# Weekly Report

###### tags:`CAID`

## imagenet
* SAF `done`
* variation `跑完一次，或許要再多跑幾次`
* SAF + variation `ongoing`
> TODO
> 整理 & 分析

## 3/15
* 改 journal

## 3/8
* journal 中文版 `100%`
* journal 最終版 `90%`
* 實驗 ImageNet 部分 `20%`
* hotspot 交接
* 跑離校流程

## 3/1
* 整理口委問題
* 交接文件

## 2/15
* prepare oral PPT

## 2/8
* journal 轉碩論 & latex 格式

## 2/1
* thesis(方法)

## 1/18
* thesis(實驗)

## 1/11
* thesis(翻譯)

## 12/28
* thesis(experiment)

## 12/21
* thesis(experiment)

## 12/14
* thesis

## 11/30
* thesis

## 11/16
* prepare presentation of group meeting

## 11/9
* thesis

## 10/5
* thesis (experiment)

## 9/27
* thesis

## 9/21
* thesis

## 6/27
* survey paper for the control group

## 6/20
* experiment for control group
* Fault-free: A Fault-resilient Deep Neural Network Accelerator based on Realistic ReRAM Devices
* ITT-RNA: Imperfection Tolerable Training for RRAM-Crossbar-Based Deep Neural-Network Accelerator

## 6/13
* SOC meeting
* paper survey for control group

## 6/6
* prepare SOC monthly meeting
* prepare slide for paper presentation
* research
* analyzation for different finetune methods
* experiment for proposed method(both SAF & variation)

## 5/30
* experiment for proposed method
* Low-Cost and Effective Fault-Tolerance Enhancement Techniques for Emerging Memories-Based Deep Neural Networks `DAC'21`

## 5/23
* experiment for proposed method
* Handling stuck-at-faults in memristor crossbar arrays using matrix transformations
* 科技部摘要 (100%)

## 5/16
* extension method for variation
* experiment
* paper survey for control group
* prepare CIM mid-monthly meeting
* 科技部報告 abstract (40%)

## 5/9
* experiment for control group(variation)
* Unary Coding and Variation-Aware Optimal Mapping Scheme for Reliable ReRAM-based Neuromorphic Computing
* CIM monthly meeting

## 5/2
* prepare CIM meeting
* psuedo code for my method

## 4/25
* paper reading
* Bit-Aware Fault-Tolerant Hybrid Retraining and Remapping Schemes or RRAM-Based Computing-in-Memory Systems `ITCAS'22`
* prepare slide for group meeting
* experiment for SAF

## 4/11
* experiment
* prepare CIM meeting
* 科技部報告 `85%`

## 3/28
- method `40%`
- 科技部報告 `10%`

## 3/21
- survey paper for control group
- organize my method
- organize code for RRAM simulator & HotSpot

## 3/14
- paper slide for group meeting
- extension of my method
- paper survey associate to RRAM variation

## 3/7
- proposed method(advanved)
- SOC meeting
- ITT-RNA Imperfection Tolerable Training for RRAM-Crossbar-Based Deep Neural-Network Accelerator `TCAD'21`

## 2/21
* experiment (method for large variation)
* Evaluating the Impact of Process Variation on RRAMs `LATS'21`

## 2/14
* prepare SOC meeting
* analyze experiment result for large variation

## 2/7
* paper survey for control group
* modified method for large variation

## 1/24
* prepare `1/27` SOC meeting
* research
* experiment for proposed method
* run more benchmark
* VLSI
* final project demo

## 1/10
* run experiment
* VLSI project
* prepare paper presentaion for group meeting

## 1/3
* VLSI hw4 `40%`
* research
* proposed method
* run experiment
* paper variation hypothesis survey

> 2022

---

> 2021

## 12/27
* VLSI hw4 `30%`
* research
* paper survey
* flow of proposed method

## 12/20
* VLSI hw4 `10%`
* research
* study RRAM variation
* analysis variation distribution on crossbar

## 12/13
* VLSI hw4 `5%`
* PattPIM: A Practical ReRAM-Based DNN Accelerator by Reusing Weight Pattern Repetitions `DAC'20`
* proposed method `ongoing`

## 12/6
* VLSI hw3
* modified Simulator
* prepare side for group meeting
* Tolerating Stuck-at Fault and Variation in Resistive Edge Inference Engine via Weight Mapping `GLSVLSI'21`
* ReRAM-Sharing: Fine-Grained Weight Sharing for ReRAM-Based Deep Neural Network Accelerator `ISCAS'21`

## 11/29
* VLSI hw3 `ongoing`
* TIME : A Training-in-memory Architecture for Memristor-based Deep Neural Networks `DAC'17`
* Tolerating Stuck-at Fault and Variation in Resistive Edge Inference Engine via Weight Mapping `GLSVLSI'21`

## 11/22
* VLSI hw2, hw3`ongoing`
* R-Accelerator: A Reconfigurable Accelerator with RRAM Based Logic Contraction and Resource Optimization for Application Specific Computing

## 11/15
* VLSI hw2
* modified simulator

## 11/8
* VLSI hw2 `40%`
* TIMELY: Pushing Data Movements and Interfaces in PIM Accelerators Towards Local and in Time Domain `2020 ISCA`

## 11/1
* VLSI hw2 `25%`
* work on RRAM simulator

## 10/25
* VLSI hw2 `ongoing`
* prepare slide of group meeting
* organized proposed work
* `ISAAC`, `TIMELY`, `PRIME` PIM arch study

## 10/18
* VLSI hw1 `DONE`
* VLSI hw2 (AXI protocol study) `5%`
* Fault Tolerance in Neuromorphic Computing Systems `2019 ASPDAC`
* Tolerating Stuck-at Fault and Variation in Resistive Edge Inference Engine via Weight Mapping `2021 GLSVLSI`
* RRAM simulator tool

## 10/4
* VLSI hw1 `90%`
* ITT-RNA: Imperfection Tolerable Training for RRAM-Crossbar-Based Deep Neural-Network Accelerator `TCAD(2020)`
* RRAM-Based Neuromorphic Hardware Reliability Improvement by Self-Healing and Error Correction `ITC-ASIA(2018)`

## 9/27
* VLSI hw1`2021` `65%`
* research
* Re2PIM: A Reconfigurable ReRAM-Based PIM Design for Variable-Sized Vector-Matrix Multiplication
* Compute-in-RRAM with Limited On-chip Resources
* [Intern work](https://hackmd.io/@hsieh22/2021intern)

## 9/13
* VLSI hw1 `50%`
* research
* CSRO-Based Reconfigurable True Random Number Generator Using RRAM `100%`
* Reconfigurable In-Memory Computing with Resistive Memory Crossbar `100%`
* RRAM-based Reconfigurable In-Memory Computing Architecture with Hybrid Routing `100%`

## 9/6
* Intern

## 6/28
* Course work
* DIC hw5`40%`
* NASA lab16`100%`
* Final exam
* CO
* DIC
* NA/SA
* Research
* NVSim code review(20%)
* study memory circuit design

## 6/21
* Course work
* prepare final
* NA/SA
* DIC
* CO
* NA/SA lab11
* research
* NVsim
* PRIME code review
## 6/7
* Course work
* DIC hw4 (100%)
* NA/SA lab14(90%)
* NA/SA lab11(5%)
* Kneron AI dongle (100%)
* Research
* reconfigurable RRAM-based accelerator
* RRAM-based RNN, LSTM accelerator

## 5/31
* Course work
* NA/SA lab11 (100%)
* DIC hw4 (90%)
* Research
* PRIME architecture(70%)
* Reconfigurable In-Memory Computing with Resistive Memory Crossbar (60%)

## 5/24
* Course work
* aichip midtern
* reference format modified(100%)
* NA/SA lab12(50%)
* research
* PRIME architecture(60%)
* 科技部報告(100%)
* paper reading
* Reconfigurable In-Memory Computing with Resistive Memory Crossbar (50%)

## 5/17
* Course work
* prepare aichip midtern
* paper format modified
* Prepare presentation for group meeting([link](https://reurl.cc/8yd8KR ))
* Research
* 科技部報告

## 5/10
* Course work
* NA/SA lab7(done)
* AIChip hw1(done)
* prepare midtern(DIC)
* papaer reading
* Fault-Tolerant Training with On-Line Fault Detection for RRAM-Based Neural Computing Systems

## 4/26
* midterm
* NA/SA
* AIChip
* AIChip hw1

## 4/19
* Prepare midterm
* NA/SA
* AIChip
* hotspot

## 4/12
* prepare presentation for group meeting
* Computation-Oriented Fault-Tolerance Schemes for RRAM Computing Systems
* DIC hw1, hw2
* NA/SA lab6

## 3/29
* Course work
* NA/SA lab5
* IC Contest
* paper reading
* NVSim_A Circuit-Level Performance, Energy, and Area Model for Emerging Nonvolatile Memory

## 3/22
* paper reading
* A Thermal-aware Optimization Framework for ReRAM-based Deep Neural Network Acceleration
* prepare presentation for group meeting
* course work
* NA/SA lab4
* ICC 2017

## 3/15
* ICC 2019
* CPU
* course work
* CA/NA lab3

## 3/8
* ICC 2016
* CPU (doing)
* paper reading
* a survey of accelerator architecture for deep neural networks

## 2/22
* paper reading
* Thermal-aware Optimizations of ReRAM-based Neuromorphic Computing Systems

## 2/8
* TSRI course
* Cell-Based IC Physical Design and Verification with Innovus
* e-learning
* CO homework
* hw1: traffic light
* hw2: mips
* hw4: cache
* IC Contest
* 2020
* Paper reading
* HR^3^AM: A Heat Resilient Design for RRAM-based Neuromorphic Computing

## 1/25
* TSRI course
* Logic synthesis with design compiler

## 1/20
* course work
* Course final project

* TSRI - Design Compiler

## 1/3
* course work
* CA project
* Advanced computer system prokect
* OPCV hw

> 2021

---

> 2020

## 12/28
* course work
* CA hw
* DLIC report
* openCV hw
* prepare presentation for group meeting

## 12/21
* course work
* DLIC hw6
* CVDL hw2-2
* CA quiz review
* paper reading
* Efficient Inference Engine on Compressed Deep Neural Network ( presentation for DLIC )

## 12/14
* course work
* DLIC hw6
* CVDL hw2
* review DLIC midterm
* code tracing

## 12/7
* course work
* DLIC hw5
* CA review
* code tracing
* paper reading
* Efficient Inference Engine on Compressed Deep Neural Network

## 11/30
* course homework
* DLIC hw5
* code tracing
* paper reading
* Throughput-Optimized OpenCL-based FPGA Accelerator

## 11/23
* course homework
* DLIC hw5 (doing)
* CA hw3
* paper reading
* code tracing
* Accelerator-friendly Neural-network Training : Learning Variations and Defects in RRAM Crossbar

## 11/16
* course homework
* DLIC hw4
* Computer Architecture
* course review
* Computer Architecture

## 11/9
* course homework
* Computer System
* CVDL hw1-5
* paper reading
* Accelerator-friendly Neural-network Training : Learning Variations and Defects in RRAM Crossbar

## 11/2
* homework
* DLIC hw3
* CVDL hw1
* ML video

## 10/26
* homework
* Computer architecture hw2
* Computer system
* machine learning
* ~ lecture 10

## 10/19
* homework
- DLIC hw2 (done)
- Computer System
- Computer Vision hw1

## 10/12
* prepare paper for group meeting
* DeepFace : Closing the Gap to Human-Level Performance in Face Verification
* do homework
* Computer System hw2
* DLIC hw2

## 10/5
* read paper
- Rescuing Memristor-based Computing with Non-linear Resistance Levels
* do homework
- Computer Architecture hw1
- Computer System hw2
- DLIC hw1 (done)

## 9/28
* machine learning(~ gradient descent)
* do homework
- Computer Architecture hw1
- DLIC hw1
- Computer System

## 9/21
* maching learning(gradient descent)
* doing homework

## 9/14
* views 承達's thesis
* course review

## 9/7
* machine learning video watching
* views 承達's thesis

## 8/24
* DIC video watching(control unit)
* DIC hw3 (done)
* EDA unit5
* machine learning video watching(regression)

## 8/17
* DIC video watching(~0519)
* EDA Unit1 , Unit3

## 8/10
* DIC video watching(~0428)
* DIC hw3（60%)
* problem 3,6

## 8/3
* DIC video watching , hw2
* ICPC programming , problem 2.5.7.8

## 7/27
* DIC video watching
* ICPC programming

## 7/20
* DIC video
* ACM/ICPC programming