Add files using upload-large-folder tool

cell_cycle/my_conf/dataset/diabetic_retinopathy/diabetic_retinopathy_full_circle_augs_2560_crop_inference.py

@@ -0,0 +1,9 @@
+from dataclasses import replace
+
+from GaussianProxy.conf.dataset.diabetic_retinopathy.diabetic_retinopathy_full_circle_augs_2560_crop_inference import (
+    dataset,
+)
+
+dataset = replace(
+    dataset, path="/projects/static2dynamic/datasets/DiabeticRetinopathy/prepared_dataset_full_circle_augmented"
+)

cell_cycle/my_conf/dataset/diabetic_retinopathy/diabetic_retinopathy_fully_ordered_inference.py

@@ -0,0 +1,15 @@
+from dataclasses import replace
+
+from GaussianProxy.conf.dataset.diabetic_retinopathy.diabetic_retinopathy_inference import dataset
+from GaussianProxy.utils.data import ContinuousTimeImageDataset
+
+assert dataset.dataset_params is not None
+updated_ds_params = replace(dataset.dataset_params, dataset_class=ContinuousTimeImageDataset)
+
+dataset = replace(
+    dataset,
+    fully_ordered=True,
+    path="/projects/static2dynamic/datasets/DiabeticRetinopathy/prepared_dataset/train",
+    path_to_single_parquet="/projects/static2dynamic/datasets/DiabeticRetinopathy/prepared_dataset/diabetic_retinopathy__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet",
+    dataset_params=updated_ds_params,
+)

cell_cycle/my_conf/dataset/ependymal_context/ependymal_context.yaml

@@ -0,0 +1,16 @@
+name: ependymal_context
+path: /projects/static2dynamic/datasets/ependymal/prepared_dataset_context
+data_shape: [3, 256, 256]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8

cell_cycle/my_conf/dataset/ependymal_context/ependymal_context_avg_t_fully_ordered_inference.py

@@ -0,0 +1,31 @@
+from dataclasses import replace
+from pathlib import Path
+
+from GaussianProxy.conf.dataset.ependymal_context.ependymal_context_fully_ordered_inference import dataset
+from GaussianProxy.conf.training_conf import DatasetParams
+from GaussianProxy.utils.data import ContinuousTimeImageDataset
+
+
+# separate gt needs a workaround for sorting_func
+def sorting_func(subdir: Path):
+    if subdir.name == "all_imgs":
+        return 0
+    else:
+        raise ValueError(f"unexpected subdir: {subdir}")
+
+
+ds_params = DatasetParams(
+    file_extension="png",
+    key_transform=str,
+    sorting_func=sorting_func,
+    dataset_class=ContinuousTimeImageDataset,
+)
+
+dataset = replace(
+    dataset,
+    name="ependymal_context_avg_t_separate_gt_fully_ordered",
+    path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_context_avg_pt_from_01_noised_facebook_dinov2-with-registers-giant/all_imgs",
+    path_to_single_parquet="/projects/static2dynamic/datasets/ependymal/prepared_dataset_context_avg_pt_from_01_noised_facebook_dinov2-with-registers-giant/ependymal_context__continuous_time_predictions__avg_t_preds.parquet",
+    dataset_params=ds_params,
+    separate_gt_starting_class_path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_context/0"
+)

cell_cycle/my_conf/dataset/ependymal_context/ependymal_context_fully_ordered_inference.py

@@ -0,0 +1,9 @@
+from dataclasses import replace
+
+from GaussianProxy.conf.dataset.ependymal_context.ependymal_context_fully_ordered_inference import dataset
+
+dataset = replace(
+    dataset,
+    path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_context",
+    path_to_single_parquet="/projects/static2dynamic/datasets/ependymal/ependymal_context__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet",
+)

cell_cycle/my_conf/dataset/ependymal_context/ependymal_context_inference.py

@@ -0,0 +1,32 @@
+from torch import float32
+from torchvision.transforms import Compose, ConvertImageDtype, Normalize
+
+from GaussianProxy.conf.training_conf import DataSet, DatasetParams
+from GaussianProxy.utils.data import ImageDataset
+
+DEFINITION = 256
+NUMBER_OF_CHANNELS = 3
+
+transforms = Compose(
+    transforms=[
+        ConvertImageDtype(float32),
+        Normalize(mean=[0.5] * NUMBER_OF_CHANNELS, std=[0.5] * NUMBER_OF_CHANNELS),
+    ]
+)
+
+ds_params = DatasetParams(
+    file_extension="png",
+    key_transform=str,
+    sorting_func=lambda subdir: int(subdir.name),
+    dataset_class=ImageDataset,
+)
+
+dataset = DataSet(
+    name="ependymal_context",
+    data_shape=(NUMBER_OF_CHANNELS, DEFINITION, DEFINITION),
+    transforms=transforms,
+    expected_initial_data_range=(0, 255),
+    dataset_params=ds_params,
+    path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_context",
+    selected_dists=["0", "3", "5", "7", "9", "16", "30"],  # ignore .REMOVED_IMAGES/16 (imaging artifacts)
+)

cell_cycle/my_conf/dataset/ependymal_cutout/ependymal_cutout.yaml

@@ -0,0 +1,17 @@
+name: ependymal_cutout
+path: /projects/static2dynamic/datasets/ependymal/prepared_dataset_crop
+data_shape: [ 3, 256, 256 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [ 0, 255 ]
+expected_dtype: torch.uint8
+selected_dists: [ 1, 2, 3, 4, 5, 6 ] # 0 is the trash class!

cell_cycle/my_conf/dataset/ependymal_cutout/ependymal_cutout_01_noised_separate_gt_fully_ordered_inference.py

@@ -0,0 +1,19 @@
+from dataclasses import replace
+
+from GaussianProxy.utils.data import ContinuousTimeImageDataset
+from my_conf.dataset.ependymal_cutout.ependymal_cutout_01_noised_separate_gt_inference import dataset
+
+assert dataset.dataset_params is not None
+dataset_params = replace(
+    dataset.dataset_params,
+    dataset_class=ContinuousTimeImageDataset,
+)
+
+dataset = replace(
+    dataset,
+    fully_ordered=True,
+    path_to_single_parquet="/projects/static2dynamic/datasets/ependymal/ependymal_cutout_01_noised__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet",
+    dataset_params=dataset_params,
+    selected_dists=["all_imgs"],
+    separate_gt_starting_class_path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_noised_0.1_crop/ground_truths/1",
+)

cell_cycle/my_conf/dataset/ependymal_cutout/ependymal_cutout_01_noised_separate_gt_inference.py

@@ -0,0 +1,18 @@
+from dataclasses import replace
+from pathlib import Path
+
+from GaussianProxy.conf.dataset.ependymal_cutout.ependymal_cutout_inference import dataset
+
+# note the data organization is special with separate ground truth folder
+assert dataset.dataset_params is not None
+dataset_params = replace(
+    dataset.dataset_params,
+    sorting_func=lambda subdir: str(subdir.name) if isinstance(subdir, Path) else str(subdir),
+)  # subdirs are "ground_truth" and "all_imgs" now...
+
+dataset = replace(
+    dataset,
+    name="ependymal_cutout_01_noised",
+    path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_noised_0.1_crop/",
+    dataset_params=dataset_params,
+)

cell_cycle/my_conf/dataset/ependymal_cutout/ependymal_cutout_03_noised_separate_gt_fully_ordered_inference.py

@@ -0,0 +1,19 @@
+from dataclasses import replace
+
+from GaussianProxy.utils.data import ContinuousTimeImageDataset
+from my_conf.dataset.ependymal_cutout.ependymal_cutout_03_noised_separate_gt_inference import dataset
+
+assert dataset.dataset_params is not None
+dataset_params = replace(
+    dataset.dataset_params,
+    dataset_class=ContinuousTimeImageDataset,
+)
+
+dataset = replace(
+    dataset,
+    fully_ordered=True,
+    path_to_single_parquet="/projects/static2dynamic/datasets/ependymal/ependymal_cutout_03_noised__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet",
+    dataset_params=dataset_params,
+    selected_dists=["all_imgs"],
+    separate_gt_starting_class_path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_noised_0.3_crop/ground_truths/1",
+)

cell_cycle/my_conf/dataset/ependymal_cutout/ependymal_cutout_white_bg_separate_gt_fully_ordered_inference.py

@@ -0,0 +1,19 @@
+from dataclasses import replace
+
+from GaussianProxy.utils.data import ContinuousTimeImageDataset
+from my_conf.dataset.ependymal_cutout.ependymal_cutout_white_bg_separate_gt_inference import dataset
+
+assert dataset.dataset_params is not None
+dataset_params = replace(
+    dataset.dataset_params,
+    dataset_class=ContinuousTimeImageDataset,
+)
+
+dataset = replace(
+    dataset,
+    fully_ordered=True,
+    path_to_single_parquet="/projects/static2dynamic/datasets/ependymal/ependymal_cutout_white_bg__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet",
+    dataset_params=dataset_params,
+    selected_dists=["all_imgs"],
+    separate_gt_starting_class_path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_white_bg_crop/ground_truths/1",
+)

cell_cycle/my_conf/dataset/ependymal_cutout/ependymal_cutout_white_bg_separate_gt_inference.py

@@ -0,0 +1,16 @@
+from dataclasses import replace
+
+from GaussianProxy.conf.dataset.ependymal_cutout.ependymal_cutout_inference import dataset
+
+# note the data organization is special with separate ground truth folder
+assert dataset.dataset_params is not None
+dataset_params = replace(
+    dataset.dataset_params, sorting_func=lambda subdir: str(subdir.name)
+)  # subdirs are "ground_truth" and "all_imgs" now...
+
+dataset = replace(
+    dataset,
+    name="ependymal_cutout_white_bg",
+    path="/projects/static2dynamic/datasets/ependymal/prepared_dataset_white_bg_crop/",
+    dataset_params=dataset_params,
+)

cell_cycle/my_conf/dataset/human_embryo/human_embryo_fully_ordered_inference.py

@@ -0,0 +1,15 @@
+from dataclasses import replace
+
+from GaussianProxy.conf.dataset.human_embryo.human_embryo_inference import dataset
+from GaussianProxy.utils.data import ContinuousTimeImageDataset1D
+
+assert dataset.dataset_params is not None
+ds_params = replace(dataset.dataset_params, dataset_class=ContinuousTimeImageDataset1D)
+
+dataset = replace(
+    dataset,
+    path="/projects/imagesets3/2022_Gomez/reformated_phases/phases",
+    fully_ordered=True,
+    path_to_single_parquet="/projects/imagesets3/2022_Gomez/reformated_phases/human_embryo__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet",
+    dataset_params=ds_params,
+)

cell_cycle/my_conf/dataset/human_embryo/human_embryo_inference.py

@@ -0,0 +1,5 @@
+from dataclasses import replace
+
+from GaussianProxy.conf.dataset.human_embryo.human_embryo_inference import dataset
+
+dataset = replace(dataset, path="/projects/imagesets3/2022_Gomez/reformated_phases/phases")

docetaxel/my_conf/dataset/BBBC021/BBBC021_196_docetaxel_fully_ordered.yaml

@@ -0,0 +1,18 @@
+name: BBBC021_196_docetaxel_fully_ordered
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/BBBC021/196x196/docetaxel
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/BBBC021/196x196/BBBC021_196_docetaxel__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [3, 196, 196]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8
+fully_ordered: true

docetaxel/my_conf/dataset/BBBC021/BBBC021_196_docetaxel_hard_aug_inference.py

@@ -0,0 +1,4 @@
+from GaussianProxy.conf.dataset.BBBC021_196_hard_aug_inference import dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/BBBC021/196x196/docetaxel_hard_augmented"
+dataset.name += "_docetaxel"

docetaxel/my_conf/dataset/ChromaLive6h/ChromaLive6h_3ch_png.yaml

@@ -0,0 +1,19 @@
+name: chromaLive6h_3ch_png_patches_380px
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/20231017ChromaLive_6hr_4ch/MIP_normalized/paired_dataset/patches
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 128
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ] # move to [-1:1]
+    std: [ 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [ 0, 255 ]
+expected_dtype: torch.uint8
+selected_dists:

docetaxel/my_conf/dataset/ChromaLive6h/ChromaLive6h_4ch_tif.yaml

@@ -0,0 +1,19 @@
+name: chromaLive6h_4ch_tif_patches_380px
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/20231017ChromaLive_6hr_4ch/MIP_normalized/paired_dataset/patches
+data_shape: [ 4, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 128
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5, 0.5 ] # move to [-1:1]
+    std: [ 0.5, 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [ 0, 65536 ]
+expected_dtype: torch.uint16
+selected_dists: [ 'time_1', 'time_3', 'time_5', 'time_7', 'time_9', 'time_11', 'time_13' ]

docetaxel/my_conf/dataset/ChromaLive6h/chromalive6h_3ch_png_hard_aug_inference.py

@@ -0,0 +1,33 @@
+from torch import float32
+from torchvision.transforms import Compose, ConvertImageDtype, Normalize, Resize
+
+from GaussianProxy.conf.training_conf import DataSet, DatasetParams
+from GaussianProxy.utils.data import ImageDataset
+
+DEFINITION = 128
+NUMBER_OF_CHANNELS = 3
+
+transforms = Compose(
+    transforms=[
+        Resize(DEFINITION),
+        ConvertImageDtype(float32),
+        Normalize(mean=[0.5] * NUMBER_OF_CHANNELS, std=[0.5] * NUMBER_OF_CHANNELS),
+    ]
+)
+
+ds_params = DatasetParams(
+    file_extension="png",
+    key_transform=str,
+    sorting_func=lambda subdir: int(subdir.name.split("_")[1]),
+    dataset_class=ImageDataset,
+)
+
+dataset = DataSet(
+    name="chromaLive6h_3ch_png_patches_380px_hard_aug",
+    data_shape=(NUMBER_OF_CHANNELS, DEFINITION, DEFINITION),
+    transforms=transforms,
+    selected_dists=None,  # not used
+    expected_initial_data_range=(0, 255),
+    dataset_params=ds_params,
+    path="/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/20231017ChromaLive_6hr_4ch/MIP_normalized/paired_dataset/patches_hard_augmented",
+)

docetaxel/my_conf/dataset/ChromaLive6h/chromalive6h_3ch_png_inference.py

@@ -0,0 +1,33 @@
+from torch import float32
+from torchvision.transforms import Compose, ConvertImageDtype, Normalize, Resize
+
+from GaussianProxy.conf.training_conf import DataSet, DatasetParams
+from GaussianProxy.utils.data import ImageDataset
+
+DEFINITION = 128
+NUMBER_OF_CHANNELS = 3
+
+transforms = Compose(
+    transforms=[
+        Resize(DEFINITION),
+        ConvertImageDtype(float32),
+        Normalize(mean=[0.5] * NUMBER_OF_CHANNELS, std=[0.5] * NUMBER_OF_CHANNELS),
+    ]
+)
+
+ds_params = DatasetParams(
+    file_extension="png",
+    key_transform=str,
+    sorting_func=lambda subdir: int(subdir.name.split("_")[1]),
+    dataset_class=ImageDataset,
+)
+
+dataset = DataSet(
+    name="chromaLive6h_3ch_png_patches_380px",
+    data_shape=(NUMBER_OF_CHANNELS, DEFINITION, DEFINITION),
+    transforms=transforms,
+    selected_dists=None,  # not used
+    expected_initial_data_range=(0, 255),
+    dataset_params=ds_params,
+    path="/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/20231017ChromaLive_6hr_4ch/MIP_normalized/paired_dataset/patches",
+)

docetaxel/my_conf/dataset/ChromaLiveTL24h/ChromaLiveTL24h.yaml

@@ -0,0 +1,19 @@
+name: chromalive_tl_24h_380px
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/20230920ChromaLiveTL_24hr4ch/ch_4_3_1___norm_whole_ds_per_channel_per_zslice_0_99perc___patches_380
+data_shape: [3, 256, 256]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 256
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5] # move to [-1:1]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8
+selected_dists: ['time_1', 'time_7', 'time_13', 'time_19', 'time_25', 'time_31', 'time_37', 'time_43', 'time_49', 'time_55', 'time_61', 'time_67', 'time_73', 'time_79', 'time_85', 'time_91', 'time_97', 'time_103', 'time_109', 'time_115', 'time_121', 'time_127', 'time_133', 'time_139', 'time_145']

docetaxel/my_conf/dataset/Jurkat/Jurkat.yaml

@@ -0,0 +1,16 @@
+name: Jurkat
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/Jurkat/rgb_images_all_cell_cycles
+data_shape: [3, 66, 66]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8

docetaxel/my_conf/dataset/Jurkat/Jurkat_fully_ordered.yaml

@@ -0,0 +1,18 @@
+name: Jurkat_fully_ordered_dinov2_regs_giant_ds_preproc
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/Jurkat/rgb_images_all_cell_cycles
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/Jurkat/Jurkat__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [3, 66, 66]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8
+fully_ordered: true

docetaxel/my_conf/dataset/Jurkat/Jurkat_inference_hard_aug.py

@@ -0,0 +1,3 @@
+from GaussianProxy.conf.dataset.Jurkat_inference import Jurkat_inference as dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/Jurkat/rgb_images_all_cell_cycles_hard_augmented"

docetaxel/my_conf/dataset/NASH_fibrosis/NASH_fibrosis.yaml

@@ -0,0 +1,24 @@
+name: NASH_fibrosis
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/prepared_data/fibrosis
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  # Convert to float32 (and normalize to [0, 1]) before resizing
+  - _target_: torchvision.transforms.ConvertImageDtype
+    dtype: ${torch_dtype:float32}
+  # Random crop from 299x299 to 192x192, then resize to 128x128
+  - _target_: torchvision.transforms.RandomCrop
+    size: 192
+  - _target_: torchvision.transforms.Resize
+    size: 128
+  # Normalize to [-1, 1]
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  # Random 8x square augmentations
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+selected_dists:
+expected_initial_data_range: [ 0, 255 ]

docetaxel/my_conf/dataset/NASH_fibrosis/NASH_fibrosis_fully_ordered.yaml

@@ -0,0 +1,26 @@
+name: NASH_fibrosis_fully_ordered_dinov2_regs_giant_ds_preproc
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/fibrosis
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/NASH_fibrosis__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  # Convert to float32 (and normalize to [0, 1]) before resizing
+  - _target_: torchvision.transforms.ConvertImageDtype
+    dtype: ${torch_dtype:float32}
+  # Random crop from 299x299 to 192x192, then resize to 128x128
+  - _target_: torchvision.transforms.RandomCrop
+    size: 192
+  - _target_: torchvision.transforms.Resize
+    size: 128
+  # Normalize to [-1, 1]
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  # Random 8x square augmentations
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+selected_dists:
+expected_initial_data_range: [ 0, 255 ]
+fully_ordered: true

docetaxel/my_conf/dataset/NASH_fibrosis/NASH_fibrosis_inference.py

@@ -0,0 +1,3 @@
+from GaussianProxy.conf.dataset.NASH_fibrosis_inference import dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/prepared_data/fibrosis"

docetaxel/my_conf/dataset/NASH_steatosis/NASH_steatosis.yaml

@@ -0,0 +1,24 @@
+name: NASH_steatosis
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/steatosis
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  # Convert to float32 (and normalize to [0, 1]) before resizing
+  - _target_: torchvision.transforms.ConvertImageDtype
+    dtype: ${torch_dtype:float32}
+  # Random crop from 299x299 to 192x192, then resize to 128x128
+  - _target_: torchvision.transforms.RandomCrop
+    size: 192
+  - _target_: torchvision.transforms.Resize
+    size: 128
+  # Normalize to [-1, 1]
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  # Random 8x square augmentations
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+selected_dists:
+expected_initial_data_range: [ 0, 255 ]

docetaxel/my_conf/dataset/NASH_steatosis/NASH_steatosis_fully_ordered.yaml

@@ -0,0 +1,26 @@
+name: NASH_steatosis_fully_ordered_dinov2_regs_giant_ds_preproc
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/steatosis
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/NASH_steatosis__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  # Convert to float32 (and normalize to [0, 1]) before resizing
+  - _target_: torchvision.transforms.ConvertImageDtype
+    dtype: ${torch_dtype:float32}
+  # Random crop from 299x299 to 192x192, then resize to 128x128
+  - _target_: torchvision.transforms.RandomCrop
+    size: 192
+  - _target_: torchvision.transforms.Resize
+    size: 128
+  # Normalize to [-1, 1]
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  # Random 8x square augmentations
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+selected_dists:
+expected_initial_data_range: [ 0, 255 ]
+fully_ordered: true

docetaxel/my_conf/dataset/NASH_steatosis/NASH_steatosis_inference.py

@@ -0,0 +1,3 @@
+from GaussianProxy.conf.dataset.NASH_steatosis_inference import dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/NASH/prepared_data/steatosis"

docetaxel/my_conf/dataset/biotine/biotine_png_128.yaml

@@ -0,0 +1,19 @@
+name: biotine_png
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/biotine/3_channels_min_99_perc_normalized_rgb_stacks_png/patches_255
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 128
+  # ConvertImageDtype also scales to [0; 1] (from the *implicit* expected range that depends on the incoming dtype...)
+  - _target_: torchvision.transforms.ConvertImageDtype
+    dtype: ${torch_dtype:float32}
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+selected_dists:
+expected_initial_data_range: [ 0, 255 ]

docetaxel/my_conf/dataset/biotine/biotine_png_128_hard_aug.yaml

@@ -0,0 +1,16 @@
+name: biotine_png_hard_aug
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/biotine/3_channels_min_99_perc_normalized_rgb_stacks_png/patches_255_hard_augmented
+data_shape: [ 3, 128, 128 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 128
+  # ConvertImageDtype also scales to [0; 1] (from the *implicit* expected range that depends on the incoming dtype...)
+  - _target_: torchvision.transforms.ConvertImageDtype
+    dtype: ${torch_dtype:float32}
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+selected_dists:
+expected_initial_data_range: [ 0, 255 ]

docetaxel/my_conf/dataset/biotine/biotine_png_128_hard_aug_inference.py

@@ -0,0 +1,3 @@
+from GaussianProxy.conf.dataset.biotine_png_128_hard_aug_inference import dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/biotine/3_channels_min_99_perc_normalized_rgb_stacks_png/patches_255_hard_augmented"

docetaxel/my_conf/dataset/biotine/biotine_png_128_inference.py

@@ -0,0 +1,3 @@
+from GaussianProxy.conf.dataset.biotine_png_128_inference import dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/biotine/3_channels_min_99_perc_normalized_rgb_stacks_png/patches_255"

docetaxel/my_conf/dataset/biotine/biotine_png_256.yaml

@@ -0,0 +1,18 @@
+name: biotine_png
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/biotine/3_channels_min_99_perc_normalized_rgb_stacks_png/patches_255
+data_shape: [ 3, 256, 256 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32}
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 256
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+selected_dists: [ 1, 5, 10, 15, 19 ]
+expected_initial_data_range: [ 0, 255 ]

docetaxel/my_conf/dataset/diabetic_retinopathy/diabetic_retinopathy.yaml

@@ -0,0 +1,18 @@
+name: diabetic_retinopathy
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/DiabeticRetinopathy/prepared_dataset/train
+data_shape: [3, 256, 256]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.transforms.Resize
+    size: 256 # single int => image resized to (size * aspect_ratio, size) or (size, size * aspect_ratio) with aspect_ratio >= 1 preserved
+  - _target_: torchvision.transforms.v2.CenterCrop
+    size: 256 # square centered crop
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8

docetaxel/my_conf/dataset/diabetic_retinopathy/diabetic_retinopathy_fully_ordered.yaml

@@ -0,0 +1,20 @@
+name: diabetic_retinopathy_fully_ordered
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/DiabeticRetinopathy/train
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/DiabeticRetinopathy/diabetic_retinopathy__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [3, 256, 256]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+    - _target_: torchvision.transforms.transforms.Resize
+      size: 256 # single int => image resized to (size * aspect_ratio, size) or (size, size * aspect_ratio) with aspect_ratio >= 1 preserved
+    - _target_: torchvision.transforms.v2.CenterCrop
+      size: 256 # square centered crop
+    - _target_: torchvision.transforms.RandomHorizontalFlip
+    - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+      dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+    - _target_: torchvision.transforms.Normalize
+      mean: [0.5, 0.5, 0.5]
+      std: [0.5, 0.5, 0.5]
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8
+fully_ordered: true

docetaxel/my_conf/dataset/diabetic_retinopathy/diabetic_retinopathy_hard_aug_inference.py

@@ -0,0 +1,4 @@
+from GaussianProxy.conf.dataset.diabetic_retinopathy_inference import diabetic_retinopathy_inference as dataset
+
+dataset.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/DiabeticRetinopathy/prepared_dataset/train_hard_augmented"
+dataset.name = "diabetic_retinopathy_inference_hard_augmented"

docetaxel/my_conf/dataset/diabetic_retinopathy/diabetic_retinopathy_inference.py

@@ -0,0 +1,3 @@
+from GaussianProxy.conf.dataset.diabetic_retinopathy_inference import diabetic_retinopathy_inference
+
+diabetic_retinopathy_inference.path = "/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/DiabeticRetinopathy/prepared_dataset/train"

docetaxel/my_conf/dataset/ependymal_context/ependymal_context.yaml

@@ -0,0 +1,16 @@
+name: ependymal_context
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/prepared_dataset_context
+data_shape: [3, 256, 256]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8

docetaxel/my_conf/dataset/ependymal_context/ependymal_context_fully_ordered.yaml

@@ -0,0 +1,18 @@
+name: ependymal_context_fully_ordered
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/prepared_dataset_context
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/ependymal_context__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [3, 256, 256]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8
+fully_ordered: true

docetaxel/my_conf/dataset/ependymal_context/ependymal_context_inference.py

@@ -0,0 +1,32 @@
+from torch import float32
+from torchvision.transforms import Compose, ConvertImageDtype, Normalize
+
+from GaussianProxy.conf.training_conf import DataSet, DatasetParams
+from GaussianProxy.utils.data import ImageDataset
+
+DEFINITION = 256
+NUMBER_OF_CHANNELS = 3
+
+transforms = Compose(
+    transforms=[
+        ConvertImageDtype(float32),
+        Normalize(mean=[0.5] * NUMBER_OF_CHANNELS, std=[0.5] * NUMBER_OF_CHANNELS),
+    ]
+)
+
+ds_params = DatasetParams(
+    file_extension="png",
+    key_transform=str,
+    sorting_func=lambda subdir: int(subdir.name),
+    dataset_class=ImageDataset,
+)
+
+dataset = DataSet(
+    name="ependymal_context",
+    data_shape=(NUMBER_OF_CHANNELS, DEFINITION, DEFINITION),
+    transforms=transforms,
+    selected_dists=None,
+    expected_initial_data_range=(0, 255),
+    dataset_params=ds_params,
+    path="/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/prepared_dataset_context",
+)

docetaxel/my_conf/dataset/ependymal_cutout/ependymal_cutout.yaml

@@ -0,0 +1,17 @@
+name: ependymal_cutout
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/prepared_dataset_crop
+data_shape: [ 3, 256, 256 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [ 0, 255 ]
+expected_dtype: torch.uint8
+selected_dists: [ 1, 2, 3, 4, 5, 6 ] # 0 is the trash class!

docetaxel/my_conf/dataset/ependymal_cutout/ependymal_cutout_fully_ordered.yaml

@@ -0,0 +1,19 @@
+name: ependymal_cutout_fully_ordered
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/prepared_dataset_crop
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/ependymal_cutout__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+data_shape: [ 3, 256, 256 ]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [ 0.5, 0.5, 0.5 ]
+    std: [ 0.5, 0.5, 0.5 ]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [ 0, 255 ]
+expected_dtype: torch.uint8
+selected_dists: [ 1, 2, 3, 4, 5, 6 ] # 0 is the trash class!
+fully_ordered: true

docetaxel/my_conf/dataset/ependymal_cutout/ependymal_cutout_inference.py

@@ -0,0 +1,32 @@
+from torch import float32
+from torchvision.transforms import Compose, ConvertImageDtype, Normalize
+
+from GaussianProxy.conf.training_conf import DataSet, DatasetParams
+from GaussianProxy.utils.data import ImageDataset
+
+DEFINITION = 256
+NUMBER_OF_CHANNELS = 3
+
+transforms = Compose(
+    transforms=[
+        ConvertImageDtype(float32),
+        Normalize(mean=[0.5] * NUMBER_OF_CHANNELS, std=[0.5] * NUMBER_OF_CHANNELS),
+    ]
+)
+
+ds_params = DatasetParams(
+    file_extension="png",
+    key_transform=str,
+    sorting_func=lambda subdir: int(subdir.name),
+    dataset_class=ImageDataset,
+)
+
+dataset = DataSet(
+    name="ependymal_cutout",
+    data_shape=(NUMBER_OF_CHANNELS, DEFINITION, DEFINITION),
+    transforms=transforms,
+    selected_dists=[1, 2, 3, 4, 5, 6],  # 0 is the trash class!
+    expected_initial_data_range=(0, 255),
+    dataset_params=ds_params,
+    path="/lustre/fsn1/projects/rech/icr/ufc43hj/datasets/ependymal/prepared_dataset_crop",
+)

docetaxel/my_conf/net/net_256_3_20M.py

@@ -0,0 +1,22 @@
+from GaussianProxy.conf.training_conf import UNet2DConditionModelConfig, TimeEncoderConfig
+
+cross_attn_dim = 64
+
+net = UNet2DConditionModelConfig(
+    sample_size=256,
+    in_channels=3,
+    out_channels=3,
+    down_block_types=("DownBlock2D", "CrossAttnDownBlock2D", "CrossAttnDownBlock2D"),
+    up_block_types=("CrossAttnUpBlock2D", "CrossAttnUpBlock2D", "UpBlock2D"),
+    block_out_channels=(64, 128, 224),
+    layers_per_block=2,
+    act_fn="silu",
+    cross_attention_dim=cross_attn_dim,
+)
+
+time_encoder = TimeEncoderConfig(
+    encoding_dim=128,
+    time_embed_dim=cross_attn_dim,
+    flip_sin_to_cos=True,
+    downscale_freq_shift=1,
+)

docetaxel/my_conf/scheduler/DDIM_3k_vpred_tresh_leading.json

@@ -0,0 +1,19 @@
+{
+    "_class_name": "DDIMScheduler",
+    "_diffusers_version": "0.32.2",
+    "beta_end": 0.02,
+    "beta_schedule": "linear",
+    "beta_start": 0.0001,
+    "clip_sample": false,
+    "clip_sample_range": 1.0,
+    "dynamic_thresholding_ratio": 0.995,
+    "num_train_timesteps": 3000,
+    "prediction_type": "v_prediction",
+    "rescale_betas_zero_snr": false,
+    "sample_max_value": 1.0,
+    "set_alpha_to_one": true,
+    "steps_offset": 0,
+    "thresholding": true,
+    "timestep_spacing": "leading",
+    "trained_betas": null
+}

docetaxel_skip_half_doses/dynamic/scheduler_config.json

@@ -0,0 +1,19 @@
+{
+  "_class_name": "DDIMScheduler",
+  "_diffusers_version": "0.35.2",
+  "beta_end": 0.02,
+  "beta_schedule": "linear",
+  "beta_start": 0.0001,
+  "clip_sample": false,
+  "clip_sample_range": 1.0,
+  "dynamic_thresholding_ratio": 0.995,
+  "num_train_timesteps": 3000,
+  "prediction_type": "v_prediction",
+  "rescale_betas_zero_snr": false,
+  "sample_max_value": 1.0,
+  "set_alpha_to_one": true,
+  "steps_offset": 0,
+  "thresholding": true,
+  "timestep_spacing": "leading",
+  "trained_betas": null
+}

docetaxel_skip_half_doses/my_conf/dataset/BBBC021/BBBC021_196_docetaxel.yaml

@@ -0,0 +1,16 @@
+name: BBBC021_196_docetaxel
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/BBBC021/196x196/docetaxel
+data_shape: [3, 196, 196]
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8

docetaxel_skip_half_doses/my_conf/dataset/BBBC021/BBBC021_196_docetaxel_fully_ordered.yaml

@@ -0,0 +1,23 @@
+name: BBBC021_196_docetaxel_fully_ordered
+
+path: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/BBBC021/196x196/docetaxel
+path_to_single_parquet: /lustre/fsn1/projects/rech/icr/ufc43hj/datasets/BBBC021/196x196/BBBC021_196_docetaxel__continuous_time_predictions__facebook_dinov2-with-registers-giant_dataset_preproc.parquet
+
+data_shape: [3, 196, 196]
+
+transforms:
+  _target_: torchvision.transforms.transforms.Compose
+  transforms:
+  - _target_: torchvision.transforms.ConvertImageDtype # this also scales to [0; 1]
+    dtype: ${torch_dtype:float32} # passed dtype must be accessible as a "torch" attribute
+  - _target_: torchvision.transforms.Normalize
+    mean: [0.5, 0.5, 0.5]
+    std: [0.5, 0.5, 0.5]
+  - _target_: torchvision.transforms.RandomHorizontalFlip
+  - _target_: torchvision.transforms.RandomVerticalFlip
+  - _target_: GaussianProxy.utils.data.RandomRotationSquareSymmetry
+
+expected_initial_data_range: [0, 255]
+expected_dtype: torch.uint8
+
+fully_ordered: true