docs: 0.6.0 release notes + uv-first README

CHANGELOG: 0.6.0 entry covering Bokeh-in-Colab (issue #109), Plot2d
(issue #137), max_cols=1 (PR #147), groups starting at different
steps (PR #149), Neptune integration removal, and the py3.10 floor.

README:
- Update CI badges to point at the renamed workflows (tests.yml,
  lint.yml) and link each badge to its workflow page
- Switch the install section to uv-first ('uv add livelossplot'),
  with pip kept as the alternative; add a 'Run without installing'
  section covering 'uv run --with' for one-off scripts, inline
  script dependencies, and Jupyter
- Replace the dead 'git+git://' source-install URL with the working
  'git+https://' form
- Fix the master->main branch in the header GIF link

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: stared

CHANGELOG.md

@@ -7,23 +7,22 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
-## [0.5.7] - 2026-05-04
+## [0.6.0] - 2026-05-04
 
-### Fixed
+### Changed
 
-- `BokehPlot` now renders and live-updates inside Google Colab. Colab blocks Jupyter Comms, so `push_notebook` was silently dead; the plot is now embedded via `<iframe srcdoc>` and refreshed through an IPython `display_id` handle ([issue #109](https://github.com/stared/livelossplot/issues/109)).
-- `Plot2d` works again as a `PlotLosses` output. Removed buggy `super().__init__(self)` in `LossSubplot`/`Plot1D` and reworked `Plot2d.send` to manage its own figure/axes and use the configured `predict` callback ([issue #137](https://github.com/stared/livelossplot/issues/137), PR #138).
-- `MatplotlibPlot` no longer errors when `max_cols=1` (PR #147).
-- Groups whose metrics start at different steps now plot correctly (PR #149).
-- CI: install `tensorboard` explicitly — TensorFlow 2.16+ no longer bundles it, which broke `external_test_tensorboard.py`.
+- Minimum Python is now 3.10.
 
 ### Removed
 
-- Neptune.AI support (`NeptuneLogger`, `to_neptune`, and the related example/test) — Neptune is no longer maintained.
+- Neptune.AI integration — Neptune is no longer maintained.
 
-### Chore
+### Fixed
 
-- Default branch renamed `master` → `main`; build badges updated.
+- `BokehPlot` now works in Google Colab ([issue #109](https://github.com/stared/livelossplot/issues/109)).
+- `Plot2d` works again as a `PlotLosses` output ([issue #137](https://github.com/stared/livelossplot/issues/137)).
+- `MatplotlibPlot` no longer errors when `max_cols=1` (PR #147).
+- Groups whose metrics start at different steps now plot correctly (PR #149).
 
 ## [0.5.6] - 2025-01-03
 

README.md

@@ -4,15 +4,15 @@
 ![PyPI status](https://img.shields.io/pypi/status/livelossplot.svg)
 ![MIT license - PyPI](https://img.shields.io/pypi/l/livelossplot.svg)
 ![Python version - PyPI](https://img.shields.io/pypi/pyversions/livelossplot.svg)
-[![GitHub Actions](https://img.shields.io/github/actions/workflow/status/stared/livelossplot/pythonpackage.yml?branch=main)](https://github.com/stared/livelossplot/actions)
-[![GitHub Actions: Integrations](https://img.shields.io/github/actions/workflow/status/stared/livelossplot/external_packages.yml?branch=main&label=integrations)](https://github.com/stared/livelossplot/actions)
-[![GitHub Actions: Linting](https://img.shields.io/github/actions/workflow/status/stared/livelossplot/flake8_yapf.yml?branch=main&label=linting)](https://github.com/stared/livelossplot/actions)
+[![Tests](https://img.shields.io/github/actions/workflow/status/stared/livelossplot/tests.yml?branch=main&label=tests)](https://github.com/stared/livelossplot/actions/workflows/tests.yml)
+[![Integrations](https://img.shields.io/github/actions/workflow/status/stared/livelossplot/external_packages.yml?branch=main&label=integrations)](https://github.com/stared/livelossplot/actions/workflows/external_packages.yml)
+[![Lint](https://img.shields.io/github/actions/workflow/status/stared/livelossplot/lint.yml?branch=main&label=lint)](https://github.com/stared/livelossplot/actions/workflows/lint.yml)
 [![Downloads](http://pepy.tech/badge/livelossplot)](https://pepy.tech/project/livelossplot)
 [![Twitter @pmigdal](https://img.shields.io/twitter/follow/pmigdal)](https://twitter.com/pmigdal)
 
 Don't train deep learning models blindfolded! Be impatient and look at each epoch of your training!
 
-([RECENT CHANGES](CHANGELOG.md), [EXAMPLES IN COLAB](https://colab.research.google.com/github/stared/livelossplot), [CODE](https://github.com/stared/livelossplot))
+([API DOCS](https://p.migdal.pl/livelossplot/), [RECENT CHANGES](CHANGELOG.md), [EXAMPLES IN COLAB](https://colab.research.google.com/github/stared/livelossplot), [CODE](https://github.com/stared/livelossplot))
 
 A live training loss plot in [Jupyter Notebook](http://jupyter.org/) for [Keras](https://keras.io/), [PyTorch](http://pytorch.org/) and other frameworks. An open-source Python package by [Piotr Migdał](https:///), [Bartłomiej Olechno](https://github.com/Bartolo1024/) and [others](https://github.com/stared/livelossplot/graphs/contributors). **Open for collaboration!** (Some tasks are as simple as writing code docstrings, so - no excuses! :))
 
@@ -26,23 +26,54 @@ model.fit(X_train, Y_train,
           verbose=0)
 ```
 
-![Animated fig for livelossplot tracking log-loss and accuracy](https://raw.githubusercontent.com/stared/livelossplot/master/livelossplot.gif)
+![Animated fig for livelossplot tracking log-loss and accuracy](https://raw.githubusercontent.com/stared/livelossplot/main/livelossplot.gif)
 
 - (The most FA)Q: Why not TensorBoard?
 - A: Jupyter Notebook compatibility (for exploration and teaching). The simplicity of use.
 
 ## Installation
 
-To install [this version from PyPI](https://pypi.org/project/livelossplot/), type:
+### Add to your project
 
 ```bash
+uv add livelossplot
+# or
 pip install livelossplot
 ```
 
-To get the newest one from this repo (note that we are in the alpha stage, so there may be frequent updates), type:
+[On PyPI](https://pypi.org/project/livelossplot/). Requires Python 3.10+.
+
+### Run without installing
+
+For a one-off script, [uv](https://docs.astral.sh/uv/) can run it in an ephemeral environment:
+
+```bash
+uv run --with livelossplot script.py
+```
+
+Or pin the dependencies inside the script itself, then `uv run script.py`:
+
+```python
+# /// script
+# requires-python = ">=3.10"
+# dependencies = ["livelossplot", "torch"]
+# ///
+from livelossplot import PlotLosses
+# ...
+```
+
+For Jupyter:
+
+```bash
+uv run --with livelossplot --with jupyterlab jupyter lab
+```
+
+### From the latest main
 
 ```bash
-pip install git+git://github.com/stared/livelossplot.git
+uv add "livelossplot @ git+https://github.com/stared/livelossplot.git"
+# or
+pip install git+https://github.com/stared/livelossplot.git
 ```
 
 ## Examples