Making use of LightningDataModule and simplification of data loading

This merge request adds LightningDataModule to better organize the code and make better use of lightning's features. This centralizes common tasks such as DataLoader creation and application of transforms into a base class to be inherited from.

Data loading was also simplified by removing custom Sample classes and maker functions, and the addition of RuntimeDataset and CachedDataset.

Remaining tasks:

• (@dcarron) Create a common/default DataModule for the shenzhen dataset that takes protocols and transforms as parameters to avoid copying code for each protocol configuration
• (@dcarron) Add typehints to ShenzenDataModule
• (@dcarron) Investigate issue where training a new model with ElasticDeformation as a transform converges more slowly if the data is not cached.
• (@andre.anjos) Update documentation on ShenzenDataModule
• (@biosignal) Update all datasets, using Shenzhen as a reference
  • (@mdelitroz) Montgomery
  • (@mdelitroz) Hivtb
  • (@andre.anjos) Indian
  • (@andre.anjos) Padchest
  • (@mdelitroz) Tbpoc
  • (@andre.anjos) tbx11_simplified -> renamed as tbx11k, protocol v1 (uses original dataset organisation)
  • (@andre.anjos) tbx11_simplified_v2 -> renamed as tbx11k, protocol v2 (uses original dataset organisation)
  • (@andre.anjos) mc_ch -> renamed as montgomery-shenzhen
  • (@andre.anjos) mc_ch_in -> renamed as montgomery-shenzhen-indian
  • (@andre.anjos) mc_ch_in_11k -> renamed as montgomery-shenzhen-indian-tbx11k-v1
  • (@andre.anjos) mc_ch_in_11k_v2 -> renamed as montgomery-shenzhen-indian-tbx11k-v2
  • (@andre.anjos) mc_ch_in_pc -> renamed as montgomery-shenzhen-indian-padchest
  • (@andre.anjos) nih_cxr14_re -> renamed as nih-cxr14 (n.b.: multi-class dataset, radiological findings)
  • (@andre.anjos) nih_cxr14_re_pc -> renamed as nih-cxr14-padchest
• (@dcarron) Update models
• (@dcarron) Update evaluation scripts
• (@dcarron) Update support for extra_validation datasets
• (@biosignal) Update full documentation
• (@biosignal) Update unit tests

Addresses the following issues:

• Closes #14 (closed)
• Closes #18 (closed)
• Closes #19 (closed)
• Closes #21 (closed)
• Closes #29 (closed)

src/ptbench/data/hivtb/fold-0.json

+{
+  "train": [
+    ["HIV-TB_Algorithm_study_X-rays/413.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/250.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/230.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/318.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/208.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/211.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/282.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/302.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/331.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/214.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/342.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/268.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/398.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/252.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/378.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/418.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/381.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/272.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/231.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/236.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/482.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/330.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/490.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/262.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/280.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/222.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/404.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/223.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/289.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/278.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/473.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/351.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/257.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/325.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/294.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/217.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/322.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/340.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/348.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/285.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/406.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/303.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/480.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/326.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/462.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/494.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/476.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/352.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/402.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/437.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/498.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/390.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/306.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/492.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/430.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/349.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/339.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/409.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/354.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/400.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/499.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/299.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/281.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/460.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/469.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/243.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/426.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/391.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/314.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/350.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/297.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/321.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/497.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/271.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/258.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/377.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/245.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/410.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/333.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/399.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/401.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/407.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/312.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/408.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/387.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/344.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/233.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/327.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/317.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/493.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/355.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/438.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/275.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/376.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/461.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/205.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/276.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/440.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/479.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/389.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/225.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/305.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/491.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/363.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/485.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/270.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/481.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/290.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/244.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/484.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/218.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/379.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/471.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/357.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/247.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/464.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/267.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/463.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/235.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/216.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/224.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/397.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/329.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/364.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/369.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/416.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/472.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/313.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/347.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/450.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/495.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/489.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/356.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/298.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/295.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/396.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/359.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/207.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/362.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/219.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/394.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/392.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/256.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/361.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/212.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/323.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/213.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/478.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/315.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/227.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/365.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/421.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/474.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/385.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/287.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/403.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/368.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/338.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/202.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/229.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/311.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/380.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/483.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/353.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/405.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/341.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/467.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/332.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/259.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/487.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/374.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/316.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/263.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/386.BMP", 0]
+  ],
+  "validation": [
+    ["HIV-TB_Algorithm_study_X-rays/279.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/221.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/367.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/393.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/432.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/210.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/309.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/370.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/343.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/411.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/383.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/434.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/308.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/209.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/457.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/288.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/301.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/372.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/241.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/412.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/366.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/334.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/371.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/284.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/375.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/242.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/273.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/265.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/269.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/310.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/237.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/428.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/249.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/283.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/454.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/496.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/360.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/255.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/477.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/465.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/324.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/486.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/384.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/358.BMP", 1]
+  ],
+  "test": [
+    ["HIV-TB_Algorithm_study_X-rays/199.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/201.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/215.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/238.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/251.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/253.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/266.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/286.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/291.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/292.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/293.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/296.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/304.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/319.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/320.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/335.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/336.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/337.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/346.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/388.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/395.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/415.BMP", 0],
+    ["HIV-TB_Algorithm_study_X-rays/468.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/475.BMP", 1],
+    ["HIV-TB_Algorithm_study_X-rays/488.BMP", 1]
+  ]
+}