Hi, I'am trying to open the dataset of Flower classification in google colab using the code (https://www.kaggle.com/guitaricet/super-minimalistic-starter) used in many codes in Notebooks (the code that open the database is almost the same in every Notebook) and i got an error in this line :
\n\nfrom kaggle_datasets import KaggleDatasets
\n\nThe error : \nModuleNotFoundError: No module named 'kaggle_datasets'
\n\nI have seen some tutorials to import datasets in google Colab and I successfully imported the files of the data, but still I have the error above, and I am choosing to run in TPU but the code
\n\nprint('Running on TPU ', tpu.master())
\n\nDosen't work.
\n\nI have seen that there is the same problem here https://www.kaggle.com/c/flower-classification-with-tpus/discussion/137429 but i didn\"t understand the solution proposed because I'm new at this.
", "messages": [ { "id": "801657", "postDate": "04/08/2020 17:10:04", "content": "Hi, I'am trying to open the dataset of Flower classification in google colab using the code (https://www.kaggle.com/guitaricet/super-minimalistic-starter) used in many codes in Notebooks (the code that open the database is almost the same in every Notebook) and i got an error in this line :
\n\nfrom kaggle_datasets import KaggleDatasets
\n\nThe error : \nModuleNotFoundError: No module named 'kaggle_datasets'
\n\nI have seen some tutorials to import datasets in google Colab and I successfully imported the files of the data, but still I have the error above, and I am choosing to run in TPU but the code
\n\nprint('Running on TPU ', tpu.master())
\n\nDosen't work.
\n\nI have seen that there is the same problem here https://www.kaggle.com/c/flower-classification-with-tpus/discussion/137429 but i didn\"t understand the solution proposed because I'm new at this.
", "rawMarkdown": "Hi, I'am trying to open the dataset of Flower classification in google colab using the code (https://www.kaggle.com/guitaricet/super-minimalistic-starter) used in many codes in Notebooks (the code that open the database is almost the same in every Notebook) and i got an error in this line : \n\n\nfrom kaggle_datasets import KaggleDatasets \n\nThe error : \nModuleNotFoundError: No module named 'kaggle_datasets'\n\n I have seen some tutorials to import datasets in google Colab and I successfully imported the files of the data, but still I have the error above, and I am choosing to run in TPU but the code\n\n print('Running on TPU ', tpu.master())\n\nDosen't work.\n\nI have seen that there is the same problem here https://www.kaggle.com/c/flower-classification-with-tpus/discussion/137429 but i didn\"t understand the solution proposed because I'm new at this.", "votes": null }, { "id": "801691", "postDate": "04/08/2020 17:53:41", "content": "You have to move the data to your own GCS bucket.\nActually, if you run KaggleDatasets.get_gcs_path() on Kaggle, you will get the name of the GCS bucket that Kaggle uses for the data. You can use that for a couple of days. But it's a cache and it will eventually be recycled. If you want your Colab to be stable, use your own GCS bucket.
", "rawMarkdown": "You have to move the data to your own GCS bucket.\nActually, if you run KaggleDatasets.get_gcs_path() on Kaggle, you will get the name of the GCS bucket that Kaggle uses for the data. You can use that for a couple of days. But it's a cache and it will eventually be recycled. If you want your Colab to be stable, use your own GCS bucket.", "votes": null }, { "id": "802366", "postDate": "04/09/2020 12:58:07", "content": "Thank you, I just found out that Kaggle has an online compiler, and it's working very well with the TPU
", "rawMarkdown": "Thank you, I just found out that Kaggle has an online compiler, and it's working very well with the TPU", "votes": null } ], "comments": [ { "id": 801691, "author_name": "mgorner", "author_url": "", "post_date": "04/08/2020 17:53:41", "content": "You have to move the data to your own GCS bucket.\nActually, if you run KaggleDatasets.get_gcs_path() on Kaggle, you will get the name of the GCS bucket that Kaggle uses for the data. You can use that for a couple of days. But it's a cache and it will eventually be recycled. If you want your Colab to be stable, use your own GCS bucket.
", "votes": null, "replies": [ { "id": 802366, "author_name": "idoune", "author_url": "", "post_date": "04/09/2020 12:58:07", "content": "Thank you, I just found out that Kaggle has an online compiler, and it's working very well with the TPU
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "801657": "Hi, I'am trying to open the dataset of Flower classification in google colab using the code (https://www.kaggle.com/guitaricet/super-minimalistic-starter) used in many codes in Notebooks (the code that open the database is almost the same in every Notebook) and i got an error in this line : \n\n\nfrom kaggle_datasets import KaggleDatasets \n\nThe error : \nModuleNotFoundError: No module named 'kaggle_datasets'\n\n I have seen some tutorials to import datasets in google Colab and I successfully imported the files of the data, but still I have the error above, and I am choosing to run in TPU but the code\n\n print('Running on TPU ', tpu.master())\n\nDosen't work.\n\nI have seen that there is the same problem here https://www.kaggle.com/c/flower-classification-with-tpus/discussion/137429 but i didn\"t understand the solution proposed because I'm new at this.", "801691": "You have to move the data to your own GCS bucket.\nActually, if you run KaggleDatasets.get_gcs_path() on Kaggle, you will get the name of the GCS bucket that Kaggle uses for the data. You can use that for a couple of days. But it's a cache and it will eventually be recycled. If you want your Colab to be stable, use your own GCS bucket.", "802366": "Thank you, I just found out that Kaggle has an online compiler, and it's working very well with the TPU" }, "source": "meta" }