Closed wurenzhong closed 6 years ago
Hello~lishen, I have download the CBIS-DDSM dataset from The Cancer Imaging Archive, it's show in the picture:
![]()
But it havn't the ".csv" file,does this file need to be created by ourselves? But we just know the information about the patient, side, CC or MLO, there's no more information. As shown in the above picture, the test and train sets are scattered throughout the folder, and they contain many subfolders, I don't know how to set the parameter of "train_dir", "test_dir", "val_dir".
So~I have two requests: 1:Would you please provide the ".csv" file? 2:Would you tell us, does we need to sort out the data by ourselves and integrate all the images into the "test_set", "train_set", "val_set" three folders?
Thank you for your time, best wishes for you!
I just start working on this... i have the same question.. from where i get .csv files
@bintykhany
You can get the .csv from https://wiki.cancerimagingarchive.net/display/Public/CBIS-DDSM#188b4eabd7024bc68ab11c4d4150cb01
@wurenzhong Thanks a lot..
I am also a beginner, please help me answer the question 2: "2:Would you tell us, does we need to sort out the data by ourselves and integrate all the images into the "test_set", "train_set", "val_set" three folders?" Thank you.
@RanChiVo Yes, these data sets need to be organized by ourselves. First, you can write a "for" loop to rename all the image names (e.g. "00000 ") to folder names (e.g. " mass-test_p_00707_right_cc ") and put them in the same folder. Then, divide them into "test_set", "train_set", "val_set" based on the ".csv" files. Finally, assign the file address.
I will try your way that helped me. I hope I can handle it. “That’s very kind of you. Thank you.”
@wurenzhong
Follow the author's instructions in the example_model_test file.
"On your machine, you shall have folders like the following:
./full_test_1152x896 ./full_test_1152x896/neg ./full_test_1152x896/pos The "neg" being the negative images and "pos" being the positive images."
I have to rely on the test.csv file to classify the image of which class, right? neg or pos. Can you explain it clearly to me? Thank you very much.
@RanChiVo The "test_set", "train_set" and "val_set" folders all have "neg', "pos" subfolders, and images are in subfolders. Sorry, It's been so long since I ran this code that I kind of forgot. Data sets can be divided according to the website of https://wiki.cancerimagingarchive.net/display/Public/CBIS-DDSM#188b4eabd7024bc68ab11c4d4150cb01
did you run the rename and creation of "test_set", "train_set", "val_set" code ?
@wurenzhong Hi, I have the same question, what is the meaning of "neg" and "pos"? Because I want to test on my own dataset.
@Tianshuo-Xu “neg” and "pos" are labels, negative and positive abbreviations
@wurenzhong I still dont understand the positive and negative, is the positive means benign? and negative means malignant? Can you give a pathology explanation for "pos" and "neg"? Thanks!
@ Tianshuo-Xu positive means malignant, and negative means benign.
@wurenzhong Thanks a lot for your explanation! I run the model using a single MG image, and the model output two points like that: [0.0003127959, 0.99998164]. What are this two points mean?
@Tianshuo-Xu It means loss and accuracy
I just downloaded this dataset but the files are named eg 1-1.dcm and in the csv file are named differently. Anyone solved this?
Hello~lishen, I have download the CBIS-DDSM dataset from The Cancer Imaging Archive, it's show in the picture:
But it havn't the ".csv" file,does this file need to be created by ourselves? But we just know the information about the patient, side, CC or MLO, there's no more information. As shown in the above picture, the test and train sets are scattered throughout the folder, and they contain many subfolders, I don't know how to set the parameter of "train_dir", "test_dir", "val_dir".
So~I have two requests: 1:Would you please provide the ".csv" file? 2:Would you tell us, does we need to sort out the data by ourselves and integrate all the images into the "test_set", "train_set", "val_set" three folders?
Thank you for your time, best wishes for you!