Functions — github.com/amazon-science/patchcore-inspection

hub / github.com/amazon-science/patchcore-inspection / functions

Functions132 in github.com/amazon-science/patchcore-inspection

⨍ Functions 132◇ Types & classes 22 ↳ Endpoints 3

↓ 19 callersMethodrun

Returns distances and indices of nearest neighbour search. Args: query_features: Features to retrieve. index

src/patchcore/common.py:66

↓ 9 callersMethodfit

Adds features to the FAISS search index. Args: features: Array of size NxD.

src/patchcore/common.py:50

↓ 9 callersMethodmerge

(self, features: list)

src/patchcore/common.py:124

↓ 8 callersMethodpredict

(self, data)

src/patchcore/patchcore.py:178

↓ 6 callersMethodload

( self, backbone, layers_to_extract_from, device, input_shape,

src/patchcore/patchcore.py:25

↓ 5 callersFunction_dummy_features

(feature_dimension)

test/test_sampler.py:8

↓ 4 callersMethod__init__

FAISS Nearest neighbourhood search. Args: on_gpu: If set true, nearest neighbour searches are done on GPU. num_worker

src/patchcore/common.py:15

↓ 4 callersMethod_embed

Returns feature embeddings for images.

src/patchcore/patchcore.py:91

↓ 4 callersFunction_standard_patchcore

(image_dimension)

test/test_patchcore.py:43

↓ 4 callersMethodload

(self, filename: str)

src/patchcore/common.py:91

↓ 3 callersMethod__init__

(self, percentage: float)

src/patchcore/sampler.py:17

↓ 3 callersMethod_compute_batchwise_differences

Computes batchwise Euclidean distances using PyTorch.

src/patchcore/sampler.py:81

↓ 3 callersMethod_detection_file

(folder, prepend="")

src/patchcore/common.py:353

↓ 3 callersFunction_dummy_constant_dataloader

(number_of_examples, shape_of_examples)

test/test_patchcore.py:15

↓ 3 callersMethodload_from_path

( self, load_path: str, device: torch.device, nn_method: patchcore.common.Fais

src/patchcore/patchcore.py:256

↓ 3 callersMethodsave

(self, filename: str)

src/patchcore/common.py:88

↓ 2 callersMethod_create_index

(self, dimension)

src/patchcore/common.py:43

↓ 2 callersFunction_dummy_various_dataloader

(number_of_examples, shape_of_examples)

test/test_patchcore.py:28

↓ 2 callersMethod_index_file

(folder, prepend="")

src/patchcore/common.py:357

↓ 2 callersMethod_index_to_gpu

(self, index)

src/patchcore/common.py:29

↓ 2 callersMethod_params_file

(filepath, prepend="")

src/patchcore/patchcore.py:231

↓ 2 callersMethod_predict

Infer score and mask for a batch of images.

src/patchcore/patchcore.py:203

↓ 2 callersMethod_train

(self, _index, _features)

src/patchcore/common.py:63

↓ 2 callersMethodreset_index

(self)

src/patchcore/common.py:94

↓ 2 callersMethodsave_to_path

(self, save_path: str, prepend: str = "")

src/patchcore/patchcore.py:234

↓ 2 callersMethodunpatch_scores

(self, x, batchsize)

src/patchcore/patchcore.py:310

↓ 1 callersFunction_approximate_greedycoreset_sampler_with_reduction

( sampling_percentage, johnsonlindenstrauss_dim )

test/test_patchcore.py:72

↓ 1 callersMethod_compute_greedy_coreset_indices

Runs iterative greedy coreset selection. Args: features: [NxD] input feature bank to sample.

src/patchcore/sampler.py:91

↓ 1 callersFunction_dummy_constant_features

(number_of_examples, feature_dimension)

test/test_sampler.py:17

↓ 1 callersFunction_dummy_features

(number_of_examples, shape_of_examples)

test/test_patchcore.py:9

↓ 1 callersFunction_dummy_images

(number_of_examples, image_shape)

test/test_patchcore.py:33

↓ 1 callersFunction_dummy_various_features

(number_of_examples, shape_of_examples)

test/test_patchcore.py:20

↓ 1 callersMethod_fill_memory_bank

Computes and sets the support features for SPADE.

src/patchcore/patchcore.py:155

↓ 1 callersMethod_gpu_cloner_options

(self)

src/patchcore/common.py:26

↓ 1 callersMethod_index_to_cpu

(self, index)

src/patchcore/common.py:38

↓ 1 callersMethod_load

(filename: str)

src/patchcore/common.py:368

↓ 1 callersMethod_predict_dataloader

This function provides anomaly scores/maps for full dataloaders.

src/patchcore/patchcore.py:183

↓ 1 callersMethod_reduce

(features)

src/patchcore/common.py:140

↓ 1 callersMethod_reduce_features

(self, features)

src/patchcore/sampler.py:52

↓ 1 callersMethod_restore_type

(self, features: torch.Tensor)

src/patchcore/sampler.py:33

↓ 1 callersMethod_save

(filename, features)

src/patchcore/common.py:361

↓ 1 callersMethod_store_type

(self, features: Union[torch.Tensor, np.ndarray])

src/patchcore/sampler.py:28

↓ 1 callersMethodconvert_to_segmentation

(self, patch_scores)

src/patchcore/common.py:192

↓ 1 callersMethodfeature_dimensions

Computes the feature dimensions for all layers given input_shape.

src/patchcore/common.py:270

↓ 1 callersMethodfit

PatchCore training. This function computes the embeddings of the training data and fills the memory bank of SPADE.

src/patchcore/patchcore.py:147

↓ 1 callersMethodget_image_data

(self)

src/patchcore/datasets/mvtec.py:115

↓ 1 callersFunctionimage_transform

(image)

bin/run_patchcore.py:147

↓ 1 callersFunctionmain

(**kwargs)

bin/load_and_evaluate_patchcore.py:27

↓ 1 callersFunctionmain

(**kwargs)

bin/run_patchcore.py:30

↓ 1 callersFunctionmask_transform

(mask)

bin/run_patchcore.py:159

↓ 1 callersMethodpatchify

Convert a tensor into a tensor of respective patches. Args: x: [torch.Tensor, bs x c x w x h] Returns: x: [tor

src/patchcore/patchcore.py:282

↓ 1 callersMethodrun

Subsamples features using Greedy Coreset. Args: features: [N x D]

src/patchcore/sampler.py:62

↓ 1 callersMethodrun

Randomly samples input feature collection. Args: features: [N x D]

src/patchcore/sampler.py:178

↓ 1 callersMethodsave

( self, save_folder: str, save_features_separately: bool = False, prepend: str

src/patchcore/common.py:372

↓ 1 callersMethodscore

(self, x)

src/patchcore/patchcore.py:313

Method__call__

(self, module, input, output)

src/patchcore/common.py:285

Method__getitem__

(self, idx)

src/patchcore/datasets/mvtec.py:91

Method__init__

PatchCore anomaly detection class.

src/patchcore/patchcore.py:20

Method__init__

(self, patchsize, stride=None)

src/patchcore/patchcore.py:278

Method__init__

Merges feature embedding by name.

src/patchcore/common.py:121

Method__init__

(self, input_dims, output_dim)

src/patchcore/common.py:146

Method__init__

(self, preprocessing_dim)

src/patchcore/common.py:164

Method__init__

(self, target_dim)

src/patchcore/common.py:174

Method__init__

(self, device, target_size=224)

src/patchcore/common.py:187

Method__init__

(self, backbone, layers_to_extract_from, device)

src/patchcore/common.py:214

Method__init__

(self, hook_dict, layer_name: str, last_layer_to_extract: str)

src/patchcore/common.py:278

Method__init__

Neearest-Neighbourhood Anomaly Scorer class. Args: n_nearest_neighbours: [int] Number of nearest neighbours used to

src/patchcore/common.py:297

Method__init__

Greedy Coreset sampling base class.

src/patchcore/sampler.py:40

Method__init__

Approximate Greedy Coreset sampling base class.

src/patchcore/sampler.py:119

Method__init__

(self, percentage: float)

src/patchcore/sampler.py:175

Method__init__

Args: source: [str]. Path to the MVTec data folder. classname: [str or None]. Name of MVTec class that should be

src/patchcore/datasets/mvtec.py:41

Method__len__

(self)

src/patchcore/datasets/mvtec.py:112

Method_compute_greedy_coreset_indices

Runs approximate iterative greedy coreset selection. This greedy coreset implementation does not require computation of the full N x

src/patchcore/sampler.py:130

Method_create_index

(self, dimension)

src/patchcore/common.py:109

Method_detach

(features)

src/patchcore/patchcore.py:94

Function_dummy_image_random_dataloader

(number_of_examples, image_shape)

test/test_patchcore.py:38

Method_gpu_cloner_options

(self)

src/patchcore/common.py:104

Method_image_to_features

(input_image)

src/patchcore/patchcore.py:159

Function_load_patchcore_from_path

(load_path)

test/test_patchcore.py:61

Method_reduce

(features)

src/patchcore/common.py:131

Method_train

(self, index, features)

src/patchcore/common.py:101

Functioncompute_and_store_final_results

Store computed results as CSV file. Args: results_path: [str] Where to store result csv. results: [List[List]] List of lists cont

src/patchcore/utils.py:127

Functioncompute_imagewise_retrieval_metrics

Computes retrieval statistics (AUROC, FPR, TPR). Args: anomaly_prediction_weights: [np.array or list] [N] Assignment weights

src/patchcore/metrics.py:6

Functioncompute_pixelwise_retrieval_metrics

Computes pixel-wise statistics (AUROC, FPR, TPR) for anomaly segmentations and ground truth segmentation masks. Args: anomaly_se

src/patchcore/metrics.py:28

Functioncreate_storage_folder

( main_folder_path, project_folder, group_folder, mode="iterate" )

src/patchcore/utils.py:77

Functiondataset

( name, data_path, subdatasets, batch_size, resize, imagesize, num_workers, augment )

bin/load_and_evaluate_patchcore.py:247

Functiondataset

( name, data_path, subdatasets, train_val_split, batch_size, resize, imagesize,

bin/run_patchcore.py:343

Methodembed

(self, data)

src/patchcore/patchcore.py:79

Methodfit

Calls the fit function of the nearest neighbour method. Args: detection_features: [list of np.arrays] [[bs x d_i]

src/patchcore/common.py:316

Functionfix_seeds

Fixed available seeds for reproducibility. Args: seed: [int] Seed value. with_torch: Flag. If true, torch-related seeds are fixed

src/patchcore/utils.py:109

Methodforward

(self, features)

src/patchcore/common.py:156

Methodforward

(self, features)

src/patchcore/common.py:168

Methodforward

Returns reshaped and average pooled features.

src/patchcore/common.py:178

Methodforward

(self, images)

src/patchcore/common.py:259

Functionget_dataloaders

(seed)

bin/run_patchcore.py:357

Functionget_dataloaders_iter

(seed)

bin/load_and_evaluate_patchcore.py:253

Functionget_patchcore

(input_shape, sampler, device)

bin/run_patchcore.py:284

Functionget_patchcore_iter

(device)

bin/load_and_evaluate_patchcore.py:205

Functionget_sampler

(device)

bin/run_patchcore.py:322

Functionimage_transform

(image)

bin/load_and_evaluate_patchcore.py:122

next →1–100 of 132, ranked by callers