The DNN+NeuroSim framework was developed by Prof. Shimeng Yu's group (Georgia Institute of Technology). The model is made publicly available on a non-commercial basis. Copyright of the model is maintained by the developers, and the model is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International Public License
This is the released version 2.1 (Aug 8, 2020) for the tool.
🌟 This V2.1 has improved following estimation:
1. Calibrate FinFET technology library: temperature-related features.
2. Calibrate FinFET technology layout features.
3. Include FeFET polarization during weight-update.
🌟 This version has also added new features into inference accuracy estimation:
1. Introduce VSA-bsaed MLSA (in addition to the original CSA-based MLSA)
2. Introduce SAR ADC
👉 👉 👉 In "Param.cpp", to switch ADC mode:
SARADC = false; // false: MLSA // true: sar ADC
currentMode = true; // false: MLSA use VSA // true: MLSA use CSA
For estimation of inference engine, please visit released V1.3 DNN+NeuroSim V1.3
In V2.1, we currently only support Pytorch wrapper, where users are able to define network structures, parameter precisions and hardware non-ideal properties. With the integrated NeuroSim which takes real traces from wrapper, the framework can support hierarchical organization from device level to circuit level, to chip level and to algorithm level, enabling instruction-accurate evaluation on both accuracy and hardware performance of on-chip training accelerator.
The default example is VGG-8 for CIFAR-10 in this framework: 1. Users can modify weight/error/gradient/activation precisions for training, while some adjusts of "beta" value in function "scale_limit" from file "wage_initializer.py" could be necessary, corresponding values have been provided as example. 2. We temporally only provide the option where: weight/error/gradient precision equals to cell-precision, i.e. one-cell-per-synapse scheme for training evaluation.
Due to additional functions (of non-ideal properties) being implemented in the framework, please expect ~12 hours simulation time for whole training process (default network VGG-8 for CIFAR-10, with 256 epochs).
Developers: Xiaochen Peng 👭 Shanshi Huang.
This research is supported by NSF CAREER award, NSF/SRC E2CDA program, and ASCENT, one of the SRC/DARPA JUMP centers.
If you use the tool or adapt the tool in your work or publication, you are required to cite the following reference:
X. Peng, S. Huang, H. Jiang, A. Lu and S. Yu, ※DNN+NeuroSim V2.0: An End-to-End Benchmarking Framework for Compute-in-Memory Accelerators for Training, § IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, doi: 10.1109/TCAD.2020.3043731, 2020.
X. Peng, S. Huang, Y. Luo, X. Sun and S. Yu, ※DNN+NeuroSim: An End-to-End Benchmarking Framework for Compute-in-Memory Accelerators with Versatile Device Technologies, § IEEE International Electron Devices Meeting (IEDM), 2019.
If you have logistic questions or comments on the model, please contact 👨 Prof. Shimeng Yu, and if you have technical questions or comments, please contact 👩 Xiaochen Peng or 👩 Shanshi Huang.
- Manual:
Documents/DNN NeuroSim V2.1 Manual.pdf
- Nonlinearity-to-A table:
Documents/Nonlinearity-NormA.htm
- MATLAB fitting script:
Documents/nonlinear_fit.m
- DNN_NeuroSim wrapped by Pytorch:
Training_pytorch
- NeuroSim under Pytorch Inference:
Training_pytorch/NeuroSIM
- Get the tool from GitHub
git clone https://github.com/neurosim/DNN_NeuroSim_V2.1.git
- Set up hardware parameters in NeuroSim Core and compile the Code
make
-
Set up hardware constraints in Python wrapper (train.py)
-
Run Pytorch wrapper (integrated with NeuroSim)
-
A list of simulation results are expected as below:
- Input activity of every layer for each epoch:
input_activity.csv
- Weight distribution parameters (mean and std) of every layer for each epoch:
weight_dist.csv
- Delta weight distribution parameters (mean and std) of every layer for each epoch:
delta_dist.csv
- Estimation of average loss and accuracy for each epoch:
PythonWrapper_Output.csv
- Estimation of on-chip training system for each epoch:
NeuroSim_Output.csv
- Detailed breakdowns of estimation of on-chip training system for each epoch:
NeuroSim_Results_Each_Epoch/NeuroSim_Breakdown_Epoch_0.csv
NeuroSim_Results_Each_Epoch/NeuroSim_Breakdown_Epoch_1.csv ... ...
NeuroSim_Results_Each_Epoch/NeuroSim_Breakdown_Epoch_256.csv
- For the usage of this tool, please refer to the user manual.
- X. Peng, S. Huang, H. Jiang, A. Lu and S. Yu, ※DNN+NeuroSim V2.0: An End-to-End Benchmarking Framework for Compute-in-Memory Accelerators for On-chip Training, § IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, doi: 10.1109/TCAD.2020.3043731, 2020.
- X. Peng, S. Huang, Y. Luo, X. Sun and S. Yu, ※DNN+NeuroSim: An End-to-End Benchmarking Framework for Compute-in-Memory Accelerators with Versatile Device Technologies, § IEEE International Electron Devices Meeting (IEDM), 2019.
- X. Peng, R. Liu, S. Yu, ※Optimizing weight mapping and data flow for convolutional neural networks on RRAM based processing-in-memory architecture, § IEEE International Symposium on Circuits and Systems (ISCAS), 2019.
- P.-Y. Chen, S. Yu, ※Technological benchmark of analog synaptic devices for neuro-inspired architectures, § IEEE Design & Test, 2019.
- P.-Y. Chen, X. Peng, S. Yu, ※NeuroSim: A circuit-level macro model for benchmarking neuro-inspired architectures in online learning, § IEEE Trans. CAD, 2018.
- X. Sun, S. Yin, X. Peng, R. Liu, J.-S. Seo, S. Yu, ※XNOR-RRAM: A scalable and parallel resistive synaptic architecture for binary neural networks,§ ACM/IEEE Design, Automation & Test in Europe Conference (DATE), 2018.
- P.-Y. Chen, X. Peng, S. Yu, ※NeuroSim+: An integrated device-to-algorithm framework for benchmarking synaptic devices and array architectures, § IEEE International Electron Devices Meeting (IEDM), 2017.
- P.-Y. Chen, S. Yu, ※Partition SRAM and RRAM based synaptic arrays for neuro-inspired computing,§ IEEE International Symposium on Circuits and Systems (ISCAS), 2016.
- P.-Y. Chen, D. Kadetotad, Z. Xu, A. Mohanty, B. Lin, J. Ye, S. Vrudhula, J.-S. Seo, Y. Cao, S. Yu, ※Technology-design co-optimization of resistive cross-point array for accelerating learning algorithms on chip,§ IEEE Design, Automation & Test in Europe (DATE), 2015.
- S. Wu, et al., ※Training and inference with integers in deep neural networks,§ arXiv: 1802.04680, 2018.
- github.com/boluoweifenda/WAGE
- github.com/stevenygd/WAGE.pytorch
- github.com/aaron-xichen/pytorch-playground