We Keep Coding

Why does my python loop return Key Error : 0 when I change the input dataframe?

I'm trying to do iterative calculation that will store the result of each iteration by append into a dataframe
however when I try to change the input dataframe into something else, I got the key error : 0
here are my complete code
d = []
df_it = df_ofr
i = 0
last_col = len(df_it.iloc[:,3:].columns) - 1
print("User Group : " + df_it[['user_type'][0]][0] + " " + df_it[['user_status'][0]][0])
for column in df_it.iloc[:,3:]:
if i > 0 :
if i < last_col: # 1 step conversion
convert_baseline = df_it[[column][0]][0]
convert_variant_a = df_it[[column][0]][1]
elif i == last_col: # end to end conversion
convert_baseline = df_it[[column][0]][0]
convert_variant_a = df_it[[column][0]][1]
lead_baseline = step_1_baseline
lead_variant_a = step_1_variant_a
#perform proportion z test
test_stat, p_value = proportions_ztest([convert_baseline,convert_variant_a], [lead_baseline,lead_variant_a], alternative='smaller')
#perform bayesian ab test
#initialize a test
test = BinaryDataTest()
#add variant using aggregated data
test.add_variant_data_agg("Baseline", totals=lead_baseline, positives=convert_baseline)
test.add_variant_data_agg("Variant A", totals=lead_variant_a, positives=convert_variant_a)
bay_result = test.evaluate(seed=99)
#append result
d.append(
{
'Convert into': column,
'# Users Baseline': lead_baseline,
'# Users Variant A': lead_variant_a,
'% CVR Baseline' : convert_baseline / lead_baseline,
'% CVR Variant A' : convert_variant_a / lead_variant_a,
'Z Test Stat' : test_stat,
'P-Value' : p_value,
'Prob Baseline being the Best' : bay_result[0]['prob_being_best'],
'Prob Variant A being the Best' : bay_result[1]['prob_being_best']
}
)
elif i == 0:
step_1_baseline = df_it[[column][0]][0]
step_1_variant_a = df_it[[column][0]][1]
i = i+1
lead_baseline = df_it[[column][0]][0]
lead_variant_a = df_it[[column][0]][1]
pd.DataFrame(d)
the one that I'm trying to change is this part
df_it = df_ofr
thanks for your help, really appreciate it
I'm trying to do iterative calculation that will store the result of each iteration by append into a dataframe

Python Chess Data (FEN) into Stockfish for Python

I am trying to use stockfish to evaluate a chess position using FEN notation all in Python. I am mainly using two libraries (pgnToFen I found on github here: https://github.com/SindreSvendby/pgnToFen and Stockfish the MIT licensed one here: https://github.com/zhelyabuzhsky/stockfish). After many bugs I have reached problem after problem. Stockfish not only can't analyse this FEN position (3b2k1/1p3pp1/8/3pP1P1/pP3P2/P2pB3/6K1/8 b f3 -) but it infinitely loops! "No worries!" and thought changing the source code would be accomplishable. Changed to _put(), but basically I am unable to put dummy values in because stdin.flush() won't execute once I give it those values! Meaning I don't even think I can skip to the next row in my dataframe. :( The code I changed is below.
def _put(self, command: str, tmp_time) -> None:
if not self.stockfish.stdin:
raise BrokenPipeError()
self.stockfish.stdin.write(f"{command}\n")
try:
self.stockfish.stdin.flush()
except:
if command != "quit":
self.stockfish.stdin.write('isready\n')
try:
time.sleep(tmp_time)
self.stockfish.stdin.flush()
except:
#print ('Imma head out', file=sys.stderr)
raise ValueError('Imma head out...')
#sys.stderr.close()
def get_evaluation(self) -> dict:
"""Evaluates current position
Returns:
A dictionary of the current advantage with "type" as "cp" (centipawns) or "mate" (checkmate in)
"""
evaluation = dict()
fen_position = self.get_fen_position()
if "w" in fen_position: # w can only be in FEN if it is whites move
compare = 1
else: # stockfish shows advantage relative to current player, convention is to do white positive
compare = -1
self._put(f"position {fen_position}", 5)
self._go()
x=0
while True:
x=x+1
text = self._read_line()
#print(text)
splitted_text = text.split(" ")
if splitted_text[0] == "info":
for n in range(len(splitted_text)):
if splitted_text[n] == "score":
evaluation = {
"type": splitted_text[n + 1],
"value": int(splitted_text[n + 2]) * compare,
}
elif splitted_text[0] == "bestmove":
return evaluation
elif x == 500:
evaluation = {
"type": 'cp',
"value": 10000,
}
return evaluation
and last but not least change to the init_ contructor below:
self._stockfish_major_version: float = float(self._read_line().split(" ")[1])
And the code where I am importing this code to is below, this is where errors pop up.
import pandas as pd
import re
import nltk
import numpy as np
from stockfish import Stockfish
import os
import sys
sys.path.insert(0, r'C:\Users\path\to\pgntofen')
import pgntofen
#nltk.download('punkt')
#Changed models.py for major version line 39 in stockfish from int to float
stockfish = Stockfish(r"C:\Users\path\to\Stockfish.exe")
file = r'C:\Users\path\to\selenium-pandas output.csv'
chunksize = 10 ** 6
for chunk in pd.read_csv(file, chunksize=chunksize):
for index, row in chunk.iterrows():
FullMovesStr = str(row['FullMoves'])
FullMovesStr = FullMovesStr.replace('+', '')
if "e.p" in FullMovesStr:
row.to_csv(r'C:\Users\MyName\Logger.csv', header=None, index=False, mode='a')
print('Enpassant')
continue
tokens = nltk.word_tokenize(FullMovesStr)
movelist = []
for tokenit in range(len(tokens)):
if "." in str(tokens[tokenit]):
try:
tokenstripped = re.sub(r"[0-9]+\.", "", tokens[tokenit])
token = [tokenstripped, tokens[tokenit+1]]
movelist.append(token)
except:
continue
else:
continue
DFMoves = pd.DataFrame(movelist, columns=[['WhiteMove', 'BlackMove']])
DFMoves['index'] = row['index']
DFMoves['Date'] = row['Date']
DFMoves['White'] = row['White']
DFMoves['Black'] = row['Black']
DFMoves['W ELO'] = row['W ELO']
DFMoves['B ELO'] = row['B ELO']
DFMoves['Av ELO'] = row['Av ELO']
DFMoves['Event'] = row['Event']
DFMoves['Site'] = row['Site']
DFMoves['ECO'] = row['ECO']
DFMoves['Opening'] = row['Opening']
pd.set_option('display.max_rows', DFMoves.shape[0]+1)
print(DFMoves[['WhiteMove', 'BlackMove']])
seqmoves = []
#seqmovesBlack = []
evalmove = []
pgnConverter = pgntofen.PgnToFen()
#stockfish.set_fen_position("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
#rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
for index, row in DFMoves.iterrows():
try:
stockfish.set_fen_position("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
except:
evalmove.append("?")
continue
#stockfish.set_fen_position("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
pgnConverter.resetBoard()
WhiteMove = str(row['WhiteMove'])
BlackMove = str(row['BlackMove'])
if index == 0:
PGNMoves1 = [WhiteMove]
seqmoves.append(WhiteMove)
#seqmoves.append(BlackMove)
else:
seqmoves.append(WhiteMove)
#seqmoves.append(BlackMove)
PGNMoves1 = seqmoves.copy()
#print(seqmoves)
try:
pgnConverter.pgnToFen(PGNMoves1)
fen = pgnConverter.getFullFen()
except:
break
try:
stockfish.set_fen_position(fen)
print(stockfish.get_board_visual())
evalpos = stockfish.get_evaluation()
evalmove.append(evalpos)
except:
pass
try:
stockfish.set_fen_position("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
except:
evalmove.append("?")
continue
pgnConverter.resetBoard()
if index == 0:
PGNMoves2 = [WhiteMove, BlackMove]
seqmoves.append(BlackMove)
else:
seqmoves.append(BlackMove)
PGNMoves2 = seqmoves.copy()
try:
pgnConverter.pgnToFen(PGNMoves2)
fen = pgnConverter.getFullFen()
except:
break
try:
stockfish.set_fen_position(fen)
print(stockfish.get_board_visual())
evalpos = stockfish.get_evaluation()
print(evalpos)
evalmove.append(evalpos)
except:
pass
#DFMoves['EvalWhite'] = evalwhite
#DFMoves['EvalBlack'] = evalblack
print(evalmove)
So the detailed question is getting stockfish.get_evalution() to just skip, or better yet fix the problem, for this ( 3b2k1/1p3pp1/8/3pP1P1/pP3P2/P2pB3/6K1/8 b f3 - ) FEN position. I have been working on this problem for quite a while so any insight into this would be very much appreciated.
My specs are Windows 10, Python 3.9, Processor:Intel(R) Core(TM) i9-10980XE CPU # 3.00GHz 3.00 GHz and RAM is 64.0 GB.
Thanks :)

Ok. It seems your fen is invalid (3b2k1/1p3pp1/8/3pP1P1/pP3P2/P2pB3/6K1/8 b f3 -). So check that. And python-chess (https://python-chess.readthedocs.io/en/latest/index.html) library allows you to use FEN AND chess engines. So, pretty cool no ? Here is an example of theses two fantastics tools :
import chess
import chess.engine
import chess.pgn
pgn = open("your_pgn_file.pgn")
game = chess.pgn.read_game(pgn)
engine = chess.engine.SimpleEngine.popen_uci("your_stockfish_path.exe")
# Iterate through all moves, play them on a board and analyse them.
board = game.board()
for move in game.mainline_moves():
board.push(move)
print(engine.analyse(board, chess.engine.Limit(time=0.1))["score"])

What is the most efficient way to a multiple variable in dictionary in python?

this my code, i'm looking, is other way to code this in most efficient way?
i have multiple variables and inserted to the dictionary.
please feel to suggest and other options like array and etc will do.
def momentEndSpan(span_type,max_combo,length):
if "simply supported" == span_type:
q = max_combo
force = {}
RA = {"PA" : q*length/2}
RB = {"PB" : q*length/2}
RA_moment = {"MA" : 0}
R_mid_moment = {"Mmid": (q*math.pow(length,2))/8 }
RB_moment = { "MB" : 0}
force.update(RA)
force.update(RB)
force.update(RA_moment)
force.update(R_mid_moment)
force.update(RB_moment)
return force
elif "one end continuous" == span_type:
q = max_combo
x = (3/8)*length
force = {}
RA = {"Phinge" : 3*q*length/8}
RB = {"Pfixed" : 5*q*length/8}
RA_moment = {"Mhinge" : 0}
R_mid_moment = {"Mmid": (q*math.pow(length,2))*(9/128) }
RB_moment = { "MB" : -1*(q*math.pow(length,2))/8 }
force.update(RA)
force.update(RB)
force.update(RA_moment)
force.update(R_mid_moment)
force.update(RB_moment)
return force
Thank you very much

The "More Pythonic" way is to create one dictionary and update once.
q = max_combo
force = {}
if "simply supported" == span_type:
new = {"PA" : q*length/2,
"PB" : q*length/2,
"MA" : 0, "Mmid": (q*math.pow(length,2))/8,
"MB" : 0}
elif "one end continuous" == span_type:
x = (3/8)*length
new = {"Phinge" : 3*q*length/8,
"Pfixed" : 5*q*length/8,
"Mhinge" : 0,
"Mmid": (q*math.pow(length,2))*(9/128),
"MB" : -1*(q*math.pow(length,2))/8 }
force.update(new)
Also, note that if the force dictionary doesn't contain any previously defined items you can simply return the new and/or just continue to update the new in your next operations if there are any. Or just use name force instead of new.
q = max_combo
if "simply supported" == span_type:
force = {...}
elif "one end continuous" == span_type:
x = (3/8)*length
force = {...}

Function return empty dataframe

I would like to make a loop to create a dataframe that gathers the lines of an input dataframe, which have common points.
My problem : When I apply the function, the output dataframe is empty...
yet with a print (output) in the loop, we see that the program works .. I do not understand, i tried to change return position but that doesn't work
Thank you in advance for your help !
def group (dataframe, identifiant, output):
for i in range(len(identifiant)):
ident = identifiant.loc[i,"IDCTV"]
# print(ident)
for j in range(len(dataframe)):
if dataframe.loc[j,"IDCONTREVENANT"] == ident:
di = dataframe.loc[j, "DATE_INFRACTION"]
nt = dataframe.loc[j,"NOTRAIN"]
genre = dataframe.loc[j,"CODEETATCIVIL"]
age = dataframe.loc[j,"AGE"]
# print(di, nt, genre, age)
for k in range(len(dataframe)):
if k != j :
if dataframe.loc[k,"DATE_INFRACTION"] == di and dataframe.loc[k,"NOTRAIN"] == nt:
idgroup = dataframe.loc[k,"IDCONTREVENANT"]
genreidgroup = dataframe.loc[k,"CODEETATCIVIL"]
ageidgroup = dataframe.loc[k,"AGE"]
output = output.append({ "IDREF" : ident ,"CODEETATCIVILREF" : genre,"AGEREF" : age ,"IDCTV" : idgroup,"CODEETATCIVILCTV" : genreidgroup,"AGECTV" : ageidgroup}, ignore_index = True)
print(output)
return output
group(df,IDCTV,df_groups)
print(df_groups)

I think you want to change
group(df,IDCTV,df_groups)
to
df_groups = group(df,IDCTV,df_groups)
Right now you're calling the group funciton and doing all that calculation, but you're not saving the output anywhere. So when you run print(df_groups) it prints out whatever it was before you called the function.

Index similar entries in Python

I have a column of data (easily imported from Google Docs thanks to gspread) that I'd like to intelligently align. I ingest entries into a dictionary. Input can include email, twitter handle or a blog URL. For example:
mike.j#gmail.com
#mikej45
j.mike#world.eu
_http://tumblr.com/mikej45
Right now, the "dumb" version is:
def NomineeCount(spreadsheet):
worksheet = spreadsheet.sheet1
nominees = worksheet.col_values(6) # F = 6
unique_nominees = {}
for c in nominees:
pattern = re.compile(r'\s+')
c = re.sub(pattern, '', c)
if unique_nominees.has_key(c) == True: # If we already have the name
unique_nominees[c] += 1
else:
unique_nominees[c] = 1
# Print out the alphabetical list of nominees with leading vote count
for w in sorted(unique_nominees.keys()):
print string.rjust(str(unique_nominees[w]), 2)+ " " + w
return nominees
What's an efficient(-ish) way to add in some smarts during the if process?

You can try with defaultdict:
from collections import defaultdict
unique_nominees = defaultdict(lambda: 0)
unique_nominees[c] += 1