Chris Hayduk ChrisHayduk
ChrisHayduk
/ merge_qlora_with_quantized_model.py
Last active
April 10, 2024 00:36
Merging QLoRA weights with quantized model
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| """ | |
| The code below combines approaches published by both @eugene-yh and @jinyongyoo on Github. | |
| Thanks for the contributions guys! | |
| """ | |
| import torch | |
| import peft |
ChrisHayduk
/ CNN Neural Net
Created
May 1, 2016 20:12
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| MultiLayerConfiguration.Builder builder = new NeuralNetConfiguration.Builder() | |
| .seed(123) | |
| .iterations(iterations) | |
| .regularization(true).l2(.0005) //Arbitrary | |
| .learningRate(0.005) | |
| .weightInit(WeightInit.XAVIER) | |
| .gradientNormalization(GradientNormalization.ClipElementWiseAbsoluteValue) | |
| .optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT) | |
| .updater(Updater.NESTEROVS).momentum(0.5) | |
| .list(6) |