INSTALL.md

Installation

Requirements

• Linux
• Python 3.0 +
• GCC ( G++ ) 4.8.5 or higher
• Miniconda Linux 64-bit

We have tested the following version of OS and software:

• OS: CentOS Linux release 7.5.1804 (Core)
• GCC ( G++ ): 4.8.5
• AWS EC2 Instance: 1.5.0

Install Samples(DigitRec) on AWS

a. Install HeteroCL from my repo using the aws branch.

git clone -b aws https://github.com/ybai62868/heterocl.git

b. Clone the aws-fpga tool repository.

git clone https://github.com/aws/aws-fpga.git 

c. Install SDAccel development environment

source envir.sh

Note: You should put the envir.sh outside the aws-fpga folder and envir.sh can be found in docs folder.

d. Clone the DigitRec Lab repository.

git clone https://github.com/ybai62868/lab_digitrec_aws.git

Note: The default location for this repo is /home/centos/src/project_data/lab_digitrec_aws in AWS. If your folder structure is different, you may need to change the correpsonding paths in some files. But I recommend you use this structure.

lab_digitrec_aws
├── harness
├── solution
│   ├── digitrec
│   │   ├── digitrec_main.py
│   │   ├── digitrec_data.py
│   │   ├── digitrec_aws.py
│   │   ├── data
│   │   |   ├── testing_set.dat
│   │   |   ├── training_set_0.dat
│   │   |   ├── training_set_1.dat
│   │   |   ├── training_set_2.dat
│   │   |   ├── training_set_3.dat
│   │   |   ├── training_set_4.dat
│   │   |   ├── training_set_5.dat
│   │   |   ├── training_set_6.dat
│   │   |   ├── training_set_7.dat
│   │   |   ├── training_set_8.dat
│   │   |   ├── training_set_9.dat
│   ├── src
│   │   ├── host
│   │   │   ├── utils.cpp
│   │   │   ├── utils.h
│   │   ├── kernel
│   ├── Makefile
│   ├── run_hw.sh
│   ├── run_sw.sh
│   ├── README.md