Recent changes to bugs

UnstablePeriod

2021-11-24T11:49:23.648000Z

ta-lib is a great project and I'm using the abstraction interface to calculate all indicators for a bunch of stocks. I am a little confused about how the unstablePeriod is working. I would like to update indicators instead of rerunning it when I get new data for price. I have it somewhat working things like on balance volume get reset even with the unstablePeriod set to about 100 globally. I read that you can set the start index equal to the end index and walk through the data for each data point but I can't seem to get that to work. was it meant that the start index should be set to the end index of the previous data set of price? or the start index should be set to to itself esentially and do that for all points. i wrote it a few years ago and had some documentation but cant find where it is located. some links were inactive now. any place i can go to for resources on this? Thanks.

#108 Issues building Docker Image

2021-01-25T20:10:07.760000Z

ta-lib is a library written in C. It's not a python module and can't be loaded into python directly. You need a python wrapper around ta-lib. By "import talib" you are trying to find and load such wrapper. The python wrapper is another project and hosted here: https://github.com/mrjbq7/ta-lib
It could be build on machine or installed with help of "pip install ta-lib". So you should have both - c library and python wrapper for it on machine.

#108 Issues building Docker Image

2021-01-25T13:25:56.436000Z

Digging deeper into the build inside the config.log, I see:

and

configure:20612: checking for pow
configure:20668: gcc -o conftest -g -O2 conftest.c -lpthread -ldl >&5
conftest.c:69:6: warning: conflicting types for built-in function 'pow' [-Wbuiltin-declaration-mismatch]
char pow ();
^~~
/usr/bin/ld: /tmp/cctOKw0g.o: in function main': /ta-lib/conftest.c:80: undefined reference topow'
collect2: error: ld returned 1 exit status

Does that help ?

Thank you

Issues building Docker Image

2021-01-25T12:56:55.892000Z

Hello,
I'm trying to build a Docker image with talib - which does build.
But when I run it, I see:

Exception has occurred: ModuleNotFoundError
No module named 'talib'
File "C:\Users\Kasar\source\temp\test.py", line 12, in <module>
import talib</module>

My Dockerfile is standard boiler plate
- except section 'Setup Talib' (which I borrowed from a forum)

#Dockerfile

FROM python:3.8

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
COPY requirements.txt .
RUN python -m pip install -r requirements.txt

# Setup TA-Lib
RUN wget http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz && \
tar -xzf ta-lib-0.4.0-src.tar.gz && \
cd ta-lib/ && \
./configure --prefix=/usr && \
make && \
make install

WORKDIR /app
COPY . /app

RUN useradd appuser && chown -R appuser /app
USER appuser

CMD ["python", "test.py"]

Can anyone help debug what's going wrong please,
Thank you

CCI should use

2020-12-09T20:32:37.393000Z

There is a bug in CCI - for some data it returns wrong results. For ex.:

import talib
import numpy as np

timeperiod = 8
data = [1.1] *timeperiod

lows = np.asarray(data)
highs = lows.copy()
closes = lows.copy()

talib_ret = talib.CCI(highs, lows, closes, timeperiod=timeperiod)
print("talib", talib_ret[-1])
>>>> talib 66.66666666666667

while 0.0 is expected at talib_ret[-1].

That's because TA_CCI() contains:

      if( (tempReal != 0.0) && (tempReal2 != 0.0) )
      {
         outReal[outIdx++] = tempReal/(0.015*(tempReal2/optInTimePeriod));
      }
      else
         outReal[outIdx++] = 0.0;

while it should be:

      if( (!TA_IS_ZERO(tempReal)) && (!TA_IS_ZERO(tempReal2)) )
      {
         outReal[outIdx++] = tempReal/(0.015*(tempReal2/optInTimePeriod));
      }
      else
         outReal[outIdx++] = 0.0;

where TA_IS_ZERO() compares absolute argument value with 1e-14 instead of 0.0.
Otherwise TA_CCI() will return:

>>> 2.220446049250313e-16 / (0.015*(1.5543122344752192e-15 / 7))
66.66666666666667

Related discussion is here

#84 MACD Lookback function off for signal values of one

2020-11-19T17:41:30.016000Z

Just for the reference. There is a discussion abut this problem here. In the end I've decided to change the check from optInTimePeriod < 2 to optInTimePeriod < 1 for following indicators: SMA, EMA, WMA, DEMA, TEMA, TRIMA, KAMA, T3. Bcs the problem affects MACDEXT function too that can call any of MA indicators. And also added checks for LOOKBACK_CALL(MA or EMA) return value in MACD and STOCH indicators.

#106 HT_TRENDLINE writes to, but doesn't read, the smoothPrice circular buffer

2020-06-29T11:48:55.735000Z

Correction: src/ta_func/ta_HT_TRENDLINE.c.

HT_TRENDLINE writes to, but doesn't read, the smoothPrice circular buffer

2020-06-29T11:42:22.066000Z

In src/ta_HT_TRENDLINE.c, the smoothPrice buffer is populated with values, but those values are never actually read. The comments indicate that the smoothPrice is being used, but instead the direct input is being used (and averaged over).

     /* idx is used to iterate for up to 50 of the last
       * value of smoothPrice.
       */
      idx = today;
      tempReal = 0.0;
      for( i=0; i < DCPeriodInt; i++ )
         tempReal += inReal[idx--];

I believe that this is either a case of:
1. a fault in the calculation, or
2. an incorrect comment and unnecessary use of processing time and resources to maintain the smoothPrice.

At this point I am unclear which of the above applies here.

#105 getting incorrect outputs for several indicators

2018-05-29T19:37:34.622000Z

I was able to get a much closer result using this
https://github.com/kylejusticemagnuson/pyti/blob/master/pyti/chande_momentum_oscillator.py

still no luck figuring out what's happening with the ta-lib CMO calc

heres a comparison of the 2 the bottom being the pyti version
(apologies for the poor labeling)

Tested against MFI and got very similar results but still slightly different but reasonable

(sorry for all the images if theres a cleaner way to format this i'm open to suggestions)

getting incorrect outputs for several indicators

2018-05-29T19:36:49.328000Z

Hello, I've spent the past few days trying to figure out if i'm making some mistake or if its an issue with ta-lib. ( I'm using the python talib wrapper)
My CMO inputs have been relatively accurate when close to 0 but the minimums and maximums have been significantly different.
I scaled the values up to deal with rounding errors and tried pulling 6,000 candles to deal with the unstable period.
If anyone could help me out it would be much appreciated.

here's my code and the CMO is plotted at the bottom,
https://gist.github.com/K-Kit/511107c4d617fb11d4330cfa2bb433e9

and here's a snapshot of the tradingview for roughly the same time period

The min and max of the last ~24 hours should have been +/- 80 but its outputting +/-46

edit: I've also tried normalizing the data, just tried reading over the C code and i don't see what could cause the issue although I'm not very experienced in C

It seems to definitely be more than CMO, I just tested CCI and am getting nonsensical values:
When I check the minimum value returned by CCI over the past day I'm getting -317.68921
however the lowest the CCI should be able to go is -100

It appears to me that it crosses the 0 line at all the same points, however the farther from 0 the more off it gets