CivArchive
    Novak Djokovic (Tennis Player) - Embedding 100k steps - 40k steps
    Preview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefined

    About: Novak Djokovic is a Serbian professional tennis player. He is currently ranked world No. 1 in men's singles by the Association of Tennis Professionals.

    How to use: Put the file into "embeddings" folder and prompt "novakdjokovic" (without quotes) in positive prompt. You can adjust it as (novakdjokovic:0.9) or (novakdjokovic:1.1) to decrease or increase likeness.

    About the file: The embedding has been trained to 40k steps and after that, all the way to 100k steps with 63 images of Novak Djokovic found on the internet.

    Releasing the results as two versions. Training data and all other embedding versions are included in the training data file.

    Connect with us: Follow our company profile here on Civitai for more quality tools coming soon or visit our website to learn more about us.

    License: No commercial use (real person), for educational purposes only.

    Description

    40k steps version, trained in 1st go with settings bellow:

    "num_of_dataset_images": 63,

        "num_vectors_per_token": 8,

        "embedding_name": "novakdjokovic",

        "learn_rate": "0.05:100, 0.02:200, 0.01:1000, 0.005:2000, 0.002:5000, 0.001:2000, 0.0005",

        "batch_size": 2,

        "gradient_step": 2,

        "training_width": 512,

        "training_height": 512,

        "steps": 40000,

        "clip_grad_mode": "disabled",

        "clip_grad_value": "0.1",

        "latent_sampling_method": "deterministic",

        "create_image_every": 500,

        "save_embedding_every": 500,

    TextualInversion
    SD 1.5

    Details

    Downloads
    299
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    3/4/2023
    Updated
    8/9/2025
    Deleted
    8/7/2025
    Trigger Words:
    novakdjokovic

    Files

    novakdjokovic.pt

    Mirrors

    CivitAI (1 mirrors)