Angainor Development
Angainor Development
No LVGL touchscreen demo does work with MaixPy Amigo. For once, the i2c pins are not right https://github.com/sipeed/MaixPy_scripts/blob/6ac4f4e97d596d2132b0be0b1fc6bf4139b2fda7/application/lvgl/lvgl_button.py#L11 These should be 24 and 27 for Amigo. Then, there is a...
I have a nRF52840 MDK USB Dongle with case, with UF2 bootloader. It enters bootloader (press button + insert, but with fixed green led, not blinking red) bootloader info on...
@miketeachman published an i2s module targeting i2s audio on esp32, see https://github.com/micropython/micropython/pull/4471 and related examples: https://github.com/miketeachman/micropython-esp32-i2s-examples This would make a perfect addition to this nice T-Watch 2020 Micropython fork.
Continue training from a previous checkpoint or LoRA adapter, without the need to merge into full model. In answer to https://github.com/tloen/alpaca-lora/issues/92
Thanks for the "Fire" module discover, I'll give it a try, much more straightforward than "Click" I used so far. https://github.com/tloen/alpaca-lora/blob/88e1130cb552916eab36d0b5e1a19df4685c1e99/finetune.py#L45 I would suggest keeping the previous settings as default:...
Did someone test Packing with ConstantLengthDataset? Just heard of it there https://huggingface.co/blog/stackllama  Could be better suited than --group_by_len
The readme does list several models, with minimal info. Some are localized, but also the default alpaca ones can use different settings than the default. While the default LoRA settings...
Stumbled upon this on an earlier version trying to Quantize a Falcon-7b (no success so far) This model has non 256 multiples of features and was triggering this error. However,...
There is a confusion in the train code between iter_nums (micro batch) and step_count (batch). The meta parameters, intervals or warmup_iter, refers either to the first one (like log_interval, max_iters)...
Hi, Thanks a lot for this clear and fat-free code base! I'm training Falcon-7B with adapters-v2 and an Alpaca-formated dataset of mine. As usual, I'm trying to max out the...