TC-Python Installation Guide

Thermo-Calc Version 2021b

 About the TC-Python Documentation

About the TC-Python Documentation

The TC-Python help is available on the Thermo-Calc website where you can also
access additional resources. You can alternatively access the locally installed
documentation from the Help menu in Thermo-Calc.

Opening the Locally Installed Documentation

1. In Thermo-Calc from the menu, select Help → Manuals Folder.

You can also navigate to the Manuals folder outside of Thermo-Calc and
based on your operating system installation location.

About the TC-Python Documentation ǀ 2 of 16

 About the TC-Python Documentation

2. In the window that opens, double-click to open the TC-Python folder.

3. Double-click the Index HTML file (or any of the HTML files).

4. The API reference documentation opens as a searchable HTML file in your local
browser.

Opening the Locally Installed Documentation ǀ 3 of 16

 Which TC-Python Installation Should I Choose?

Which TC-Python Installation Should I Choose?

A license is required to run TC-Python. Also remember to review the System

Requirements page on our website.

There are these options to install TC-Python.

See the Videos about TC-Python page for additional learning resources.

Install with the Bundled Python-interpreter

This is a straightforward installation where TC-Python is preinstalled together with some
popular Python packages.
This is the recommended option for new users to TC-Python who may only want to run the
examples. The available preinstalled packages (e.g. numpy, scipy, scikit-learn, and
matplotlib) can also be sufficient for many advanced users.

"Installing TC-Python with a Bundled Python-interpreter" on the next page.

Install with a Python-interpreter of Your Choice

This installation requires additional steps and is the recommended option for more
advanced use and has full flexibility.

"Installing TC-Python Into a Python-interpreter of Your Choice" on page 8.

Which TC-Python Installation Should I Choose? ǀ 4 of 16

 Installing TC-Python with a Bundled Python-interpreter

Installing TC-Python with a Bundled Python-

interpreter
TC-Python can be installed using a Python-interpreter that is bundled to Thermo-Calc. This
interpreter has TC-Python preinstalled together with some popular Python packages. This is
the recommended option for new users to TC-Python who may only want to run the
examples. The available preinstalled packages (e.g. numpy, scipy, scikit-learn, and
matplotlib) can also be sufficient for many advanced users.

The bundled interpreter package is also used to run the Property Models in
Thermo-Calc. Do not make changes to interpreter packages or this can stop
Thermo-Calc from working correctly. To fix issues caused by changes to an
interpreter, completely uninstall Thermo-Calc and start over, either with the
bundled interpreter package or using interpreters of your choice.

If you are planning to do more advanced calculations that require other

external packages then you may want to follow these instructions instead:
"Installing TC-Python Into a Python-interpreter of Your Choice" on page 8

Limitations

The following examples cannot be run with the bundled Python-interpreter as

additional packages are required: pyex_M_01_Input_from_file.py (pandas) and
pyex_M_02_Output_to_file.py (pandas, lxml, h5py).

The bundled Python 3.7.2 interpreter contains the following major packages.

If you have recommendations for additional packages to include (see below),

send an email to Thermo-Calc Support.

Package Version

matplotlib 3.3.2

numpy 1.19.2

scikit-learn 0.23.2

Installing TC-Python with a Bundled Python-interpreter ǀ 5 of 16

 Installing TC-Python with a Bundled Python-interpreter

Package Version

scipy 1.5.2

TC-Python 2021b

Step 1. Install an IDE (Integrated Development Environment)

Any editor can be used to write the Python code, but an IDE is recommended, e.g. PyCharm.
These instructions are based on the use of PyCharm.
Use of an IDE gives you access to code completion, which helps when you use the API as it
gives you the available methods on the objects you are working with.

1. Navigate to the PyCharm website: https://www.jetbrains.com/pycharm/download.

2. Click to choose your OS and then click Download. You can use the Community
version of Pycharm.
3. Follow the instructions. It is recommended you keep all the defaults.

For Mac installations, you also need to set some environment variables as
described below in "macOS: Setting Environment Variables" on page 14.

Step 2. Configure PyCharm to Use the Bundled Python-interpreter

Open PyCharm and configure the interpreter:

1. Go the menu File→Settings.

2. Navigate in the tree to Project: YourProjectName and choose Project Interpreter.

3. Click on the settings symbol close to the Project Interpreter dropdown menu and
choose Add.

4. Select System Interpreter and add the bundled Thermo-Calc Python 3 interpreter,

Step 1. Install an IDE (Integrated Development Environment) ǀ 6 of 16

 Installing TC-Python with a Bundled Python-interpreter

which is located in a folder based on the operating system:

Operating
Path to the bundled Python-interpreter
system

Windows C:\Program Files\Thermo-Calc\2021b\python\python.exe

Linux /home/UserName/Thermo-Calc/2021b/python/bin/python3

macOS /Applications/Thermo-Calc-2021b.app/Contents/Resources/python/bin/python3

5. Select the interpreter and confirm.

Step 3. Run a TC-Python Example

It is recommended that you open one or more of the included examples to both check that
the installation has worked and to start familiarizing yourself with the code.
Open the TC-Python Project in PyCharm

When you first open the TC-Python project and examples, it can take a few
moments for the Pycharm IDE to index before some of the options are available.
1. Open PyCharm and then choose File → Open. The first time you open the project
you will need to navigate to the path of the TC-Python installation based on your
operating system.
2. Click the Examples folder and then click OK.
3. From any subfolder:
l Double-click to open an example file to examine the code.
l Right-click an example and choose Run.

If you are not following the recommended approach and create a new project
(File→New Project…), you need to consider that by default the options to
choose the interpreter are hidden within the Create Project window. For this,
click Project Interpreter: New Virtual Environment and in most cases choose
your System Interpreter containing the Python bundled to Thermo-Calc instead
of the default New Virtual Environment.

Step 3. Run a TC-Python Example ǀ 7 of 16

 Installing TC-Python Into a Python-interpreter of Your Choice

Installing TC-Python Into a Python-interpreter of

Your Choice
TC-Python can be installed using a Python-interpreter of your choice. This is the
recommended option for any more advanced usage and provides full flexibility.

If you are new to TC-Python and want to mainly run the examples, then it is
recommended you install the bundled Python-interpreter. The available
preinstalled packages (e.g. numpy, scipy, scikit-learn, and matplotlib) can also
be sufficient for many advanced users. See "Installing TC-Python with a Bundled
Python-interpreter" on page 5.

Step 1. Install a Python Distribution

If you already have a Python distribution installation, version 3.5 or higher, skip
this step.

These instructions are based on using the Anaconda platform for the Python distribution.
Install version 3.5 or higher to be able to work with TC-Python, although it is recommended
that you use the most recent version.
Install Anaconda

1. Navigate to the Anaconda website: https://www.anaconda.com/download/.

2. Click to choose your OS (operating system) and then click Download. Follow the
instructions. It is recommended you keep all the defaults.

Step 2. Install Thermo-Calc and the TC-Python SDK

1. Install Thermo-Calc.
2. When the installation is complete, open the TC-Python folder that includes the *.whl
file needed for the next step. There is also an Examples folder with Python files you
can use in the IDE to understand and work with TC-Python.

Operating
Path to the TC-Python folder
system

Windows C:\Users\UserName\Documents\Thermo-Calc\2021b\SDK\TC-Python

Linux /home/UserName/Thermo-Calc/2021b/SDK/TC-Python

Installing TC-Python Into a Python-interpreter of Your Choice ǀ 8 of 16

 Installing TC-Python Into a Python-interpreter of Your Choice

Operating
Path to the TC-Python folder
system

macOS /Users/Shared/Thermo-Calc/2021b/SDK/TC-Python

Step 3. Install TC-Python

"Updating to a Newer Version" on page 13

On Windows, it is recommended you use the Anaconda distribution prompt (i.e. Anaconda),
especially if you have other Python installations.

Do not use Virtual Environments unless you have a good reason to do so. Also
see "Fixing Potential Issues with the Environment" on page 15.

1. Open the command line. For example, in Anaconda on a Windows OS, go to Start →
Anaconda → Anaconda Prompt.

2. At the command line, enter the following. Make sure there are no spaces at the end
of the string or in the folder name or it will not run:
pip install <path to the TC-Python folder>/TC_Python-<version>-
py3-none-any.whl

On Linux, and depending on the interpreter, pip3 is typically used.

3. Press <Enter>. When the process is completed, there is a confirmation that

TC-Python is installed.

Step 3. Install TC-Python ǀ 9 of 16

 Installing TC-Python Into a Python-interpreter of Your Choice

If your computer is located behind a proxy-server, the default pip-

command will fail with a network connection error. In that case you
need to install the dependency py4j in a special configuration:

pip install -proxy user:password@proxy_ip:port py4j

Go to the TC-Python Help Troubleshooting section for detailed information.

This can be found on our website or from the installed documentation
where you can search for "pip install fails".

Step 4. Install an IDE (Integrated Development Environment)

Any editor can be used to write the Python code, but an IDE is recommended, e.g. PyCharm.
These instructions are based on the use of PyCharm.
Use of an IDE gives you access to code completion, which helps when you use the API as it
gives you the available methods on the objects you are working with.

1. Navigate to the PyCharm website: https://www.jetbrains.com/pycharm/download.

2. Click to choose your OS and then click Download. You can use the Community
version of Pycharm.
3. Follow the instructions. It is recommended you keep all the defaults.

For Mac installations, you also need to set some environment variables as
described below in "macOS: Setting Environment Variables" on page 14.

Step 5. Open the IDE and Run a TC-Python Example

After you complete all the software installations and set up the pip install path, you are
ready to open the IDE to start working with TC-Python.
It is recommended that you open one or more of the included examples to both check that
the installation has worked and to start familiarizing yourself with the code.
Open the TC-Python Project in PyCharm

When you first open the TC-Python project and examples, it can take a few
moments for the Pycharm IDE to index before some of the options are available.

Step 4. Install an IDE (Integrated Development Environment) ǀ 10 of 16

 Installing TC-Python Into a Python-interpreter of Your Choice

1. Open PyCharm and then choose File → Open. The first time you open the project
you will need to navigate to the path of the TC-Python installation based on your
operating system.
2. Click the Examples folder and then click OK.
3. From any subfolder:
l Double-click to open an example file to examine the code.
l Right-click an example and choose Run.

Step 5. Open the IDE and Run a TC-Python Example ǀ 11 of 16

 Thermo-Calc Documentation Set

Installation Default Directory: TC-Python

These are the default paths to the TC-Python folder based on your operating system.

Operating
Path to the TC-Python folder
system

Windows C:\Users\UserName\Documents\Thermo-Calc\2021b\SDK\TC-Python

Linux /home/UserName/Thermo-Calc/2021b/SDK/TC-Python

macOS /Users/Shared/Thermo-Calc/2021b/SDK/TC-Python

For a bundled Python-interpreter installation, these are the paths to this location based on
your operating system.

Operating
Path to the bundled Python-interpreter
system

Windows C:\Program Files\Thermo-Calc\2021b\python\python.exe

Linux /home/UserName/Thermo-Calc/2021b/python/bin/python3

macOS /Applications/Thermo-Calc-2021b.app/Contents/Resources/python/bin/python3

Which TC-Python Installation Should I Choose? ǀ 12 of 16

 Thermo-Calc Documentation Set

Updating to a Newer Version

If you are using the bundled Python-interpreter automatically installed with

Thermo-Calc, these steps are not required.

These steps are required if you have previously followed the instructions in
"Installing TC-Python Into a Python-interpreter of Your Choice" on page 8.

When updating to a newer version of Thermo-Calc, you always need to also install the latest
version of TC-Python. It is not sufficient to run the installer of Thermo-Calc (except for
bundled versions of the Python interpreter).
pip install <path to the TC-Python folder>/TC_Python-<version>-py3-
none-any.whl

In case of problems you may wish to uninstall the previous version of TC-Python in advance:
pip uninstall TC-Python
pip install <path to the TC-Python folder>/TC_Python-<version>-py3-
none-any.whl

However, this should not normally be required.

On Linux typically pip3 is used.

You can check the currently installed version of TC-Python by running:

pip show TC-Python

Which TC-Python Installation Should I Choose? ǀ 13 of 16

 Thermo-Calc Documentation Set

macOS: Setting Environment Variables

In order to use TC-Python on Mac you need to set some environment variables.
TC21B_HOME=/Applications/Thermo-Calc-2021b.app/Contents/Resources

If you use a license server:

LSHOST=<name-of-the-license-server>

If you have a node-locked license:

LSHOST= NO-NET
LSERVRC=/Users/Shared/Thermo-Calc/lservrc

In PyCharm, you can add environment variables in the configurations.

Select Run → Edit Configurations to open the Run/Debug Configurations window. Enter
the environment variable(s) by clicking the button to the right of the Environment Variables
text field. Now the environment variables(s) will be set for each new configuration by
default. The same way for configuring the environment variables can be used on other
operating systems as if necessary.

Which TC-Python Installation Should I Choose? ǀ 14 of 16

 Thermo-Calc Documentation Set

Fixing Potential Issues with the Environment

In most cases you should run TC-Python within your global Python 3 interpreter and not use
Virtual Environments unless you have a good reason to do so.
If there are problems with the interpreter settings, for example, a common problem when
you first use TC-Python is the message No module named tc_python, which you can resolve
in the Settings window as follows:

1. Go the menu File → Settings.

2. Navigate in the tree to <Project.YourProjectName> and choose Project Interpreter.
3. Click on the settings symbol close to the Project Interpreter drop-down menu and
choose Add.
4. Now choose System Interpreter and add your existing Python 3 interpreter.
5. Select your added interpreter and confirm.

If you are not following the recommended approach and create a new project
(File→New Project…), you need to consider that by default the options to
choose the interpreter are hidden within the Create Project window. For this,
click Project Interpreter: New Virtual Environment and in most cases choose
your System Interpreter containing the Python bundled to Thermo-Calc instead
of the default New Virtual Environment.

If you really need to use a Virtual Environment, it is recommended you review

the hints provided in the Best Practices section in the full TC-Python HTML help
or PDF.

Which TC-Python Installation Should I Choose? ǀ 15 of 16

 Thermo-Calc Documentation Set

TC-Python Diagnostic Script

A diagnostic script is available to help you if you have problems running TC-Python.

It is important to run the diagnostic script EXACTLY the same way as you run
your TC-Python script (in the same IDE, same project, same Python
environment, same Jupyter notebook, etc).

The script automatically prints useful suggestions that, in many cases, will help you to fix the
problem yourself. If you are unable to solve the problem, contact us for support. You can
find the script in these locations:

l Miscellaneous example (diagnostics.py)

l In the Troubleshooting section of the TC-Python API Reference Documentation. Also
refer to the Best Practices in the same guide to see if there are any other solutions or
tips.
l On our website.

Which TC-Python Installation Should I Choose? ǀ 16 of 16