=========================
*Compatible with NxSDK 0.7.0 - 0.8.0*
**Added**
- Allow ``LIF.min_voltage`` to have effect. The exact minimum voltage on the
chip is highly affected by discritization (since the chip only allows
minimum voltages in powers of two), but this will at least provide something
in the ballpark.
(`169 <https://github.com/nengo/nengo-loihi/pull/169>`__)
- Population spikes can now be used to send information more efficiently
to the chip. Population spikes are necessary for larger models
like those using CIFAR-10 data.
(`161 <https://github.com/nengo/nengo-loihi/pull/161>`__)
**Changed**
- PES learning in Nengo Loihi more closely matches learning in core Nengo.
(`139 <https://github.com/nengo/nengo-loihi/pull/139>`__)
- Learning in the emulator more closely matches learning on hardware.
(`139 <https://github.com/nengo/nengo-loihi/pull/139>`__)
- The neurons used to transmit decoded values on-chip can be configured.
By default, we use ten pairs of heterogeneous neurons per dimension.
(`132 <https://github.com/nengo/nengo-loihi/pull/132>`_)
- Internal classes and functions have been reorganized and refactored.
See the pull request for more details.
(`159 <https://github.com/nengo/nengo-loihi/pull/159>`_)
- Simulator now gives a warning if the user requests a progress bar, instead
of an error. This avoids potential problems in ``nengo_gui`` and elsewhere.
(`187 <https://github.com/nengo/nengo-loihi/pull/187>`_)
- Nengo Loihi now supports NxSDK version 0.8.0.
Versions 0.7.0 and 0.7.5 are still supported.
(`188 <https://github.com/nengo/nengo-loihi/pull/188>`__)
**Fixed**
- We integrate current (U) and voltage (V) more accurately now by accounting
for rounding during the decay process. This integral is used when
discretizing weights and firing thresholds. This change significantly
improves accuracy for many networks, but in particular dynamical systems
like integrators.
(`124 <https://github.com/nengo/nengo-loihi/pull/124>`_,
`114 <https://github.com/nengo/nengo-loihi/issues/114>`_)
- Ensure things in the build and execution happen in a consistent order from
one build/run to the next (by using ``OrderedDict``, which is deterministic,
instead of ``dict``, which is not). This makes debugging easier and seeding
consistent.
(`151 <https://github.com/nengo/nengo-loihi/pull/151>`_)
- Probes that use snips on the chip (when running with ``precompute=False``)
now deal with negative values correctly.
(`169 <https://github.com/nengo/nengo-loihi/pull/124>`_,
`141 <https://github.com/nengo/nengo-loihi/issues/141>`_)
- Filtering for probes on the chip
is guaranteed to use floating-point now (so that the filtered output
is correct, even if the underlying values are integers).
(`169 <https://github.com/nengo/nengo-loihi/pull/124>`_,
`141 <https://github.com/nengo/nengo-loihi/issues/141>`_)
- Neuron (spike) probes can now be filtered with ``synapse`` objects.
(`182 <https://github.com/nengo/nengo-loihi/issues/182>`__,
`183 <https://github.com/nengo/nengo-loihi/pull/180>`__)