Batch generate YouTube playlist thumbnails with custom text using Python

Vijay Bhaskar — Tue, 28 Dec 2021 07:27:58 +0000

vijjusri14 / posterMark

Batch generate posters with custom text

Useful to generate YouTube playlist thumbnails

Folder Structure

assets  
│
└─── fonts
│       Satisfy.ttf
│       ShadowIntoLight.ttf
│
└─── images
        poster.png

Place your TrueTypeFonts in the fonts directory and background images in the images directory

Settings

Update settings.json to customize the output

{
    "OutputPath": "out",
    "ImagesPath": "./assets/images/",
    "BackgroundImage": "poster.png",
    "FontsPath": "./assets/fonts/",
    "Title": {
        "Font": "ShadowsIntoLight.ttf",
        "Size": 300,
        "Color": "(254, 27, 28)",
        "Position": "(100, 50)"
    },
    "Subtitle": {
        "Font": "Satisfy.ttf",
        "Size": 150,
        "Color": "(85, 30, 138)",
        "Position": "(100, 500)"
    }
}

You can also change the background image, title and subtitle fonts

The color is represented by (R, G, B) values and position by (X, Y)

The fonts used here are Satisfy and ShadowsIntoLight from Google fonts

Posters

Update fileList.json to specify the titles and subtitles for the posters

{
  "Sample": {
    "Title": "PosterMark",
    "Subtitle": "Batch generate posters with custom text"
  }
}

Key of the each json object will be the file name of the poster. Here in the above example, the file name will be Sample.png

Usage

Install Python3

Clone this repository

Create a virtual environment(venv) in the cloned repository directory and activate the venv

python3 -m venv .\venv

Install pip packages

pip install -r .\requirements.txt

Update settings.json and fileList.json as required

Run python script

python posterMark.py

An OpenAI Gym docker that can render on Windows

Vijay Bhaskar — Tue, 28 Dec 2021 07:08:01 +0000

vijjusri14 / OpenAI-Gym-Docker

An OpenAI Gym docker that can render on Windows

This image starts from the jupyter/tensorflow-notebook, and has atari_py installed. Therefore, many environments can be played.

This image is made for running OpenAI Gym on Windows. But in general, it works on other OS like Linux, MacOS etc. as well.

Prerequisite

Description

Dockerfile: Dockerfile to build the OpenAI Gym image
example: Some example notebooks for testing
example/env_render.ipynb: Test Gym environments rendering
example/18_reinforcement_learning.ipynb: This is a copy from Chapter 18 in Géron, Aurélien's book: Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow. Source code is here in GitHub.

Build the image

docker build -t openai_gym_docker:v1.0 .

Create and run the instance of the container

Get the Network InterfaceAlias name using the below command

netsh interface ipv4 show interface

Customize the number of CPUs and Memory in the below command according to your system specifications. Also update Network InterfaceAlias from previous command

docker run -itd --name=openai-gym -p 8888:8888 --cpus='3' --memory='6g' --add-host=dockerhost:$(Get-NetIPAddress -AddressFamily IPv4 -InterfaceAlias 'Wi-Fi' | Select-Object -ExpandProperty 'IPAddress';) -e DISPLAY=dockerhost:0.0 openai_gym_docker:v1.0

Get Jupyter token

docker exec -it openai-gym bash -c 'jupyter notebook list'

Once the docker instance starts running, open browser and visit http://localhost:8888 and login with the token from above.

Jupyter will promt you to set new password for future logins.

Create/Import Python Notebooks and start exploring Reinforcement Learning 🤖

Stop the container

docker stop openai-gym

Start the container

docker start openai-gym

DEV Community: Vijay Bhaskar

Batch generate YouTube playlist thumbnails with custom text using Python

vijjusri14 / posterMark

Batch generate posters with custom text

Folder Structure

Settings

Posters

Usage

An OpenAI Gym docker that can render on Windows

vijjusri14 / OpenAI-Gym-Docker

An OpenAI Gym docker that can render on Windows

Prerequisite

Description

Build the image

Create and run the instance of the container

Get Jupyter token

Stop the container

Start the container