the words are in json file . i want to use those words for evaluation of my model . but i dont know how to deal with the json file . where can i get the converted jpg bangla handwritten words?
\nor can anyone help me to read those images for my testing data ?
https://data.mendeley.com/datasets/r43wkvdk4w/1
", "messages": [ { "id": "1139464", "postDate": "01/05/2021 12:10:32", "content": "the words are in json file . i want to use those words for evaluation of my model . but i dont know how to deal with the json file . where can i get the converted jpg bangla handwritten words?
\nor can anyone help me to read those images for my testing data ?
https://data.mendeley.com/datasets/r43wkvdk4w/1
", "rawMarkdown": "the words are in json file . i want to use those words for evaluation of my model . but i dont know how to deal with the json file . where can i get the converted jpg bangla handwritten words?\nor can anyone help me to read those images for my testing data ?\n\nhttps://data.mendeley.com/datasets/r43wkvdk4w/1", "votes": null }, { "id": "1139724", "postDate": "01/05/2021 15:25:39", "content": "please download the train_images128.zip file attached here in this kernel : https://www.kaggle.com/mobassir/resized-rgb-images-for-faster-training?select=train_images128.zip
\nor you can directly use this data : https://www.kaggle.com/pestipeti/bengaliai
let me know if it helps or not,thank you
", "rawMarkdown": "please download the train_images128.zip file attached here in this kernel : https://www.kaggle.com/mobassir/resized-rgb-images-for-faster-training?select=train_images128.zip\nor you can directly use this data : https://www.kaggle.com/pestipeti/bengaliai\n\nlet me know if it helps or not,thank you", "votes": null }, { "id": "1140279", "postDate": "01/05/2021 22:44:46", "content": "Interesting dataset!
\nHere's some code that extracts a single image and label. You can build on that to create a dataset of individual words.
\nhttps://www.kaggle.com/reasat/extract-word-image-and-label
Mobassir, thanks for sharing , but the dataset contains character level images , i already use those to create words . now want to evaluate my model on handwritten words . but , cannot find any jpg format word level datset .
", "rawMarkdown": "Mobassir, thanks for sharing , but the dataset contains character level images , i already use those to create words . now want to evaluate my model on handwritten words . but , cannot find any jpg format word level datset .", "votes": null }, { "id": "1140588", "postDate": "01/06/2021 05:50:17", "content": "TahsinReasat, really appreciated . thank ypou so much , i really need it . hope it will work for me
", "rawMarkdown": "TahsinReasat, really appreciated . thank ypou so much , i really need it . hope it will work for me", "votes": null }, { "id": "1144341", "postDate": "01/08/2021 12:05:53", "content": "I saw your code. will you please help me to build a test data pipeline for CNN model?, i am having trouble with the json file . thanks in advance
", "rawMarkdown": "I saw your code. will you please help me to build a test data pipeline for CNN model?, i am having trouble with the json file . thanks in advance", "votes": null }, { "id": "1144640", "postDate": "01/08/2021 15:32:29", "content": "Sure, ask questions in the notebook or the discussion section of the dataset. https://www.kaggle.com/reasat/banglawriting
", "rawMarkdown": "Sure, ask questions in the notebook or the discussion section of the dataset. https://www.kaggle.com/reasat/banglawriting", "votes": null }, { "id": "1144659", "postDate": "01/08/2021 15:44:54", "content": "okay , thank you so much
", "rawMarkdown": "okay , thank you so much", "votes": null } ], "comments": [ { "id": 1139724, "author_name": "mobassir", "author_url": "", "post_date": "01/05/2021 15:25:39", "content": "please download the train_images128.zip file attached here in this kernel : https://www.kaggle.com/mobassir/resized-rgb-images-for-faster-training?select=train_images128.zip
\nor you can directly use this data : https://www.kaggle.com/pestipeti/bengaliai
let me know if it helps or not,thank you
", "votes": null, "replies": [] }, { "id": 1140279, "author_name": "reasat", "author_url": "", "post_date": "01/05/2021 22:44:46", "content": "Interesting dataset!
\nHere's some code that extracts a single image and label. You can build on that to create a dataset of individual words.
\nhttps://www.kaggle.com/reasat/extract-word-image-and-label
I saw your code. will you please help me to build a test data pipeline for CNN model?, i am having trouble with the json file . thanks in advance
", "votes": null, "replies": [] }, { "id": 1144640, "author_name": "reasat", "author_url": "", "post_date": "01/08/2021 15:32:29", "content": "Sure, ask questions in the notebook or the discussion section of the dataset. https://www.kaggle.com/reasat/banglawriting
", "votes": null, "replies": [] }, { "id": 1144659, "author_name": "shapna", "author_url": "", "post_date": "01/08/2021 15:44:54", "content": "okay , thank you so much
", "votes": null, "replies": [] } ] }, { "id": 1140583, "author_name": "shapna", "author_url": "", "post_date": "01/06/2021 05:48:40", "content": "Mobassir, thanks for sharing , but the dataset contains character level images , i already use those to create words . now want to evaluate my model on handwritten words . but , cannot find any jpg format word level datset .
", "votes": null, "replies": [] }, { "id": 1140588, "author_name": "shapna", "author_url": "", "post_date": "01/06/2021 05:50:17", "content": "TahsinReasat, really appreciated . thank ypou so much , i really need it . hope it will work for me
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1139464": "the words are in json file . i want to use those words for evaluation of my model . but i dont know how to deal with the json file . where can i get the converted jpg bangla handwritten words?\nor can anyone help me to read those images for my testing data ?\n\nhttps://data.mendeley.com/datasets/r43wkvdk4w/1", "1139724": "please download the train_images128.zip file attached here in this kernel : https://www.kaggle.com/mobassir/resized-rgb-images-for-faster-training?select=train_images128.zip\nor you can directly use this data : https://www.kaggle.com/pestipeti/bengaliai\n\nlet me know if it helps or not,thank you", "1140279": "Interesting dataset!\nHere's some code that extracts a single image and label. You can build on that to create a dataset of individual words.\nhttps://www.kaggle.com/reasat/extract-word-image-and-label", "1140583": "Mobassir, thanks for sharing , but the dataset contains character level images , i already use those to create words . now want to evaluate my model on handwritten words . but , cannot find any jpg format word level datset .", "1140588": "TahsinReasat, really appreciated . thank ypou so much , i really need it . hope it will work for me", "1144341": "I saw your code. will you please help me to build a test data pipeline for CNN model?, i am having trouble with the json file . thanks in advance", "1144640": "Sure, ask questions in the notebook or the discussion section of the dataset. https://www.kaggle.com/reasat/banglawriting", "1144659": "okay , thank you so much" }, "source": "meta" }