diff --git a/tests/test_downsampling.py b/tests/test_downsampling.py index f807554..b25393f 100644 --- a/tests/test_downsampling.py +++ b/tests/test_downsampling.py @@ -1,5 +1,7 @@ +import numpy import os +from uips.options import UIPSOptions, stepOptions from uips import UIPS_INPUT_DIR from uips.wrapper import ( downsample_dataset_from_input, @@ -7,7 +9,7 @@ ) -def test_nf_input(): +def test_nf_input_file(): input_file = os.path.join(UIPS_INPUT_DIR, "input_test") downsample_dataset_from_input_file(input_file) @@ -17,6 +19,39 @@ def test_bins_input(): downsample_dataset_from_input_file(input_file) +def test_nf_input(): + + step1_options = stepOptions( + nEpochs = 1, + batch_size_train = 2048, + nWorkingData = 1e4, + learning_rate = 2e-4, + nCouplingLayer = 2, + num_bins = 3, + hidden_features = 12, + num_blocks = 3 + ) + step2_options = step1_options + downsampler_options = UIPSOptions( + num_pdf_iter = 2, + pdfMethod = "NormalizingFlow", + use_gpu = False, + printTiming = True, + nDatReduced = 1e5, + preShuffled = True, + scalerFile = "scaler.npz", + nSamples_list = [1e3], + computeDistanceCriterion = True, + prefixDownsampledData = "downSampledData", + data_freq_adjustment = 1, + nWorkingDataAdjustment = -1, + ) + + dataFile = "../data/combustion2DToDownsampleSmall.npy" + dataset = np.load(dataFile) + downsample_dataset_from_input(downsampler_options, dataset) + + if __name__ == "__main__": test_nf_input() test_bins_input() diff --git a/tutorials/gaussian_2d_example.ipynb b/tutorials/gaussian_2d_example.ipynb new file mode 100644 index 0000000..11c1de1 --- /dev/null +++ b/tutorials/gaussian_2d_example.ipynb @@ -0,0 +1,202 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "from uips.options import UIPSOptions, stepOptions\n", + "from uips import UIPS_INPUT_DIR\n", + "from uips.wrapper import downsample_dataset_from_input\n", + "import numpy as np\n", + "\n", + "import matplotlib.pyplot as plt\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Dataset has 100000 samples of dimension 2\n", + "WARNING: Only 100000 samples, this may not work\n", + "LOAD DATA ... DONE!\n", + "RESCALE DATA ... DONE!\n", + "RANDOM: \n", + "\t nSample 1000 mean dist = 0.0662, std dist = 0.0728\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\t nSample 100000 mean dist = 0.0068, std dist = 0.0109\n", + "There are 2494 trainable parameters in this model.\n", + "Loss 2.5486 Step 980 / 980 |██████████████████████████████████████████████████| 100.0% Complete\n", + "Time Train : 44.28s\n", + "Eval Step 4 / 4 |██████████████████████████████████████████████████| 100.0% Complete\n", + "Time Eval : 0.86s\n", + "TRAIN ITER 0\n", + "\t nSample 1000 mean dist = 0.1046, std dist = 0.0780\n", + "\t nSample 100000 mean dist = 0.0068, std dist = 0.0109\n", + "There are 2806 trainable parameters in this model.\n", + "Loss 2.5585 Step 980 / 980 |██████████████████████████████████████████████████| 100.0% Complete\n", + "Time Train : 55.51s\n", + "Eval Step 4 / 4 |██████████████████████████████████████████████████| 100.0% Complete\n", + "Time Eval : 0.87s\n", + "TRAIN ITER 1\n", + "\t nSample 1000 mean dist = 0.1044, std dist = 0.0789\n", + "WARNING: Possible failure of the sampling procedure for nSample : 1000\n", + "\t nSample 100000 mean dist = 0.0068, std dist = 0.0109\n", + "\n", + "\n", + "For sample 1000 use downSampledData_1000_it0.npz\n", + "\n", + "\n" + ] + } + ], + "source": [ + "ndat = int(1e5)\n", + "nSampl = 1000\n", + "ds = np.random.multivariate_normal([0, 0], np.eye(2), size=ndat)\n", + "\n", + "step1_options = stepOptions(\n", + " nEpochs = 5,\n", + " batch_size_train = 2048,\n", + " nWorkingData = ndat,\n", + " learning_rate = 2e-4,\n", + " nCouplingLayer = 2,\n", + " num_bins = 4,\n", + " hidden_features = 12,\n", + " num_blocks = 3\n", + ")\n", + "step2_options = stepOptions(\n", + " nEpochs = 20,\n", + " batch_size_train = 2048,\n", + " nWorkingData = ndat,\n", + " learning_rate = 2e-4,\n", + " nCouplingLayer = 2,\n", + " num_bins = 8,\n", + " hidden_features = 12,\n", + " num_blocks = 3\n", + ")\n", + "\n", + "downsampler_options = UIPSOptions(\n", + " stepOptionsList=[step1_options, step2_options],\n", + " num_pdf_iter = 2,\n", + " pdfMethod = \"NormalizingFlow\",\n", + " use_gpu = False,\n", + " printTiming = True,\n", + " nDatReduced = -1,\n", + " nDimReduced = -1,\n", + " preShuffled = True,\n", + " scalerFile = \"scaler.npz\",\n", + " nSamples_list = [nSampl],\n", + " computeDistanceCriterion = True,\n", + " prefixDownsampledData = \"downSampledData\",\n", + " data_freq_adjustment = 1,\n", + " nWorkingDataAdjustment = -1,\n", + " dataFile = \"../data/combustion2DToDownsampleSmall.npy\",\n", + " batch_size_eval = 2.5e4,\n", + " seed = 42\n", + ")\n", + "\n", + "downsample_ds = downsample_dataset_from_input(downsampler_options, ds)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "plt.scatter(ds[:,0], ds[:,1], alpha=0.1)" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "plt.scatter(downsample_ds[1000][1].data[:,0], downsample_ds[1000][1].data[:,1])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "mfix-exa", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.13" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/tutorials/uips_normal.py b/tutorials/uips_normal.py index 3fc7e22..14fab95 100644 --- a/tutorials/uips_normal.py +++ b/tutorials/uips_normal.py @@ -19,7 +19,7 @@ inpt["nEpochs"] = f"5 20" inpt["nSamples"] = f"{nSampl}" -best_files = downsample_dataset_from_input(inpt) +best_files = downsample_dataset_from_input(inpt if par.irank == par.iroot: diff --git a/uips/inputs/downsampler.py b/uips/inputs/downsampler.py new file mode 100644 index 0000000..e4b2461 --- /dev/null +++ b/uips/inputs/downsampler.py @@ -0,0 +1,27 @@ +import os +import torch +from ..options import UIPSOptions + +class DownSampler(): + def __init__(self, options: UIPSOptions): + os.environ["KMP_DUPLICATE_LIB_OK"] = "True" + self.options = options + + self.configure_torch() + self.configure_numpy() + + def configure_torch(self): + """ Configure torch based on DownSamplerOptions + """ + + if self.options.use_gpu: + device = torch.device("cuda") + torch.set_default_dtype(torch.cuda.float32) + else: + device = torch.device("cpu") + torch.set_default_dtype(torch.float32) + torch.manual_seed(int(inpt["seed"]) + par.irank) + + def configure_numpy(): + numpy.random.seed(self.options.seed) + diff --git a/uips/options.py b/uips/options.py new file mode 100644 index 0000000..ab11e1f --- /dev/null +++ b/uips/options.py @@ -0,0 +1,66 @@ +from dataclasses import dataclass, field, asdict +import os +import typing as t +import enum + +class PDFMethodEnum(enum.Enum): + NORMALIZINGFLOW = enum.auto() + BINS = enum.auto() + + def lookup(self, name): + for member in self: + if member.name.lower() == name.lower(): + return member + raise ValueError(f"PDFMethodEnum has no method {name}") + +@dataclass +class stepOptions: + # Number of epochs for each normalizing flow training + nEpochs: int = 5 + # Batch size for normalizing flow training + batch_size_train: int = 25_000 + # Subset of data on which training is done + nWorkingData: int = 10_000 + data_freq_adjustment: int = 1 + # Learning rate during normalizing flow training + learning_rate: float = 2e-4 + num_bins: int = 4 + hidden_features: int = 12 + nCouplingLayer: int = 2 + num_blocks: int = 3 + +@dataclass +class UIPSOptions: + num_pdf_iter: int = 2 + stepOptionsList: t.List[stepOptions] = field(default_factory = lambda : [stepOptions(), stepOptions()]) + pdfMethod: str = "NormalizingFlow" + batch_size_eval: int = 2.5e4 + seed: int = 42 + printTiming: bool = False + nDatReduced: int = -1 + nDimReduced: int = -1 + dimList: t.Optional[int] = None + dataFile: "os.PathLike[str]" = "./datafile.npy" + preShuffled: bool = True + scalerFile: "os.PathLike[str]" = "scaler.npz" + nSamples_list: t.List[int] = field(default_factory=lambda : [1e3, 1e4, 1e5]) + nSampleCriterionLimit: int = 1e5 + computeDistanceCriterion: bool = True + prefixDownsampledData: str = "downSampledData" + use_gpu: bool = True + num_pdf_bins: int = 20 + data_freq_adjustment: int = 1 + # Subset of data used to adjust the sampling probability to the desired number of samples + # If nWorkingDataAdjustment < 0, all data with the prescribed frequency is used + nWorkingDataAdjustment: int = -1 + + @property + def nSamples(self) -> t.List[int]: + return [int(n) for n in self.nSamples_list] + + def __post_init__(self): + if len(self.stepOptionsList) != self.num_pdf_iter: + raise Exception("Length of stepOptions is less than the number of pdf iterations") + + def stepOptions_as_list(self, field): + return [asdict(step_option)[field] for step_option in self.stepOptionsList] \ No newline at end of file diff --git a/uips/sampler.py b/uips/sampler.py index 9705af6..e4f1816 100644 --- a/uips/sampler.py +++ b/uips/sampler.py @@ -39,7 +39,7 @@ def checkParamListLength(listToCheck, targetLength, name): def makeParamList(strEntry, fun, inpt, pdf_iter): - num_pdf_iter = int(inpt["num_pdf_iter"]) + num_pdf_iter = inpt.num_pdf_iter param_list_inpt = [fun(n) for n in inpt[strEntry].split()] checkParamListLength(param_list_inpt, num_pdf_iter, strEntry) if len(param_list_inpt) == 1 and num_pdf_iter > 1: @@ -66,7 +66,7 @@ def computeDistanceToClosestNeighbor(data): def rescaleData(np_data, inpt): - scaler = np.load(inpt["scalerFile"]) + scaler = np.load(inpt.scalerFile) np_data_rescaled = np_data.copy() np_data_rescaled = (np_data_rescaled - scaler["minVal"]) / ( 0.125 * (scaler["maxVal"] - scaler["minVal"]) @@ -79,14 +79,11 @@ def createFlow(dim, pdf_iter, inpt): base_transform_type = "spline" grad_norm_clip_value = float(5) - hidden_features_list = makeParamList( - "hidden_features", str2int, inpt, pdf_iter - ) - num_blocks_list = makeParamList("num_blocks", str2int, inpt, pdf_iter) - num_bins_list = makeParamList("num_bins", str2int, inpt, pdf_iter) - num_coupling_layer_list = makeParamList( - "nCouplingLayer", str2int, inpt, pdf_iter - ) + hidden_features_list = inpt.stepOptions_as_list("hidden_features") + + num_blocks_list = inpt.stepOptions_as_list("num_blocks") + num_bins_list = inpt.stepOptions_as_list("num_bins") + num_coupling_layer_list = inpt.stepOptions_as_list("nCouplingLayer") transform = transforms.CompositeTransform( [ @@ -147,16 +144,12 @@ def trainFlow(np_data, flow, pdf_iter, inpt): # Timer times = time.time() - learning_rate_list = makeParamList( - "learning_rate", str2float, inpt, pdf_iter - ) - num_epochs_list = makeParamList("nEpochs", str2int, inpt, pdf_iter) - batch_size_list = makeParamList( - "batch_size_train", str2int, inpt, pdf_iter - ) + learning_rate_list = inpt.stepOptions_as_list("learning_rate") + num_epochs_list = inpt.stepOptions_as_list("nEpochs") + batch_size_list = inpt.stepOptions_as_list("batch_size_train") - BATCH_SIZE = batch_size_list[pdf_iter] - EPOCHS = num_epochs_list[pdf_iter] + BATCH_SIZE = batch_size_list[-1] + EPOCHS = num_epochs_list[-1] grad_norm_clip_value = float(5) # create optimizer @@ -168,7 +161,7 @@ def trainFlow(np_data, flow, pdf_iter, inpt): return last_loss os.environ["KMP_DUPLICATE_LIB_OK"] = "True" - use_gpu = (inpt["use_gpu"] == "True") and (torch.cuda.is_available()) + use_gpu = (inpt.use_gpu) and (torch.cuda.is_available()) # Perform training on GPU if possible if use_gpu: @@ -232,7 +225,7 @@ def trainFlow(np_data, flow, pdf_iter, inpt): # Timer timee = time.time() - printTiming = inpt["printTiming"] == "True" + printTiming = inpt.printTiming if printTiming: par.printRoot(f"Time Train : {timee - times:.2f}s") @@ -248,7 +241,7 @@ def trainBinPDF(np_data, pdf_iter, inpt): return None, None # Train np_data_rescaled = rescaleData(np_data, inpt) - H, edges = np.histogramdd(np_data_rescaled, bins=int(inpt["num_pdf_bins"])) + H, edges = np.histogramdd(np_data_rescaled, bins=inpt.num_pdf_bins) logProb = np.log(1e-16 + H / np.sum(H)) os.makedirs("TrainingLog", exist_ok=True) @@ -260,7 +253,7 @@ def trainBinPDF(np_data, pdf_iter, inpt): # Timer timee = time.time() - printTiming = inpt["printTiming"] == "True" + printTiming = inpt.printTiming if printTiming: par.printRoot(f"Time Train : {timee - times:.2f}s") @@ -359,7 +352,7 @@ def evalLogProbNF(flow, np_data_to_downsample, pdf_iter, inpt): # Evaluation flow.eval() log_density_np = [] - BATCH_SIZE = int(float(inpt["batch_size_eval"])) + BATCH_SIZE = inpt.batch_size_eval to_downsample_loader = makePytorchData( np_data_to_downsample, BATCH_SIZE, inpt, shuffle=False ) @@ -389,7 +382,7 @@ def evalLogProbNF(flow, np_data_to_downsample, pdf_iter, inpt): # Timer timee = time.time() - printTiming = inpt["printTiming"] == "True" + printTiming = inpt.printTiming if printTiming: par.printRoot(f"Time Eval : {par.allmaxScalar(timee - times):.2f}s") @@ -429,7 +422,7 @@ def evalLogProbBIN(np_data_to_downsample, pdf_iter, inpt): # Timer timee = time.time() - printTiming = inpt["printTiming"] == "True" + printTiming = inpt.printTiming if printTiming: par.printRoot(f"Time Eval : {par.allmaxScalar(timee - times):.2f}s") @@ -476,13 +469,10 @@ def downSample( inpt, ): # Mode of adjustment of the sampling probability - nWorkingDataAdjustment = int(float(inpt["nWorkingDataAdjustment"])) + nWorkingDataAdjustment = inpt.nWorkingDataAdjustment if nWorkingDataAdjustment < 0: use_serial_adjustment = False - try: - data_freq_adjustment = int(inpt["data_freq_adjustment"]) - except KeyError: - data_freq_adjustment = 1 + data_freq_adjustment = inpt.data_freq_adjustment else: use_serial_adjustment = True diff --git a/uips/utils/dataUtils.py b/uips/utils/dataUtils.py index dae5362..1a20427 100644 --- a/uips/utils/dataUtils.py +++ b/uips/utils/dataUtils.py @@ -33,33 +33,34 @@ def checkData(shape, N, d, nWorkingData, nWorkingDataAdjustment, useNF): # @profile -def prepareData(inpt): +def prepareData(inpt, dataset = None): # Set parameters from input - dataFile = find_data(inpt["dataFile"]) - preShuffled = inpt["preShuffled"] == "True" - scalerFile = inpt["scalerFile"] - nWorkingDatas = [int(float(n)) for n in inpt["nWorkingData"].split()] + if dataset is None: + # Load the dataset but don't read it just yet + dataFile = find_data(inpt.dataFile) + dataset = np.load(dataFile, mmap_mode="r") + + preShuffled = inpt.preShuffled + scalerFile = inpt.scalerFile + nWorkingDatas = inpt.stepOptions_as_list("nWorkingData") if len(nWorkingDatas) == 1: - nWorkingDatas = nWorkingDatas * int(inpt["num_pdf_iter"]) - nWorkingDataAdjustment = int(float(inpt["nWorkingDataAdjustment"])) + nWorkingDatas = nWorkingDatas * inpt.num_pdf_iter + nWorkingDataAdjustment = int(inpt.nWorkingDataAdjustment) # Reduce Data try: - dimList = [int(n) for n in inpt["dimList"].split()] + dimList = [int(n) for n in inpt.dimList.split()] except: dimList = None try: - nDimReduced = int(float(inpt["nDimReduced"])) + nDimReduced = inpt.nDimReduced except: nDimReduced = -1 try: - nDatReduced = int(float(inpt["nDatReduced"])) + nDatReduced = inpt.nDatReduced except: nDatReduced = -1 - # Load the dataset but don't read it just yet - dataset = np.load(dataFile, mmap_mode="r") - # Check that dataset shape make sense if nDatReduced > 0: nFullData = min(dataset.shape[0], nDatReduced) @@ -72,7 +73,7 @@ def prepareData(inpt): else: nDim = dataset.shape[1] if par.irank == par.iroot: - useNF = inpt["pdf_method"].lower() == "normalizingflow" + useNF = inpt.pdfMethod.lower() == "normalizingflow" checkData( dataset.shape, nFullData, diff --git a/uips/wrapper.py b/uips/wrapper.py index ae10966..ec14071 100644 --- a/uips/wrapper.py +++ b/uips/wrapper.py @@ -1,6 +1,9 @@ import os import sys import time +import typing as t + +from dataclasses import dataclass import numpy as np import torch @@ -13,6 +16,10 @@ from uips.utils.plotFun import * from uips.utils.torchutils import get_num_parameters +@dataclass +class DownsampledDataset: + data: np.array + dataIndices: t.List def downsample_dataset_from_input_file(inpt_file): inpt_file = find_input(inpt_file) @@ -20,33 +27,36 @@ def downsample_dataset_from_input_file(inpt_file): return downsample_dataset_from_input(inpt) -def downsample_dataset_from_input(inpt): - use_normalizing_flow = inpt["pdf_method"].lower() == "normalizingflow" - use_bins = inpt["pdf_method"].lower() == "bins" +def downsample_dataset_from_input(inpt, dataset): + use_normalizing_flow = inpt.pdfMethod.lower() == "normalizingflow" + use_bins = inpt.pdfMethod == "bins" # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # ~~~~ Parameters to save # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # List of sample size - nSamples = [int(float(n)) for n in inpt["nSamples"].split()] + nSamples = inpt.nSamples # Data size used to adjust the sampling probability - nWorkingDataAdjustment = int(float(inpt["nWorkingDataAdjustment"])) + nWorkingDataAdjustment = inpt.nWorkingDataAdjustment if nWorkingDataAdjustment < 0: use_serial_adjustment = False else: use_serial_adjustment = True # Data size used to learn the data probability - nWorkingDatas = [int(float(n)) for n in inpt["nWorkingData"].split()] + nWorkingDatas = inpt.stepOptions_as_list("nWorkingData") + if len(nWorkingDatas) == 1: - nWorkingDatas = nWorkingDatas * int(inpt["num_pdf_iter"]) + nWorkingDatas = nWorkingDatas * inpt.num_pdf_iter + for nWorkingData in nWorkingDatas: if not nWorkingData in nSamples: nSamples += [nWorkingData] + # Do we compute the neighbor distance criterion - computeCriterion = inpt["computeDistanceCriterion"] == "True" + computeCriterion = inpt.computeDistanceCriterion try: - nSampleCriterionLimit = int(inpt["nSampleCriterionLimit"]) + nSampleCriterionLimit = int(inpt.nSampleCriterionLimit) except: nSampleCriterionLimit = int(1e5) @@ -56,7 +66,7 @@ def downsample_dataset_from_input(inpt): os.environ["KMP_DUPLICATE_LIB_OK"] = "True" use_gpu = ( - (inpt["use_gpu"] == "True") + inpt.use_gpu and (torch.cuda.is_available()) and (par.irank == par.iroot) ) @@ -69,14 +79,14 @@ def downsample_dataset_from_input(inpt): device = torch.device("cpu") torch.set_default_dtype(torch.float32) # REPRODUCIBILITY - torch.manual_seed(int(inpt["seed"]) + par.irank) - np.random.seed(int(inpt["seed"]) + par.irank) + torch.manual_seed(inpt.seed + par.irank) + np.random.seed(inpt.seed + par.irank) # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # ~~~~ Prepare Data and scatter across processors # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - data_to_downsample_, dataInd_, working_data, nFullData = prepareData(inpt) + data_to_downsample_, dataInd_, working_data, nFullData = prepareData(inpt, dataset) dim = data_to_downsample_.shape[1] @@ -96,9 +106,9 @@ def downsample_dataset_from_input(inpt): ) # Prepare arrays used for sanity checks - meanCriterion = np.zeros((int(inpt["num_pdf_iter"]), len(nSamples))) - stdCriterion = np.zeros((int(inpt["num_pdf_iter"]), len(nSamples))) - flow_nll_loss = np.zeros(int(inpt["num_pdf_iter"])) + meanCriterion = np.zeros((inpt.num_pdf_iter, len(nSamples))) + stdCriterion = np.zeros((inpt.num_pdf_iter, len(nSamples))) + flow_nll_loss = np.zeros(inpt.num_pdf_iter) # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # ~~~~ Downsample @@ -106,7 +116,7 @@ def downsample_dataset_from_input(inpt): data_for_pdf_est = working_data - for pdf_iter in range(int(inpt["num_pdf_iter"])): + for pdf_iter in range(inpt.num_pdf_iter): if use_normalizing_flow: # Create the normalizing flow flow = sampler.createFlow(dim, pdf_iter, inpt) @@ -157,6 +167,9 @@ def downsample_dataset_from_input(inpt): par.printRoot(f"TRAIN ITER {pdf_iter}") + if par.irank == par.iroot: + output_datasets = {n:{} for n in nSamples} + for inSample, nSample in enumerate(nSamples): # Downsample ( @@ -188,7 +201,7 @@ def downsample_dataset_from_input(inpt): f"\t nSample {nSample} mean dist = {mean:.4f}, std dist = {std:.4f}" ) - if pdf_iter == int(inpt["num_pdf_iter"]) - 1: + if pdf_iter == inpt.num_pdf_iter - 1: # Last pdf iter : Root proc saves downsampled data, and checks the outcome sampler.checkProcedure( meanCriterion[:, inSample], @@ -196,13 +209,11 @@ def downsample_dataset_from_input(inpt): randomCriterion[inSample], ) if par.irank == par.iroot: - np.savez( - f"{inpt['prefixDownsampledData']}_{nSample}_it{pdf_iter}.npz", - data=downSampledData, - indices=downSampledIndices, + output_datasets[nSample][pdf_iter] = DownsampledDataset( + downSampledData, downSampledIndices ) - if not (pdf_iter == int(inpt["num_pdf_iter"]) - 1): + if not (pdf_iter == inpt.num_pdf_iter - 1): # Prepare data for the next training iteration ( downSampledData, @@ -226,17 +237,17 @@ def downsample_dataset_from_input(inpt): if np.amax(meanCriterion) > 0: print("\n") maxCrit = np.argmax(meanCriterion, axis=0) - nSamples_asked = [int(float(n)) for n in inpt["nSamples"].split()] + nSamples_asked = inpt.nSamples for iSample, nSample in enumerate(nSamples_asked): print( - f"For sample {nSample} use {inpt['prefixDownsampledData']}_{nSample}_it{maxCrit[iSample]}.npz" + f"For sample {nSample} use {inpt.prefixDownsampledData}_{nSample}_it{maxCrit[iSample]}.npz" ) best_files[nSample] = ( - f"{inpt['prefixDownsampledData']}_{nSample}_it{maxCrit[iSample]}.npz" + f"{inpt.prefixDownsampledData}_{nSample}_it{maxCrit[iSample]}.npz" ) print("\n") - return best_files + return output_datasets # if par.irank==par.iroot: