Closed Jenifer-Anjali closed 4 years ago
Hey,
For the 1st issue, could you please show me your Info.txt file ? (It should normally be a copy/paste of the positions of tumors given here : http://peipa.essex.ac.uk/info/mias.html - starting from mdb001 G CIRC B 535 425 197
...)
For the 2nd issue, make sure that the package maskrcnn_benchmark is correctly installed : https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/INSTALL.md#windows-10
Thank you
Hello, As you mentioned i edited the info.txt file and tried to run the generate_COCO_annotations.py file but i get this key error. I have attached the info.txt file and the snapshots of the error. Also i have included the directory structure of the code, kindly help me solve this problem. Info.txt
Here is the correct Info.txt file you should use : https://github.com/delmalih/MIAS-mammography-obj-detection/blob/master/utils/Info.txt
Hello sir, This is the file i have used already but still i'm getting "index 0 is out of bounds for axis 0 with size 0" when i try to map it. Exactly at line no 134 in generate_COCO_annotations.py and similar code of lines. I also tried removing the headers but the again i get error REFNUM and so on. I guess its because Info.txt file can't be read. Also could you please look at my directory structure and correct me if something's wrong.
When i tried to run this command i get Info.txt is not found python generate_COCO_annotations.py --images ../mias-db/ --annotations ../mias-db/Info.txt --output ../mias-db/COCO--aug_fact 20 --train_val_split 0.9
It works fine for me. Make sure :
python generate_COCO_annotations.py -i ./mias-db -a ./utils/Info.txt -o ../mias-db/COCO --aug_fact 20 --train_val_split 0.9
Oh I used python 3.7 let me try this and let you know how it goes.
Thanks
@Jenifer-Anjali
in windows system,at line num 67 in generate_COCO_annotations.py should be changed as this:
imgname, = os.path.splitext(os.path.basename(img_path))
@Jenifer-Anjali in windows system,at line num 67 in generate_COCO_annotations.py should be changed as this: imgname, = os.path.splitext(os.path.basename(img_path))
Thank you.
hey, i tried implementing the faster-r-cnn code for reference purpose in windows but when i try to Generate COCO or VOC augmented data code, i get index out of bounds error at line 68 to 75 in generate_COCO_annotations.py and also that i get import error as "no module maskrcnn_benchmark found" while i try to implement train.py even though i tried correcting the path.
Can you please help me solve the issue? Thankyou
Also i have attached the images for your reference.