Can't work well on Colab

[yuya-0411]

Hello.

I'm interested in this system. I tried running it on google colaboratory though, I got the error below.

Step3

How to solve this?

Thank you for your attention.

[williamyang1991]

I find a related issue https://github.com/learnables/learn2learn/issues/310#issuecomment-1037515196

this happens when Google drive limits the number of downloads of a file. Rsyncing a local copy works, else wait a few hours for the limit to reset.

And the possible solution https://github.com/learnables/learn2learn/issues/310#issuecomment-1051009734

A temporal solution: 1 Download the saved models directly from the google drive link I provide in the github page with your web browser. 2 Move them to the folder the network look for the saved model in the collab and rename it based on the code. 3 Load the state dict again

[AK391]

@williamyang1991 to avoid google drive issues, would also suggest hosting the models on the huggingface hub

Some of the benefits of sharing your models through the Hub would be:

• versioning, commit history and diffs
• repos provide useful metadata about their tasks, languages, metrics, etc that make them discoverable
• multiple features from TensorBoard visualizations, PapersWithCode integration, and more
• wider reach of your work to the ecosystem

you can learn more here: https://huggingface.co/docs/hub/models-uploading

[williamyang1991]

Hi @AK391 , Yes, Liming (SoraEndless) has just told me about it :) We plan to share our models on the huggingface just like DualStyleGAN. However, we're extremely busy next week. So once we have spare time, we will work on it.

[yuya-0411]

@williamyang1991 @AK391 Thank you for sharing the ways to solve this issue.

I hope that sharing will work well. Before it, I will try what @williamyang1991 suggested first.

thank you

[yuya-0411]

@williamyang1991

Along with your suggestion, I understand the number of files which will be downloaded is the cause for this problem. so, I did the below instead of using my drive.

def get_download_model_command(file_id, file_name):
    """ Get wget download command for downloading the desired model and save to directory ../checkpoint/. """
    current_directory = os.getcwd()
    save_path = MODEL_DIR
    if not os.path.exists(save_path):
        os.makedirs(save_path)
    url = r"""wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id={FILE_ID}' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id={FILE_ID}" -O {SAVE_PATH}/{FILE_NAME} && rm -rf /tmp/cookies.txt""".format(FILE_ID=file_id, FILE_NAME=file_name, SAVE_PATH=save_path)
    return url

MODEL_PATHS = {
    "encoder": {"id": "1NgI4mPkboYvYw3MWcdUaQhkr0OWgs9ej", "name": "encoder.pt"},
    "faceparsing": {"id": "1jY0mTjVB8njDh6e0LP_2UxuRK3MnjoIR", "name": "faceparsing.pth"},
    # "arcane_exstyle": {"id": "1TC67wRJkdmNRZTqYMUEFkrhWRKKZW40c", "name": "exstyle_code.npy"},
    # "caricature_exstyle": {"id": "1xr9sx_WmRYJ4qHGTtdVQCSxSo4HP3-ip", "name": "exstyle_code.npy"},
    # "cartoon_exstyle": {"id": "1BuCeLk3ASZcoHlbfT28qNru4r5f-hErr", "name": "exstyle_code.npy"},
    "pixar_exstyle": {"id": "1yTaKuSrL7I0i0RYEEK5XD6GI-y5iNUbj", "name": "exstyle_code.npy"},
    # "arcane000": {"id": "1pF4fJ8acmawMsjjXo4HXRIOXeZR8jLVh", "name": "generator.pt"},
    # "arcane077": {"id": "16rLTF2oC0ZeurnM6hjrfrc8BxtW8P8Qf", "name": "generator.pt"},
    # "caricature039": {"id": "1C1E4WEoDWzl0nAxR9okKffFmlMOENbeF", "name": "generator.pt"},
    # "caricature068": {"id": "1B1ko1x8fX2aJ4BYCL12AnknVAi3qQc8W", "name": "generator.pt"},
    # "cartoon026": {"id": "1YJYODh_vEyUrL0q02okjcicpJhdYY8An", "name": "generator.pt"},
    # "cartoon299": {"id": "101qMUMfcI2qDxEbfCBt5mOg2aSqdTaIt", "name": "generator.pt"},
    "pixar052": {"id": "16j_l1x0DD0PjwO8YdplAk69sh3-v95rr", "name": "generator.pt"},
    # "cartoon": {"id": "11s0hwhZWTLacMAzZH4OU-o3Qkp54h30J", "name": "generator.pt"},
}

I wanted to try pixar-like transformation, so I changed the menu for downloading a little bit.

I succeeded. it's interesting!!!

thank you

[zast57]

Which lines to change in the colab to say that it will fetch the data in my personal google drive instead of yours?

[williamyang1991]

Which lines to change in the colab to say that it will fetch the data in my personal google drive instead of yours?

@zast57 for example, if the file in your google driver is shared by the link https://drive.google.com/file/d/1YJYODh_vEyUrL0q02okjcicpJhdYY8An/view?usp=sharing (this is cartoon26 model in my drive) Then you just need to change the id of this line of the colab into 1YJYODh_vEyUrL0q02okjcicpJhdYY8An

"cartoon026": {"id": "1YJYODh_vEyUrL0q02okjcicpJhdYY8An", "name": "generator.pt"},

for example, if your link is https://drive.google.com/file/d/ABCDEFG/view?usp=sharing Then the line should be changed into

"cartoon026": {"id": "ABCDEFG", "name": "generator.pt"},

[zast57]

@williamyang1991 Ok :). It s perfect Thank you :)