I was trying to use imageio for video frames processing. I think in kaggle offline docker image imageio is available, however, imageio-ffmpeg is not available in kaggle offline kernel? Could you please add this plugin. This is very crucial for video processing. Or does anyone know any libraries for video frames processing which is available in kaggle offline kernel execution?\nCan we include any libraries in the offline kaggle kernel execution?
", "messages": [ { "id": "698253", "postDate": "12/19/2019 01:38:06", "content": "I was trying to use imageio for video frames processing. I think in kaggle offline docker image imageio is available, however, imageio-ffmpeg is not available in kaggle offline kernel? Could you please add this plugin. This is very crucial for video processing. Or does anyone know any libraries for video frames processing which is available in kaggle offline kernel execution?\nCan we include any libraries in the offline kaggle kernel execution?
", "rawMarkdown": "I was trying to use imageio for video frames processing. I think in kaggle offline docker image imageio is available, however, imageio-ffmpeg is not available in kaggle offline kernel? Could you please add this plugin. This is very crucial for video processing. Or does anyone know any libraries for video frames processing which is available in kaggle offline kernel execution?\nCan we include any libraries in the offline kaggle kernel execution?", "votes": null }, { "id": "699653", "postDate": "12/20/2019 18:01:25", "content": "I am no good at github so not sure of the process but requests for additional libraries probably should be made with a pull request or a request in right discussion topic.
\n\nGo to the support tab to find this info.
\n\n\n\n\nThank you for the suggestion! We're always interested to hear about packages and modules that would make great additions to our Notebooks environment. You can request new packages by submitting a pull request:
\n
Python - https://github.com/Kaggle/docker-python\nR - https://github.com/Kaggle/docker-rstats
\n\nOr, you can post your request on our Product Feedback forum: https://www.kaggle.com/product-feedback>
", "rawMarkdown": "I am no good at github so not sure of the process but requests for additional libraries probably should be made with a pull request or a request in right discussion topic.\n\nGo to the support tab to find this info.\n\n> Thank you for the suggestion! We're always interested to hear about packages and modules that would make great additions to our Notebooks environment. You can request new packages by submitting a pull request:\n\nPython - https://github.com/Kaggle/docker-python\nR - https://github.com/Kaggle/docker-rstats\n\nOr, you can post your request on our Product Feedback forum: https://www.kaggle.com/product-feedback>", "votes": null }, { "id": "699754", "postDate": "12/20/2019 21:19:30", "content": "This kernel shows how to use imageio-ffmpeg in your notebooks.
", "rawMarkdown": "[This kernel](https://www.kaggle.com/minhtam/fake-detect-basic) shows how to use imageio-ffmpeg in your notebooks.", "votes": null }, { "id": "699769", "postDate": "12/20/2019 21:54:37", "content": "Thanks a lot, I was searching for this 😀
", "rawMarkdown": "Thanks a lot, I was searching for this 😀", "votes": null }, { "id": "746712", "postDate": "02/15/2020 12:51:30", "content": "Hi,Human Analog. I recently met a problem while using ffmpeg to extract the wav file from mp4. Do I have to write the code like\n command = f\"ffmpeg-git-20200114-amd64-static/ffmpeg -i {file} -ab 192000 -ac 2 -ar 44100 -vn {file[-14:-4]}.{output_format} \". \nI mean I didnt write the whole path like \"../working/ffmpegxxxxxx\", and I got error in the end. Will it be diiferent if I write \"../working/ffmpegxxxxxx\" and change the output path like'../working/xxxx.wav'. My code ran well in public validation set, but got error in public test set after I submitted. Can you show me the correct way to use ffmpeg?
I am no good at github so not sure of the process but requests for additional libraries probably should be made with a pull request or a request in right discussion topic.
\n\nGo to the support tab to find this info.
\n\n\n\n\nThank you for the suggestion! We're always interested to hear about packages and modules that would make great additions to our Notebooks environment. You can request new packages by submitting a pull request:
\n
Python - https://github.com/Kaggle/docker-python\nR - https://github.com/Kaggle/docker-rstats
\n\nOr, you can post your request on our Product Feedback forum: https://www.kaggle.com/product-feedback>
", "votes": null, "replies": [] }, { "id": 699754, "author_name": "humananalog", "author_url": "", "post_date": "12/20/2019 21:19:30", "content": "This kernel shows how to use imageio-ffmpeg in your notebooks.
", "votes": null, "replies": [ { "id": 699769, "author_name": "beingmiakashs", "author_url": "", "post_date": "12/20/2019 21:54:37", "content": "Thanks a lot, I was searching for this 😀
", "votes": null, "replies": [] }, { "id": 746712, "author_name": "huangliuyu", "author_url": "", "post_date": "02/15/2020 12:51:30", "content": "Hi,Human Analog. I recently met a problem while using ffmpeg to extract the wav file from mp4. Do I have to write the code like\n command = f\"ffmpeg-git-20200114-amd64-static/ffmpeg -i {file} -ab 192000 -ac 2 -ar 44100 -vn {file[-14:-4]}.{output_format} \". \nI mean I didnt write the whole path like \"../working/ffmpegxxxxxx\", and I got error in the end. Will it be diiferent if I write \"../working/ffmpegxxxxxx\" and change the output path like'../working/xxxx.wav'. My code ran well in public validation set, but got error in public test set after I submitted. Can you show me the correct way to use ffmpeg?