# How to load Python packages in PythonCall? They've already been installed by Conda in REPL

**URL:** <https://discourse.julialang.org/t/how-to-load-python-packages-in-pythoncall-theyve-already-been-installed-by-conda-in-repl/104176>\
**Category:** General Usage\
**Created:** [September 23, 2023, 2:30pm UTC](https://discourse.julialang.org/t/how-to-load-python-packages-in-pythoncall-theyve-already-been-installed-by-conda-in-repl/104176 "2023-09-23T14:30:07Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![MaxM](https://avatars.discourse-cdn.com/v4/letter/m/85e7bf/32.png) [@MaxM](https://discourse.julialang.org/u/MaxM)\
**Post date:** [September 23, 2023, 2:30pm UTC](https://discourse.julialang.org/t/how-to-load-python-packages-in-pythoncall-theyve-already-been-installed-by-conda-in-repl/104176/1 "2023-09-23T14:30:08Z")

</div>

In the Julia REPL, I just ran `Conda.add("pandas")` and it ran as Anaconda prompt usually ran, concluding with “All requested packages already installed.” However, the following code then gives me an error (in a Jupyter Notebook).

```julia
using PythonCall
pypandas=pyimport("pandas") 

```

Python: ModuleNotFoundError: No module named ‘pandas’

On the other hand, pyimport(“math”) works just fine. How do I get PythonCall to recognise installed packages?

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [September 23, 2023, 3:54pm UTC](https://discourse.julialang.org/t/how-to-load-python-packages-in-pythoncall-theyve-already-been-installed-by-conda-in-repl/104176/3 "2023-09-23T15:54:48Z")

</div>

> [@yolhan\_mannes](#):
>
> You can link Pycall to your python repo

@MaxM is using PythonCall, not PyCall.

> [@MaxM](#):
>
> `using PythonCall`

PythonCall uses a different version of Python than the one installed by Conda.jl, by default. See the [PythonCall manual’s “Configuration” section](https://juliapy.github.io/PythonCall.jl/stable/pythoncall/#pythoncall-config). In particular, it uses CondaPkg.jl, not Conda.jl, by default.

See the configuration manual above on how to make PythonCall use a different Python installation. For example, you should be able to make it use Conda.jl via something like

```julia-auto
import Conda
ENV["JULIA_CONDAPKG_BACKEND"] = "Current"
ENV["JULIA_CONDAPKG_EXE"] = Conda.conda
ENV["JULIA_PYTHONCALL_EXE"] = joinpath(Conda.BINDIR, "python3")

using PythonCall

```

However, if you are using PythonCall I would generally recommend using CondaPkg.jl rather than Conda.jl, as the former is what PythonCall is designed for.