![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F479538%2F6b4103de902e0d51c966548a9977fc97%2FIEEE.PNG?generation=1562970749167435&alt=media\")
Post links to your external data sources here.
", "messages": [ { "id": 563161, "postDate": "2019-06-28T01:16:16.817Z", "content": "Post links to your external data sources here.
", "rawMarkdown": "Post links to your external data sources here.", "votes": 29 }, { "id": 585741, "postDate": "2019-07-28T01:08:26.883Z", "content": "Old competition data from these datasets:
\n\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nPretrained EfficientNet models from here:
\n\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch
\n\nCadeine pretrained models:
\n\nhttps://www.kaggle.com/jesucristo/pretrained-models-cadene
\n\nPyTorch pretrain models (from various datasets):
\n\nhttps://www.kaggle.com/pytorch
", "rawMarkdown": "Old competition data from these datasets:\n\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n\nPretrained EfficientNet models from here: \n\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch\n\nCadeine pretrained models:\n\nhttps://www.kaggle.com/jesucristo/pretrained-models-cadene\n\nPyTorch pretrain models (from various datasets):\n\nhttps://www.kaggle.com/pytorch", "votes": 5, "replies": [ { "id": 605140, "postDate": "2019-08-22T05:21:43.763Z", "content": "I tried using the combined https://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images dataset and I actually got worse results.
\n\nCombining the two sets results in quite a large class imbalance. It may improve with some sort of oversampling, still need to test that out though
", "rawMarkdown": "I tried using the combined https://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images dataset and I actually got worse results. \n\nCombining the two sets results in quite a large class imbalance. It may improve with some sort of oversampling, still need to test that out though" } ] }, { "id": 575634, "postDate": "2019-07-15T18:42:30.007Z", "content": "Instagram pretrained weights for resnext:\nhttps://github.com/facebookresearch/WSL-Images
", "rawMarkdown": "Instagram pretrained weights for resnext:\nhttps://github.com/facebookresearch/WSL-Images", "votes": 5, "replies": [ { "id": 575666, "postDate": "2019-07-15T19:34:48.260Z", "content": "Wow, didn't know that. Thank you so much for introducing us to that. :)
", "rawMarkdown": "Wow, didn't know that. Thank you so much for introducing us to that. :)", "votes": 2 } ] }, { "id": 566186, "postDate": "2019-07-01T21:56:22.220Z", "content": "Subject matter data: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/582710/Grading_definitions_for_referrable_disease_2017_new_110117.pdf \nThis is what human screeners are supposed to look for when grading diabetic retinopathy, at least in the UK. The physicians and scientists at Aravind Eye Hospital might be using a different grading system, but I assume that they use essentially the same features to do so.
", "rawMarkdown": "Subject matter data: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/582710/Grading_definitions_for_referrable_disease_2017_new_110117.pdf \nThis is what human screeners are supposed to look for when grading diabetic retinopathy, at least in the UK. The physicians and scientists at Aravind Eye Hospital might be using a different grading system, but I assume that they use essentially the same features to do so.", "votes": 6 }, { "id": 573887, "postDate": "2019-07-12T22:41:18.940Z", "content": "Diabetic Retinopathy: Segmentation and Grading Challenge\nworkshop at IEEE International Symposium on Biomedical Imaging (ISBI-2018)
\n\nIEEE competition home page
\nhttps://idrid.grand-challenge.org/Home/
Direct Link to download data (NOTE: need to subscribe IEEE)\nhttps://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
\n\n\n\n\nB. Disease Grading: it consists of\n 1. Original color fundus images (516 images divided into train set (413 images) and test set (103 images) - JPG Files)\n 2. Groundtruth Labels for Diabetic Retinopathy and Diabetic Macular Edema Severity Grade (Divided into train and test set - CSV File)
\n
Reference: https://res.mdpi.com/data/data-03-00025/article_deploy/data-03-00025.pdf?filename=&attachment=1
\n\n
@maxwell110 Are you using this data to pretrain your model? Is it allowed to use external data?
", "rawMarkdown": "@maxwell110 Are you using this data to pretrain your model? Is it allowed to use external data?\n", "votes": -1 } ] }, { "id": 599343, "postDate": "2019-08-14T21:00:50.050Z", "content": "Some relevant datasets and models:\n* IDRiD dataset https://www.mdpi.com/2306-5729/3/3/25\n* DeepSeeNet https://github.com/ncbi-nlp/DeepSeeNet\n* https://github.com/getsanjeev/retina-features\n* https://github.com/sidharthramesh/Retinet\n* https://github.com/Connor323/Eye-Fundus-Image-Segmentation
\n\nAlso what others have already posted, including old competition data and messidor
", "rawMarkdown": "Some relevant datasets and models:\n* IDRiD dataset https://www.mdpi.com/2306-5729/3/3/25\n* DeepSeeNet https://github.com/ncbi-nlp/DeepSeeNet\n* https://github.com/getsanjeev/retina-features\n* https://github.com/sidharthramesh/Retinet\n* https://github.com/Connor323/Eye-Fundus-Image-Segmentation\n\nAlso what others have already posted, including old competition data and messidor\n", "votes": 4, "replies": [ { "id": 599825, "postDate": "2019-08-15T11:34:52.450Z", "content": "How to download the IDRiD dataset ?
", "rawMarkdown": "How to download the IDRiD dataset ?" }, { "id": 600284, "postDate": "2019-08-15T22:44:26.510Z", "content": "https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
", "rawMarkdown": "https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid" }, { "id": 600289, "postDate": "2019-08-15T23:11:30.147Z", "content": "Found some more relevant data:
\n\nMessidor 1 and 2\nhttps://github.com/mikevoets/jama16-retina-replication\nhttp://www.adcis.net/en/third-party/messidor/\nMessidor 2 images can be downloaded from here https://medicine.uiowa.edu/eye/abramoff, I have yet to figure out how these images map to the ratings though. Please share if you figure that out (it's not very useful without that)\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades
", "rawMarkdown": "Found some more relevant data:\n\n**Messidor 1 and 2**\nhttps://github.com/mikevoets/jama16-retina-replication\nhttp://www.adcis.net/en/third-party/messidor/\nMessidor 2 images can be downloaded from here https://medicine.uiowa.edu/eye/abramoff, I have yet to figure out how these images map to the ratings though. Please share if you figure that out (it's not very useful without that)\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades\n\n", "votes": 2 }, { "id": 600312, "postDate": "2019-08-16T00:58:16.463Z", "content": "thanks！\nI've downloaded it.
", "rawMarkdown": "thanks！\nI've downloaded it.", "votes": 1 }, { "id": 612590, "postDate": "2019-08-29T22:11:36.050Z", "content": "Some more pretrained models
\n\nhttps://pytorch.org/hub/facebookresearch_WSL-Images_resnext/\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/mikevoets/jama16-retina-replication
", "rawMarkdown": "Some more pretrained models\n\nhttps://pytorch.org/hub/facebookresearch_WSL-Images_resnext/\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/mikevoets/jama16-retina-replication\n" } ] }, { "id": 563188, "postDate": "2019-06-28T02:24:21.683Z", "content": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data", "votes": 3, "replies": [ { "id": 564775, "postDate": "2019-06-30T01:04:30.137Z", "content": "Hi! Would you, please, explain, how to get these data. I don't see download button??
", "rawMarkdown": "Hi! Would you, please, explain, how to get these data. I don't see download button??" }, { "id": 565388, "postDate": "2019-06-30T22:01:24.307Z", "content": "You can download the individual files I think, otherwise try using the kaggle-api CLI tool found here
", "rawMarkdown": "You can download the individual files I think, otherwise try using the kaggle-api CLI tool found [here](https://github.com/Kaggle/kaggle-api)" }, { "id": 565405, "postDate": "2019-06-30T22:49:11.633Z", "content": "Hi there,
\n\ncan we use this data in a submission Kernel? My Kernels are failing to load, when I try to include this dataset
", "rawMarkdown": "Hi there,\n\ncan we use this data in a submission Kernel? My Kernels are failing to load, when I try to include this dataset" }, { "id": 565414, "postDate": "2019-06-30T23:13:33.123Z", "content": "@kostapal \nThat topic is discussed in this thread.
\nhttps://www.kaggle.com/c/aptos2019-blindness-detection/discussion/97774#latest-565278
Can I use the external data for training and submit the result of that model ? here, i am the first time in kaggle competition.
", "rawMarkdown": "Can I use the external data for training and submit the result of that model ? here, i am the first time in kaggle competition. " }, { "id": 566115, "postDate": "2019-07-01T20:09:47.903Z", "content": "@maxwell110 \nThanks a lot!
", "rawMarkdown": "@maxwell110 \nThanks a lot!" }, { "id": 569462, "postDate": "2019-07-06T17:23:06.767Z", "content": "Click \"Late submission\" and accept rules if you have not. Download options will be activated post this.
", "rawMarkdown": "Click \"Late submission\" and accept rules if you have not. Download options will be activated post this." }, { "id": 575218, "postDate": "2019-07-15T07:21:36.983Z", "content": "Have you used the data to train your model? And if yes, was it helpful? Because this is around 80 GB data with around 35,000 labeled images. You can practically train deep model from random initialisation. But is it worth trying?
", "rawMarkdown": "Have you used the data to train your model? And if yes, was it helpful? Because this is around 80 GB data with around 35,000 labeled images. You can practically train deep model from random initialisation. But is it worth trying?" }, { "id": 589461, "postDate": "2019-07-31T23:31:06.143Z", "content": "82GB data, thanks a lot!
", "rawMarkdown": "82GB data, thanks a lot!" } ] }, { "id": 607797, "postDate": "2019-08-25T22:12:59.750Z", "content": "messidor-2 images are no longer avaiable in the links shared here (https://www.medicmind.tech/resources-2).
\n\nThey are available at https://medicine.uiowa.edu/eye/abramoff, but the image names used here and in https://www.kaggle.com/google-brain/messidor2-dr-grades are different.
\n\nIf you guys have both the images and the annotations with the same names, can you pls share it, perhaps as a dataset.
", "rawMarkdown": "messidor-2 images are no longer avaiable in the links shared here (https://www.medicmind.tech/resources-2).\n\nThey are available at https://medicine.uiowa.edu/eye/abramoff, but the image names used here and in https://www.kaggle.com/google-brain/messidor2-dr-grades are different.\n\nIf you guys have both the images and the annotations with the same names, can you pls share it, perhaps as a dataset.\n", "votes": 1, "replies": [ { "id": 608274, "postDate": "2019-08-26T15:21:02.740Z", "content": "according my finding, messidor-1 has serveral zip file, name like Base1-1 to Base3-4.\nMessidor-2 file name is different, start with IMGxxxxx.
\n\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades \nBase on the above label, some files' label are missing. But i'm more doubt on is the Adjudicated label accurate or not.
", "rawMarkdown": "according my finding, messidor-1 has serveral zip file, name like Base1-1 to Base3-4.\nMessidor-2 file name is different, start with IMGxxxxx.\n\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades \nBase on the above label, some files' label are missing. But i'm more doubt on is the Adjudicated label accurate or not." } ] }, { "id": 572047, "postDate": "2019-07-10T11:47:40.267Z", "content": "The Messidor Database
\n\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F479538%2F4bc5b9178f0548fe0c7e039558bf324c%2FMedic-Mind.jpg?generation=1562759141735281&alt=media\")
https://www.medicmind.tech/resources-2/
\n\nI'm not sure this dataset is public or not (need a register to download images)
\n\nGrading\nhttps://github.com/mikevoets/jama16-retina-replication
", "rawMarkdown": "**The Messidor Database**\n\n\n\nhttps://www.medicmind.tech/resources-2/\n\nI'm not sure this dataset is public or not (need a register to download images)\n\n**Grading**\nhttps://github.com/mikevoets/jama16-retina-replication", "votes": 1, "replies": [ { "id": 573166, "postDate": "2019-07-11T21:48:09.337Z", "content": "Hey Maxwell, were you able to access to the Messidor 2 Database by any chance, the link on the MedicMind page seems to be no longer working.
", "rawMarkdown": "Hey Maxwell, were you able to access to the Messidor 2 Database by any chance, the link on the MedicMind page seems to be no longer working. ", "votes": 1 }, { "id": 573202, "postDate": "2019-07-12T00:01:12.080Z", "content": "\n\nI think there are 2 type of Messidor Datasets in the page.
\n\nThe Messidor Database (left side of the figure)\nsize: 1200, available
Messidor 2 Database (right side of the figure)\nsize: 1748, maybe NOT available and NOT public data
You can download 1. The Messidor Databse from the button inside the green.\n( Need to register )
\n\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F479538%2F974e5a35128ffc0d1ff1b2da7785f936%2Fmessidor.PNG?generation=1562889455192813&alt=media\")
Thanks for the detailed explanation @maxwell110 .
", "rawMarkdown": "Thanks for the detailed explanation @maxwell110 . " }, { "id": 579958, "postDate": "2019-07-19T13:00:15.267Z", "content": "https://drive.google.com/drive/folders/1VPCvVsPgrfPNIl932xgU3XC_WFLUsXJR\nhttp://www.it.lut.fi/project/imageret/diaretdb0/\nhttp://www.it.lut.fi/project/imageret/diaretdb1/index.html\nhttp://cecas.clemson.edu/~ahoover/stare/
", "rawMarkdown": "https://drive.google.com/drive/folders/1VPCvVsPgrfPNIl932xgU3XC_WFLUsXJR\nhttp://www.it.lut.fi/project/imageret/diaretdb0/\nhttp://www.it.lut.fi/project/imageret/diaretdb1/index.html\nhttp://cecas.clemson.edu/~ahoover/stare/", "votes": 6 }, { "id": 584434, "postDate": "2019-07-26T00:58:36.593Z", "content": "Hi @bloodaxe Thank you so much for sharing the dataset !\nfor the first link (google drive) any label for those images? I only saw a csv file with rename information.
", "rawMarkdown": "Hi @bloodaxe Thank you so much for sharing the dataset !\nfor the first link (google drive) any label for those images? I only saw a csv file with rename information.", "votes": 1 }, { "id": 594640, "postDate": "2019-08-08T08:34:59.417Z", "content": "First link points to ORIGA dataset. I was not able to find labels for it.
", "rawMarkdown": "First link points to ORIGA dataset. I was not able to find labels for it." } ] }, { "id": 569765, "postDate": "2019-07-07T09:23:06.540Z", "content": "Is it possible to use this data https://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images instead of the dataset provided for the competition?
", "rawMarkdown": "Is it possible to use this data https://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images instead of the dataset provided for the competition?", "votes": 1, "replies": [ { "id": 571972, "postDate": "2019-07-10T09:39:32.683Z", "content": "Have you used the data and improve your score?
", "rawMarkdown": "Have you used the data and improve your score?" }, { "id": 572380, "postDate": "2019-07-10T21:02:48.010Z", "content": "I haven't started working on it yet. I'm just studying the problem and explore the data.
", "rawMarkdown": "I haven't started working on it yet. I'm just studying the problem and explore the data." } ] }, { "id": 565518, "postDate": "2019-07-01T03:50:48.393Z", "content": "Previous competition data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nWhich is taken from \nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "Previous competition data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nWhich is taken from \nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data", "votes": 1 }, { "id": 563887, "postDate": "2019-06-28T19:07:01.053Z", "content": "In addition to previous mentions, may use https://idrid.grand-challenge.org/ which is another yet small dataset of diabetic retinopathy. This one has leisure-level segmentation labeling which is interesting.
", "rawMarkdown": "In addition to previous mentions, may use https://idrid.grand-challenge.org/ which is another yet small dataset of diabetic retinopathy. This one has leisure-level segmentation labeling which is interesting. ", "votes": 1, "replies": [ { "id": 563906, "postDate": "2019-06-28T19:13:32.633Z", "content": "also any of the pretrained (on ImageNet) models in https://keras.io/applications/
", "rawMarkdown": "also any of the pretrained (on ImageNet) models in https://keras.io/applications/ " }, { "id": 565428, "postDate": "2019-06-30T23:53:06.650Z", "content": "how you guys use pretrain model, it seems like we cannot set internet access when submmit
", "rawMarkdown": "how you guys use pretrain model, it seems like we cannot set internet access when submmit", "votes": 1 }, { "id": 566191, "postDate": "2019-07-01T22:02:04.327Z", "content": "Hey ziming, per the Kernel Requirements listed in the Overview tab, \"You can still train a model offline, upload it as a dataset, and use the kernel exclusively to perform inference.\"
", "rawMarkdown": "Hey ziming, per the Kernel Requirements listed in the Overview tab, \"You can still train a model offline, upload it as a dataset, and use the kernel exclusively to perform inference.\" ", "votes": 2 }, { "id": 568537, "postDate": "2019-07-05T04:42:54.163Z", "content": "Does it mean that a transfer learning model can not be used for final submission?
", "rawMarkdown": "Does it mean that a transfer learning model can not be used for final submission?" }, { "id": 573164, "postDate": "2019-07-11T21:47:09.617Z", "content": "Hey Ankit, to the the best of my understanding a transfer learning model can be used for the final submission on the condition that you can get it to predict inside a Kaggle kernel.
", "rawMarkdown": "Hey Ankit, to the the best of my understanding a transfer learning model can be used for the final submission on the condition that you can get it to predict inside a Kaggle kernel. " } ] }, { "id": 563315, "postDate": "2019-06-28T06:12:56.553Z", "content": "Just in case:\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://pytorch.org/docs/stable/torchvision/models.html
", "rawMarkdown": "Just in case:\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://pytorch.org/docs/stable/torchvision/models.html", "votes": 1 }, { "id": 606625, "postDate": "2019-08-23T20:01:40.680Z", "content": "Data:
\n\nPrevious Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMessidor https://www.medicmind.tech/resources-2\nPretrained:
\n\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://www.kaggle.com/mhiro2/pytorch-pretrained-models
", "rawMarkdown": "Data:\n\nPrevious Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMessidor https://www.medicmind.tech/resources-2\nPretrained:\n\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://www.kaggle.com/mhiro2/pytorch-pretrained-models\n", "votes": 2 }, { "id": 605891, "postDate": "2019-08-22T22:54:43.320Z", "content": "Data:\n- Previous Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data\n- IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- Messidor https://www.medicmind.tech/resources-2\n- RIGA https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en (may use)
\n\nPretrained:\n- https://github.com/lukemelas/EfficientNet-PyTorch
", "rawMarkdown": "Data:\n- Previous Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data\n- IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- Messidor https://www.medicmind.tech/resources-2\n- RIGA https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en (may use)\n\nPretrained:\n- https://github.com/lukemelas/EfficientNet-PyTorch", "votes": 2 }, { "id": 601218, "postDate": "2019-08-17T07:59:09.900Z", "content": "Fundus image data without DR grade annotations for unsupervised learning:\n- PALM: https://palm.grand-challenge.org/ (800 images)\n- REFUGE: https://refuge.grand-challenge.org (1200 images)\n- riga: https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en (289 images)
", "rawMarkdown": "Fundus image data without DR grade annotations for unsupervised learning:\n- PALM: https://palm.grand-challenge.org/ (800 images)\n- REFUGE: https://refuge.grand-challenge.org (1200 images)\n- riga: https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en (289 images)\n", "votes": 2 }, { "id": 600189, "postDate": "2019-08-15T19:05:38.720Z", "content": "Datasets:\n- IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- Messidor https://www.medicmind.tech/resources-2\n- previous competition https://www.kaggle.com/c/diabetic-retinopathy-detection
\n\nModels:\n- DenseNet https://arxiv.org/pdf/1608.06993.pdf\n- EfficientNet https://arxiv.org/pdf/1905.11946v1.pdf\n- ResNet https://arxiv.org/pdf/1512.03385.pdf\n- Inception-v4 https://arxiv.org/pdf/1602.07261.pdf\n- custom, ensembles
\n\nFrameworks:\n- PyTorch https://pytorch.org
", "rawMarkdown": "Datasets:\n- IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- Messidor https://www.medicmind.tech/resources-2\n- previous competition https://www.kaggle.com/c/diabetic-retinopathy-detection\n\nModels:\n- DenseNet https://arxiv.org/pdf/1608.06993.pdf\n- EfficientNet https://arxiv.org/pdf/1905.11946v1.pdf\n- ResNet https://arxiv.org/pdf/1512.03385.pdf\n- Inception-v4 https://arxiv.org/pdf/1602.07261.pdf\n- custom, ensembles\n\nFrameworks:\n- PyTorch https://pytorch.org", "votes": 2 }, { "id": 563929, "postDate": "2019-06-28T19:33:25.740Z", "content": "DenseNet weights: https://www.kaggle.com/xhlulu/densenet-keras\nMobileNet weights: https://www.kaggle.com/xhlulu/mobilenet-v2-keras-weights
", "rawMarkdown": "DenseNet weights: https://www.kaggle.com/xhlulu/densenet-keras\nMobileNet weights: https://www.kaggle.com/xhlulu/mobilenet-v2-keras-weights", "votes": 2 }, { "id": 563196, "postDate": "2019-06-28T02:41:25.760Z", "content": "may use https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "may use https://www.kaggle.com/c/diabetic-retinopathy-detection/data", "votes": 2 }, { "id": 588113, "postDate": "2019-07-30T07:14:24.567Z", "content": "prepare for later edits :) (so far I use only what people already mentioned here)
\n\njust in case it's not listed:\nhttps://github.com/rwightman/pytorch-image-models (incl. any weights which he uses)\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/facebookresearch/ResNeXt
", "rawMarkdown": "prepare for later edits :) (so far I use only what people already mentioned here)\n\njust in case it's not listed:\nhttps://github.com/rwightman/pytorch-image-models (incl. any weights which he uses)\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/facebookresearch/ResNeXt\n\n\n", "votes": -1 }, { "id": 582502, "postDate": "2019-07-23T09:02:17.583Z", "content": "could i use these external data to train my model? thanks
", "rawMarkdown": "could i use these external data to train my model? thanks", "votes": -1 }, { "id": 569127, "postDate": "2019-07-06T05:49:10.240Z", "content": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
", "rawMarkdown": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized", "votes": -1 }, { "id": 623494, "postDate": "2019-09-11T02:51:48.800Z", "content": "Data: Previous Competition\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection\nPretrained Model:\nhttps://github.com/lukemelas/EfficientNet-PyTorch
", "rawMarkdown": "Data: Previous Competition\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection\nPretrained Model:\nhttps://github.com/lukemelas/EfficientNet-PyTorch" }, { "id": 621747, "postDate": "2019-09-08T21:50:38.823Z", "content": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nPretrained effientnets: https://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5\nmodel : https://www.kaggle.com/ratan123/install
", "rawMarkdown": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nPretrained effientnets: https://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5\nmodel : https://www.kaggle.com/ratan123/install" }, { "id": 621217, "postDate": "2019-09-08T10:38:20.763Z", "content": "https://www.kaggle.com/yeshila/basemodels\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch\nhttps://www.kaggle.com/chanhu/efficientnet\nhttps://www.kaggle.com/rishabhiitbhu/pretrainedmodels
", "rawMarkdown": "https://www.kaggle.com/yeshila/basemodels\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch\nhttps://www.kaggle.com/chanhu/efficientnet\nhttps://www.kaggle.com/rishabhiitbhu/pretrainedmodels" }, { "id": 621003, "postDate": "2019-09-08T06:24:48.683Z", "content": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images
\n\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch
", "rawMarkdown": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\n\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch" }, { "id": 620701, "postDate": "2019-09-07T21:27:10.033Z", "content": "(I don't know if it's required for everyone to put the external data they use here, or it's just for sharing information. it's my first competition, posting it just in case :D)\nData: \nPrevious Competition:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID:\nhttps://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nPretrained effientnets:\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch
", "rawMarkdown": "(I don't know if it's required for everyone to put the external data they use here, or it's just for sharing information. it's my first competition, posting it just in case :D)\nData: \nPrevious Competition:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID:\nhttps://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nPretrained effientnets:\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch" }, { "id": 620691, "postDate": "2019-09-07T21:12:10.373Z", "content": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMESSIDOR: http://www.adcis.net/en/third-party/messidor/\nPretrained\nEfficientNet-PyTorch: https://github.com/lukemelas/EfficientNet-PyTorch\nCadene models: https://www.kaggle.com/jesucristo/pretrained-models-cadene
", "rawMarkdown": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMESSIDOR: http://www.adcis.net/en/third-party/messidor/\nPretrained\nEfficientNet-PyTorch: https://github.com/lukemelas/EfficientNet-PyTorch\nCadene models: https://www.kaggle.com/jesucristo/pretrained-models-cadene" }, { "id": 620682, "postDate": "2019-09-07T20:39:42.140Z", "content": "Previous Competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "Previous Competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data" }, { "id": 620639, "postDate": "2019-09-07T19:04:09.913Z", "content": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
Pretrained\nhttps://github.com/lukemelas/EfficientNet-PyTorch
1) https://www.kaggle.com/c/diabetic-retinopathy-detection
", "rawMarkdown": "1) https://www.kaggle.com/c/diabetic-retinopathy-detection" }, { "id": 620225, "postDate": "2019-09-07T07:08:03.530Z", "content": "https://www.kaggle.com/wok4711/diabetic-retinopathy-net-en
", "rawMarkdown": "https://www.kaggle.com/wok4711/diabetic-retinopathy-net-en" }, { "id": 619795, "postDate": "2019-09-06T15:22:58.007Z", "content": "Pretrained\n- https://www.kaggle.com/hmendonca/efficientnet-pytorch\n- https://www.kaggle.com/pytorch/resnet50\n- https://www.kaggle.com/pytorch/resnet34\n- https://www.kaggle.com/pytorch/densenet121
\n\nPrevious competition data:\n- https://www.kaggle.com/raghaw/retinopathy-15-1\n- https://www.kaggle.com/raghaw/retinopathy-15-2
", "rawMarkdown": "Pretrained\n- https://www.kaggle.com/hmendonca/efficientnet-pytorch\n- https://www.kaggle.com/pytorch/resnet50\n- https://www.kaggle.com/pytorch/resnet34\n- https://www.kaggle.com/pytorch/densenet121\n\nPrevious competition data:\n- https://www.kaggle.com/raghaw/retinopathy-15-1\n- https://www.kaggle.com/raghaw/retinopathy-15-2" }, { "id": 618824, "postDate": "2019-09-05T14:37:38.953Z", "content": "Data:\nPrevious Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data | https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nPretrained weights:\nhttps://www.kaggle.com/keras/\nhttps://www.kaggle.com/xhlulu/densenet-keras\nhttps://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5
", "rawMarkdown": "Data:\nPrevious Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data | https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n\nPretrained weights:\nhttps://www.kaggle.com/keras/\nhttps://www.kaggle.com/xhlulu/densenet-keras\nhttps://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5\n\n" }, { "id": 618116, "postDate": "2019-09-04T21:11:59.210Z", "content": "Data:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nPretrained models:\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch
", "rawMarkdown": "Data:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n\nPretrained models:\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch" }, { "id": 618040, "postDate": "2019-09-04T18:52:35.540Z", "content": "https://github.com/lukemelas/EfficientNet-PyTorch\n'http://storage.googleapis.com/public-models/efficientnet-b0-08094119.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b1-dbc7070a.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b2-27687264.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b3-c8376fa2.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b4-e116e8b3.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth',\n'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b6-c76e70fd.pth',\n'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b7-dcc49843.pth'
", "rawMarkdown": "https://github.com/lukemelas/EfficientNet-PyTorch\n'http://storage.googleapis.com/public-models/efficientnet-b0-08094119.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b1-dbc7070a.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b2-27687264.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b3-c8376fa2.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b4-e116e8b3.pth',\n'http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth',\n'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b6-c76e70fd.pth',\n'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b7-dcc49843.pth'" }, { "id": 617788, "postDate": "2019-09-04T13:31:13.540Z", "content": "https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/wok4711/diabetic-retinopathy-2015-en\nhttps://www.kaggle.com/wok4711/diabetic-retinopathy-2019-en
", "rawMarkdown": "https://www.kaggle.com/igorkrashenyi/pytorch-model-zoo\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/wok4711/diabetic-retinopathy-2015-en\nhttps://www.kaggle.com/wok4711/diabetic-retinopathy-2019-en" }, { "id": 617713, "postDate": "2019-09-04T12:05:29.193Z", "content": "same for me
", "rawMarkdown": "same for me" }, { "id": 617675, "postDate": "2019-09-04T11:25:12.917Z", "content": "It seems all external data I use are cited here...
", "rawMarkdown": "It seems all external data I use are cited here..." }, { "id": 616401, "postDate": "2019-09-03T05:33:26.327Z", "content": "https://www.kaggle.com/drhabib/starter-kernel-for-0-79\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://yadi.sk/d/-8AWymOPyVZns\nhttps://www.kaggle.com/haowumelbourne/wide-resnet-50-2-bottleneck-linear\nhttps://github.com/szagoruyko/wide-residual-networks
", "rawMarkdown": "https://www.kaggle.com/drhabib/starter-kernel-for-0-79\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://yadi.sk/d/-8AWymOPyVZns\nhttps://www.kaggle.com/haowumelbourne/wide-resnet-50-2-bottleneck-linear\nhttps://github.com/szagoruyko/wide-residual-networks" }, { "id": 616293, "postDate": "2019-09-03T01:06:15.660Z", "content": "https://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/NVIDIA/apex\nhttps://www.kaggle.com/valmmm/aptos2019\nhttps://www.kaggle.com/valmmm/aptos2015
", "rawMarkdown": "https://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/NVIDIA/apex\nhttps://www.kaggle.com/valmmm/aptos2019\nhttps://www.kaggle.com/valmmm/aptos2015" }, { "id": 615099, "postDate": "2019-09-01T13:04:37.733Z", "content": "https://www.kaggle.com/chopinforest1986/efficientnetb0b7-keras-weights
", "rawMarkdown": "https://www.kaggle.com/chopinforest1986/efficientnetb0b7-keras-weights" }, { "id": 614825, "postDate": "2019-09-01T05:52:22.373Z", "content": "DATA:\n- https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nWeights:\n- https://www.kaggle.com/gaborfodor/keras-pretrained-models
", "rawMarkdown": "DATA:\n- https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n\nWeights:\n- https://www.kaggle.com/gaborfodor/keras-pretrained-models" }, { "id": 614694, "postDate": "2019-08-31T22:55:19.947Z", "content": "'efficientnet-b0': 'http://storage.googleapis.com/public-models/efficientnet-b0-08094119.pth',\n 'efficientnet-b1': 'http://storage.googleapis.com/public-models/efficientnet-b1-dbc7070a.pth',\n 'efficientnet-b2': 'http://storage.googleapis.com/public-models/efficientnet-b2-27687264.pth',\n 'efficientnet-b3': 'http://storage.googleapis.com/public-models/efficientnet-b3-c8376fa2.pth',\n 'efficientnet-b4': 'http://storage.googleapis.com/public-models/efficientnet-b4-e116e8b3.pth',\n 'efficientnet-b5': 'http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth',\n 'efficientnet-b6': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b6-c76e70fd.pth',\n 'efficientnet-b7': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b7-dcc49843.pth'\nand pytourch model zoo - pretrained
", "rawMarkdown": "'efficientnet-b0': 'http://storage.googleapis.com/public-models/efficientnet-b0-08094119.pth',\n 'efficientnet-b1': 'http://storage.googleapis.com/public-models/efficientnet-b1-dbc7070a.pth',\n 'efficientnet-b2': 'http://storage.googleapis.com/public-models/efficientnet-b2-27687264.pth',\n 'efficientnet-b3': 'http://storage.googleapis.com/public-models/efficientnet-b3-c8376fa2.pth',\n 'efficientnet-b4': 'http://storage.googleapis.com/public-models/efficientnet-b4-e116e8b3.pth',\n 'efficientnet-b5': 'http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth',\n 'efficientnet-b6': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b6-c76e70fd.pth',\n 'efficientnet-b7': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b7-dcc49843.pth'\nand pytourch model zoo - pretrained" }, { "id": 614693, "postDate": "2019-08-31T22:54:57.260Z", "content": "'efficientnet-b0': 'http://storage.googleapis.com/public-models/efficientnet-b0-08094119.pth',\n 'efficientnet-b1': 'http://storage.googleapis.com/public-models/efficientnet-b1-dbc7070a.pth',\n 'efficientnet-b2': 'http://storage.googleapis.com/public-models/efficientnet-b2-27687264.pth',\n 'efficientnet-b3': 'http://storage.googleapis.com/public-models/efficientnet-b3-c8376fa2.pth',\n 'efficientnet-b4': 'http://storage.googleapis.com/public-models/efficientnet-b4-e116e8b3.pth',\n 'efficientnet-b5': 'http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth',\n 'efficientnet-b6': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b6-c76e70fd.pth',\n 'efficientnet-b7': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b7-dcc49843.pth'
\n\nAnd pytourch model zoo pre-trained.
", "rawMarkdown": "'efficientnet-b0': 'http://storage.googleapis.com/public-models/efficientnet-b0-08094119.pth',\n 'efficientnet-b1': 'http://storage.googleapis.com/public-models/efficientnet-b1-dbc7070a.pth',\n 'efficientnet-b2': 'http://storage.googleapis.com/public-models/efficientnet-b2-27687264.pth',\n 'efficientnet-b3': 'http://storage.googleapis.com/public-models/efficientnet-b3-c8376fa2.pth',\n 'efficientnet-b4': 'http://storage.googleapis.com/public-models/efficientnet-b4-e116e8b3.pth',\n 'efficientnet-b5': 'http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth',\n 'efficientnet-b6': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b6-c76e70fd.pth',\n 'efficientnet-b7': 'http://storage.googleapis.com/public-models/efficientnet/efficientnet-b7-dcc49843.pth'\n\nAnd pytourch model zoo pre-trained." }, { "id": 614600, "postDate": "2019-08-31T18:16:33.237Z", "content": "previous competetion data\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data
network and pretrained \nhttps://github.com/qubvel/efficientnet
Previous competition data: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nEfficientNet: https://www.kaggle.com/chanhu/efficientnet\nCadene models: https://www.kaggle.com/jesucristo/pretrained-models-cadene
", "rawMarkdown": "Previous competition data: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nEfficientNet: https://www.kaggle.com/chanhu/efficientnet\nCadene models: https://www.kaggle.com/jesucristo/pretrained-models-cadene" }, { "id": 612608, "postDate": "2019-08-29T22:38:44.337Z", "content": "Pretrained models:\n https://www.kaggle.com/hmendonca/efficientnet-pytorch\n https://www.kaggle.com/pytorch/\n https://www.kaggle.com/iafoss/pytorch-pretrained-models\n https://github.com/facebookresearch/WSL-Images
\n\nData:\n https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n http://www.adcis.net/en/third-party/messidor/\n https://www.kaggle.com/google-brain/messidor2-dr-grades
\n\nLibs:\n https://www.kaggle.com/taindow/earlystoppingpytorch\n https://www.kaggle.com/taindow/nvidia-apex
", "rawMarkdown": "Pretrained models:\n https://www.kaggle.com/hmendonca/efficientnet-pytorch\n https://www.kaggle.com/pytorch/\n https://www.kaggle.com/iafoss/pytorch-pretrained-models\n https://github.com/facebookresearch/WSL-Images\n \nData:\n https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n http://www.adcis.net/en/third-party/messidor/\n https://www.kaggle.com/google-brain/messidor2-dr-grades\n \nLibs:\n https://www.kaggle.com/taindow/earlystoppingpytorch\n https://www.kaggle.com/taindow/nvidia-apex" }, { "id": 612452, "postDate": "2019-08-29T19:53:32.807Z", "content": "Previous Competition data:\n- https://www.kaggle.com/c/diabetic-retinopathy-detection/data
\n\nPretrained models:\n- https://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5\n- https://www.kaggle.com/keras/inceptionresnetv2\n- https://www.kaggle.com/xhlulu/densenet-keras\n- http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth\n- https://github.com/Cadene/pretrained-models.pytorch
", "rawMarkdown": "Previous Competition data:\n- https://www.kaggle.com/c/diabetic-retinopathy-detection/data\n\nPretrained models:\n- https://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5\n- https://www.kaggle.com/keras/inceptionresnetv2\n- https://www.kaggle.com/xhlulu/densenet-keras\n- http://storage.googleapis.com/public-models/efficientnet-b5-586e6cc6.pth\n- https://github.com/Cadene/pretrained-models.pytorch" }, { "id": 611785, "postDate": "2019-08-29T12:39:19.130Z", "content": "previous competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nwsl images: https://github.com/facebookresearch/WSL-Images\nefficientnet pytorch: https://github.com/lukemelas/EfficientNet-PyTorch\nnvidia apex: https://github.com/NVIDIA/apex
", "rawMarkdown": "previous competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nwsl images: https://github.com/facebookresearch/WSL-Images\nefficientnet pytorch: https://github.com/lukemelas/EfficientNet-PyTorch\nnvidia apex: https://github.com/NVIDIA/apex" }, { "id": 611412, "postDate": "2019-08-29T08:45:08.833Z", "content": "Data from previous competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nImageNet-pretrained models available in torchvision, pretrainedmodels Python package, EfficientNet
", "rawMarkdown": "Data from previous competition: [https://www.kaggle.com/c/diabetic-retinopathy-detection/data](https://www.kaggle.com/c/diabetic-retinopathy-detection/data)\nImageNet-pretrained models available in torchvision, pretrainedmodels Python package, EfficientNet" }, { "id": 610435, "postDate": "2019-08-28T19:46:31.323Z", "content": "I may use the following items:\nData from the old competition: \nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nOld competition data from these datasets:\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nAny of the pretrained models (on ImageNet) in https://keras.io/applications/\nand any of the pretrained weights available on\nhttps://www.kaggle.com/dimitreoliveira/keras-notop
", "rawMarkdown": "I may use the following items:\nData from the old competition: \nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nOld competition data from these datasets:\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n\nAny of the pretrained models (on ImageNet) in https://keras.io/applications/\nand any of the pretrained weights available on\nhttps://www.kaggle.com/dimitreoliveira/keras-notop\n" }, { "id": 610408, "postDate": "2019-08-28T19:15:10.533Z", "content": "Data\nPrev competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMessidor: https://www.medicmind.tech/resources-2
\n\nPretrained:
\n\nTesting a few:\n- DenseNet https://arxiv.org/pdf/1608.06993.pdf\n- EfficientNet https://arxiv.org/pdf/1905.11946v1.pdf\n- ResNet https://arxiv.org/pdf/1512.03385.pdf\n- Inception-v4 https://arxiv.org/pdf/1602.07261.pdf\n- custom, ensembles
", "rawMarkdown": "Data\nPrev competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMessidor: https://www.medicmind.tech/resources-2\n\nPretrained:\n\nTesting a few:\n- DenseNet https://arxiv.org/pdf/1608.06993.pdf\n- EfficientNet https://arxiv.org/pdf/1905.11946v1.pdf\n- ResNet https://arxiv.org/pdf/1512.03385.pdf\n- Inception-v4 https://arxiv.org/pdf/1602.07261.pdf\n- custom, ensembles" }, { "id": 609689, "postDate": "2019-08-28T02:19:02.803Z", "content": "https://odir2019.grand-challenge.org/
", "rawMarkdown": "https://odir2019.grand-challenge.org/" }, { "id": 609138, "postDate": "2019-08-27T12:44:22.150Z", "content": "Previous competition data resized:\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nPretrained efficient-net models(B0-B5):\nhttps://github.com/qubvel/efficientnet
", "rawMarkdown": "Previous competition data resized:\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n\nPretrained efficient-net models(B0-B5):\nhttps://github.com/qubvel/efficientnet" }, { "id": 608717, "postDate": "2019-08-27T05:24:55.567Z", "content": "I may use this:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data - IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid - Messidor https://www.medicmind.tech/resources-2 - RIGA https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en \nPretrained: - https://github.com/lukemelas/EfficientNet-PyTorch
", "rawMarkdown": "I may use this:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data - IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid - Messidor https://www.medicmind.tech/resources-2 - RIGA https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en \nPretrained: - https://github.com/lukemelas/EfficientNet-PyTorch" }, { "id": 608268, "postDate": "2019-08-26T15:16:33.307Z", "content": "Data from old competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "Data from old competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data" }, { "id": 606908, "postDate": "2019-08-24T09:39:44.407Z", "content": "https://github.com/pytorch/vision\nhttps://github.com/rwightman/pytorch-image-models\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "https://github.com/pytorch/vision\nhttps://github.com/rwightman/pytorch-image-models\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data" }, { "id": 604229, "postDate": "2019-08-21T07:11:39.283Z", "content": "How do you access external data once you have added it to a kernel ?\nI added the older competition dataset to this kernel, https://www.kaggle.com/chrisfs/new-data-fast-ai-starter-with-resnet-152\nbut I can't seem to address it. I have tried to load the trainLabels.csv file using\n'../input/aptos2019-blindness-detection/retinopathy_train_2015/trainLabels.csv'\nand\n'../input/aptos2019-blindness-detection/rescaled_train_896/trainLabels.csv'\nand I got a 'file not found' error for all. \nWhat is the correct path name to access that added dataset ?\nThanks for any help in advance,
\n\nChris
", "rawMarkdown": "How do you access external data once you have added it to a kernel ?\nI added the older competition dataset to this kernel, https://www.kaggle.com/chrisfs/new-data-fast-ai-starter-with-resnet-152\nbut I can't seem to address it. I have tried to load the trainLabels.csv file using\n'../input/aptos2019-blindness-detection/retinopathy_train_2015/trainLabels.csv'\nand\n'../input/aptos2019-blindness-detection/rescaled_train_896/trainLabels.csv'\nand I got a 'file not found' error for all. \nWhat is the correct path name to access that added dataset ?\nThanks for any help in advance,\n\nChris ", "replies": [ { "id": 604416, "postDate": "2019-08-21T11:05:48.377Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ] }, { "id": 602284, "postDate": "2019-08-18T21:20:35.263Z", "content": "Data:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
\n\nPretrained models:\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/facebookresearch/FixRes\nhttps://github.com/facebookresearch/WSL-Images
", "rawMarkdown": "Data:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n\nPretrained models:\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/facebookresearch/FixRes\nhttps://github.com/facebookresearch/WSL-Images" }, { "id": 600814, "postDate": "2019-08-16T15:38:19.307Z", "content": "https://www.kaggle.com/hmendonca/efficientnet-pytorch\nhttps://www.kaggle.com/chanhu/efficientnet\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
", "rawMarkdown": "https://www.kaggle.com/hmendonca/efficientnet-pytorch\nhttps://www.kaggle.com/chanhu/efficientnet\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized" }, { "id": 599106, "postDate": "2019-08-14T13:57:27.900Z", "content": "data\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/discussion/16149\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades
\n\npretrained model\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch
", "rawMarkdown": "data\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/discussion/16149\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades\n\n\npretrained model\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch" }, { "id": 598071, "postDate": "2019-08-13T04:30:49.540Z", "content": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data \nhttps://pytorch.org/docs/stable/torchvision/models.html
", "rawMarkdown": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data \nhttps://pytorch.org/docs/stable/torchvision/models.html" }, { "id": 597088, "postDate": "2019-08-11T19:38:19.317Z", "content": "Previous competition data:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nThis competition data + Previous competition data Dataset:\nhttps://www.kaggle.com/yasufuminakama/aptos-train-dataset\nPretrained models:\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/Cadene/pretrained-models.pytorch\n[update(may use)]
\nIDRID:\nhttps://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
\nMessidor:\n https://www.medicmind.tech/resources-2
https://www.kaggle.com/c/diabetic-retinopathy-detection/data \nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://pytorch.org/docs/stable/torchvision/models.html
", "rawMarkdown": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data \nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://pytorch.org/docs/stable/torchvision/models.html" }, { "id": 584323, "postDate": "2019-07-25T18:28:53.643Z", "content": "https://github.com/qubvel/efficientnet\nhttps://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
", "rawMarkdown": "https://github.com/qubvel/efficientnet\nhttps://github.com/tensorflow/tpu/tree/master/models/official/efficientnet" }, { "id": 582417, "postDate": "2019-07-23T06:47:28.167Z", "content": "Previous competition data (resized version)\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
", "rawMarkdown": "Previous competition data (resized version)\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\n", "replies": [ { "id": 596847, "postDate": "2019-08-11T11:58:15.297Z", "content": "Do you think this previous data is useful?
", "rawMarkdown": "Do you think this previous data is useful?", "votes": 2 } ] }, { "id": 581797, "postDate": "2019-07-22T12:13:35.493Z", "content": "https://www.kaggle.com/google-brain/messidor2-dr-grades
", "rawMarkdown": "https://www.kaggle.com/google-brain/messidor2-dr-grades\n" }, { "id": 581002, "postDate": "2019-07-21T08:28:30.500Z", "content": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "rawMarkdown": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data" }, { "id": 578011, "postDate": "2019-07-17T09:26:05.383Z", "content": "Pretrained EfficientNet weights: https://github.com/qubvel/efficientnet
", "rawMarkdown": "Pretrained EfficientNet weights: https://github.com/qubvel/efficientnet", "replies": [ { "id": 578012, "postDate": "2019-07-17T09:28:50.697Z", "content": "Similar resource: https://github.com/titu1994/keras-efficientnets
", "rawMarkdown": "Similar resource: https://github.com/titu1994/keras-efficientnets" } ] }, { "id": 577922, "postDate": "2019-07-17T07:44:26.303Z", "content": "https://www.kaggle.com/c/intel-mobileodt-cervical-cancer-screening/discussion/30134 using this thread for pre trained weights for different models ,like densenet, res50, inceptionv3 , inceptionv4
", "rawMarkdown": "https://www.kaggle.com/c/intel-mobileodt-cervical-cancer-screening/discussion/30134 using this thread for pre trained weights for different models ,like densenet, res50, inceptionv3 , inceptionv4\n" }, { "id": 573854, "postDate": "2019-07-12T20:40:56.047Z", "content": "Also using: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
", "rawMarkdown": "Also using: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized" }, { "id": 572143, "postDate": "2019-07-10T14:29:16.750Z", "content": "https://github.com/creafz/pytorch-cnn-finetune
", "rawMarkdown": "https://github.com/creafz/pytorch-cnn-finetune" }, { "id": 569911, "postDate": "2019-07-07T14:43:31.803Z", "content": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
", "rawMarkdown": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized" }, { "id": 569673, "postDate": "2019-07-07T06:11:57.290Z", "content": "https://github.com/EthanRosenthal/spacecutter\nhttps://github.com/fabianp/mord
", "rawMarkdown": "https://github.com/EthanRosenthal/spacecutter\nhttps://github.com/fabianp/mord" }, { "id": 568778, "postDate": "2019-07-05T12:40:14.393Z", "content": "may use: keras pretrained weights https://www.kaggle.com/gmhost/keras-pretrain-model-weights
", "rawMarkdown": "may use: keras pretrained weights https://www.kaggle.com/gmhost/keras-pretrain-model-weights" }, { "id": 567684, "postDate": "2019-07-03T20:44:06.063Z", "content": "RDN Pre-trained weights: https://github.com/idealo/image-super-resolution
", "rawMarkdown": "RDN Pre-trained weights: https://github.com/idealo/image-super-resolution" }, { "id": 621215, "postDate": "2019-09-08T10:37:15.740Z", "rawMarkdown": "", "isDeleted": true }, { "id": 605720, "postDate": "2019-08-22T16:48:52.077Z", "rawMarkdown": "", "isDeleted": true }, { "id": 604452, "postDate": "2019-08-21T12:01:26.037Z", "content": "https://www.isi.uu.nl/Research/Databases/DRIVE/
", "rawMarkdown": "https://www.isi.uu.nl/Research/Databases/DRIVE/", "isDeleted": true }, { "id": 600291, "postDate": "2019-08-15T23:23:01.263Z", "rawMarkdown": "", "isDeleted": true }, { "id": 590461, "postDate": "2019-08-02T08:09:27.173Z", "content": "Other pytorch models https://www.kaggle.com/iafoss/pytorch-pretrained-models
", "rawMarkdown": "Other pytorch models https://www.kaggle.com/iafoss/pytorch-pretrained-models", "isDeleted": true }, { "id": 588918, "postDate": "2019-07-31T08:17:03.160Z", "rawMarkdown": "", "isDeleted": true }, { "id": 586376, "postDate": "2019-07-29T04:54:42.720Z", "content": "Pretrained resnet18 https://www.kaggle.com/darshak/fastai-vision-pretrained
", "rawMarkdown": "Pretrained resnet18 https://www.kaggle.com/darshak/fastai-vision-pretrained", "isDeleted": true }, { "id": 565588, "postDate": "2019-07-01T06:32:28.853Z", "rawMarkdown": "", "isDeleted": true }, { "id": 564847, "postDate": "2019-06-30T04:36:40.397Z", "rawMarkdown": "", "isDeleted": true }, { "id": 563985, "postDate": "2019-06-28T21:06:49.710Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 585741, "author_name": "Lex Toumbourou", "author_url": "", "post_date": "2019-07-28T01:08:26.883000", "content": "Old competition data from these datasets:
\n\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
\n\nPretrained EfficientNet models from here:
\n\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch
\n\nCadeine pretrained models:
\n\nhttps://www.kaggle.com/jesucristo/pretrained-models-cadene
\n\nPyTorch pretrain models (from various datasets):
\n\nhttps://www.kaggle.com/pytorch
", "votes": 5, "replies": [ { "id": 605140, "author_name": "Antonio De Perio", "author_url": "", "post_date": "2019-08-22T05:21:43.763000", "content": "I tried using the combined https://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images dataset and I actually got worse results.
\n\nCombining the two sets results in quite a large class imbalance. It may improve with some sort of oversampling, still need to test that out though
", "votes": 0, "replies": [] } ] }, { "id": 575634, "author_name": "Tom Aindow", "author_url": "", "post_date": "2019-07-15T18:42:30.007000", "content": "Instagram pretrained weights for resnext:\nhttps://github.com/facebookresearch/WSL-Images
", "votes": 5, "replies": [ { "id": 575666, "author_name": "Rishabh Agrahari", "author_url": "", "post_date": "2019-07-15T19:34:48.260000", "content": "Wow, didn't know that. Thank you so much for introducing us to that. :)
", "votes": 2, "replies": [] } ] }, { "id": 566186, "author_name": "Charles Laurin", "author_url": "", "post_date": "2019-07-01T21:56:22.220000", "content": "Subject matter data: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/582710/Grading_definitions_for_referrable_disease_2017_new_110117.pdf \nThis is what human screeners are supposed to look for when grading diabetic retinopathy, at least in the UK. The physicians and scientists at Aravind Eye Hospital might be using a different grading system, but I assume that they use essentially the same features to do so.
", "votes": 6, "replies": [] }, { "id": 573887, "author_name": "Maxwell", "author_url": "", "post_date": "2019-07-12T22:41:18.940000", "content": "Diabetic Retinopathy: Segmentation and Grading Challenge\nworkshop at IEEE International Symposium on Biomedical Imaging (ISBI-2018)
\n\nIEEE competition home page
\nhttps://idrid.grand-challenge.org/Home/
Direct Link to download data (NOTE: need to subscribe IEEE)\nhttps://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
\n\n\n\n\nB. Disease Grading: it consists of\n 1. Original color fundus images (516 images divided into train set (413 images) and test set (103 images) - JPG Files)\n 2. Groundtruth Labels for Diabetic Retinopathy and Diabetic Macular Edema Severity Grade (Divided into train and test set - CSV File)
\n
Reference: https://res.mdpi.com/data/data-03-00025/article_deploy/data-03-00025.pdf?filename=&attachment=1
\n\n
@maxwell110 Are you using this data to pretrain your model? Is it allowed to use external data?
", "votes": -1, "replies": [] } ] }, { "id": 599343, "author_name": "Guido Zuidhof", "author_url": "", "post_date": "2019-08-14T21:00:50.050000", "content": "Some relevant datasets and models:\n* IDRiD dataset https://www.mdpi.com/2306-5729/3/3/25\n* DeepSeeNet https://github.com/ncbi-nlp/DeepSeeNet\n* https://github.com/getsanjeev/retina-features\n* https://github.com/sidharthramesh/Retinet\n* https://github.com/Connor323/Eye-Fundus-Image-Segmentation
\n\nAlso what others have already posted, including old competition data and messidor
", "votes": 4, "replies": [ { "id": 599825, "author_name": "Yangfan", "author_url": "", "post_date": "2019-08-15T11:34:52.450000", "content": "How to download the IDRiD dataset ?
", "votes": 0, "replies": [] }, { "id": 600284, "author_name": "Guido Zuidhof", "author_url": "", "post_date": "2019-08-15T22:44:26.510000", "content": "https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
", "votes": 0, "replies": [] }, { "id": 600289, "author_name": "Guido Zuidhof", "author_url": "", "post_date": "2019-08-15T23:11:30.147000", "content": "Found some more relevant data:
\n\nMessidor 1 and 2\nhttps://github.com/mikevoets/jama16-retina-replication\nhttp://www.adcis.net/en/third-party/messidor/\nMessidor 2 images can be downloaded from here https://medicine.uiowa.edu/eye/abramoff, I have yet to figure out how these images map to the ratings though. Please share if you figure that out (it's not very useful without that)\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades
", "votes": 2, "replies": [] }, { "id": 600312, "author_name": "Yangfan", "author_url": "", "post_date": "2019-08-16T00:58:16.463000", "content": "thanks！\nI've downloaded it.
", "votes": 1, "replies": [] }, { "id": 612590, "author_name": "Guido Zuidhof", "author_url": "", "post_date": "2019-08-29T22:11:36.050000", "content": "Some more pretrained models
\n\nhttps://pytorch.org/hub/facebookresearch_WSL-Images_resnext/\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://github.com/mikevoets/jama16-retina-replication
", "votes": 0, "replies": [] } ] }, { "id": 563188, "author_name": "Kele", "author_url": "", "post_date": "2019-06-28T02:24:21.683000", "content": "https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "votes": 3, "replies": [ { "id": 564775, "author_name": "Serge", "author_url": "", "post_date": "2019-06-30T01:04:30.137000", "content": "Hi! Would you, please, explain, how to get these data. I don't see download button??
", "votes": 0, "replies": [] }, { "id": 565388, "author_name": "Guido Zuidhof", "author_url": "", "post_date": "2019-06-30T22:01:24.307000", "content": "You can download the individual files I think, otherwise try using the kaggle-api CLI tool found here
", "votes": 0, "replies": [] }, { "id": 565405, "author_name": "Konstantin Palagachev", "author_url": "", "post_date": "2019-06-30T22:49:11.633000", "content": "Hi there,
\n\ncan we use this data in a submission Kernel? My Kernels are failing to load, when I try to include this dataset
", "votes": 0, "replies": [] }, { "id": 565414, "author_name": "Maxwell", "author_url": "", "post_date": "2019-06-30T23:13:33.123000", "content": "@kostapal \nThat topic is discussed in this thread.
\nhttps://www.kaggle.com/c/aptos2019-blindness-detection/discussion/97774#latest-565278
Can I use the external data for training and submit the result of that model ? here, i am the first time in kaggle competition.
", "votes": 0, "replies": [] }, { "id": 566115, "author_name": "Konstantin Palagachev", "author_url": "", "post_date": "2019-07-01T20:09:47.903000", "content": "@maxwell110 \nThanks a lot!
", "votes": 0, "replies": [] }, { "id": 569462, "author_name": "sabarinath", "author_url": "", "post_date": "2019-07-06T17:23:06.767000", "content": "Click \"Late submission\" and accept rules if you have not. Download options will be activated post this.
", "votes": 0, "replies": [] }, { "id": 575218, "author_name": "Amey Patil", "author_url": "", "post_date": "2019-07-15T07:21:36.983000", "content": "Have you used the data to train your model? And if yes, was it helpful? Because this is around 80 GB data with around 35,000 labeled images. You can practically train deep model from random initialisation. But is it worth trying?
", "votes": 0, "replies": [] }, { "id": 589461, "author_name": "elsa zhou", "author_url": "", "post_date": "2019-07-31T23:31:06.143000", "content": "82GB data, thanks a lot!
", "votes": 0, "replies": [] } ] }, { "id": 607797, "author_name": "Ravi Kumar Vadapalli", "author_url": "", "post_date": "2019-08-25T22:12:59.750000", "content": "messidor-2 images are no longer avaiable in the links shared here (https://www.medicmind.tech/resources-2).
\n\nThey are available at https://medicine.uiowa.edu/eye/abramoff, but the image names used here and in https://www.kaggle.com/google-brain/messidor2-dr-grades are different.
\n\nIf you guys have both the images and the annotations with the same names, can you pls share it, perhaps as a dataset.
", "votes": 1, "replies": [ { "id": 608274, "author_name": "CC Joshua ", "author_url": "", "post_date": "2019-08-26T15:21:02.740000", "content": "according my finding, messidor-1 has serveral zip file, name like Base1-1 to Base3-4.\nMessidor-2 file name is different, start with IMGxxxxx.
\n\nhttps://www.kaggle.com/google-brain/messidor2-dr-grades \nBase on the above label, some files' label are missing. But i'm more doubt on is the Adjudicated label accurate or not.
", "votes": 0, "replies": [] } ] }, { "id": 572047, "author_name": "Maxwell", "author_url": "", "post_date": "2019-07-10T11:47:40.267000", "content": "The Messidor Database
\n\n
https://www.medicmind.tech/resources-2/
\n\nI'm not sure this dataset is public or not (need a register to download images)
\n\nGrading\nhttps://github.com/mikevoets/jama16-retina-replication
", "votes": 1, "replies": [ { "id": 573166, "author_name": "Kaan Donbekci", "author_url": "", "post_date": "2019-07-11T21:48:09.337000", "content": "Hey Maxwell, were you able to access to the Messidor 2 Database by any chance, the link on the MedicMind page seems to be no longer working.
", "votes": 1, "replies": [] }, { "id": 573202, "author_name": "Maxwell", "author_url": "", "post_date": "2019-07-12T00:01:12.080000", "content": "\n\nI think there are 2 type of Messidor Datasets in the page.
\n\nThe Messidor Database (left side of the figure)\nsize: 1200, available
Messidor 2 Database (right side of the figure)\nsize: 1748, maybe NOT available and NOT public data
You can download 1. The Messidor Databse from the button inside the green.\n( Need to register )
\n\n
Thanks for the detailed explanation @maxwell110 .
", "votes": 0, "replies": [] }, { "id": 579958, "author_name": "Eugene Khvedchenya", "author_url": "", "post_date": "2019-07-19T13:00:15.267000", "content": "https://drive.google.com/drive/folders/1VPCvVsPgrfPNIl932xgU3XC_WFLUsXJR\nhttp://www.it.lut.fi/project/imageret/diaretdb0/\nhttp://www.it.lut.fi/project/imageret/diaretdb1/index.html\nhttp://cecas.clemson.edu/~ahoover/stare/
", "votes": 6, "replies": [] }, { "id": 584434, "author_name": "CC Joshua ", "author_url": "", "post_date": "2019-07-26T00:58:36.593000", "content": "Hi @bloodaxe Thank you so much for sharing the dataset !\nfor the first link (google drive) any label for those images? I only saw a csv file with rename information.
", "votes": 1, "replies": [] }, { "id": 594640, "author_name": "Eugene Khvedchenya", "author_url": "", "post_date": "2019-08-08T08:34:59.417000", "content": "First link points to ORIGA dataset. I was not able to find labels for it.
", "votes": 0, "replies": [] } ] }, { "id": 569765, "author_name": "Nicola Macchiarulo", "author_url": "", "post_date": "2019-07-07T09:23:06.540000", "content": "Is it possible to use this data https://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images instead of the dataset provided for the competition?
", "votes": 1, "replies": [ { "id": 571972, "author_name": "哈尔的移动城堡", "author_url": "", "post_date": "2019-07-10T09:39:32.683000", "content": "Have you used the data and improve your score?
", "votes": 0, "replies": [] }, { "id": 572380, "author_name": "Nicola Macchiarulo", "author_url": "", "post_date": "2019-07-10T21:02:48.010000", "content": "I haven't started working on it yet. I'm just studying the problem and explore the data.
", "votes": 0, "replies": [] } ] }, { "id": 565518, "author_name": "Prashant Kikani", "author_url": "", "post_date": "2019-07-01T03:50:48.393000", "content": "Previous competition data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nWhich is taken from \nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "votes": 1, "replies": [] }, { "id": 563887, "author_name": "Kaan Donbekci", "author_url": "", "post_date": "2019-06-28T19:07:01.053000", "content": "In addition to previous mentions, may use https://idrid.grand-challenge.org/ which is another yet small dataset of diabetic retinopathy. This one has leisure-level segmentation labeling which is interesting.
", "votes": 1, "replies": [ { "id": 563906, "author_name": "Kaan Donbekci", "author_url": "", "post_date": "2019-06-28T19:13:32.633000", "content": "also any of the pretrained (on ImageNet) models in https://keras.io/applications/
", "votes": 0, "replies": [] }, { "id": 565428, "author_name": "ziming", "author_url": "", "post_date": "2019-06-30T23:53:06.650000", "content": "how you guys use pretrain model, it seems like we cannot set internet access when submmit
", "votes": 1, "replies": [] }, { "id": 566191, "author_name": "Kaan Donbekci", "author_url": "", "post_date": "2019-07-01T22:02:04.327000", "content": "Hey ziming, per the Kernel Requirements listed in the Overview tab, \"You can still train a model offline, upload it as a dataset, and use the kernel exclusively to perform inference.\"
", "votes": 2, "replies": [] }, { "id": 568537, "author_name": "Ankit Bansal", "author_url": "", "post_date": "2019-07-05T04:42:54.163000", "content": "Does it mean that a transfer learning model can not be used for final submission?
", "votes": 0, "replies": [] }, { "id": 573164, "author_name": "Kaan Donbekci", "author_url": "", "post_date": "2019-07-11T21:47:09.617000", "content": "Hey Ankit, to the the best of my understanding a transfer learning model can be used for the final submission on the condition that you can get it to predict inside a Kaggle kernel.
", "votes": 0, "replies": [] } ] }, { "id": 563315, "author_name": "Andrey Lukyanenko", "author_url": "", "post_date": "2019-06-28T06:12:56.553000", "content": "Just in case:\nhttps://github.com/Cadene/pretrained-models.pytorch\nhttps://pytorch.org/docs/stable/torchvision/models.html
", "votes": 1, "replies": [] }, { "id": 606625, "author_name": "Borys Tymchenko", "author_url": "", "post_date": "2019-08-23T20:01:40.680000", "content": "Data:
\n\nPrevious Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data\nIDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMessidor https://www.medicmind.tech/resources-2\nPretrained:
\n\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://www.kaggle.com/mhiro2/pytorch-pretrained-models
", "votes": 2, "replies": [] }, { "id": 605891, "author_name": "oct_path", "author_url": "", "post_date": "2019-08-22T22:54:43.320000", "content": "Data:\n- Previous Competition https://www.kaggle.com/c/diabetic-retinopathy-detection/data\n- IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- Messidor https://www.medicmind.tech/resources-2\n- RIGA https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en (may use)
\n\nPretrained:\n- https://github.com/lukemelas/EfficientNet-PyTorch
", "votes": 2, "replies": [] }, { "id": 601218, "author_name": "JaeminSon", "author_url": "", "post_date": "2019-08-17T07:59:09.900000", "content": "Fundus image data without DR grade annotations for unsupervised learning:\n- PALM: https://palm.grand-challenge.org/ (800 images)\n- REFUGE: https://refuge.grand-challenge.org (1200 images)\n- riga: https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z?locale=en (289 images)
", "votes": 2, "replies": [] }, { "id": 600189, "author_name": "Adrian Brodzik", "author_url": "", "post_date": "2019-08-15T19:05:38.720000", "content": "Datasets:\n- IDRID https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\n- Messidor https://www.medicmind.tech/resources-2\n- previous competition https://www.kaggle.com/c/diabetic-retinopathy-detection
\n\nModels:\n- DenseNet https://arxiv.org/pdf/1608.06993.pdf\n- EfficientNet https://arxiv.org/pdf/1905.11946v1.pdf\n- ResNet https://arxiv.org/pdf/1512.03385.pdf\n- Inception-v4 https://arxiv.org/pdf/1602.07261.pdf\n- custom, ensembles
\n\nFrameworks:\n- PyTorch https://pytorch.org
", "votes": 2, "replies": [] }, { "id": 563929, "author_name": "xhlulu", "author_url": "", "post_date": "2019-06-28T19:33:25.740000", "content": "DenseNet weights: https://www.kaggle.com/xhlulu/densenet-keras\nMobileNet weights: https://www.kaggle.com/xhlulu/mobilenet-v2-keras-weights
", "votes": 2, "replies": [] }, { "id": 563196, "author_name": "Yiheng Wang", "author_url": "", "post_date": "2019-06-28T02:41:25.760000", "content": "may use https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "votes": 2, "replies": [] }, { "id": 588113, "author_name": "steelrose", "author_url": "", "post_date": "2019-07-30T07:14:24.567000", "content": "prepare for later edits :) (so far I use only what people already mentioned here)
\n\njust in case it's not listed:\nhttps://github.com/rwightman/pytorch-image-models (incl. any weights which he uses)\nhttps://github.com/lukemelas/EfficientNet-PyTorch\nhttps://github.com/facebookresearch/ResNeXt
", "votes": -1, "replies": [] }, { "id": 582502, "author_name": "Virtual Machine", "author_url": "", "post_date": "2019-07-23T09:02:17.583000", "content": "could i use these external data to train my model? thanks
", "votes": -1, "replies": [] }, { "id": 569127, "author_name": "jionie", "author_url": "", "post_date": "2019-07-06T05:49:10.240000", "content": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
", "votes": -1, "replies": [] }, { "id": 623494, "author_name": "Quan Qiu", "author_url": "", "post_date": "2019-09-11T02:51:48.800000", "content": "Data: Previous Competition\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection\nPretrained Model:\nhttps://github.com/lukemelas/EfficientNet-PyTorch
", "votes": 0, "replies": [] }, { "id": 621747, "author_name": "Taegwan Kim", "author_url": "", "post_date": "2019-09-08T21:50:38.823000", "content": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nPretrained effientnets: https://www.kaggle.com/ratthachat/efficientnet-keras-weights-b0b5\nmodel : https://www.kaggle.com/ratan123/install
", "votes": 0, "replies": [] }, { "id": 621217, "author_name": "Wang Xinliang", "author_url": "", "post_date": "2019-09-08T10:38:20.763000", "content": "https://www.kaggle.com/yeshila/basemodels\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch\nhttps://www.kaggle.com/chanhu/efficientnet\nhttps://www.kaggle.com/rishabhiitbhu/pretrainedmodels
", "votes": 0, "replies": [] }, { "id": 621003, "author_name": "Sabbir Ahmed", "author_url": "", "post_date": "2019-09-08T06:24:48.683000", "content": "https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nhttps://www.kaggle.com/benjaminwarner/resized-2015-2019-blindness-detection-images
\n\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch
", "votes": 0, "replies": [] }, { "id": 620701, "author_name": "Sirong Huang", "author_url": "", "post_date": "2019-09-07T21:27:10.033000", "content": "(I don't know if it's required for everyone to put the external data they use here, or it's just for sharing information. it's my first competition, posting it just in case :D)\nData: \nPrevious Competition:\nhttps://www.kaggle.com/c/diabetic-retinopathy-detection/data\nhttps://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID:\nhttps://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nPretrained effientnets:\nhttps://www.kaggle.com/hmendonca/efficientnet-pytorch
", "votes": 0, "replies": [] }, { "id": 620691, "author_name": "Tahsin Mostafiz", "author_url": "", "post_date": "2019-09-07T21:12:10.373000", "content": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized\nIDRID: https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid\nMESSIDOR: http://www.adcis.net/en/third-party/messidor/\nPretrained\nEfficientNet-PyTorch: https://github.com/lukemelas/EfficientNet-PyTorch\nCadene models: https://www.kaggle.com/jesucristo/pretrained-models-cadene
", "votes": 0, "replies": [] }, { "id": 620682, "author_name": "Winter is here", "author_url": "", "post_date": "2019-09-07T20:39:42.140000", "content": "Previous Competition: https://www.kaggle.com/c/diabetic-retinopathy-detection/data
", "votes": 0, "replies": [] }, { "id": 620639, "author_name": "Saad Mahmud", "author_url": "", "post_date": "2019-09-07T19:04:09.913000", "content": "Data\nPrevious Competition: https://www.kaggle.com/tanlikesmath/diabetic-retinopathy-resized
Pretrained\nhttps://github.com/lukemelas/EfficientNet-PyTorch