What’s New in PyCharm 2024.3.1

File creation without project setup

If you prefer not to worry about interpreter settings or project configuration, PyCharm can take care of them for you. The IDE now offers a one-click option that allows you to jump straight into the editor. You can start coding and executing immediately with a single Python script or Jupyter notebook, bypassing project setup and saving time.

Auto-import suggestions in code completion

PyCharm’s code completion now offers new suggestions, such as the option to auto-import class and function names from libraries that have not yet been imported into your code. This streamlined feature ensures smoother workflows and faster access to the tools you need, with no manual imports required!

Ability to comment out several cells at once

Sometimes, you want to isolate a part of the code in your notebook to see how the output would look without it, and sometimes, this code expands to multiple cells. For such cases, you can quickly comment out several cells at once. To do so, simply switch to Command mode, select the target cells with Shift+Up/Down, and then press Cmd+/ (macOS) or Ctrl+/ (Windows/Linux).

Structure view

With the enhanced Structure view, you can navigate through your notebooks at a glance. PyCharm automatically generates it from the headings of the Markdown cells, and you can also add Python cells with commented first lines to it from the View options menu.

From the Structure view, you can move around the notebook, run specific sections, and see whether a given section is running at any given time.

Integrated TensorBoard support

Enhance your machine learning experiments with TensorBoard integration directly in Jupyter notebooks. Effortlessly track and visualize key metrics like loss and accuracy, inspect model graphs, and monitor changes in weights and biases over time. Dive deeper with tools for embedding projection, data visualization (images, text, and audio), and TensorFlow profiling – all inside PyCharm.

After chatting with AI Assistant about a specific problem, you may want to move the results to a dedicated notebook. Simply prompt AI Assistant or use the new /create-jupyter-file command in the chat to create a notebook that contains the associated code. PyCharm will also add the notebook to your project directory.