Quantra_py environment

Installations & Python Packages
1

Hi , I am unable to set up quantra_py environment , getting the below error
(base) C:\Users\shreeharshaa>conda activate quantra_py

EnvironmentNameNotFound: Could not find conda environment: quantra_py
You can list all discoverable environments with conda info --envs.

kindly help in resolving this

2

Hello,

It seems that the quantra environment was not installed correctly and hence you are getting the error.

You can first run the following code in the anaconda prompt

conda env list

If you see the name quantra_py, it means it wasn’t properly installed and you should uninstall it using the following line of code

conda remove --name quantra_py --all

Once done, you can install a new virtual environment. (Or if you don’t see the name, then you can go ahead with a new installation.)

Use the following line of code

conda env create -f https://raw.githubusercontent.com/quantra-go-algo/Quantra_Environment/main/environment_windows_ml.yml

This should create a new virtual environment.

3

I am trying to set up environment with ML libraries using conda env create -f https://quantra.quantinsti.com/downloads/machinelearning/environment.yml

Per the Anaconda prompt, this is failing due to conflicts within the packages (the environment does not get resolved).

I downloaded the yml and noticed the few potential issues that I have listed below. I tried solving for the issues, but it is not working so far. Happy to share the exact error message dump I get when I try running the remote file.

  • Redundant Dependencies:
  • You have listed several packages both under dependencies and within pip. Since pip installs should only be used for packages not available through conda, remove duplicates from pip when they exist in dependencies.
  • Inconsistent Package Names:
  • arch-py in dependencies is installed as arch in pip. It should be arch in both places.
  • Potential Conflicts:
  • tensorflow==2.6.0 is quite outdated. If you need TensorFlow, consider tensorflow==2.15.0 (latest stable) to avoid compatibility issues.
  • numpy==1.19.5 is also old. Some of your other packages (e.g., pandas_profiling) may require a newer version.
  • pyfolio and pyfolio-reloaded may conflict as they are forks of each other.
  • Obsolete or Deprecated Packages:
  • nsepy==0.8 is no longer actively maintained.
  • tweepy==3.10.0 is very old; the latest version is 4.x.
  • yfinance==0.1.63 is quite outdated.
4

Got reply from support and that works for me.

For creating the machine learning environment, kindly uninstall the existing virtual environment and install the machine learning environment using the following line in the anaconda prompt as one line.

conda env create -f https://raw.githubusercontent.com/quantra-go-algo/Quantra_Environment/main/environment_windows_ml.yml

But if you are not working on machine learning algorithms, then the general environment will be fine as well.

For installing the general environment without machine learning:

conda env create -f https://quantra.quantinsti.com/downloads/general/environment.yml