In [21]:
plt.figure()
sns.scatterplot(x='ra',y='redshift',data=s)
Out[21]:
<Axes: xlabel='ra', ylabel='redshift'>
In this notebook, we perform exploratory data analysis in preparation for tomorrow's k-means clustering. The goal is to look for associations in floating-point data, which might represent clusters.
import matplotlib.pyplot as plt
import numpy as np
import warnings
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
from sklearn.model_selection import train_test_split, cross_val_score
import os
os.chdir("/home/wln/Documents/python_programs/Astronomy_Datasets")
s = pd.read_csv("sloan_survey.csv")
This is where we import the libraries we need to perform the rest of our analyisis, as well as collecting our data from the Sloan survey csv file.
s.columns
Index(['objid', 'ra', 'dec', 'u', 'g', 'r', 'i', 'z', 'run', 'rerun', 'camcol',
'field', 'specobjid', 'class', 'redshift', 'plate', 'mjd', 'fiberid'],
dtype='object')
s.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 10000 entries, 0 to 9999 Data columns (total 18 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 objid 10000 non-null float64 1 ra 10000 non-null float64 2 dec 10000 non-null float64 3 u 10000 non-null float64 4 g 10000 non-null float64 5 r 10000 non-null float64 6 i 10000 non-null float64 7 z 10000 non-null float64 8 run 10000 non-null int64 9 rerun 10000 non-null int64 10 camcol 10000 non-null int64 11 field 10000 non-null int64 12 specobjid 10000 non-null float64 13 class 10000 non-null object 14 redshift 10000 non-null float64 15 plate 10000 non-null int64 16 mjd 10000 non-null int64 17 fiberid 10000 non-null int64 dtypes: float64(10), int64(7), object(1) memory usage: 1.4+ MB
Here we observe that we have 10 floating point columns, and seven integer columns. 5 of the floating point columns handle the spectral qualities of the image, and two handle right ascension and declination, respectively. There is also a column for redshift and two for id values. The integer columns relate to CCD information.
print(s['run'].unique())
print(len(s['run'].unique()))
[ 752 756 308 727 745 1035 1045 1140 1231 1332 1334 1302 1239 1119 1331 1345 1350 1404 1412 1336 1402 1411 1356] 23
There were 23 different imaging runs represented in this sample of 10000 objects
print(s['rerun'].unique())
[301]
Only one of the imaging runs was rerun.
print(s['camcol'].unique())
[4 2 1 5 6 3]
There were 6 different camera columns used.
print(s['field'].unique())
print(len(s['field'].unique()))
[267 268 269 270 271 272 273 274 275 276 277 278 279 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 502 503 504 505 506 508 509 510 511 512 515 516 517 518 519 520 521 522 104 105 110 111 112 113 114 115 116 25 26 28 29 90 91 92 93 95 74 70 71 66 103 538 539 567 568 571 573 575 576 579 580 411 414 415 416 417 418 420 421 422 424 425 426 427 428 430 431 432 435 437 438 439 440 442 443 280 281 282 283 284 285 286 287 288 289 244 245 246 249 251 252 253 254 255 256 257 258 259 260 261 262 263 314 315 316 318 319 320 321 322 323 324 325 326 523 524 525 526 527 528 529 530 531 581 583 584 585 591 607 608 167 169 536 540 563 569 570 577 588 590 598 602 606 205 446 197 199 200 201 202 203 207 208 209 210 211 213 214 215 216 217 218 219 220 290 291 292 221 222 223 224 225 226 227 444 445 448 451 452 453 454 455 456 457 458 459 460 461 463 464 465 466 467 469 317 447 532 534 535 537 544 559 560 561 562 564 565 566 572 574 587 533 541 542 543 545 546 102 107 108 117 118 119 120 121 122 125 126 127 128 129 228 229 230 265 266 327 329 330 331 333 334 335 336 338 339 340 341 343 344 345 346 347 348 349 185 186 187 189 192 193 195 196 198 206 212 231 232 233 234 235 236 237 238 239 240 241 242 470 471 474 475 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 593 595 596 597 600 605 609 163 165 247 248 250 547 549 550 551 552 553 554 555 556 557 558 138 139 140 141 142 143 144 145 146 147 148 151 152 153 154 155 156 158 161 162 328 332 342 350 351 130 131 132 133 134 135 136 137 337 352 353 355 356 357 358 360 361 362 363 364 365 366 367 368 495 496 497 498 499 500 501 507 513 264 164 166 168 170 171 172 173 174 175 177 178 184 123 124 149 150 582 586 589 592 594 599 601 243 354 369 370 371 372 578 373 374 375 376 377 378 379 380 381 190 191 194 11 12 13 15 16 17 18 19 359 47 14 157 159 604 46 382 383 384 385 386 387 388 389 391 393 394 395 603 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 80 82 84 85 86 88 89 94 97 98 548 188 33 23 32 160 179 180 181 182 20 21 22 24 27 52 53 54 390 392 396 176 183 626 627 628 631 632 633 634 635 636 637 638 639 640 642 643 644 645 646 648 649 650 652 653 654 655 656 657 659 204 441 30 31 34 35 36 37 38 399 413 419 423 429 661 662 663 664 665 666 667 668 669 670 671 673 674 675 676 677 679 680 681 682 683 397 398 400 401 402 55 56 57 58 59 60 62 63 64 65 67 68 69 72 75 76 77 78 81 83 433 434 436 39 630 762 763 764 765 766 767 61 40 41 42 43 48 49 51 449 641 647 651 660 672 678 684 686 687 688 689 690 692 694 696 697 698 699 700 701 703 704 705 707 710 711 44 45 50 685 691 693 695 702 709 407 410 412 99 100 101 106 109 450 403 404 408 409 96 405 462 468 79 87 472 473 476 477 478 768 73 406] 703
There were 703 fields in this run.
plt.figure()
sns.scatterplot(x='ra',y='redshift',data=s)
<Axes: xlabel='ra', ylabel='redshift'>
In this plot of right ascension vs. redshift, we see that there are very few objects with a redshift greater than three, and potential clusters may exist on this graph.
plt.figure()
sns.scatterplot(x='dec',y='redshift', data=s)
<Axes: xlabel='dec', ylabel='redshift'>
We see similar features in this plot of declination vs. redshift.
plt.figure()
sns.scatterplot(x='ra', y='dec', data=s)
<Axes: xlabel='ra', ylabel='dec'>
This is a plot of declination vs right ascension. There are two large clusters of objects, as can be seen in the left of the graph.
In this analysis, three clusters appear to exist in all three graphs. Hence, a value of n_clusters = 3 will start tomorrow's k-means clustering.