Closed KalvinThien closed 2 months ago
Hi, do you mean the Seed ID from a generated image?
Hi, do you mean the Seed ID from a generated image?
That's right, using the Seed ID will help create a photo that is then designed in the same way, just changing the prompt to change the scene. It will help a lot in creating photos, and turning photos into videos. Because I think it will come in handy when creating videos in bulk.
I've also designed a small piece of code to create a bunch of Leonardo accounts for free (150 creadits/day). It will help a lot when copying the seed ID and transferring it and the photos that I need. To maintain consistency.
I will implement it in the next update.
I will implement it in the next update.
Nice 👌
I will implement it in the next update.
Thanks for that
Also, I noticed that you are using a random 1 number mechanic user_agent available randomly, it is only suitable for a small number of users. You can use this module to create different user agents, which will be convenient and the code is more compact :
this is the part I use to create videos from photos, I hope it will help you integrate into leonardoWrapper to be able to call to create videos from photos :
def generate_and_download_video(
leonardo_page, prompt, file_suffix, image_locator, save_dir
):
leonardo_page.page.get_by_placeholder("Type a prompt ...").click()
leonardo_page.page.get_by_placeholder("Type a prompt ...").fill(prompt)
leonardo_page.page.get_by_role("button", name="Generate 2").click()
image = leonardo_page.page.locator(image_locator)
image.wait_for(state="visible", timeout=20000)
image.hover()
image.get_by_label("Generate Motion video").click()
leonardo_page.page.locator(
"footer > div > button.chakra-button.css-17emeou"
).click()
video = leonardo_page.page.locator(
"div.css-1uhvlas > div:nth-child(2) > div.css-hoyejk > div > div:nth-child(3) > div > div:nth-child(2) > div:nth-child(1) > div > div > div:nth-child(2)"
)
video.wait_for(state="visible", timeout=600000)
video.hover()
leonardo_page.page.locator(
"div.css-1uhvlas > div:nth-child(2) > div.css-hoyejk > div > div:nth-child(3) > div > div:nth-child(2) > div:nth-child(1)"
).hover()
Path(SAVE_PATH + save_dir + "/").mkdir(parents=True, exist_ok=True)
with leonardo_page.page.expect_download() as download_info:
leonardo_page.page.locator(
"div.css-1uhvlas > div:nth-child(2) > div.css-hoyejk > div > div:nth-child(3) > div > div:nth-child(2) > div:nth-child(1) > div > div > div.css-1lpsa6b > div.css-1sg7dwz> div > button"
).click()
download = download_info.value
download.save_as(os.path.join(save_dir, file_suffix))```
We are working request based and everything is working
And we will add video generation soon
thanks for that
The project is so great, can you add a way to get the Seed ID? I want when creating the first image, the subsequent images will use the seed id to maintain the consistency of the image afterwards. Is this possible? Thank you