Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
188 changes: 188 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -332,8 +332,196 @@ For example, **set CUDA_VISIBLE_DEVICES=X,Y,Z,... & python cloudclassify-gpu.py*

---

# **For Ubuntu on Windows**

## **Preparing an Ubuntu Virtual Environment In a Windows Host OS**
**Step 1: Install Virtualbox**
Virtual box is a tool which allows you to run different operating systems virtually on your host operating system. This is referred to as a virtual machine.

In order to install Virtualbox, follow this [link](https://www.virtualbox.org/wiki/Downloads)

Under "VirtualBox 6.1.28 platform packages" select the windows host download link.

Open VirtualBox

**Step 2: Download Ubuntu ISO file**
For the purposes of this document I will be utilizing ubuntu 20.04, although 18.04 is recommended.

Follow this [link](https://ubuntu.com/#download)

Under the Download button select 20.04 LTS

**Step 3: Set Up Ubuntu virtual machine in virtualbox**
Open virtualbox

Click the "New" button

Provide the name of your new machine (I recommend the name Ubuntu 20.04)

Set Machine folder location(your choice)

Ensure the Type and version fields are correct (Type: Linux ; Version: Ubuntu(64-bit))

Select Next

Select memory size (Anywhere inside the green area of the scale)

Select Next

Select "Create a virtual hard disk now"
Select Create

Select VDI(Virtual Disk Image)

Select Next

Select Dynamically Allocated (Allows the OS to expand disk size if required)

Select Next

Choose File Location

Choose virtual Hard disk size(I recommend around 100Gb)

Click Create

**Step 4: Configure Settings for Ubuntu Virtual Machine**

Select your machine

Select Settings

Under General, click advanced settings. Select Bidirectional for both shared clipboard and drag and drop fields. This will allow you to share files between host machine and ubuntu machine.

Under System, click processor. Choose the number of CPUs you wish to use. (if no preference leave as default)

Under storage Click empty. Under attributes, by optical drive click Disk dropdown and select choose a disk file. Provide location for ISO file downloaded in step 2.

Select OK.

**Step 5: Open your virtual machine**
Select your machine and press start.

Select preferred language and click install ubuntu

Select Keyboard Layout and Language, Click Continue

Under "What apps would you like to start with?" Select normal installation

Under "Other Options" select download updates while installing ubuntu and Install third party software for graphics and Wi-Fi hardware and additional media formats

Click continue

Under Installation Type, Select Erase Disk and install ubuntu

Click Install Now

Click continue

Select your Location

Enter your preferred Name, Username and Password

Click Continue.

Wait for installation to complete.

Click Restart Now.

Log in to the user you created

Follow Setup Instructions provided by the OS.

**Step 6: (Optional) Install Virtual Box Guest Additions**
This software will allow your virtualbox virtual machine to properly maximize on your screen

Open Terminal

Give command:
```
sudo apt install build-essential dkms linux-headers-$(uname -r)
```
Provide your password

Enter y, press enter

Close terminal

Select devices on virtualbox window

Select Insert Guest additions CD Image

Select Run

Provide your password, click authenticate

Restart ubuntu operating system



# **For Ubuntu**


* **Step 1**: Create Conda Environment

Enter This command to create the conda environment:

```
conda create -n cloud-env python=3.8
```

* **Step 2**: Activate the Conda Environment

To activate the conda environment you created, enter this command:

```
conda activate cloud-env
```

Followed by this command to install pip to the conda environment:

```
conda install pip
```

* **Step 3**: Download the Training Data and Required python packages

Download the training data at this [link](https://drive.google.com/file/d/13W_etsy1A4czDTDHrPVliOImAfTlRSze/view) and place it in a folder named cloudpackage on your desktop

[Download the zip](https://github.com/stccenter/CloudClassification/archive/refs/heads/main.zip) containing your script and place the contents of the zip in the cloudpackage folder

[Download the zip](https://drive.google.com/file/d/0B7XkCwpI5KDYNlNUTTlSS21pQmM/edit?resourcekey=0-wjGZdNAUop6WykTtMip30g)containing the bin file and unzip the contents into the cloudpackage folder

To change to the desktop directory, Enter the Command:


```
cd Desktop
```

To change to the directory to your folder, Enter the Command:

```
cd cloudpackage
```

Install the required packages using the command:

```
pip install -r requirements_cpu.txt
```

* **Step 4**: Run the Script

To run the Script enter the command:

```
python cloudclassify-cpu.py
```


Estimated Time: 2hr 48min

## **GPU-based implementation**

**Step 1** The AWS Deep Learning AMI comes with different versions of CUDA. Please switch to the correct CUDA version, **10.2**, by using the following commands:
Expand Down