Original subtitles

In this video, we'll show you how to install YOLO v seven CPU mode on windows.

CPU mode is intended for computers or laptops equipped with an nvidia GPU.

Why should you use TPU?

The reason for this is the GPUs can accelerate the computational process for deep learning resulting

in maximum performance.

The first step is to create an environment, press windows and type in condom.

Click on End Condo Navigator.

If it's already open click environments.

Click create to start creating environments, then enter the environments name.

In this example, we name it Your life is 7 to 4.

You can v.

In Python, select Python 3.10, then click create.

Wait until the environment is finished.

It will look like this when it is finished.

Then open the anaconda prompt.

Address the Windows key, then type in condom.

Click on any kind of prompt.

After that, activate the environment, use the commands, activate.

Your life is 7 to 4.

You can feel.

Then press enter.

If it looks like this, it means that the environment is active.

Then navigate to the directory where the seven source code will be safe.

In this example, it will be stored in a directory.

Use the command D, then press enter.

The next step is to clone the YOLO V7 repository.

First launched a browser.

Then visit the following repository.

The original repository has been forced into this repository to clone copy the repository link using

click code, then click the following button.

If it has been copied back again to the prompt.

Then execute the command.

Declan.

Then trace the repository link.

Press enter.

Wait until the cloning process is finished.

If it is completed, the results will be saved in all of these seven folder.

Since the name of the owner of his seven folder to all of his seven CPU use commands.

When YOLO.

Five seven.

YOLO five seven CPU.

This is the root directory of the seven source code.

Then navigate to the root folder of the role of seven GPU using command city YOLO.

574 you press enter.

Following that, we install patrols which can run on the GPU.

The first step is to launch a browser and navigate to the following URL.

Continue to scroll down.

On titles build select stable version.

Under your OS select windows in packages, select PIP in language, Select Python at just the compute

platform to the installed version of queued.

If nothing matches, choose the first thing that is lower than the installed version of queued.

In this example queue the 11.6.

Here is the installation command.

Copy the command.

With a block like this.

Then right click and select copy.

We turn to the prompt after that.

Then pace the command to install patrols by pressing control fee.

Press enter.

Wait until the installation is finished.

If it looks like this, the installation is finished.

Next install python requirements using the command chip install.

That's are requirements.

If you don't text, press enter.

Wait until the installation is finished.

To test the installation, we will perform object detection on the image.

However, before we begin the detection, we must first download all of these seven weights.

We launched the browser.

Scroll down after the.

Here are some examples of weights that can be used.

We will download with you all of these seven in this example.

Wait until the download is finished.

When you're finished, go to the downloads folder and open the file.

Next move the words to the root folder of the seven Dpu.

Click the weights file, then press control X.

Place in the root folder of 70 CPU by pressing the control button.

YOLO V seven includes a number of sample images that can be used for detection.

Sample image located in the inference.

Images folder.

In this example, we will detect objects on the image.

Horses duct taped.

We return to the Anaconda prompt to perform the detection, execute the command python, detect the

PI.

That's, that's it's, you know, five, seven.

That's their source.

Inference images sources dot jpg press enter.

Wait for the detection process to finish.

When finished, the detection results will be saved in the folder listed below.

To see it.

We open Windows Explorer.

Then go back to the root for the rule of seven TPU, then navigate to the folder containing the detection

results.

The result of object detection using all of seven is shown below.

In the detections and training section.

We will use CPU mode to maximise the performance of the whole of seven.

See you in the next video.