File size: 41,411 Bytes
9bf26a6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 |
'''
* Project : Screenipy
* Author : Pranjal Joshi
* Created : 28/04/2021
* Description : Class for analyzing and validating stocks
'''
import sys
import math
import numpy as np
import pandas as pd
import joblib
import keras
import time
import classes.Utility as Utility
from copy import copy
from advanced_ta import LorentzianClassification
from classes.Utility import isGui
from sklearn.preprocessing import StandardScaler
from scipy.signal import argrelextrema
from scipy.stats import linregress
from classes.ColorText import colorText
from classes.SuppressOutput import SuppressOutput
from classes.ScreenipyTA import ScreenerTA
try:
import chromadb
CHROMA_AVAILABLE = True
except:
CHROMA_AVAILABLE = False
# Exception for newly listed stocks with candle nos < daysToLookback
class StockDataNotAdequate(Exception):
pass
# Exception for only downloading stock data and not screening
class DownloadDataOnly(Exception):
pass
# Exception for stocks which are not newly listed when screening only for Newly Listed
class NotNewlyListed(Exception):
pass
# This Class contains methods for stock analysis and screening validation
class tools:
def __init__(self, configManager) -> None:
self.configManager = configManager
# Private method to find candle type
# True = Bullish, False = Bearish
def getCandleType(self, dailyData):
return bool(dailyData['Close'].iloc[0] >= dailyData['Open'].iloc[0])
# Preprocess the acquired data
def preprocessData(self, data, daysToLookback=None):
if daysToLookback is None:
daysToLookback = self.configManager.daysToLookback
if self.configManager.useEMA:
sma = ScreenerTA.EMA(data['Close'],timeperiod=50)
lma = ScreenerTA.EMA(data['Close'],timeperiod=200)
data.insert(6,'SMA',sma)
data.insert(7,'LMA',lma)
else:
sma = data.rolling(window=50).mean()
lma = data.rolling(window=200).mean()
data.insert(6,'SMA',sma['Close'])
data.insert(7,'LMA',lma['Close'])
vol = data.rolling(window=20).mean()
rsi = ScreenerTA.RSI(data['Close'], timeperiod=14)
data.insert(8,'VolMA',vol['Volume'])
data.insert(9,'RSI',rsi)
data = data[::-1] # Reverse the dataframe
# data = data.fillna(0)
# data = data.replace([np.inf, -np.inf], 0)
fullData = data
trimmedData = data.head(daysToLookback)
return (fullData, trimmedData)
# Validate LTP within limits
def validateLTP(self, data, screenDict, saveDict, minLTP=None, maxLTP=None):
if minLTP is None:
minLTP = self.configManager.minLTP
if maxLTP is None:
maxLTP = self.configManager.maxLTP
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
recent = data.head(1)
pct_change = (data[::-1]['Close'].pct_change(fill_method=None) * 100).iloc[-1]
if pct_change > 0.2:
pct_change = colorText.GREEN + (" (%.1f%%)" % pct_change) + colorText.END
elif pct_change < -0.2:
pct_change = colorText.FAIL + (" (%.1f%%)" % pct_change) + colorText.END
else:
pct_change = colorText.WARN + (" (%.1f%%)" % pct_change) + colorText.END
ltp = round(recent['Close'].iloc[0],2)
saveDict['LTP'] = str(ltp)
verifyStageTwo = True
if self.configManager.stageTwo and len(data) > 250:
yearlyLow = data.head(250).min()['Close']
yearlyHigh = data.head(250).max()['Close']
if ltp < (2 * yearlyLow) or ltp < (0.75 * yearlyHigh):
verifyStageTwo = False
if(ltp >= minLTP and ltp <= maxLTP and verifyStageTwo):
screenDict['LTP'] = colorText.GREEN + ("%.2f" % ltp) + pct_change + colorText.END
return True
screenDict['LTP'] = colorText.FAIL + ("%.2f" % ltp) + pct_change + colorText.END
return False
# Validate if share prices are consolidating
def validateConsolidation(self, data, screenDict, saveDict, percentage=10):
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
hc = data.describe()['Close']['max']
lc = data.describe()['Close']['min']
if ((hc - lc) <= (hc*percentage/100) and (hc - lc != 0)):
screenDict['Consolidating'] = colorText.BOLD + colorText.GREEN + "Range = " + str(round((abs((hc-lc)/hc)*100),1))+"%" + colorText.END
else:
screenDict['Consolidating'] = colorText.BOLD + colorText.FAIL + "Range = " + str(round((abs((hc-lc)/hc)*100),1)) + "%" + colorText.END
saveDict['Consolidating'] = str(round((abs((hc-lc)/hc)*100),1))+"%"
return round((abs((hc-lc)/hc)*100),1)
# Validate Moving averages and look for buy/sell signals
def validateMovingAverages(self, data, screenDict, saveDict, maRange=2.5):
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
recent = data.head(1)
if(recent['SMA'].iloc[0] > recent['LMA'].iloc[0] and recent['Close'].iloc[0] > recent['SMA'].iloc[0]):
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + 'Bullish' + colorText.END
saveDict['MA-Signal'] = 'Bullish'
elif(recent['SMA'].iloc[0] < recent['LMA'].iloc[0]):
screenDict['MA-Signal'] = colorText.BOLD + colorText.FAIL + 'Bearish' + colorText.END
saveDict['MA-Signal'] = 'Bearish'
elif(recent['SMA'].iloc[0] == 0):
screenDict['MA-Signal'] = colorText.BOLD + colorText.WARN + 'Unknown' + colorText.END
saveDict['MA-Signal'] = 'Unknown'
else:
screenDict['MA-Signal'] = colorText.BOLD + colorText.WARN + 'Neutral' + colorText.END
saveDict['MA-Signal'] = 'Neutral'
smaDev = data['SMA'].iloc[0] * maRange / 100
lmaDev = data['LMA'].iloc[0] * maRange / 100
open, high, low, close, sma, lma = data['Open'].iloc[0], data['High'].iloc[0], data['Low'].iloc[0], data['Close'].iloc[0], data['SMA'].iloc[0], data['LMA'].iloc[0]
maReversal = 0
# Taking Support 50
if close > sma and low <= (sma + smaDev):
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + '50MA-Support' + colorText.END
saveDict['MA-Signal'] = '50MA-Support'
maReversal = 1
# Validating Resistance 50
elif close < sma and high >= (sma - smaDev):
screenDict['MA-Signal'] = colorText.BOLD + colorText.FAIL + '50MA-Resist' + colorText.END
saveDict['MA-Signal'] = '50MA-Resist'
maReversal = -1
# Taking Support 200
elif close > lma and low <= (lma + lmaDev):
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + '200MA-Support' + colorText.END
saveDict['MA-Signal'] = '200MA-Support'
maReversal = 1
# Validating Resistance 200
elif close < lma and high >= (lma - lmaDev):
screenDict['MA-Signal'] = colorText.BOLD + colorText.FAIL + '200MA-Resist' + colorText.END
saveDict['MA-Signal'] = '200MA-Resist'
maReversal = -1
# For a Bullish Candle
if self.getCandleType(data):
# Crossing up 50
if open < sma and close > sma:
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + 'BullCross-50MA' + colorText.END
saveDict['MA-Signal'] = 'BullCross-50MA'
maReversal = 1
# Crossing up 200
elif open < lma and close > lma:
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + 'BullCross-200MA' + colorText.END
saveDict['MA-Signal'] = 'BullCross-200MA'
maReversal = 1
# For a Bearish Candle
elif not self.getCandleType(data):
# Crossing down 50
if open > sma and close < sma:
screenDict['MA-Signal'] = colorText.BOLD + colorText.FAIL + 'BearCross-50MA' + colorText.END
saveDict['MA-Signal'] = 'BearCross-50MA'
maReversal = -1
# Crossing up 200
elif open > lma and close < lma:
screenDict['MA-Signal'] = colorText.BOLD + colorText.FAIL + 'BearCross-200MA' + colorText.END
saveDict['MA-Signal'] = 'BearCross-200MA'
maReversal = -1
return maReversal
# Validate if volume of last day is higher than avg
def validateVolume(self, data, screenDict, saveDict, volumeRatio=2.5):
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
recent = data.head(1)
if recent['VolMA'].iloc[0] == 0: # Handles Divide by 0 warning
saveDict['Volume'] = "Unknown"
screenDict['Volume'] = colorText.BOLD + colorText.WARN + "Unknown" + colorText.END
return True
ratio = round(recent['Volume'].iloc[0]/recent['VolMA'].iloc[0],2)
saveDict['Volume'] = str(ratio)+"x"
if(ratio >= volumeRatio and ratio != np.nan and (not math.isinf(ratio)) and (ratio != 20)):
screenDict['Volume'] = colorText.BOLD + colorText.GREEN + str(ratio) + "x" + colorText.END
return True
screenDict['Volume'] = colorText.BOLD + colorText.FAIL + str(ratio) + "x" + colorText.END
return False
# Find accurate breakout value
def findBreakout(self, data, screenDict, saveDict, daysToLookback):
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
recent = data.head(1)
data = data[1:]
hs = round(data.describe()['High']['max'],2)
hc = round(data.describe()['Close']['max'],2)
rc = round(recent['Close'].iloc[0],2)
if np.isnan(hc) or np.isnan(hs):
saveDict['Breaking-Out'] = 'BO: Unknown'
screenDict['Breaking-Out'] = colorText.BOLD + colorText.WARN + 'BO: Unknown' + colorText.END
return False
if hs > hc:
if ((hs - hc) <= (hs*2/100)):
saveDict['Breaking-Out'] = str(hc)
if rc >= hc:
screenDict['Breaking-Out'] = colorText.BOLD + colorText.GREEN + "BO: " + str(hc) + " R: " + str(hs) + colorText.END
return True and self.getCandleType(recent)
screenDict['Breaking-Out'] = colorText.BOLD + colorText.FAIL + "BO: " + str(hc) + " R: " + str(hs) + colorText.END
return False
noOfHigherShadows = len(data[data.High > hc])
if(daysToLookback/noOfHigherShadows <= 3):
saveDict['Breaking-Out'] = str(hs)
if rc >= hs:
screenDict['Breaking-Out'] = colorText.BOLD + colorText.GREEN + "BO: " + str(hs) + colorText.END
return True and self.getCandleType(recent)
screenDict['Breaking-Out'] = colorText.BOLD + colorText.FAIL + "BO: " + str(hs) + colorText.END
return False
saveDict['Breaking-Out'] = str(hc) + ", " + str(hs)
if rc >= hc:
screenDict['Breaking-Out'] = colorText.BOLD + colorText.GREEN + "BO: " + str(hc) + " R: " + str(hs) + colorText.END
return True and self.getCandleType(recent)
screenDict['Breaking-Out'] = colorText.BOLD + colorText.FAIL + "BO: " + str(hc) + " R: " + str(hs) + colorText.END
return False
else:
saveDict['Breaking-Out'] = str(hc)
if rc >= hc:
screenDict['Breaking-Out'] = colorText.BOLD + colorText.GREEN + "BO: " + str(hc) + colorText.END
return True and self.getCandleType(recent)
screenDict['Breaking-Out'] = colorText.BOLD + colorText.FAIL + "BO: " + str(hc) + colorText.END
return False
# Validate 'Inside Bar' structure for recent days
def validateInsideBar(self, data, screenDict, saveDict, chartPattern=1, daysToLookback=5):
orgData = data
daysToLookback = int(daysToLookback)
for i in range(daysToLookback, round(daysToLookback*0.5)-1, -1):
if i == 2:
return 0 # Exit if only last 2 candles are left
if chartPattern == 1:
if "Up" in saveDict['Trend'] and ("Bull" in saveDict['MA-Signal'] or "Support" in saveDict['MA-Signal']):
data = orgData.head(i)
refCandle = data.tail(1)
if (len(data.High[data.High > refCandle.High.item()]) == 0) and (len(data.Low[data.Low < refCandle.Low.item()]) == 0) and (len(data.Open[data.Open > refCandle.High.item()]) == 0) and (len(data.Close[data.Close < refCandle.Low.item()]) == 0):
screenDict['Pattern'] = colorText.BOLD + colorText.WARN + ("Inside Bar (%d)" % i) + colorText.END
saveDict['Pattern'] = "Inside Bar (%d)" % i
return i
else:
return 0
else:
if "Down" in saveDict['Trend'] and ("Bear" in saveDict['MA-Signal'] or "Resist" in saveDict['MA-Signal']):
data = orgData.head(i)
refCandle = data.tail(1)
if (len(data.High[data.High > refCandle.High.item()]) == 0) and (len(data.Low[data.Low < refCandle.Low.item()]) == 0) and (len(data.Open[data.Open > refCandle.High.item()]) == 0) and (len(data.Close[data.Close < refCandle.Low.item()]) == 0):
screenDict['Pattern'] = colorText.BOLD + colorText.WARN + ("Inside Bar (%d)" % i) + colorText.END
saveDict['Pattern'] = "Inside Bar (%d)" % i
return i
else:
return 0
return 0
# Validate if recent volume is lowest of last 'N' Days
def validateLowestVolume(self, data, daysForLowestVolume):
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
if daysForLowestVolume is None:
daysForLowestVolume = 30
data = data.head(daysForLowestVolume)
recent = data.head(1)
if((recent['Volume'].iloc[0] <= data.describe()['Volume']['min']) and recent['Volume'].iloc[0] != np.nan):
return True
return False
# validate if RSI is within given range
def validateRSI(self, data, screenDict, saveDict, minRSI, maxRSI):
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
rsi = int(data.head(1)['RSI'].iloc[0])
saveDict['RSI'] = rsi
if(rsi >= minRSI and rsi <= maxRSI) and (rsi <= 70 and rsi >= 30):
screenDict['RSI'] = colorText.BOLD + colorText.GREEN + str(rsi) + colorText.END
return True
screenDict['RSI'] = colorText.BOLD + colorText.FAIL + str(rsi) + colorText.END
return False
# Find out trend for days to lookback
def findTrend(self, data, screenDict, saveDict, daysToLookback=None,stockName=""):
if daysToLookback is None:
daysToLookback = self.configManager.daysToLookback
data = data.head(daysToLookback)
data = data[::-1]
data = data.set_index(np.arange(len(data)))
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
with SuppressOutput(suppress_stdout=True,suppress_stderr=True):
data['tops'] = data['Close'].iloc[list(argrelextrema(np.array(data['Close']), np.greater_equal, order=1)[0])]
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
try:
try:
if len(data) < daysToLookback:
raise StockDataNotAdequate
slope,c = np.polyfit(data.index[data.tops > 0], data['tops'][data.tops > 0], 1)
except Exception as e:
slope,c = 0,0
angle = np.rad2deg(np.arctan(slope))
if (angle == 0):
screenDict['Trend'] = colorText.BOLD + colorText.WARN + "Unknown" + colorText.END
saveDict['Trend'] = 'Unknown'
elif (angle <= 30 and angle >= -30):
screenDict['Trend'] = colorText.BOLD + colorText.WARN + "Sideways" + colorText.END
saveDict['Trend'] = 'Sideways'
elif (angle >= 30 and angle < 61):
screenDict['Trend'] = colorText.BOLD + colorText.GREEN + "Weak Up" + colorText.END
saveDict['Trend'] = 'Weak Up'
elif angle >= 60:
screenDict['Trend'] = colorText.BOLD + colorText.GREEN + "Strong Up" + colorText.END
saveDict['Trend'] = 'Strong Up'
elif (angle <= -30 and angle > -61):
screenDict['Trend'] = colorText.BOLD + colorText.FAIL + "Weak Down" + colorText.END
saveDict['Trend'] = 'Weak Down'
elif angle <= -60:
screenDict['Trend'] = colorText.BOLD + colorText.FAIL + "Strong Down" + colorText.END
saveDict['Trend'] = 'Strong Down'
except np.linalg.LinAlgError:
screenDict['Trend'] = colorText.BOLD + colorText.WARN + "Unknown" + colorText.END
saveDict['Trend'] = 'Unknown'
return saveDict['Trend']
# Find if stock is validating volume spread analysis
def validateVolumeSpreadAnalysis(self, data, screenDict, saveDict):
try:
data = data.head(2)
try:
# Check for previous RED candles
# Current candle = 0th, Previous Candle = 1st for following logic
if data.iloc[1]['Open'] >= data.iloc[1]['Close']:
spread1 = abs(data.iloc[1]['Open'] - data.iloc[1]['Close'])
spread0 = abs(data.iloc[0]['Open'] - data.iloc[0]['Close'])
lower_wick_spread0 = max(data.iloc[0]['Open'], data.iloc[0]['Close']) - data.iloc[0]['Low']
vol1 = data.iloc[1]['Volume']
vol0 = data.iloc[0]['Volume']
if spread0 > spread1 and vol0 < vol1 and data.iloc[0]['Volume'] < data.iloc[0]['VolMA'] and data.iloc[0]['Close'] <= data.iloc[1]['Open'] and spread0 < lower_wick_spread0 and data.iloc[0]['Volume'] <= int(data.iloc[1]['Volume']*0.75):
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + 'Supply Drought' + colorText.END
saveDict['Pattern'] = 'Supply Drought'
return True
if spread0 < spread1 and vol0 > vol1 and data.iloc[0]['Volume'] > data.iloc[0]['VolMA'] and data.iloc[0]['Close'] <= data.iloc[1]['Open']:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + 'Demand Rise' + colorText.END
saveDict['Pattern'] = 'Demand Rise'
return True
except IndexError:
pass
return False
except:
import traceback
traceback.print_exc()
return False
# Find if stock gaining bullish momentum
def validateMomentum(self, data, screenDict, saveDict):
try:
data = data.head(3)
for row in data.iterrows():
# All 3 candles should be Green and NOT Circuits
if row[1]['Close'].item() <= row[1]['Open'].item():
return False
openDesc = data.sort_values(by=['Open'], ascending=False)
closeDesc = data.sort_values(by=['Close'], ascending=False)
volDesc = data.sort_values(by=['Volume'], ascending=False)
try:
if data.equals(openDesc) and data.equals(closeDesc) and data.equals(volDesc):
if (data['Open'].iloc[0].item() >= data['Close'].iloc[1].item()) and (data['Open'].iloc[1].item() >= data['Close'].iloc[2].item()):
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + 'Momentum Gainer' + colorText.END
saveDict['Pattern'] = 'Momentum Gainer'
return True
except IndexError:
pass
return False
except Exception as e:
import traceback
traceback.print_exc()
return False
# Find stock reversing at given MA
def findReversalMA(self, data, screenDict, saveDict, maLength, percentage=0.015):
if maLength is None:
maLength = 20
data = data[::-1]
if self.configManager.useEMA:
maRev = ScreenerTA.EMA(data['Close'],timeperiod=maLength)
else:
maRev = ScreenerTA.MA(data['Close'],timeperiod=maLength)
data.insert(10,'maRev',maRev)
data = data[::-1].head(3)
if data.equals(data[(data.Close >= (data.maRev - (data.maRev*percentage))) & (data.Close <= (data.maRev + (data.maRev*percentage)))]) and data.head(1)['Close'].iloc[0] >= data.head(1)['maRev'].iloc[0]:
if self.configManager.stageTwo:
if data.head(1)['maRev'].iloc[0] < data.head(2)['maRev'].iloc[1] or data.head(2)['maRev'].iloc[1] < data.head(3)['maRev'].iloc[2] or data.head(1)['SMA'].iloc[0] < data.head(1)['LMA'].iloc[0]:
return False
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + f'Reversal-{maLength}MA' + colorText.END
saveDict['MA-Signal'] = f'Reversal-{maLength}MA'
return True
return False
# Find stock showing RSI crossing with RSI 9 SMA
def findRSICrossingMA(self, data, screenDict, saveDict, maLength=9):
data = data[::-1]
maRsi = ScreenerTA.MA(data['RSI'], timeperiod=maLength)
data.insert(10,'maRsi',maRsi)
data = data[::-1].head(3)
if data['maRsi'].iloc[0] <= data['RSI'].iloc[0] and data['maRsi'].iloc[1] > data['RSI'].iloc[1]:
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + f'RSI-MA-Buy' + colorText.END
saveDict['MA-Signal'] = f'RSI-MA-Buy'
return True
elif data['maRsi'].iloc[0] >= data['RSI'].iloc[0] and data['maRsi'].iloc[1] < data['RSI'].iloc[1]:
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + f'RSI-MA-Sell' + colorText.END
saveDict['MA-Signal'] = f'RSI-MA-Sell'
return True
return False
# Find IPO base
def validateIpoBase(self, stock, data, screenDict, saveDict, percentage=0.3):
listingPrice = data[::-1].head(1)['Open'].iloc[0]
currentPrice = data.head(1)['Close'].iloc[0]
ATH = data.describe()['High']['max']
if ATH > (listingPrice + (listingPrice * percentage)):
return False
away = round(((currentPrice - listingPrice)/listingPrice)*100, 1)
if((listingPrice - (listingPrice * percentage)) <= currentPrice <= (listingPrice + (listingPrice * percentage))):
if away > 0:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + f'IPO Base ({away} %)' + colorText.END
else:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + 'IPO Base ' + colorText.FAIL + f'({away} %)' + colorText.END
saveDict['Pattern'] = f'IPO Base ({away} %)'
return True
return False
# Find Conflucence
def validateConfluence(self, stock, data, screenDict, saveDict, percentage=0.1):
recent = data.head(1)
if(abs(recent['SMA'].iloc[0] - recent['LMA'].iloc[0]) <= (recent['SMA'].iloc[0] * percentage)):
difference = round(abs(recent['SMA'].iloc[0] - recent['LMA'].iloc[0])/recent['Close'].iloc[0] * 100,2)
if recent['SMA'].iloc[0] >= recent['LMA'].iloc[0]:
screenDict['MA-Signal'] = colorText.BOLD + colorText.GREEN + f'Confluence ({difference}%)' + colorText.END
saveDict['MA-Signal'] = f'Confluence ({difference}%)'
else:
screenDict['MA-Signal'] = colorText.BOLD + colorText.FAIL + f'Confluence ({difference}%)' + colorText.END
saveDict['MA-Signal'] = f'Confluence ({difference}%)'
return True
return False
# Find if stock is newly listed
def validateNewlyListed(self, data, daysToLookback):
daysToLookback = int(daysToLookback[:-1])
recent = data.head(1)
if len(data) < daysToLookback and (recent['Close'].iloc[0] != np.nan and recent['Close'].iloc[0] > 0):
return True
return False
# Find stocks approching to long term trendlines
def findTrendlines(self, data, screenDict, saveDict, percentage = 0.05):
period = int(''.join(c for c in self.configManager.period if c.isdigit()))
if len(data) < period:
return False
data = data[::-1]
data['Number'] = np.arange(len(data))+1
data_high = data.copy()
data_low = data.copy()
points = 30
''' Ignoring the Resitance for long-term purpose
while len(data_high) > points:
slope, intercept, r_value, p_value, std_err = linregress(x=data_high['Number'], y=data_high['High'])
data_high = data_high.loc[data_high['High'] > slope * data_high['Number'] + intercept]
slope, intercept, r_value, p_value, std_err = linregress(x=data_high['Number'], y=data_high['Close'])
data['Resistance'] = slope * data['Number'] + intercept
'''
while len(data_low) > points:
slope, intercept, r_value, p_value, std_err = linregress(x=data_low['Number'], y=data_low['Low'])
data_low = data_low.loc[data_low['Low'] < slope * data_low['Number'] + intercept]
slope, intercept, r_value, p_value, std_err = linregress(x=data_low['Number'], y=data_low['Close'])
data['Support'] = slope * data['Number'] + intercept
now = data.tail(1)
limit_upper = now['Support'].iloc[0].item() + (now['Support'].iloc[0].item() * percentage)
limit_lower = now['Support'].iloc[0].item() - (now['Support'].iloc[0].item() * percentage)
if limit_lower < now['Close'].iloc[0].item() < limit_upper and slope > 0.15:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + 'Trendline-Support' + colorText.END
saveDict['Pattern'] = 'Trendline-Support'
return True
''' Plots for debugging
import matplotlib.pyplot as plt
fig, ax1 = plt.subplots(figsize=(15,10))
color = 'tab:green'
xdate = [x.date() for x in data.index]
ax1.set_xlabel('Date', color=color)
ax1.plot(xdate, data.Close, label="close", color=color)
ax1.tick_params(axis='x', labelcolor=color)
ax2 = ax1.twiny() # ax2 and ax1 will have common y axis and different x axis, twiny
ax2.plot(data.Number, data.Resistance, label="Res")
ax2.plot(data.Number, data.Support, label="Sup")
plt.legend()
plt.grid()
plt.show()
'''
return False
# Find NRx range for Reversal
def validateNarrowRange(self, data, screenDict, saveDict, nr=4):
if Utility.tools.isTradingTime():
rangeData = data.head(nr+1)[1:]
now_candle = data.head(1)
rangeData['Range'] = abs(rangeData['Close'] - rangeData['Open'])
recent = rangeData.head(1)
if recent['Range'].iloc[0] == rangeData.describe()['Range']['min']:
if self.getCandleType(recent) and now_candle['Close'].iloc[0] >= recent['Close'].iloc[0]:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + f'Buy-NR{nr}' + colorText.END
saveDict['Pattern'] = f'Buy-NR{nr}'
return True
elif not self.getCandleType(recent) and now_candle['Close'].iloc[0] <= recent['Close'].iloc[0]:
screenDict['Pattern'] = colorText.BOLD + colorText.FAIL + f'Sell-NR{nr}' + colorText.END
saveDict['Pattern'] = f'Sell-NR{nr}'
return True
return False
else:
rangeData = data.head(nr)
rangeData['Range'] = abs(rangeData['Close'] - rangeData['Open'])
recent = rangeData.head(1)
if recent['Range'].iloc[0] == rangeData.describe()['Range']['min']:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + f'NR{nr}' + colorText.END
saveDict['Pattern'] = f'NR{nr}'
return True
return False
# Validate Lorentzian Classification signal
def validateLorentzian(self, data, screenDict, saveDict, lookFor=1):
# lookFor: 1-Any, 2-Buy, 3-Sell
data = data[::-1] # Reverse the dataframe
data = data.rename(columns={'Open':'open', 'Close':'close', 'High':'high', 'Low':'low', 'Volume':'volume'})
lc = LorentzianClassification(data=data)
if lc.df.iloc[-1]['isNewBuySignal']:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + f'Lorentzian-Buy' + colorText.END
saveDict['Pattern'] = f'Lorentzian-Buy'
if lookFor != 3:
return True
elif lc.df.iloc[-1]['isNewSellSignal']:
screenDict['Pattern'] = colorText.BOLD + colorText.FAIL + f'Lorentzian-Sell' + colorText.END
saveDict['Pattern'] = f'Lorentzian-Sell'
if lookFor != 2:
return True
return False
# Validate VPC
def validateVCP(self, data, screenDict, saveDict, stockName=None, window=3, percentageFromTop=3):
try:
percentageFromTop /= 100
data.reset_index(inplace=True)
data.rename(columns={'index':'Date'}, inplace=True)
data['tops'] = data['High'].iloc[list(argrelextrema(np.array(data['High']), np.greater_equal, order=window)[0])].head(4)
data['bots'] = data['Low'].iloc[list(argrelextrema(np.array(data['Low']), np.less_equal, order=window)[0])].head(4)
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
tops = data[data.tops > 0]
bots = data[data.bots > 0]
highestTop = round(tops.describe()['High']['max'],1)
filteredTops = tops[tops.tops > (highestTop-(highestTop*percentageFromTop))]
# print(tops)
# print(filteredTops)
# print(tops.sort_values(by=['tops'], ascending=False))
# print(tops.describe())
# print(f"Till {highestTop-(highestTop*percentageFromTop)}")
if(filteredTops.equals(tops)): # Tops are in the range
lowPoints = []
for i in range(len(tops)-1):
endDate = tops.iloc[i]['Date']
startDate = tops.iloc[i+1]['Date']
lowPoints.append(data[(data.Date >= startDate) & (data.Date <= endDate)].describe()['Low']['min'])
lowPointsOrg = lowPoints
lowPoints.sort(reverse=True)
lowPointsSorted = lowPoints
ltp = data.head(1)['Close'].iloc[0]
if lowPointsOrg == lowPointsSorted and ltp < highestTop and ltp > lowPoints[0]:
screenDict['Pattern'] = colorText.BOLD + colorText.GREEN + f'VCP (BO: {highestTop})' + colorText.END
saveDict['Pattern'] = f'VCP (BO: {highestTop})'
return True
except Exception as e:
import traceback
print(traceback.format_exc())
return False
def getNiftyPrediction(self, data, proxyServer):
import warnings
warnings.filterwarnings("ignore")
# Disable GPUs as this causes wrong preds in Docker
import tensorflow as tf
physical_devices = tf.config.list_physical_devices('GPU')
try:
tf.config.set_visible_devices([], 'GPU')
visible_devices = tf.config.get_visible_devices()
for device in visible_devices:
assert device.device_type != 'GPU'
except:
pass
#
model, pkl = Utility.tools.getNiftyModel(proxyServer=proxyServer)
datacopy = copy(data[pkl['columns']])
with SuppressOutput(suppress_stderr=True, suppress_stdout=True):
data = data[pkl['columns']]
### v2 Preprocessing
for col in pkl['columns']:
data[col] = data[col].pct_change(fill_method=None) * 100
data = data.ffill().dropna()
data = data.iloc[-1]
###
data = pkl['scaler'].transform([data])
pred = model.predict(data)[0]
if pred > 0.5:
out = colorText.BOLD + colorText.FAIL + "BEARISH" + colorText.END + colorText.BOLD
sug = "Hold your Short position!"
else:
out = colorText.BOLD + colorText.GREEN + "BULLISH" + colorText.END + colorText.BOLD
sug = "Stay Bullish!"
if not Utility.tools.isClosingHour():
print(colorText.BOLD + colorText.WARN + "Note: The AI prediction should be executed After 3 PM Around the Closing hours as the Prediction Accuracy is based on the Closing price!" + colorText.END)
print(colorText.BOLD + colorText.BLUE + "\n" + "[+] Nifty AI Prediction -> " + colorText.END + colorText.BOLD + "Market may Open {} next day! {}".format(out, sug) + colorText.END)
print(colorText.BOLD + colorText.BLUE + "\n" + "[+] Nifty AI Prediction -> " + colorText.END + "Probability/Strength of Prediction = {}%".format(Utility.tools.getSigmoidConfidence(pred[0])))
if isGui():
return pred, 'BULLISH' if pred <= 0.5 else 'BEARISH', Utility.tools.getSigmoidConfidence(pred[0]), pd.DataFrame(datacopy.iloc[-1]).T
return pred
def monitorFiveEma(self, proxyServer, fetcher, result_df, last_signal, risk_reward = 3):
col_names = ['High', 'Low', 'Close', '5EMA']
data_list = ['nifty_buy', 'banknifty_buy', 'nifty_sell', 'banknifty_sell']
data_tuple = fetcher.fetchFiveEmaData()
for cnt in range(len(data_tuple)):
d = data_tuple[cnt]
d['5EMA'] = ScreenerTA.EMA(d['Close'],timeperiod=5)
d = d[col_names]
d = d.dropna().round(2)
with SuppressOutput(suppress_stderr=True, suppress_stdout=True):
if 'sell' in data_list[cnt]:
streched = d[(d.Low > d['5EMA']) & (d.Low - d['5EMA'] > 0.5)]
streched['SL'] = streched.High
validate = d[(d.Low.shift(1) > d['5EMA'].shift(1)) & (d.Low.shift(1) - d['5EMA'].shift(1) > 0.5)]
old_index = validate.index
else:
mask = (d.High < d['5EMA']) & (d['5EMA'] - d.High > 0.5) # Buy
streched = d[mask]
streched['SL'] = streched.Low
validate = d.loc[mask.shift(1).fillna(False)]
old_index = validate.index
tgt = pd.DataFrame((validate.Close.reset_index(drop=True) - ((streched.SL.reset_index(drop=True) - validate.Close.reset_index(drop=True)) * risk_reward)),columns=['Target'])
validate = pd.concat([
validate.reset_index(drop=True),
streched['SL'].reset_index(drop=True),
tgt,
],
axis=1
)
validate = validate.tail(len(old_index))
validate = validate.set_index(old_index)
if 'sell' in data_list[cnt]:
final = validate[validate.Close < validate['5EMA']].tail(1)
else:
final = validate[validate.Close > validate['5EMA']].tail(1)
if data_list[cnt] not in last_signal:
last_signal[data_list[cnt]] = final
elif data_list[cnt] in last_signal:
try:
condition = last_signal[data_list[cnt]][0]['SL'].iloc[0]
except KeyError:
condition = last_signal[data_list[cnt]]['SL'].iloc[0]
# if last_signal[data_list[cnt]] is not final: # Debug - Shows all conditions
if condition != final['SL'].iloc[0]:
# Do something with results
try:
result_df = pd.concat([
result_df,
pd.DataFrame([
[
colorText.BLUE + str(final.index[0]) + colorText.END,
colorText.BOLD + colorText.WARN + data_list[cnt].split('_')[0].upper() + colorText.END,
(colorText.BOLD + colorText.FAIL + data_list[cnt].split('_')[1].upper() + colorText.END) if 'sell' in data_list[cnt] else (colorText.BOLD + colorText.GREEN + data_list[cnt].split('_')[1].upper() + colorText.END),
colorText.FAIL + str(final.SL[0]) + colorText.END,
colorText.GREEN + str(final.Target[0]) + colorText.END,
f'1:{risk_reward}'
]
], columns=result_df.columns)
], axis=0)
result_df.reset_index(drop=True, inplace=True)
except Exception as e:
pass
# Then update
last_signal[data_list[cnt]] = [final]
result_df.drop_duplicates(keep='last', inplace=True)
result_df.sort_values(by='Time', inplace=True)
return result_df[::-1]
# Add data to vector database
def addVector(self, data, stockCode, daysToLookback):
data = data[::-1] # Reinverting preprocessedData for pct_change
data = data.pct_change(fill_method=None)
# data = data[::-1] # Do we need to invert again? No we dont - See operation after flatten
data = data[['Open', 'High', 'Low', 'Close']]
data = data.reset_index(drop=True)
data = data.dropna()
data = data.to_numpy().flatten().tolist()
data = data[(-4 * daysToLookback):] # Keep only OHLC * daysToLookback samples
if len(data) == (4 * daysToLookback):
chroma_client = chromadb.PersistentClient(path="./chromadb_store/")
collection = chroma_client.get_or_create_collection(name="nse_stocks")
collection.upsert(
embeddings=[data],
documents=[stockCode],
ids=[stockCode]
)
return data
'''
# Find out trend for days to lookback
def validateVCP(data, screenDict, saveDict, daysToLookback=ConfigManager.daysToLookback, stockName=None):
// De-index date
data.reset_index(inplace=True)
data.rename(columns={'index':'Date'}, inplace=True)
data = data.head(daysToLookback)
data = data[::-1]
data = data.set_index(np.arange(len(data)))
data = data.fillna(0)
data = data.replace([np.inf, -np.inf], 0)
data['tops'] = data['Close'].iloc[list(argrelextrema(np.array(data['Close']), np.greater_equal, order=3)[0])]
data['bots'] = data['Close'].iloc[list(argrelextrema(np.array(data['Close']), np.less_equal, order=3)[0])]
try:
try:
top_slope,top_c = np.polyfit(data.index[data.tops > 0], data['tops'][data.tops > 0], 1)
bot_slope,bot_c = np.polyfit(data.index[data.bots > 0], data['bots'][data.bots > 0], 1)
topAngle = math.degrees(math.atan(top_slope))
vcpAngle = math.degrees(math.atan(bot_slope) - math.atan(top_slope))
# print(math.degrees(math.atan(top_slope)))
# print(math.degrees(math.atan(bot_slope)))
# print(vcpAngle)
# print(topAngle)
# print(data.max()['bots'])
# print(data.max()['tops'])
if (vcpAngle > 20 and vcpAngle < 70) and (topAngle > -10 and topAngle < 10) and (data['bots'].max() <= data['tops'].max()) and (len(data['bots'][data.bots > 0]) > 1):
print("---> GOOD VCP %s at %sRs" % (stockName, top_c))
import os
os.system("echo %s >> vcp_plots\VCP.txt" % stockName)
import matplotlib.pyplot as plt
plt.scatter(data.index[data.tops > 0], data['tops'][data.tops > 0], c='g')
plt.scatter(data.index[data.bots > 0], data['bots'][data.bots > 0], c='r')
plt.plot(data.index, data['Close'])
plt.plot(data.index, top_slope*data.index+top_c,'g--')
plt.plot(data.index, bot_slope*data.index+bot_c,'r--')
if stockName != None:
plt.title(stockName)
# plt.show()
plt.savefig('vcp_plots\%s.png' % stockName)
plt.clf()
except np.RankWarning:
pass
except np.linalg.LinAlgError:
return False
'''
|