Issue with build_coordinates.py

[MatthewBM]

Hello, I installed ffn and downloaded the sample data without issue. I am trying to train the sample model and got an error.

I ran this:

(ffn) user:~/ffn-master$ runffn.sh #!/bin/bash python compute_partitions.py
--input_volume ~/third_party/neuroproof_examples/validation_sample/groundtruth.h5:stack
--output_volume ~/third_party/neuroproof_examples/validation_sample/af.h5:af
--thresholds 0.025,0.05,0.075,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9
--lom_radius 24,24,24
--min_size 10000 python build_coordinates.py
--partition_volumes ~/third_party/neuroproof_examples/validation_sample/af.h5:af
--coordinate_output ~/third_party/neuroproof_examples/validation_sample/tf_record_file
--margin 24,24,24

compute_partitions.py runs fine, but with build_coordinates I get this:

Traceback (most recent call last): File "build_coordinates.py", line 110, in app.run(main) File "/home/ncmir-lab/.conda/envs/ffn/lib/python3.6/site-packages/absl/app.py", line 300, in run _run_main(main, args) File "/home/ncmir-lab/.conda/envs/ffn/lib/python3.6/site-packages/absl/app.py", line 251, in _run_main sys.exit(main(argv)) File "build_coordinates.py", line 62, in main name, path, dataset = partvol.split(':') ValueError: not enough values to unpack (expected 3, got 2)

[MatthewBM]

I realized the home path doesn't work in this context, I have fixed it by changing: --partition_volumes ~/third_party/neuroproof_examples/validation_sample/af.h5:af to --partition_volumes validation1:/home/user/third_party/neuroproof_examples/validation_sample/af.h5:af

[mk123qwe]

I realized the home path doesn't work in this context, I have fixed it by changing: --partition_volumes ~/third_party/neuroproof_examples/validation_sample/af.h5:af to --partition_volumes validation1:/home/user/third_party/neuroproof_examples/validation_sample/af.h5:af

Can you get the results in the paper?

[MatthewBM]

I didn't reproduce the exact demo data in the paper because it's a different species than what I'm interested in but it performs great on my data.

[mk123qwe]

I didn't reproduce the exact demo data in the paper because it's a different species than what I'm interested in but it performs great on my data.

Run_inference.py can get the result of each 2D segmentation ,but how to reconstruct the result of 3D segmentation?

[MatthewBM]

I'm not sure what you're asking. FFN is a 3D solver, the output is 3D.

[mk123qwe]

I'm not sure what you're asking. FFN is a 3D solver, the output is 3D.

Output shape is (250, 250, 250)，it shows that 250(numbers) x 2D segmentation(w:250,h:250).

[mk123qwe]

I didn't reproduce the exact demo data in the paper because it's a different species than what I'm interested in but it performs great on my data.

HELLO， There is no validation set in the code. How do you determine if your training results are great?

[oradomskyi]

Hi @MatthewBM , @mk123qwe do you know how to run ffn on a dataset of about 5TB?

[MatthewBM]

Hi @FrayaMiner,

Why don't you send me an email [email protected] with a little more detail about what you're trying to do.