+ 180 to 180 or 0 to 360)
+ """
+ w, = numpy.where(longitude < minval)
+ while w.size > 0:
+ longitude[w] += 360.0
+ w, = numpy.where(longitude < minval)
+
+ w, = numpy.where(longitude > maxval)
+ while w.size > 0:
+ longitude[w] -= 360.0
+ w, = numpy.where(longitude > maxval)
+
+
+def atbound2(theta, phi):
+ """
+ Wraps theta and phi to be between -180 and 180 for theta, and 0 and 360 for phi.
+ """
+ atbound(theta, -180.0, 180.0)
+
+ w, = numpy.where(numpy.abs(theta) > 90.0)
+ if w.size > 0:
+ theta[w] = 180.0 - theta[w]
+ phi[w] += 180.0
+
+ atbound(theta, -180.0, 180.0)
+ atbound(phi, 0.0, 360.0)
+
+ w, = numpy.where(numpy.abs(theta) == 90.0)
+ if w.size > 0:
+ phi[w] = 0.0
+
+#
+# SDSS specific conversions
+#
+
+
+def eq2sdss(ra_in, dec_in, dtype='f8'):
+ """
+ NAME:
+ eq2sdss
+ PURPOSE:
+ Convert from ra, dec to the corrected clambda, ceta
+ SDSS survey coordinate system. It is corrected so that the
+ longitude eta ranges from [-180.0, 180.0] and the latitude
+ lambda ranges from [-90.0,90.0]. The standard lambda/eta
+ both range from [-180.0,180.0] which doesn't make sense.
+ NOTE: lambda is often referred to as longitude but this
+ is incorrect since it has poles at [-90,90]
+ CALLING SEQUENCE:
+ from esutil import coords
+ (clambda, ceta) = coords.eq2sdss(ra, dec, dtype='f8')
+ INPUTS:
+ ra: Equatorial latitude in degrees.
+ dec: Equatorial longitude in degrees.
+ OPTIONAL INPUTS:
+ dtype: The data type of output. Default is 'f8'. See
+ numpy.typeDict for a list of possible types.
+ dtype: The data type of output. Default is 'f8'.
+ OUTPUTS:
+ clambda: Corrected Survey longitude (actually lattitude) in degrees
+ ceta: Corrected Survey latitude (actually logitude) in degrees
+ REVISION HISTORY:
+ Written: 11-March-2006 Converted from IDL program.
+ """
+
+ # Make a copy as an array. ndmin=1 to avoid messed up scalar arrays
+ ra = numpy.array(ra_in, ndmin=1, copy=True, dtype=dtype)
+ dec = numpy.array(dec_in, ndmin=1, copy=True, dtype=dtype)
+
+ if ra.size != dec.size:
+ raise ValueError("RA, DEC must be same size")
+
+ # range checking
+ if ra.min() < 0.0 or ra.max() > 360.0:
+ raise ValueError('RA must we within [0,360]')
+ if dec.min() < -90.0 or dec.max() > 90.0:
+ raise ValueError('DEC must we within [-90,90]')
+
+ ra *= D2R
+ dec *= D2R
+ ra -= _sdsspar['node']
+
+ # generate x,y,z on unit sphere, clearing memory as we go
+ cdec = cos(dec)
+
+ x = cos(ra) * cdec
+ y = sin(ra) * cdec
+
+ ra = 0
+ cdec = 0 # mem
+
+ z = numpy.sin(dec)
+
+ dec = 0 # mem
+
+ # generate clambda, ceta
+ # do things in place to save memory
+
+ # clambda = -arcsin( x ) (not a copy clambda=x)
+ arcsin(x, x)
+ clambda = x
+ clambda *= -1
+
+
+ arctan2(z, y, z)
+ ceta = z
+ ceta -= _sdsspar['etapole']
+
+ clambda *= R2D
+ ceta *= R2D
+
+ atbound(ceta, -180.0, 180.0)
+
+ return (clambda, ceta)
+
+def sdss2eq(clambda_in, ceta_in, dtype='f8'):
+ """
+ NAME:
+ sdss2eq
+ PURPOSE:
+ Convert corrected clambda, ceta SDSS survey coordinate system t
+ equatorial coords.
+ CALLING SEQUENCE:
+ from esutil import coords
+ (ra, dec) = coords.sdss2eq(clambda, ceta, dtype='f8')
+ INPUTS:
+ clambda: Corrected Survey longitude (actually lattitude) in degrees
+ ceta: Corrected Survey latitude (actually logitude) in degrees
+ OPTIONAL INPUTS:
+ dtype: The data type of output. Default is 'f8'. See
+ numpy.typeDict for a list of possible types.
+ OUTPUTS:
+ ra: Equatorial latitude in degrees.
+ dec: Equatorial longitude in degrees.
+ REVISION HISTORY:
+ Written: 11-March-2006 Converted from IDL program.
+ """
+
+ # Make a copy as an array. ndmin=1 to avoid messed up scalar arrays
+ clambda = numpy.array(clambda_in, ndmin=1, copy=True, dtype=dtype)
+ ceta = numpy.array(ceta_in, ndmin=1, copy=True, dtype=dtype)
+
+ # range checking
+ if clambda.min() < -90.0 or clambda.max() > 90.0:
+ raise ValueError('CLAMBDA must we within [-90,90]')
+ if ceta.min() < -180.0 or ceta.max() > 180.0:
+ raise ValueError('CETA must we within [-180,180]')
+
+ clambda *= D2R
+ ceta *= D2R
+
+ x = -sin(clambda)
+ y = cos(ceta + _sdsspar['etapole']) * cos(clambda)
+ z = sin(ceta + _sdsspar['etapole']) * cos(clambda)
+
+ ra = arctan2(y, x) + _sdsspar['node']
+ dec = arcsin(z)
+
+ ra *= R2D
+ dec *= R2D
+ atbound2(dec, ra)
+
+ return (ra, dec)
+
+
+def _eq2survey(ra_in, dec_in, dtype='f8'):
+ """
+ NAME:
+ _eq2survey
+ PURPOSE:
+ Convert from ra, dec to the lambda, eta
+ SDSS survey coordinate system. Note this coordinate system is
+ not well defined. Recommend you use csurvey coords.
+ CALLING SEQUENCE:
+ from esutil import coords
+ (lambda, eta) = coords._eq2survey(ra, dec, dtype='f8')
+ INPUTS:
+ ra: Equatorial latitude in degrees.
+ dec: Equatorial longitude in degrees.
+ OPTIONAL INPUTS:
+ dtype: The data type of output. Default is 'f8'. See
+ numpy.typeDict for a list of possible types.
+ OUTPUTS:
+ lambda: SDSS Survey longitude (actually lattitude) in degrees
+ eta: SDSS Survey latitude (actually logitude) in degrees
+ REVISION HISTORY:
+ Written: 11-March-2006 Converted from IDL program.
+ """
+
+ # Make a copy as an array. ndmin=1 to avoid messed up scalar arrays
+ ra = numpy.array(ra_in, ndmin=1, copy=True, dtype=dtype)
+ dec = numpy.array(dec_in, ndmin=1, copy=True, dtype=dtype)
+
+ if ra.size != dec.size:
+ raise ValueError("RA, DEC must be same size")
+
+ # range checking
+ if ra.min() < 0.0 or ra.max() > 360.0:
+ raise ValueError('RA must we within [0,360]')
+ if dec.min() < -90.0 or dec.max() > 90.0:
+ raise ValueError('DEC must we within [-90,90]')
+
+
+
+ ra *= D2R
+ dec *= D2R
+ ra -= _sdsspar['node']
+
+ # generate x,y,z on unit sphere, clearing memory as we go
+ cdec = cos(dec)
+
+ x = cos(ra) * cdec
+ y = sin(ra) * cdec
+
+ ra = 0
+ cdec = 0 # mem
+
+ z = sin(dec)
+
+ dec = 0 # mem
+
+ # generate lam, eta
+ # do things in place to save memory
+
+ # lam = -arcsin( x ) (not a copy lam=x)
+ arcsin(x, x)
+ lam = x
+ lam *= -1
+
+
+ arctan2(z, y, z)
+ eta = z
+ eta -= _sdsspar['etapole']
+
+ lam *= R2D
+ eta *= R2D
+
+ atbound2(lam, eta)
+ atbound(eta, -180.0, 180.0)
+
+ w, = numpy.where(eta > (90.0 - _sdsspar['center_dec']))
+ if w.size > 0:
+ eta[w] -= 180.0
+ lam[w] = 180.0 - lam[w]
+
+ atbound(lam, -180.0, 180.0)
+
+ return (lam, eta)
+
+
+def dec_parse(decstring):
+ """
+ parse a colon separated string representing declination into
+ degrees.
+ parameters
+ ----------
+ decstring: string
+ DD:MM:SS.sss the value is specified in degrees, minutes, seconds
+ Only the degrees are required. Additional
+ precision (minutes, seconds) are optional in the string (i.e. "12" or
+ "12:34" or "12:34:56" are all valid input strings)
+ Corrections by Paul Ray and Dave Smith, NRL, 2013-03-19
+ """
+ dec = 0.0
+ sign = 1.0
+
+ if decstring.find("-") >= 0:
+ sign = -1.0
+ ds = decstring.split(':')
+ lds = len(ds)
+ if lds >= 1:
+ deg = abs(float(ds[0]))
+ dec += deg
+ if lds >= 2:
+ minutes = float(ds[1])
+ dec += minutes / 60.0
+ if lds >= 3:
+ sec = float(ds[2])
+ dec += sec / 3600.0
+ return dec * sign
+
+def ra_parse(rastring, hours=True):
+ """
+ parse a colon separated string representing right ascension into
+ decimal degrees.
+ parameters
+ ----------
+ rastring: string
+ "HH:MM:SS.sss" if hours is True and
+ "DD:MM:SS.sss" if hours is False (indicating that
+ the value is specified in degrees, minutes, seconds)
+ In all cases, only the hours (or degrees) are required. Additional
+ precision (minutes, seconds) are optional in the string (i.e. "12" or
+ "12:34" or "12:34:56" are all valid input strings)
+ Corrections by Paul Ray and Dave Smith, NRL, 2013-03-19
+ """
+ ra = 0.0
+
+ rs = rastring.split(':')
+ lrs = len(rs)
+ if lrs >= 1:
+ ra += float(rs[0])
+ if lrs >= 2:
+ minutes = float(rs[1])
+ ra += minutes/60.0
+ if lrs >= 3:
+ sec = float(rs[2])
+ ra += sec/3600.0
+ if hours:
+ ra *= 15
+ return ra
+
+def fitsheader2dict(hdr, ext=0):
+ """
+ Convert a fits header object into a dict. A dict provides more expected
+ interface to the data but cannot be written back to a fits file without
+ transformation.
+ """
+
+ hdict = {}
+ for key in hdr:
+ hdict[key.lower()] = hdr[key]
+
+ return hdict
+
+def shiftlon(lon_input, shift=None, wrap=True):
+ """
+ Name:
+ shiftlon
+ Calling Sequence:
+ newlon = shiftlon(longitude, wrap=True, shift=0.0)
+ Purpose:
+ Shift the value of a longitude. By default, the value is "wrapped" to
+ be [-180,180] instead of [0,360]
+ If the shift keyword is sent, then the longitude is simply shifted by
+ the input value and then constrained to be again on the [0,360) range.
+ Input:
+ A longitude or array of longitudes on the range [0,360)
+ Keywords:
+ shift:
+ If shift is sent, then lon-shift is returned, constrained to still
+ be on [0,360).
+ wrap:
+ If shift is not sent, and wrap is True, wrap the range to
+ [-180,180]
+ """
+ lon = numpy.array(lon_input, ndmin=1, copy=True, dtype='f8')
+
+ if shift is not None:
+ negshift = False
+ if shift < 0:
+ negshift = True
+
+ abs_shift = abs(shift)
+
+ # make sure in range [0,360)
+ abs_shift = abs_shift % 360.0
+
+ if negshift:
+ lon += abs_shift
+
+ w, = numpy.where(lon > 360.0)
+ if w.size > 0:
+ lon[w] -= 360.0
+ else:
+ lon -= abs_shift
+
+ w, = numpy.where(lon < 0.0)
+ if w.size > 0:
+ lon[w] += 360.0
+
+ elif wrap:
+ w, = where(lon > 180)
+ if w.size > 0:
+ lon[w] -= 360
+
+ return lon
+
+
+
+def shiftra(ra, shift=None, wrap=True):
+ """
+ Name:
+ shiftra
+ Calling Sequence:
+ newra = shiftra(ra, wrap=True, shift=0.0)
+ Purpose:
+ Shift the value of a longitude RA. By default, the value is "wrapped"
+ to be [-180,180] instead of [0,360]
+ If the shift keyword is sent, then the longitude is simply shifted by
+ the input value and then constrained to be again on the [0,360) range.
+ Input:
+ ra or any other longitude on the range [0,360)
+ Keywords:
+ shift:
+ If shift is sent, then ra-shift is returned, constrained to still
+ be on [0,360).
+ wrap:
+ If shift is not sent, and wrap is True, wrap the range to
+ [-180,180]
+ """
+ return shiftlon(ra, shift=shift, wrap=wrap)
+
+
+
+def radec2aitoff(ra, dec):
+ """
+ Take the ra/dec into aitoff coords
+ """
+
+ r2 = numpy.sqrt(2.0)
+ f = 2. * r2 / PI
+
+
+ sra = shiftra(ra)
+
+ alpha2 = sra / 2. * D2R
+ delta = dec * D2R
+
+ cdec = cos(delta)
+
+ denom = sqrt(1.0 + cdec * cos(alpha2))
+
+ x = cdec * sin(alpha2) * 2. * r2 / denom
+ y = sin(delta) * r2 / denom
+ x = x * R2D / f
+ y = y * R2D / f
+
+ return x, y
+
+def _check_range(rng, allowed):
+ if rng is None:
+ rng = allowed
+ else:
+ if not hasattr(rng, '__len__'):
+ raise ValueError("range object does not have len() method")
+
+ if rng[0] < allowed[0] or rng[1] > allowed[1]:
+ raise ValueError(f"lon_range should be within [{allowed[0]}, {allowed[1]}]")
+ return rng
+
+def randsphere(num, ra_range=None, dec_range=None, system='eq'):
+ """
+ Generate random points on the sphere
+ You can limit the range in ra and dec. To generate on a spherical cap, see
+ randcap()
+ parameters
+ ----------
+ num: integer
+ The number of randoms to generate
+ ra_range: list, optional
+ Should be within range [0,360]. Default [0,360]
+ dec_range: list, optional
+ Should be within range [-90,90]. Default [-90,90]
+ system: string
+ Default is 'eq' for the ra-dec system. Can also be 'xyz'.
+ output
+ ------
+ for system == 'eq' the return is a tuple
+ ra,dec = randsphere(...)
+ for system == 'xyz' the return is a tuple
+ x,y,z = randsphere(...)
+ examples
+ --------
+ ra,dec = randsphere(2000, ra_range=[10,35], dec_range=[-25,15])
+ x,y,z = randsphere(2000, system='xyz')
+ """
+
+ ra_range = _check_range(ra_range, [0.0, 360.0])
+ dec_range = _check_range(dec_range, [-90.0, 90.0])
+
+ ra = numpy.random.random(num)
+ ra *= (ra_range[1] - ra_range[0])
+ if ra_range[0] > 0:
+ ra += ra_range[0]
+
+ # number [-1,1)
+ cosdec_min = cos(deg2rad(90.0 + dec_range[0]))
+ cosdec_max = cos(deg2rad(90.0 + dec_range[1]))
+ v = numpy.random.random(num)
+ v *= (cosdec_max - cosdec_min)
+ v += cosdec_min
+
+ numpy.clip(v, -1.0, 1.0, v)
+ # Now this generates on [0,pi)
+ dec = numpy.arccos(v)
+
+ # convert to degrees
+ rad2deg(dec, dec)
+ # now in range [-90,90.0)
+ dec -= 90.0
+
+ if system == 'xyz':
+ x, y, z = eq2xyz(ra, dec)
+ return x, y, z
+
+ return ra, dec
+
+def randcap(nrand, ra, dec, rad, get_radius=False, dorot=False):
+ """
+ Generate random points in a sherical cap
+ parameters
+ ----------
+ nrand:
+ The number of random points
+ ra,dec:
+ The center of the cap in degrees. The ra should be within [0,360) and
+ dec from [-90,90]
+ rad: float
+ radius of the cap, same units as ra,dec
+ get_radius: bool, optional
+ if true, return radius of each point in radians
+ dorot: bool
+ If dorot is True, generate the points on the equator and rotate them to
+ be centered at the desired location. This is the default when the dec
+ is within 0.1 degrees of the pole, to avoid calculation issues
+ """
+ # generate uniformly in r**2
+ if dec >= 89.9 or dec <= -89.9:
+ dorot = True
+
+ if dorot:
+ tra, tdec = 90.0, 0.0
+ rand_ra, rand_dec, rand_r = randcap(nrand, 90.0, 0.0, rad, get_radius=True)
+ rand_ra, rand_dec = rotate(0.0, dec - tdec, 0.0, rand_ra, rand_dec)
+ rand_ra, rand_dec = rotate(ra - tra, 0.0, 0.0, rand_ra, rand_dec)
+ else:
+ rand_r = numpy.random.random(nrand)
+ rand_r = sqrt(rand_r) * rad
+
+ # put in degrees
+ numpy.deg2rad(rand_r, rand_r)
+
+ # generate position angle uniformly 0,2*PI
+ rand_posangle = numpy.random.random(nrand) * 2 * PI
+
+ theta = numpy.array(dec, dtype='f8', ndmin=1, copy=True)
+ phi = numpy.array(ra, dtype='f8', ndmin=1, copy=True)
+ theta += 90
+
+ numpy.deg2rad(theta, theta)
+ numpy.deg2rad(phi, phi)
+
+ sintheta = sin(theta)
+ costheta = cos(theta)
+
+ sinr = sin(rand_r)
+ cosr = cos(rand_r)
+
+ cospsi = cos(rand_posangle)
+ costheta2 = costheta * cosr + sintheta * sinr * cospsi
+
+ numpy.clip(costheta2, -1, 1, costheta2)
+
+ # gives [0,pi)
+ theta2 = arccos(costheta2)
+ sintheta2 = sin(theta2)
+
+ cosDphi = (cosr - costheta * costheta2) / (sintheta * sintheta2)
+
+ numpy.clip(cosDphi, -1, 1, cosDphi)
+ Dphi = arccos(cosDphi)
+
+ # note fancy usage of where
+ phi2 = numpy.where(rand_posangle > PI, phi + Dphi, phi - Dphi)
+
+ numpy.rad2deg(phi2, phi2)
+ numpy.rad2deg(theta2, theta2)
+ rand_ra = phi2
+ rand_dec = theta2 - 90.0
+
+ atbound(rand_ra, 0.0, 360.0)
+
+ if get_radius:
+ #numpy.rad2deg(rand_r, rand_r)
+ return rand_ra, rand_dec, rand_r
+
+ return rand_ra, rand_dec
+
+def randcap_brute(nrand, ra, dec, rad, get_radius=False):
+ """
+ Generate random points in a sherical cap using brute
+ force rejection sampling. This is extremely
+ slow and is used for testing purposes only.
+ parameters
+ ----------
+ nrand: int
+ The number of random points
+ ra,dec: float
+ The center of the cap in degrees. The ra should be within [0,360) and
+ dec from [-90,90]
+ rad: float
+ radius of the cap, same units as ra,dec
+ get_radius: bool, optional
+ if true, return radius of each point in radians
+ """
+
+ ora = zeros(nrand)
+ odec = zeros(nrand)
+ orad = zeros(nrand)
+
+ ngood = 0
+ nleft = nrand
+
+ while ngood < nrand:
+ tra, tdec = randsphere(nleft)
+ d = sphdist(ra, dec, tra, tdec)
+ w, = where(d <= rad)
+ if w.size > 0:
+ ora[ngood:ngood + w.size] = tra[w]
+ odec[ngood:ngood + w.size] = tdec[w]
+ orad[ngood:ngood + w.size] = d[w]
+
+ ngood += w.size
+ nleft -= w.size
+
+ if get_radius:
+ return ora, odec, orad
+ return ora, odec
+
+def rotate(phi, theta, psi, ra, dec):
+ """
+ rotation the given positions on the sphere
+ The convention is the usual zxz
+ Parameters
+ ----------
+ phi, theta, psi: numbers
+ The euler angles in zxz convention
+ ra, dec: numbers or arrays
+ positions to be rotated
+ """
+
+ if hasattr(ra, '__len__'):
+ is_scalar = False
+ else:
+ is_scalar = True
+
+ ra = numpy.array(ra, ndmin=1, copy=False)
+ dec = numpy.array(dec, ndmin=1, copy=False)
+ if ra.size != dec.size:
+ raise ValueError('ra[%d] has different size than dec[%d]' % (ra.size, dec.size))
+
+ twopi = 2.0 * PI
+ fourpi = 4.0 * PI
+
+ # use negative; rotating the points is like rotating
+ # the coord system in the opposite direction
+ phi = deg2rad(-phi)
+ theta = deg2rad(-theta)
+ psi = deg2rad(-psi)
+
+ sintheta = sin(theta)
+ costheta = cos(theta)
+
+
+ a = deg2rad(ra) - phi
+ b = deg2rad(dec)
+
+ sb = sin(b)
+ cb = cos(b)
+ cbsa = cb * sin(a)
+
+ b = -sintheta * cbsa + costheta * sb
+
+ w, = numpy.where(b > 1.0)
+ if w.size > 0:
+ b[w] = 1.0
+
+ dec_out = arcsin(b)
+
+ a = arctan2(costheta * cbsa + sintheta * sb, cb * cos(a))
+ ra_out = (a + psi + fourpi) % twopi
+
+ rad2deg(ra_out, out=ra_out)
+ rad2deg(dec_out, out=dec_out)
+
+ if is_scalar:
+ ra_out = ra_out[0]
+ dec_out = dec_out[0]
+
+ return ra_out, dec_out
+
+
+def rect_area(lon_min, lon_max, lat_min, lat_max):
+ """
+ Calculate the area of a rectangle on the sphere.
+ parameters
+ ----------
+ lon_min, lon_max, lat_min, lat_max:
+ Definition of the rectangle, in degrees
+ """
+ smax = sin(deg2rad(lat_max))
+ smin = sin(deg2rad(lat_min))
+ area = (smax - smin) * (lon_max - lon_min)
+ return numpy.abs(area) * R2D
\ No newline at end of file
diff --git a/notebooks/stamp_note_images/alerce_frontend.png b/notebooks/stamp_note_images/alerce_frontend.png
new file mode 100644
index 0000000..67415c9
Binary files /dev/null and b/notebooks/stamp_note_images/alerce_frontend.png differ
diff --git a/notebooks/stamp_note_images/betas.png b/notebooks/stamp_note_images/betas.png
new file mode 100644
index 0000000..f0822c0
Binary files /dev/null and b/notebooks/stamp_note_images/betas.png differ
diff --git a/notebooks/stamp_note_images/class_samples.png b/notebooks/stamp_note_images/class_samples.png
new file mode 100644
index 0000000..ee36534
Binary files /dev/null and b/notebooks/stamp_note_images/class_samples.png differ
diff --git a/notebooks/stamp_note_images/cm.png b/notebooks/stamp_note_images/cm.png
new file mode 100644
index 0000000..d4ad762
Binary files /dev/null and b/notebooks/stamp_note_images/cm.png differ
diff --git a/notebooks/stamp_note_images/cnn_architecture.png b/notebooks/stamp_note_images/cnn_architecture.png
new file mode 100644
index 0000000..a53f92a
Binary files /dev/null and b/notebooks/stamp_note_images/cnn_architecture.png differ
diff --git a/notebooks/stamp_note_images/cnn_table.png b/notebooks/stamp_note_images/cnn_table.png
new file mode 100644
index 0000000..96cf718
Binary files /dev/null and b/notebooks/stamp_note_images/cnn_table.png differ
diff --git a/notebooks/stamp_note_images/sn_hunter.png b/notebooks/stamp_note_images/sn_hunter.png
new file mode 100644
index 0000000..df93f0a
Binary files /dev/null and b/notebooks/stamp_note_images/sn_hunter.png differ
![](\"stamp_note_images/cnn_architecture.png\")
![](\"stamp_note_images/cnn_table.png\")
![](\"stamp_note_images/class_samples.png\")
\n",
+ "\n",
+ "Here we show examples of the five classes that are to be discriminated by using only the first detection. For each class, the triplet of images in each row are science, reference and difference images from left to right. Each row corresponds to a different candidate. The images are cropped at the center resulting in 21x21 images, and normalized. Further details about the pre-processing of the images is explained in Carrasco-Davis et al 2020.\n",
+ "\n",
+ "### Convolutional neural network architecture: \n",
+ "\n",
+ "![](\"stamp_note_images/cm.png\")
![](\"stamp_note_images/betas.png\")
![](\"stamp_note_images/sn_hunter.png\")
\n",
+ "\n",
+ "On the bottom left side, the location of each candidate in sky coordinates with respect to the Galactic plane and the ecliptic are depicted. On the bottom right side, a selection of the top candidates is listed, initially ordered by SN probability score from the stamp classifier. The list of candidates can be sorted by other parameters, and updated/refreshed to include newly ingested alerts. On the top left side, the SN candidate ID is shown as a clickable link to the [ALeRCE frontend](https://alerce.online/object/ZTF20abpvolc), with relevant metadata such as radec coordinates, magnitude, date, etc. At the bottom there are links to other sources of information, including ALeRCE, NED, TNS, and the Simbad Astronomical Database. In the middle of the figure there is a colored image from Aladin. On the top right side, the stamps of the first detection are shown, along with buttons for reporting the candidate as eventual bogus or as a possible SN.\n",
+ "\n",
+ "![](\"stamp_note_images/alerce_frontend.png\")
\n",
+ "\n",
+ "When the candidate is inspected using the [ALeRCE frontend](https://alerce.online/object/ZTF20abpvolc), the full prediction of classes by the stamp classifier is shown at the bottom in addition to the full information of the light curve available, metadata, cross matches, etc. When the light curve of the object has six or more points in one of the bands, it is classified by the light curve classifier (Sánchez-Sáez et al. 2020), with the prediction also available in the ALeRCE frontend with a more complex taxonomy."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Requirements "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from IPython.display import display\n",
+ "import sys\n",
+ "import pandas as pd\n",
+ "import numpy as np\n",
+ "import matplotlib.pyplot as plt\n",
+ "from astropy.time import Time\n",
+ "import ephem\n",
+ "from coords_utils import ec2gal\n",
+ "from tqdm import tqdm\n",
+ "import copy\n",
+ "import pickle"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# ALeRCE client "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Install and import the alerce API client https://alerce.readthedocs.io/en/latest/"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#!python3 -m pip install --user alerce\n",
+ "from alerce.api import AlerceAPI\n",
+ "client = AlerceAPI()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Plotting light curves "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We will create a simple function that plots the light curve given an object id (oid), a dataframe with detections and a dataframe with non detections. This function will be used to plot light curves of some objects predicted by the stamp classifier."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def plotLC(oid, SN_det, SN_nondet):\n",
+ " \n",
+ " fig, ax = plt.subplots(figsize = (14, 7))\n",
+ " labels = {1: 'g', 2: 'r'}\n",
+ " colors = {1: '#56E03A', 2: '#D42F4B'} # color blind friendly green and red \n",
+ " markers = {1: 'o', 2: 's'}\n",
+ " sizes = {1: 30, 2: 60}\n",
+ " \n",
+ " # loop the passbands\n",
+ " for fid in [1, 2]:\n",
+ " \n",
+ " # plot detections if available\n",
+ " mask = SN_det.fid == fid\n",
+ " if np.sum(mask) > 0:\n",
+ " # note that the detections index is candid and that we are plotting the psf corrected magnitudes\n",
+ " ax.errorbar(SN_det[mask].mjd, SN_det[mask].magpsf, \n",
+ " yerr = SN_det[mask].sigmapsf, c=colors[fid], label=labels[fid], marker=markers[fid])\n",
+ " \n",
+ " # plot non detections if available\n",
+ " mask = (SN_nondet.fid == fid) & (SN_nondet.diffmaglim > -900)\n",
+ " if np.sum(mask) > 0: \n",
+ " # non detections index is mjd\n",
+ " ax.scatter(SN_nondet[mask].index, SN_nondet[mask].diffmaglim, c=colors[fid], alpha = 0.5,\n",
+ " marker='v', label=\"lim.mag. %s\" % labels[fid], s=sizes[fid])\n",
+ " \n",
+ " ax.set_title(oid)\n",
+ " ax.set_xlabel(\"MJD\")\n",
+ " ax.set_ylabel(\"Apparent magnitude\")\n",
+ " ax.legend()\n",
+ " ax.set_ylim(ax.get_ylim()[::-1])\n",
+ "\n",
+ "def getSNdata(oid, doLC=False, dostamp=False):\n",
+ "\n",
+ " results = {\"oid\": oid}\n",
+ " \n",
+ " # query detections\n",
+ " SN_det = client.get_detections(oid, format='pandas')\n",
+ " results[\"lc_det\"] = SN_det\n",
+ " \n",
+ " # query non detections\n",
+ " SN_nondet = client.get_non_detections(oid, format='pandas')\n",
+ " results[\"lc_nondet\"] = SN_nondet\n",
+ " \n",
+ " # plot the LC\n",
+ " if doLC:\n",
+ " plotLC(oid, SN_det, SN_nondet)\n",
+ " \n",
+ " # show the first image stamp\n",
+ " if dostamp:\n",
+ " client.plot_stamp(oid) \n",
+ " \n",
+ " # return data\n",
+ " return results"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Classified Alerts "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Get objects per class\n",
+ "Here, we build a function that uses the ALeRCE client to query objects according to the predictions of the stamp classifier. In order to select the classifier, we define the query in terms of the **pcclassearly**, the classes are the following strings:\n",
+ "+ \"agn\": Active Galactic Nuclei\n",
+ "+ \"sn\": Supernova\n",
+ "+ \"vs\": Variable star\n",
+ "+ \"asteroid\"\n",
+ "+ \"bogus\"\n",
+ "\n",
+ "Using the ALeRCE client we can select the min and max number of observation per object, also sort them by firstmjd, number of observations (nobs), probability of the selected class (classearly), etc. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def get_objects_per_class(classearly=\"sn\", pclassearly=0.5, min_nobs=10, max_nobs=50, n_objects=100):\n",
+ "\n",
+ " total = n_objects\n",
+ " records_per_page = n_objects\n",
+ " page = 1\n",
+ " sortBy = \"pclassearly\"\n",
+ " min_firstmjd = Time(\"2017-11-01T00:00:00\", format=\"isot\", scale=\"utc\").mjd + 70\n",
+ "\n",
+ " params = {\n",
+ " \"records_per_pages\": records_per_page, # Number of queried objects\n",
+ " \"page\": page,\n",
+ " \"sortBy\": sortBy,\n",
+ " \"query_parameters\": {\n",
+ " \"filters\": {\n",
+ " \"pclassearly\": pclassearly, # Probability of objects for classearly\n",
+ " \"classearly\": classearly, # string or identifier of the queried class\n",
+ " \"nobs\": {\n",
+ " \"min\": min_nobs, # Min number of observations\n",
+ " \"max\": max_nobs # Max number of observations\n",
+ " },\n",
+ " \"dates\": {\n",
+ " \"firstmjd\": {\n",
+ " \"min\": min_firstmjd # Min mjd of the first detection\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " \n",
+ " objects = client.query(params, format='pandas')\n",
+ " objects.sort_values(by=\"nobs\", ascending=False, inplace=True)\n",
+ " print(objects.shape)\n",
+ " \n",
+ " return objects"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The previous function returns a pandas dataframe with a list of objects by their ZTF oid as index, and relevant information in their columns as we will see next.\n",
+ "Now, lets query some objects for each of the class available from the stamp classifier. For now, the ALeRCE client works with class ids numbers as shown in the dictionary defined below as early_classes. Here, we are querying a few thousands of objects from the ALeRCE database. We will also sort the classes by ascending number of observations in the cases of bogus and asteroids in order to get a clean examples. AGNs, VS and SNe are sorted in descending order of number of observations to appreciate better the light curves for each of these classes.\n",
+ "\n",
+ "**This query will take a few seconds**"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(10000, 43)\n",
+ "Result of a query using the ALeRCE client\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "