%reload_ext autoreload
%autoreload 2
%matplotlib inline
from fastai.vision import *
from fastai.metrics import error_rate
path = Path('Database')
path.ls()
[PosixPath('Database/GoodPLR_0.404_7581.jpg'), PosixPath('Database/GoodPLR_0.433_2440.jpg'), PosixPath('Database/GoodPLR_0.352_5594.jpg'), PosixPath('Database/GoodPLR_0.46_5072.jpg'), PosixPath('Database/GoodPLR_0.37_914.jpg'), PosixPath('Database/GoodPLR_0.404_840.jpg'), PosixPath('Database/GoodPLR_0.404_9135.jpg'), PosixPath('Database/GoodPLR_0.509_6291.jpg'), PosixPath('Database/GoodPLR_0.389_3233.jpg'), PosixPath('Database/GoodPLR_0.46_1675.jpg'), PosixPath('Database/GoodPLR_0.391_4683.jpg'), PosixPath('Database/GoodPLR_0.439_7106.jpg'), PosixPath('Database/GoodPLR_0.442_3577.jpg'), PosixPath('Database/GoodPLR_0.259_2725.jpg'), PosixPath('Database/GoodPLR_0.463_3786.jpg'), PosixPath('Database/GoodPLR_0.52_3953.jpg'), PosixPath('Database/GoodPLR_0.434_7464.jpg'), PosixPath('Database/GoodPLR_0.349_8029.jpg'), PosixPath('Database/GoodPLR_0.421_4339.jpg'), PosixPath('Database/GoodPLR_0.339_9193.jpg'), PosixPath('Database/GoodPLR_0.404_3004.jpg'), PosixPath('Database/GoodPLR_0.36_8549.jpg'), PosixPath('Database/GoodPLR_0.362_2952.jpg'), PosixPath('Database/GoodPLR_0.37_478.jpg'), PosixPath('Database/GoodPLR_0.421_1363.jpg'), PosixPath('Database/GoodPLR_0.46_2342.jpg'), PosixPath('Database/GoodPLR_0.46_6261.jpg'), PosixPath('Database/GoodPLR_0.341_3736.jpg'), PosixPath('Database/GoodPLR_0.472_1775.jpg'), PosixPath('Database/GoodPLR_0.396_2996.jpg'), PosixPath('Database/GoodPLR_0.3_7677.jpg'), PosixPath('Database/GoodPLR_0.611_8898.jpg'), PosixPath('Database/GoodPLR_0.545_1715.jpg'), PosixPath('Database/GoodPLR_0.404_4433.jpg'), PosixPath('Database/GoodPLR_0.451_9843.jpg'), PosixPath('Database/GoodPLR_0.469_621.jpg'), PosixPath('Database/GoodPLR_0.457_649.jpg'), PosixPath('Database/GoodPLR_0.451_3790.jpg'), PosixPath('Database/GoodPLR_0.527_3408.jpg'), PosixPath('Database/GoodPLR_0.412_8090.jpg'), PosixPath('Database/GoodPLR_0.439_5332.jpg'), PosixPath('Database/GoodPLR_0.509_528.jpg'), PosixPath('Database/GoodPLR_0.453_97.jpg'), PosixPath('Database/GoodPLR_0.421_5869.jpg'), PosixPath('Database/GoodPLR_0.358_9200.jpg'), PosixPath('Database/GoodPLR_0.385_8919.jpg'), PosixPath('Database/GoodPLR_0.433_4170.jpg'), PosixPath('Database/GoodPLR_0.294_7891.jpg'), PosixPath('Database/GoodPLR_0.345_4944.jpg'), PosixPath('Database/GoodPLR_0.429_3374.jpg'), PosixPath('Database/GoodPLR_0.345_1363.jpg'), PosixPath('Database/GoodPLR_0.358_2210.jpg'), PosixPath('Database/GoodPLR_0.358_3472.jpg'), PosixPath('Database/GoodPLR_0.42_186.jpg'), PosixPath('Database/GoodPLR_0.35_9730.jpg'), PosixPath('Database/GoodPLR_0.345_2904.jpg'), PosixPath('Database/GoodPLR_0.396_866.jpg'), PosixPath('Database/GoodPLR_0.321_2785.jpg'), PosixPath('Database/GoodPLR_0.472_7271.jpg'), PosixPath('Database/GoodPLR_0.383_4186.jpg'), PosixPath('Database/GoodPLR_0.256_7193.jpg'), PosixPath('Database/GoodPLR_0.5_8621.jpg'), PosixPath('Database/GoodPLR_0.389_387.jpg'), PosixPath('Database/GoodPLR_0.468_8549.jpg'), PosixPath('Database/GoodPLR_0.393_4466.jpg'), PosixPath('Database/GoodPLR_0.446_2972.jpg'), PosixPath('Database/GoodPLR_0.56_5872.jpg'), PosixPath('Database/GoodPLR_0.48_5767.jpg'), PosixPath('Database/GoodPLR_0.4_3042.jpg'), PosixPath('Database/GoodPLR_0.434_2704.jpg'), PosixPath('Database/GoodPLR_0.442_8469.jpg'), PosixPath('Database/GoodPLR_0.491_4330.jpg'), PosixPath('Database/GoodPLR_0.471_462.jpg'), PosixPath('Database/GoodPLR_0.472_7153.jpg'), PosixPath('Database/GoodPLR_0.358_9380.jpg'), PosixPath('Database/GoodPLR_0.426_4216.jpg'), PosixPath('Database/GoodPLR_0.238_1857.jpg'), PosixPath('Database/GoodPLR_0.426_4146.jpg'), PosixPath('Database/GoodPLR_0.441_5527.jpg'), PosixPath('Database/GoodPLR_0.537_4572.jpg'), PosixPath('Database/GoodPLR_0.327_121.jpg'), PosixPath('Database/GoodPLR_0.412_6757.jpg'), PosixPath('Database/GoodPLR_0.35_4755.jpg'), PosixPath('Database/GoodPLR_0.391_6700.jpg'), PosixPath('Database/GoodPLR_0.408_5190.jpg'), PosixPath('Database/GoodPLR_0.442_2803.jpg'), PosixPath('Database/GoodPLR_0.462_1921.jpg'), PosixPath('Database/GoodPLR_0.417_3765.jpg'), PosixPath('Database/GoodPLR_0.446_5492.jpg'), PosixPath('Database/GoodPLR_0.438_8798.jpg'), PosixPath('Database/GoodPLR_0.537_8547.jpg'), PosixPath('Database/GoodPLR_0.404_7475.jpg'), PosixPath('Database/GoodPLR_0.481_7662.jpg'), PosixPath('Database/GoodPLR_0.439_7177.jpg'), PosixPath('Database/GoodPLR_0.455_2387.jpg'), PosixPath('Database/GoodPLR_0.473_3151.jpg'), PosixPath('Database/GoodPLR_0.365_7357.jpg'), PosixPath('Database/GoodPLR_0.418_53.jpg'), PosixPath('Database/GoodPLR_0.442_1477.jpg'), PosixPath('Database/GoodPLR_0.527_4367.jpg'), PosixPath('Database/GoodPLR_0.478_1642.jpg'), PosixPath('Database/GoodPLR_0.404_2266.jpg'), PosixPath('Database/GoodPLR_0.46_8846.jpg'), PosixPath('Database/GoodPLR_0.468_1945.jpg'), PosixPath('Database/GoodPLR_0.449_1899.jpg'), PosixPath('Database/GoodPLR_0.509_704.jpg'), PosixPath('Database/GoodPLR_0.522_2888.jpg'), PosixPath('Database/GoodPLR_0.357_238.jpg'), PosixPath('Database/GoodPLR_0.429_3697.jpg'), PosixPath('Database/GoodPLR_0.529_9426.jpg'), PosixPath('Database/GoodPLR_0.472_7543.jpg'), PosixPath('Database/GoodPLR_0.479_3398.jpg'), PosixPath('Database/GoodPLR_0.48_7067.jpg'), PosixPath('Database/GoodPLR_0.451_9265.jpg'), PosixPath('Database/GoodPLR_0.345_9173.jpg'), PosixPath('Database/GoodPLR_0.377_7544.jpg'), PosixPath('Database/GoodPLR_0.483_8133.jpg'), PosixPath('Database/GoodPLR_0.5_5843.jpg'), PosixPath('Database/GoodPLR_0.385_4206.jpg'), PosixPath('Database/GoodPLR_0.365_9072.jpg'), PosixPath('Database/GoodPLR_0.489_666.jpg'), PosixPath('Database/GoodPLR_0.468_5861.jpg'), PosixPath('Database/GoodPLR_0.39_9028.jpg'), PosixPath('Database/GoodPLR_0.463_9787.jpg'), PosixPath('Database/models'), PosixPath('Database/GoodPLR_0.389_6505.jpg'), PosixPath('Database/GoodPLR_0.457_3367.jpg'), PosixPath('Database/GoodPLR_0.396_4852.jpg'), PosixPath('Database/GoodPLR_0.32_3518.jpg'), PosixPath('Database/GoodPLR_0.473_6279.jpg'), PosixPath('Database/GoodPLR_0.489_619.jpg'), PosixPath('Database/GoodPLR_0.357_8261.jpg'), PosixPath('Database/GoodPLR_0.4_3232.jpg'), PosixPath('Database/GoodPLR_0.446_24.jpg'), PosixPath('Database/GoodPLR_0.418_2412.jpg'), PosixPath('Database/GoodPLR_0.4_2501.jpg'), PosixPath('Database/.ipynb_checkpoints'), PosixPath('Database/GoodPLR_0.434_266.jpg'), PosixPath('Database/GoodPLR_0.4_1797.jpg'), PosixPath('Database/GoodPLR_0.315_5310.jpg'), PosixPath('Database/GoodPLR_0.462_200.jpg'), PosixPath('Database/GoodPLR_0.365_5661.jpg'), PosixPath('Database/GoodPLR_0.442_1653.jpg'), PosixPath('Database/GoodPLR_0.5_5120.jpg'), PosixPath('Database/GoodPLR_0.52_6453.jpg'), PosixPath('Database/GoodPLR_0.421_815.jpg'), PosixPath('Database/GoodPLR_0.5_2574.jpg'), PosixPath('Database/GoodPLR_0.31_7665.jpg'), PosixPath('Database/GoodPLR_0.32_791.jpg'), PosixPath('Database/GoodPLR_0.345_6004.jpg'), PosixPath('Database/GoodPLR_0.48_4147.jpg'), PosixPath('Database/GoodPLR_0.365_7663.jpg'), PosixPath('Database/GoodPLR_0.442_9326.jpg'), PosixPath('Database/GoodPLR_0.442_9435.jpg'), PosixPath('Database/GoodPLR_0.49_192.jpg'), PosixPath('Database/GoodPLR_0.39_8593.jpg'), PosixPath('Database/GoodPLR_0.42_3708.jpg'), PosixPath('Database/GoodPLR_0.479_4228.jpg'), PosixPath('Database/GoodPLR_0.426_654.jpg'), PosixPath('Database/GoodPLR_0.327_9716.jpg'), PosixPath('Database/GoodPLR_0.492_4189.jpg'), PosixPath('Database/GoodPLR_0.4_8430.jpg'), PosixPath('Database/GoodPLR_0.46_8323.jpg'), PosixPath('Database/GoodPLR_0.222_2094.jpg'), PosixPath('Database/GoodPLR_0.529_7364.jpg'), PosixPath('Database/GoodPLR_0.345_2954.jpg'), PosixPath('Database/GoodPLR_0.42_4819.jpg'), PosixPath('Database/GoodPLR_0.482_5835.jpg'), PosixPath('Database/GoodPLR_0.412_1458.jpg'), PosixPath('Database/GoodPLR_0.444_2873.jpg'), PosixPath('Database/GoodPLR_0.404_4915.jpg'), PosixPath('Database/GoodPLR_0.532_7177.jpg'), PosixPath('Database/GoodPLR_0.418_2342.jpg'), PosixPath('Database/GoodPLR_0.396_97.jpg'), PosixPath('Database/GoodPLR_0.25_2925.jpg'), PosixPath('Database/GoodPLR_0.448_3465.jpg'), PosixPath('Database/GoodPLR_0.463_8105.jpg'), PosixPath('Database/GoodPLR_0.404_3682.jpg'), PosixPath('Database/GoodPLR_0.345_8376.jpg'), PosixPath('Database/GoodPLR_0.46_3063.jpg'), PosixPath('Database/GoodPLR_0.389_470.jpg'), PosixPath('Database/GoodPLR_0.442_9570.jpg'), PosixPath('Database/GoodPLR_0.455_3528.jpg'), PosixPath('Database/GoodPLR_0.431_4040.jpg'), PosixPath('Database/GoodPLR_0.426_2715.jpg'), PosixPath('Database/GoodPLR_0.415_1981.jpg'), PosixPath('Database/GoodPLR_0.42_3102.jpg'), PosixPath('Database/GoodPLR_0.4_10.jpg'), PosixPath('Database/GoodPLR_0.418_7903.jpg'), PosixPath('Database/GoodPLR_0.421_4499.jpg'), PosixPath('Database/GoodPLR_0.489_8875.jpg')]
fnames = get_image_files(path)
fnames[:5]
[PosixPath('Database/GoodPLR_0.404_7581.jpg'), PosixPath('Database/GoodPLR_0.433_2440.jpg'), PosixPath('Database/GoodPLR_0.352_5594.jpg'), PosixPath('Database/GoodPLR_0.46_5072.jpg'), PosixPath('Database/GoodPLR_0.37_914.jpg')]
# variable a will determine the nearest digit that the float will be rounded to.
# small a will result in many classes
def round_nearest(x, a):
return round(x / a) * a
#pat = r'/([^/]+)_\d+.jpg$'
def get_float_labels(y):
return (float(y.parts[-1].split('_')[1]))
??ImageDataBunch.from_name_func
np.random.seed(42)
bs = 40
bs = 40
tfms = get_transforms(do_flip=False)
fnames = get_image_files(path)
fnames
[PosixPath('Database/GoodPLR_0.404_7581.jpg'), PosixPath('Database/GoodPLR_0.433_2440.jpg'), PosixPath('Database/GoodPLR_0.352_5594.jpg'), PosixPath('Database/GoodPLR_0.46_5072.jpg'), PosixPath('Database/GoodPLR_0.37_914.jpg'), PosixPath('Database/GoodPLR_0.404_840.jpg'), PosixPath('Database/GoodPLR_0.404_9135.jpg'), PosixPath('Database/GoodPLR_0.509_6291.jpg'), PosixPath('Database/GoodPLR_0.389_3233.jpg'), PosixPath('Database/GoodPLR_0.46_1675.jpg'), PosixPath('Database/GoodPLR_0.391_4683.jpg'), PosixPath('Database/GoodPLR_0.439_7106.jpg'), PosixPath('Database/GoodPLR_0.442_3577.jpg'), PosixPath('Database/GoodPLR_0.259_2725.jpg'), PosixPath('Database/GoodPLR_0.463_3786.jpg'), PosixPath('Database/GoodPLR_0.52_3953.jpg'), PosixPath('Database/GoodPLR_0.434_7464.jpg'), PosixPath('Database/GoodPLR_0.349_8029.jpg'), PosixPath('Database/GoodPLR_0.421_4339.jpg'), PosixPath('Database/GoodPLR_0.339_9193.jpg'), PosixPath('Database/GoodPLR_0.404_3004.jpg'), PosixPath('Database/GoodPLR_0.36_8549.jpg'), PosixPath('Database/GoodPLR_0.362_2952.jpg'), PosixPath('Database/GoodPLR_0.37_478.jpg'), PosixPath('Database/GoodPLR_0.421_1363.jpg'), PosixPath('Database/GoodPLR_0.46_2342.jpg'), PosixPath('Database/GoodPLR_0.46_6261.jpg'), PosixPath('Database/GoodPLR_0.341_3736.jpg'), PosixPath('Database/GoodPLR_0.472_1775.jpg'), PosixPath('Database/GoodPLR_0.396_2996.jpg'), PosixPath('Database/GoodPLR_0.3_7677.jpg'), PosixPath('Database/GoodPLR_0.611_8898.jpg'), PosixPath('Database/GoodPLR_0.545_1715.jpg'), PosixPath('Database/GoodPLR_0.404_4433.jpg'), PosixPath('Database/GoodPLR_0.451_9843.jpg'), PosixPath('Database/GoodPLR_0.469_621.jpg'), PosixPath('Database/GoodPLR_0.457_649.jpg'), PosixPath('Database/GoodPLR_0.451_3790.jpg'), PosixPath('Database/GoodPLR_0.527_3408.jpg'), PosixPath('Database/GoodPLR_0.412_8090.jpg'), PosixPath('Database/GoodPLR_0.439_5332.jpg'), PosixPath('Database/GoodPLR_0.509_528.jpg'), PosixPath('Database/GoodPLR_0.453_97.jpg'), PosixPath('Database/GoodPLR_0.421_5869.jpg'), PosixPath('Database/GoodPLR_0.358_9200.jpg'), PosixPath('Database/GoodPLR_0.385_8919.jpg'), PosixPath('Database/GoodPLR_0.433_4170.jpg'), PosixPath('Database/GoodPLR_0.294_7891.jpg'), PosixPath('Database/GoodPLR_0.345_4944.jpg'), PosixPath('Database/GoodPLR_0.429_3374.jpg'), PosixPath('Database/GoodPLR_0.345_1363.jpg'), PosixPath('Database/GoodPLR_0.358_2210.jpg'), PosixPath('Database/GoodPLR_0.358_3472.jpg'), PosixPath('Database/GoodPLR_0.42_186.jpg'), PosixPath('Database/GoodPLR_0.35_9730.jpg'), PosixPath('Database/GoodPLR_0.345_2904.jpg'), PosixPath('Database/GoodPLR_0.396_866.jpg'), PosixPath('Database/GoodPLR_0.321_2785.jpg'), PosixPath('Database/GoodPLR_0.472_7271.jpg'), PosixPath('Database/GoodPLR_0.383_4186.jpg'), PosixPath('Database/GoodPLR_0.256_7193.jpg'), PosixPath('Database/GoodPLR_0.5_8621.jpg'), PosixPath('Database/GoodPLR_0.389_387.jpg'), PosixPath('Database/GoodPLR_0.468_8549.jpg'), PosixPath('Database/GoodPLR_0.393_4466.jpg'), PosixPath('Database/GoodPLR_0.446_2972.jpg'), PosixPath('Database/GoodPLR_0.56_5872.jpg'), PosixPath('Database/GoodPLR_0.48_5767.jpg'), PosixPath('Database/GoodPLR_0.4_3042.jpg'), PosixPath('Database/GoodPLR_0.434_2704.jpg'), PosixPath('Database/GoodPLR_0.442_8469.jpg'), PosixPath('Database/GoodPLR_0.491_4330.jpg'), PosixPath('Database/GoodPLR_0.471_462.jpg'), PosixPath('Database/GoodPLR_0.472_7153.jpg'), PosixPath('Database/GoodPLR_0.358_9380.jpg'), PosixPath('Database/GoodPLR_0.426_4216.jpg'), PosixPath('Database/GoodPLR_0.238_1857.jpg'), PosixPath('Database/GoodPLR_0.426_4146.jpg'), PosixPath('Database/GoodPLR_0.441_5527.jpg'), PosixPath('Database/GoodPLR_0.537_4572.jpg'), PosixPath('Database/GoodPLR_0.327_121.jpg'), PosixPath('Database/GoodPLR_0.412_6757.jpg'), PosixPath('Database/GoodPLR_0.35_4755.jpg'), PosixPath('Database/GoodPLR_0.391_6700.jpg'), PosixPath('Database/GoodPLR_0.408_5190.jpg'), PosixPath('Database/GoodPLR_0.442_2803.jpg'), PosixPath('Database/GoodPLR_0.462_1921.jpg'), PosixPath('Database/GoodPLR_0.417_3765.jpg'), PosixPath('Database/GoodPLR_0.446_5492.jpg'), PosixPath('Database/GoodPLR_0.438_8798.jpg'), PosixPath('Database/GoodPLR_0.537_8547.jpg'), PosixPath('Database/GoodPLR_0.404_7475.jpg'), PosixPath('Database/GoodPLR_0.481_7662.jpg'), PosixPath('Database/GoodPLR_0.439_7177.jpg'), PosixPath('Database/GoodPLR_0.455_2387.jpg'), PosixPath('Database/GoodPLR_0.473_3151.jpg'), PosixPath('Database/GoodPLR_0.365_7357.jpg'), PosixPath('Database/GoodPLR_0.418_53.jpg'), PosixPath('Database/GoodPLR_0.442_1477.jpg'), PosixPath('Database/GoodPLR_0.527_4367.jpg'), PosixPath('Database/GoodPLR_0.478_1642.jpg'), PosixPath('Database/GoodPLR_0.404_2266.jpg'), PosixPath('Database/GoodPLR_0.46_8846.jpg'), PosixPath('Database/GoodPLR_0.468_1945.jpg'), PosixPath('Database/GoodPLR_0.449_1899.jpg'), PosixPath('Database/GoodPLR_0.509_704.jpg'), PosixPath('Database/GoodPLR_0.522_2888.jpg'), PosixPath('Database/GoodPLR_0.357_238.jpg'), PosixPath('Database/GoodPLR_0.429_3697.jpg'), PosixPath('Database/GoodPLR_0.529_9426.jpg'), PosixPath('Database/GoodPLR_0.472_7543.jpg'), PosixPath('Database/GoodPLR_0.479_3398.jpg'), PosixPath('Database/GoodPLR_0.48_7067.jpg'), PosixPath('Database/GoodPLR_0.451_9265.jpg'), PosixPath('Database/GoodPLR_0.345_9173.jpg'), PosixPath('Database/GoodPLR_0.377_7544.jpg'), PosixPath('Database/GoodPLR_0.483_8133.jpg'), PosixPath('Database/GoodPLR_0.5_5843.jpg'), PosixPath('Database/GoodPLR_0.385_4206.jpg'), PosixPath('Database/GoodPLR_0.365_9072.jpg'), PosixPath('Database/GoodPLR_0.489_666.jpg'), PosixPath('Database/GoodPLR_0.468_5861.jpg'), PosixPath('Database/GoodPLR_0.39_9028.jpg'), PosixPath('Database/GoodPLR_0.463_9787.jpg'), PosixPath('Database/GoodPLR_0.389_6505.jpg'), PosixPath('Database/GoodPLR_0.457_3367.jpg'), PosixPath('Database/GoodPLR_0.396_4852.jpg'), PosixPath('Database/GoodPLR_0.32_3518.jpg'), PosixPath('Database/GoodPLR_0.473_6279.jpg'), PosixPath('Database/GoodPLR_0.489_619.jpg'), PosixPath('Database/GoodPLR_0.357_8261.jpg'), PosixPath('Database/GoodPLR_0.4_3232.jpg'), PosixPath('Database/GoodPLR_0.446_24.jpg'), PosixPath('Database/GoodPLR_0.418_2412.jpg'), PosixPath('Database/GoodPLR_0.4_2501.jpg'), PosixPath('Database/GoodPLR_0.434_266.jpg'), PosixPath('Database/GoodPLR_0.4_1797.jpg'), PosixPath('Database/GoodPLR_0.315_5310.jpg'), PosixPath('Database/GoodPLR_0.462_200.jpg'), PosixPath('Database/GoodPLR_0.365_5661.jpg'), PosixPath('Database/GoodPLR_0.442_1653.jpg'), PosixPath('Database/GoodPLR_0.5_5120.jpg'), PosixPath('Database/GoodPLR_0.52_6453.jpg'), PosixPath('Database/GoodPLR_0.421_815.jpg'), PosixPath('Database/GoodPLR_0.5_2574.jpg'), PosixPath('Database/GoodPLR_0.31_7665.jpg'), PosixPath('Database/GoodPLR_0.32_791.jpg'), PosixPath('Database/GoodPLR_0.345_6004.jpg'), PosixPath('Database/GoodPLR_0.48_4147.jpg'), PosixPath('Database/GoodPLR_0.365_7663.jpg'), PosixPath('Database/GoodPLR_0.442_9326.jpg'), PosixPath('Database/GoodPLR_0.442_9435.jpg'), PosixPath('Database/GoodPLR_0.49_192.jpg'), PosixPath('Database/GoodPLR_0.39_8593.jpg'), PosixPath('Database/GoodPLR_0.42_3708.jpg'), PosixPath('Database/GoodPLR_0.479_4228.jpg'), PosixPath('Database/GoodPLR_0.426_654.jpg'), PosixPath('Database/GoodPLR_0.327_9716.jpg'), PosixPath('Database/GoodPLR_0.492_4189.jpg'), PosixPath('Database/GoodPLR_0.4_8430.jpg'), PosixPath('Database/GoodPLR_0.46_8323.jpg'), PosixPath('Database/GoodPLR_0.222_2094.jpg'), PosixPath('Database/GoodPLR_0.529_7364.jpg'), PosixPath('Database/GoodPLR_0.345_2954.jpg'), PosixPath('Database/GoodPLR_0.42_4819.jpg'), PosixPath('Database/GoodPLR_0.482_5835.jpg'), PosixPath('Database/GoodPLR_0.412_1458.jpg'), PosixPath('Database/GoodPLR_0.444_2873.jpg'), PosixPath('Database/GoodPLR_0.404_4915.jpg'), PosixPath('Database/GoodPLR_0.532_7177.jpg'), PosixPath('Database/GoodPLR_0.418_2342.jpg'), PosixPath('Database/GoodPLR_0.396_97.jpg'), PosixPath('Database/GoodPLR_0.25_2925.jpg'), PosixPath('Database/GoodPLR_0.448_3465.jpg'), PosixPath('Database/GoodPLR_0.463_8105.jpg'), PosixPath('Database/GoodPLR_0.404_3682.jpg'), PosixPath('Database/GoodPLR_0.345_8376.jpg'), PosixPath('Database/GoodPLR_0.46_3063.jpg'), PosixPath('Database/GoodPLR_0.389_470.jpg'), PosixPath('Database/GoodPLR_0.442_9570.jpg'), PosixPath('Database/GoodPLR_0.455_3528.jpg'), PosixPath('Database/GoodPLR_0.431_4040.jpg'), PosixPath('Database/GoodPLR_0.426_2715.jpg'), PosixPath('Database/GoodPLR_0.415_1981.jpg'), PosixPath('Database/GoodPLR_0.42_3102.jpg'), PosixPath('Database/GoodPLR_0.4_10.jpg'), PosixPath('Database/GoodPLR_0.418_7903.jpg'), PosixPath('Database/GoodPLR_0.421_4499.jpg'), PosixPath('Database/GoodPLR_0.489_8875.jpg')]
data = ImageDataBunch.from_name_func(path, fnames, get_float_labels, ds_tfms=get_transforms(), size=224, ).normalize(imagenet_stats)
data = (ImageItemList(fnames)
.random_split_by_pct(0.3)
.label_from_func(get_float_labels)
.transform(tfms, size=224)
.databunch())
data.bs = bs
learn = create_cnn(data, models.resnet50, metrics=[mean_squared_error])
data.show_batch(rows=3, figsize=(7,6))
learn = create_cnn(data, models.resnet50,metrics=error_rate)
learn.model
Sequential( (0): Sequential( (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False) (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (2): ReLU(inplace) (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False) (4): Sequential( (0): Bottleneck( (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) (downsample): Sequential( (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (2): Bottleneck( (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) ) (5): Sequential( (0): Bottleneck( (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) (downsample): Sequential( (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (2): Bottleneck( (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (3): Bottleneck( (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) ) (6): Sequential( (0): Bottleneck( (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) (downsample): Sequential( (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (2): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (3): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (4): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (5): Bottleneck( (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) ) (7): Sequential( (0): Bottleneck( (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) (downsample): Sequential( (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False) (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (1): Bottleneck( (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) (2): Bottleneck( (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (relu): ReLU(inplace) ) ) ) (1): Sequential( (0): AdaptiveConcatPool2d( (ap): AdaptiveAvgPool2d(output_size=1) (mp): AdaptiveMaxPool2d(output_size=1) ) (1): Flatten() (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (3): Dropout(p=0.25) (4): Linear(in_features=4096, out_features=512, bias=True) (5): ReLU(inplace) (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (7): Dropout(p=0.5) (8): Linear(in_features=512, out_features=5, bias=True) ) )
learn.fit_one_cycle(4)
epoch | train_loss | valid_loss | error_rate |
---|
--------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) <ipython-input-22-495233eaf2b4> in <module> ----> 1 learn.fit_one_cycle(4) ~/anaconda3/lib/python3.7/site-packages/fastai/train.py in fit_one_cycle(learn, cyc_len, max_lr, moms, div_factor, pct_start, wd, callbacks, tot_epochs, start_epoch) 20 callbacks.append(OneCycleScheduler(learn, max_lr, moms=moms, div_factor=div_factor, pct_start=pct_start, tot_epochs=tot_epochs, 21 start_epoch=start_epoch)) ---> 22 learn.fit(cyc_len, max_lr, wd=wd, callbacks=callbacks) 23 24 def lr_find(learn:Learner, start_lr:Floats=1e-7, end_lr:Floats=10, num_it:int=100, stop_div:bool=True, wd:float=None): ~/anaconda3/lib/python3.7/site-packages/fastai/basic_train.py in fit(self, epochs, lr, wd, callbacks) 176 callbacks = [cb(self) for cb in self.callback_fns] + listify(callbacks) 177 fit(epochs, self.model, self.loss_func, opt=self.opt, data=self.data, metrics=self.metrics, --> 178 callbacks=self.callbacks+callbacks) 179 180 def create_opt(self, lr:Floats, wd:Floats=0.)->None: ~/anaconda3/lib/python3.7/site-packages/fastai/utils/mem.py in wrapper(*args, **kwargs) 78 79 try: ---> 80 return func(*args, **kwargs) 81 except Exception as e: 82 if ("CUDA out of memory" in str(e) or ~/anaconda3/lib/python3.7/site-packages/fastai/basic_train.py in fit(epochs, model, loss_func, opt, data, callbacks, metrics) 93 if not data.empty_val: 94 val_loss = validate(model, data.valid_dl, loss_func=loss_func, ---> 95 cb_handler=cb_handler, pbar=pbar) 96 else: val_loss=None 97 if cb_handler.on_epoch_end(val_loss): break ~/anaconda3/lib/python3.7/site-packages/fastai/basic_train.py in validate(model, dl, loss_func, cb_handler, pbar, average, n_batch) 55 if not is_listy(yb): yb = [yb] 56 nums.append(yb[0].shape[0]) ---> 57 if cb_handler and cb_handler.on_batch_end(val_losses[-1]): break 58 if n_batch and (len(nums)>=n_batch): break 59 nums = np.array(nums, dtype=np.float32) ~/anaconda3/lib/python3.7/site-packages/fastai/callback.py in on_batch_end(self, loss) 257 "Handle end of processing one batch with `loss`." 258 self.state_dict['last_loss'] = loss --> 259 stop = np.any(self('batch_end', not self.state_dict['train'])) 260 if self.state_dict['train']: 261 self.state_dict['iteration'] += 1 ~/anaconda3/lib/python3.7/site-packages/fastai/callback.py in __call__(self, cb_name, call_mets, **kwargs) 198 def __call__(self, cb_name, call_mets=True, **kwargs)->None: 199 "Call through to all of the `CallbakHandler` functions." --> 200 if call_mets: [getattr(met, f'on_{cb_name}')(**self.state_dict, **kwargs) for met in self.metrics] 201 return [getattr(cb, f'on_{cb_name}')(**self.state_dict, **kwargs) for cb in self.callbacks] 202 ~/anaconda3/lib/python3.7/site-packages/fastai/callback.py in <listcomp>(.0) 198 def __call__(self, cb_name, call_mets=True, **kwargs)->None: 199 "Call through to all of the `CallbakHandler` functions." --> 200 if call_mets: [getattr(met, f'on_{cb_name}')(**self.state_dict, **kwargs) for met in self.metrics] 201 return [getattr(cb, f'on_{cb_name}')(**self.state_dict, **kwargs) for cb in self.callbacks] 202 ~/anaconda3/lib/python3.7/site-packages/fastai/callback.py in on_batch_end(self, last_output, last_target, **kwargs) 292 if not is_listy(last_target): last_target=[last_target] 293 self.count += last_target[0].size(0) --> 294 self.val += last_target[0].size(0) * self.func(last_output, *last_target).detach().cpu() 295 296 def on_epoch_end(self, **kwargs): ~/anaconda3/lib/python3.7/site-packages/fastai/metrics.py in error_rate(input, targs) 42 def error_rate(input:Tensor, targs:Tensor)->Rank0Tensor: 43 "1 - `accuracy`" ---> 44 return 1 - accuracy(input, targs) 45 46 def dice(input:Tensor, targs:Tensor, iou:bool=False)->Rank0Tensor: ~/anaconda3/lib/python3.7/site-packages/fastai/metrics.py in accuracy(input, targs) 27 input = input.argmax(dim=-1).view(n,-1) 28 targs = targs.view(n,-1) ---> 29 return (input==targs).float().mean() 30 31 def accuracy_thresh(y_pred:Tensor, y_true:Tensor, thresh:float=0.5, sigmoid:bool=True)->Rank0Tensor: RuntimeError: Expected object of scalar type Long but got scalar type Float for argument #2 'other'
interp = ClassificationInterpretation.from_learner(learn)
interp.plot_confusion_matrix(figsize=(12,12), dpi=60)
learn.lr_find()
LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.
learn.recorder.plot()
learn.save('PLRstage1')
learn.unfreeze()
learn.fit_one_cycle(2, max_lr=slice(3e-4,2e-2))
learn.save('stage-2')
learn.load('PLRstage1')
import os, random
randomchoice = random.choice(os.listdir("/home/ubuntu/Eyetrain/Database/"))
randomchoice
img = open_image('/home/ubuntu/Eyetrain/Database/'+randomchoice)
img
log_preds = learn.predict(img)
log_preds
log_preds[0]
e^0.268269