commit 4640e017f9c61d34cdedda2cc524ffb047f1d8d5
Author: William Casarin <jb55@jb55.com>
Date: Wed, 28 Apr 2021 08:02:52 -0700
messing around with umap and harps
Signed-off-by: William Casarin <jb55@jb55.com>
Diffstat:
4 files changed, 202 insertions(+), 0 deletions(-)
diff --git a/.envrc b/.envrc
@@ -0,0 +1 @@
+use nix
diff --git a/HARPS_RVBank_header.txt b/HARPS_RVBank_header.txt
@@ -0,0 +1,107 @@
+J/A+A/???/??? Radial velocities for HAPRS targets (Trifonov+, 2020)
+================================================================================
+A public HARPS radial velocity database corrected for systematic errors
+ T. Trifonov et al.
+ <Astron. Astrophys. ???, ??? (2019)>
+ =2020A&A...???..???T
+================================================================================
+ADC_Keywords: Radial velocities
+Keywords: methods: data analysis - planetary systems
+
+Abstract:
+
+See https://arxiv.org/pdf/2001.05942.pdf
+
+Description:
+ Time series for radial velocities and activity indicators from HARPS
+ spectrograph are presented. See Trifonov et al. (2020) for a detailed
+ description of the parameters.
+
+File Summary:
+--------------------------------------------------------------------------------
+ FileName Lrecl Records Explanations
+--------------------------------------------------------------------------------
+ReadMe 80 . This file
+--------------------------------------------------------------------------------
+
+Byte-by-byte Description of file: HARPS_RVBank.dat
+--------------------------------------------------------------------------------
+ Bytes Format Units Label Explanations
+------------------------------------------------------------------------------
+ 1-13 F13.5 d BJD Barycentric Julian date
+ 14-22 F8.3 m/s RV_mlc_nzp Radial velocity (from mlc.dat, sa, NZP and drift corrected, -pre and -post calculated separately)
+ 23-29 F6.3 m/s e_RV_mlc_nzp Radial velocity error
+
+ 14-22 F8.3 m/s RV_drs_nzp Radial velocity (from drs.dat, sa, NZP and drift corrected -pre and -post calculated separately)
+ 23-29 F6.3 m/s e_RV_drs_nzp Radial velocity error
+
+ 14-22 F8.3 m/s RV_mlc Radial velocity (from mlc.dat, sa and drift corrected, -pre and -post calculated separately)
+ 23-29 F6.3 m/s e_RV_mlc Radial velocity error
+
+ 14-22 F8.3 m/s RV_drs Radial velocity (from drs.dat, sa and drift corrected, -pre and -post calculated separately)
+ 23-29 F6.3 m/s e_RV_drs Radial velocity error
+
+ 14-22 F8.3 m/s RV_mlc_j Radial velocity (from mlc.dat, sa and drift corrected, -pre and -post calculated jointly)
+ 23-29 F6.3 m/s e_RV_mlc_j Radial velocity error
+
+ 46-54 F8.3 m/s CRX Chromatic index
+ 55-63 F8.3 m/s e_CRX Chromatic index error
+ 64-72 F8.3 m/s*km/s dLW Differential line width
+ 73-81 F8.3 m/s*km/s e_dLW Differential line width error
+ 91-99 F8.4 --- Halpha Halpha index
+100-108 F8.4 --- e_Halpha Halpha index error
+ 91-99 F8.4 --- NaD1 NaD1 index
+100-108 F8.4 --- e_NaD1 NaD1 index error
+ 91-99 F8.4 --- NaD2 NaD2 index
+100-108 F8.4 --- e_NaD2 NaD2 index error
+109-112 I3 --- f_RV [0,15] Bitwise flag (1)
+130-138 F8.3 km/s FWHM_DRS Full width half maximum
+139-147 F8.3 --- CONTRAST_DRS
+148-156 F8.3 m/s Bisector
+157-166 F8.3 km/s RVGUESS User guess for RV
+167-175 F8.3 m/s*km/s SNR_DRS Signal-to-noise ratio in order 55
+176-189 F13.5 km/s BJD_DRS Barycentric Julian date from DRS
+ 82-90 F8.3 km/s BERV Barycentric Earth radial velocity
+190-198 F8.3 km/s BERV_DRS Barycentric Earth radial velocity from DRS
+199-207 F8.3 m/s DRIFT drift measure
+208-218 F8.3 m/s E_DRIFT drift measure error
+219-227 F8.3 m/s SA Contribution from secular acceleration
+228-239 F8.3 m/s NZP_mlc NZP for BJD
+240-251 F8.3 m/s dNZP_mlc Contribution from intra-night systmatics
+252-260 F8.3 --- TMMEAN Flux weighted mean point
+261-270 F8.3 s EXPTIME Exposure time
+298-306 F8.3 m/s MLCRX ML Chromatic index (Slope over logarithmic wavelength)
+307-315 F8.3 m/s E_MLCRX error for MLCRX (slope error)
+ A --- TIMEID Identifier for file (time stamp)
+ A --- DRIFT_LAMP FP,ThAr?
+ A --- MASK DRS MASK for CCF
+ A --- PROGID Prog-ID
+ A --- PROGID Prog-PI
+ F8.2 m/s*km/s AIRMASS Airmass
+ A --- OBJAB
+ A --- THAR_FP
+ A --- DPR_TYPE
+
+
+
+Comments:
+
+1) The formating in this version is still not perfect !!! TBFixed
+2) Some entries and details might change.
+
+Note (1): The flags are bitwise where:
+ 0 - normal (reliable) spectra,
+ 1 - nosci frame, e.g. calibration files,
+ 2 - spectra taken in I2 mode,
+ 4 - spectra taken in "eggs" mode,
+ 16 - coordinates too much off,
+ 32 - spectra not within a nautical twilight (daytime),
+ 64 - spectra with too low S/N,
+ 128 - spectra with too high S/N.
+--------------------------------------------------------------------------------
+
+Acknowledgements:
+ Trifon Trifnov <trifonov@mpia.de>
+
+================================================================================
+(End) Trifon Trifnov [MPIA, Germany] 15-August-2019
diff --git a/default.nix b/default.nix
@@ -0,0 +1,6 @@
+{ pkgs ? import <nixpkgs> {} }:
+with pkgs;
+mkShell {
+ name = "umap_harps";
+ buildInputs = with python3Packages; [ scikitlearn seaborn pandas matplotlib numpy umap-learn ];
+}
diff --git a/harps_umap.py b/harps_umap.py
@@ -0,0 +1,88 @@
+#!/usr/bin/env python3
+
+import numpy as np
+from sklearn.datasets import load_digits
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import matplotlib.pyplot as plt
+import matplotlib.cm as cm
+import matplotlib.colors as colors
+import seaborn as sns
+import pandas as pd
+import umap
+
+#harps = pd.read_csv("harps-20.csv")
+#harps = pd.read_csv("harps.csv")
+harps = pd.read_csv("harps-all.csv")
+harps.head()
+harps = harps.dropna()
+harps['target'] = harps['target'].astype('category')
+
+print(harps.RVGUESS.value_counts())
+
+reducer = umap.UMAP()
+
+harps_data = harps[
+ [
+ #"RV_mlc_nzp"
+ #"e_RV_mlc_nzp"
+ #"RV_drs_nzp"
+ #,"e_RV_drs_nzp"
+ #,"RV_mlc"
+ #,"e_RV_mlc"
+ #,"RV_drs"
+ #,"e_RV_drs"
+ #,"RV_mlc_j"
+ #,"e_RV_mlc_j"
+ "CRX"
+ #,"e_CRX"
+ ,"dLW"
+ #,"e_dLW"
+ ,"Halpha"
+ #,"e_Halpha"
+ ,"NaD1"
+ #,"e_NaD1"
+ ,"NaD2"
+ #,"e_NaD2"
+ #,"f_RV"
+ #,"FWHM_DRS"
+ #,"CONTRAST_DRS"
+ #,"BIS"
+ #,"RVGUESS"
+ ,"SNR_DRS"
+ ,"BJD_DRS"
+ ,"BERV"
+ ,"BERV_DRS"
+ ,"DRIFT"
+ #,"e_DRIFT"
+ #,"SA"
+ #,"NZP_mlc"
+ #,"dNZP_mlc"
+ #,"NZP_drs"
+ ,"dNZP_drs"
+ ,"TMMEAN"
+ ,"EXPTIME"
+ ,"MLCRX"
+ #,"E_MLCRX"
+ ]
+].values.astype(np.float32)
+
+embedding = reducer.fit_transform(harps_data)
+print(embedding.shape)
+
+rv = harps.RV_mlc_nzp
+rvmin = rv.min()
+
+plt.style.use('dark_background')
+
+plt.scatter(
+ embedding[:, 0],
+ embedding[:, 1],
+ norm=colors.SymLogNorm(linthresh=0.03, linscale=0.03, base=10, vmin=rv.min(), vmax=rv.max()),
+ c=rv, cmap=cm.cool, s=1
+ )
+
+plt.gca().set_aspect('equal', 'datalim')
+plt.title('UMAP projection of the HARPS dataset', fontsize=24)
+
+plt.savefig('out.png', dpi=300)
