diff --git a/Inference_SinglePhoton.py b/Inference_SinglePhoton.py index 455ac90..b28c469 100644 --- a/Inference_SinglePhoton.py +++ b/Inference_SinglePhoton.py @@ -14,30 +14,36 @@ np.random.seed(42) torch.backends.cudnn.deterministic = True torch.backends.cudnn.benchmark = False +NX, NY = 400, 400 + configs = {} configs['SiemenStarLowerLeft'] = { - 'dataFiles': [f'/mnt/sls_det_storage/moench_data/MLXID/Samples/Measurement/2504_SOLEIL_SiemenStarClusters_MOENCH040_150V/SiemenStarLowerLeft/clusters_chunk{i}.h5' for i in range(200)], # 200 files + # 'dataFiles': [f'/mnt/sls_det_storage/moench_data/MLXID/Samples/Measurement/2504_SOLEIL_SiemenStarClusters_MOENCH040_150V/SiemenStarLowerLeft/clusters_chunk{i}.h5' for i in range(1)], # 200 files, no zeroing pixels outside the cluster + 'dataFiles': [f'/home/xie_x1/MLXID/DataProcess/Samples/SiemenStarLowerLeft/1Photon_CS3_chunk{i}.h5' for i in range(1)], # 160 files, no zeroing pixels outside the cluster 'modelVersion': '251022', + 'energy': 15, # keV 'roi': [140, 230, 120, 210], # x_min, x_max, y_min, y_max, 'noise': 0.13, # keV; for the model selection } configs['SiemenStarLowerRight'] = { - 'dataFiles': [f'/home/xie_x1/MLXID/DataProcess/Samples/SiemenStarLowerRight/1Photon_CS3_chunk{i}.h5' for i in range(320)], # 320 files + 'dataFiles': [f'/home/xie_x1/MLXID/DataProcess/Samples/SiemenStarLowerRight/1Photon_CS3_chunk{i}.h5' for i in range(320)], # 320 files. !!! zeroed pixels outside the cluster, to be fixed 'modelVersion': '251022', + 'energy': 15, # keV 'roi': [235, 345, 110, 220], # x_min, x_max, y_min, y_max, 'noise': 0.13, # keV } -task = 'SiemenStarLowerRight' +task = 'SiemenStarLowerLeft' config = configs[task] BinningFactor = 10 -numberOfAugOps = 8 +numberOfAugOps = 1 +flag_normalize = False Roi = config['roi'] X_st, X_ed, Y_st, Y_ed = Roi -mlSuperFrame = np.zeros(((Y_ed-Y_st)*BinningFactor, (X_ed-X_st)*BinningFactor)) -countFrame = np.zeros((Y_ed-Y_st, X_ed-X_st)) +mlSuperFrame = np.zeros((NY*BinningFactor, NX*BinningFactor)) +countFrame = np.zeros((NY, NX)) subpixelDistribution = np.zeros((BinningFactor, BinningFactor)) def inv0(p): return p @@ -70,7 +76,10 @@ def apply_inverse_transforms(predictions: torch.Tensor, numberOfAugOps: int) -> if __name__ == "__main__": model = models.get_model_class(config['modelVersion'])().cuda() - model.load_state_dict(torch.load(f'/home/xie_x1/MLXID/DeepLearning/Models/singlePhoton{config["modelVersion"]}_15keV_Noise{config["noise"]}keV_E300_aug8.pth', weights_only=True)) + modelName = f'singlePhoton{config["modelVersion"]}_{config["energy"]}keV_Noise{config["noise"]}keV_aug1' + if flag_normalize: + modelName += '_normalized' + model.load_state_dict(torch.load(f'/home/xie_x1/MLXID/DeepLearning/Models/{modelName}.pth', weights_only=True)) nChunks = np.ceil(len(config['dataFiles']) / 16).astype(int) for idxChunk in range(nChunks): predictions = [] @@ -83,7 +92,8 @@ if __name__ == "__main__": sampleFiles, sampleRatio=1, datasetName='Inference', - numberOfAugOps=numberOfAugOps + numberOfAugOps=numberOfAugOps, + normalize=flag_normalize ) dataLoader = torch.utils.data.DataLoader( dataset, @@ -104,71 +114,49 @@ if __name__ == "__main__": _chunk_predictions.append(outputs) predictions.extend(_chunk_predictions) - ### save samples and inferred positions - # _h5_file = h5py.File(f'InferredSamples/Chunk{idxChunk}.h5', 'w') - # dset_1Photon_clusters = _h5_file.create_dataset( - # 'clusters', (0, 5, 5), maxshape=(None, 5, 5), dtype='f4', - # chunks=True, compression='gzip' - # ) - # dset_1photon_label = _h5_file.create_dataset( - # 'labels', (0, 4), maxshape=(None, 4), dtype='f4', - # chunks=True - # ) - # _len = dataset.samples.shape[0] - # dset_1Photon_clusters.resize((_len, 5, 5)) - # dset_1photon_label.resize((_len, 4)) - # _chunk_samples = np.zeros(( _len, 5, 5), dtype=np.float32) - # _chunk_samples[:, 1:-1, 1:-1] = dataset.samples[:, 0, :, :] - # dset_1Photon_clusters[:] = _chunk_samples - - # _chunk_predictions = torch.cat(_chunk_predictions, dim=0) - # _chunk_predictions = apply_inverse_transforms(_chunk_predictions, numberOfAugOps) - # _chunk_labels = np.zeros((_len, 4), dtype=np.float32) - # _chunk_labels[:, :2] = _chunk_predictions.numpy() - # dset_1photon_label[:] = _chunk_labels - # _h5_file.close() - - # np.savez(f'InferredSamples/Chunk{idxChunk}.npz', samples=_chunk_samples, labels=_chunk_labels) - - predictions = torch.cat(predictions, dim=0) predictions = apply_inverse_transforms(predictions, numberOfAugOps) predictions += torch.tensor([1.5, 1.5]).unsqueeze(0) # adjust back to original coordinate system referencePoints = np.concatenate(referencePoints, axis=0) print(f'mean x = {torch.mean(predictions[:, 0])}, std x = {torch.std(predictions[:, 0])}') print(f'mean y = {torch.mean(predictions[:, 1])}, std y = {torch.std(predictions[:, 1])}') - absolutePositions = predictions.numpy() + referencePoints[:, :2] - 1 + absolutePositions = predictions.numpy() + referencePoints[:, :2] - hit_x = np.floor((absolutePositions[:, 0] - Roi[0]) * BinningFactor).astype(int) - hit_x = np.clip(hit_x, 0, mlSuperFrame.shape[1]-1) - hit_y = np.floor((absolutePositions[:, 1] - Roi[2]) * BinningFactor).astype(int) - hit_y = np.clip(hit_y, 0, mlSuperFrame.shape[0]-1) + hit_x = np.floor(absolutePositions[:, 0] * BinningFactor).astype(int) + hit_y = np.floor(absolutePositions[:, 1] * BinningFactor).astype(int) np.add.at(mlSuperFrame, (hit_y, hit_x), 1) - np.add.at(countFrame, ((referencePoints[:, 1] - Roi[2]).astype(int), - (referencePoints[:, 0] - Roi[0]).astype(int)), 1) + np.add.at(countFrame, ((referencePoints[:, 1] + 1).astype(int), + (referencePoints[:, 0] + 1).astype(int)), 1) ### the reference points refer to the lower-left corner of the pixel, so add 1 to get the pixel index np.add.at(subpixelDistribution, (np.floor((absolutePositions[:, 1] % 1) * BinningFactor).astype(int), np.floor((absolutePositions[:, 0] % 1) * BinningFactor).astype(int)), 1) + ### Save results and plots import os - os.makedirs(f'InferenceResults/{task}', exist_ok=True) + outputDir = f'InferenceResults/{task}/{config["modelVersion"]}_{config["energy"]}keV_Noise{config["noise"]}keV_augX{numberOfAugOps}' + if flag_normalize: + outputDir += '_normalized' + os.makedirs(outputDir, exist_ok=True) plt.clf() + mlSuperFrame = mlSuperFrame[config['roi'][2]*BinningFactor : config['roi'][3]*BinningFactor, config['roi'][0]*BinningFactor : config['roi'][1]*BinningFactor] + average = np.mean(mlSuperFrame) plt.imshow(mlSuperFrame, origin='lower', extent=[Y_st, Y_ed, X_st, X_ed]) plt.colorbar() - plt.savefig(f'InferenceResults/{task}/ML_1Photon_superFrame.png', dpi=300) - np.save(f'InferenceResults/{task}/ML_1Photon_superFrame.npy', mlSuperFrame) + plt.savefig(f'{outputDir}/1Photon_ML_superFrame.png', dpi=300) + np.save(f'{outputDir}/1Photon_ML_superFrame.npy', mlSuperFrame) plt.clf() + countFrame = countFrame[config['roi'][2] : config['roi'][3], config['roi'][0] : config['roi'][1]] plt.imshow(countFrame, origin='lower', extent=[Y_st, Y_ed, X_st, X_ed]) plt.colorbar() - plt.savefig(f'InferenceResults/{task}/count_Frame.png', dpi=300) - np.save(f'InferenceResults/{task}/count_Frame.npy', countFrame) + plt.savefig(f'{outputDir}/1Photon_count_Frame.png', dpi=300) + np.save(f'{outputDir}/1Photon_count_Frame.npy', countFrame) plt.clf() plt.imshow(subpixelDistribution, origin='lower', extent=[0, BinningFactor, 0, BinningFactor]) plt.colorbar() - plt.savefig(f'InferenceResults/{task}/subpixel_1Photon_Distribution.png', dpi=300) - np.save(f'InferenceResults/{task}/subpixel_1Photon_Distribution.npy', subpixelDistribution) \ No newline at end of file + plt.savefig(f'{outputDir}/1Photon_subpixel_Distribution.png', dpi=300) + np.save(f'{outputDir}/1Photon_subpixel_Distribution.npy', subpixelDistribution) \ No newline at end of file