{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# LPD Offline Correction #\n",
"\n",
"Author: European XFEL Data Analysis Group"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2018-12-03T15:19:56.056417Z",
"start_time": "2018-12-03T15:19:56.003012Z"
}
},
"outputs": [],
"source": [
"# Input parameters\n",
"in_folder = \"/gpfs/exfel/exp/FXE/202201/p003073/raw/\" # the folder to read data from, required\n",
"out_folder = \"/gpfs/exfel/data/scratch/schmidtp/random/LPD_test\" # the folder to output to, required\n",
"metadata_folder = '' # Directory containing calibration_metadata.yml when run by xfel-calibrate.\n",
"sequences = [-1] # Sequences to correct, use [-1] for all\n",
"modules = [-1] # Modules indices to correct, use [-1] for all, only used when karabo_da is empty\n",
"karabo_da = ['all'] # Data aggregators names to correct, use [''] for all\n",
"run = 10 # run to process, required\n",
"\n",
"# Source parameters\n",
"karabo_id = 'FXE_DET_LPD1M-1' # Karabo domain for detector.\n",
"input_source = '{karabo_id}/DET/{module_index}CH0:xtdf' # Input fast data source.\n",
"output_source = '' # Output fast data source, empty to use same as input.\n",
"\n",
"# CalCat parameters\n",
"creation_time = \"\" # The timestamp to use with Calibration DB. Required Format: \"YYYY-MM-DD hh:mm:ss\" e.g. 2019-07-04 11:02:41\n",
"cal_db_interface = '' # Not needed, compatibility with current webservice.\n",
"cal_db_timeout = 0 # Not needed, compatbility with current webservice.\n",
"cal_db_root = '/gpfs/exfel/d/cal/caldb_store'\n",
"\n",
"# Operating conditions\n",
"mem_cells = 512 # Memory cells, LPD constants are always taken with 512 cells.\n",
"bias_voltage = 250.0 # Detector bias voltage.\n",
"capacitor = '5pF' # Capacitor setting: 5pF or 50pF\n",
"photon_energy = 9.2 # Photon energy in keV.\n",
"category = 0 # Whom to blame.\n",
"\n",
"# Correction parameters\n",
"offset_corr = True # Offset correction.\n",
"rel_gain = True # Gain correction based on RelativeGain constant.\n",
"ff_map = True # Gain correction based on FFMap constant.\n",
"gain_amp_map = True # Gain correction based on GainAmpMap constant.\n",
"\n",
"# Output options\n",
"overwrite = True # set to True if existing data should be overwritten\n",
"chunks_data = 1 # HDF chunk size for pixel data in number of frames.\n",
"chunks_ids = 32 # HDF chunk size for cellId and pulseId datasets.\n",
"create_virtual_cxi_in = '' # Folder to create virtual CXI files in (for each sequence).\n",
"\n",
"# Parallelization options\n",
"sequences_per_node = 1 # Sequence files to process per node\n",
"max_nodes = 8 # Maximum number of SLURM jobs to split correction work into\n",
"num_workers = 8 # Worker processes per node, 8 is safe on 768G nodes but won't work on 512G.\n",
"num_threads_per_worker = 32 # Number of threads per worker.\n",
"\n",
"def balance_sequences(in_folder, run, sequences, sequences_per_node, karabo_da, max_nodes):\n",
" from xfel_calibrate.calibrate import balance_sequences as bs\n",
" return bs(in_folder, run, sequences, sequences_per_node, karabo_da, max_nodes=max_nodes)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2018-12-03T15:19:56.990566Z",
"start_time": "2018-12-03T15:19:56.058378Z"
}
},
"outputs": [],
"source": [
"from collections import OrderedDict\n",
"from pathlib import Path\n",
"from time import perf_counter\n",
"import gc\n",
"import re\n",
"import warnings\n",
"\n",
"import numpy as np\n",
"import h5py\n",
"\n",
"import matplotlib\n",
"matplotlib.use('agg')\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline\n",
"\n",
"from calibration_client import CalibrationClient\n",
"from calibration_client.modules import CalibrationConstantVersion\n",
"import extra_data as xd\n",
"import extra_geom as xg\n",
"import pasha as psh\n",
"\n",
"from extra_data.components import LPD1M\n",
"\n",
"from cal_tools.calcat_interface import CalCatError, LPD_CalibrationData\n",
"from cal_tools.lpdalgs import correct_lpd_frames\n",
"from cal_tools.tools import CalibrationMetadata, calcat_creation_time\n",
"from cal_tools.files import DataFile\n",
"from cal_tools.restful_config import restful_config"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Prepare environment"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"file_re = re.compile(r'^RAW-R(\\d{4})-(\\w+\\d+)-S(\\d{5})$') # This should probably move to cal_tools\n",
"\n",
"run_folder = Path(in_folder) / f'r{run:04d}'\n",
"out_folder = Path(out_folder)\n",
"out_folder.mkdir(exist_ok=True)\n",
"\n",
"output_source = output_source or input_source\n",
"\n",
"cal_db_root = Path(cal_db_root)\n",
"\n",
"metadata = CalibrationMetadata(metadata_folder or out_folder)\n",
"# Constant paths & timestamps are saved under retrieved-constants in calibration_metadata.yml\n",
"retrieved_constants = metadata.setdefault(\"retrieved-constants\", {})\n",
"\n",
"creation_time = calcat_creation_time(in_folder, run, creation_time)\n",
"print(f'Using {creation_time.isoformat()} as creation time')\n",
"\n",
"# Pick all modules/aggregators or those selected.\n",
"if karabo_da == ['all']:\n",
" if modules == [-1]:\n",
" modules = list(range(16))\n",
"\n",
" karabo_da = [f'LPD{i:02d}' for i in modules]\n",
"else:\n",
" modules = [int(x[-2:]) for x in karabo_da]\n",
" \n",
"# Pick all sequences or those selected.\n",
"if not sequences or sequences == [-1]:\n",
" do_sequence = lambda seq: True\n",
"else:\n",
" do_sequence = [int(x) for x in sequences].__contains__ \n",
" \n",
"# List of detector sources.\n",
"det_inp_sources = [input_source.format(karabo_id=karabo_id, module_index=int(da[-2:])) for da in karabo_da]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Select data to process"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data_to_process = []\n",
"\n",
"for inp_path in run_folder.glob('RAW-*.h5'):\n",
" match = file_re.match(inp_path.stem)\n",
" \n",
" if match[2] not in karabo_da or not do_sequence(int(match[3])):\n",
" continue\n",
" \n",
" outp_path = out_folder / 'CORR-R{run:04d}-{aggregator}-S{seq:05d}.h5'.format(\n",
" run=int(match[1]), aggregator=match[2], seq=int(match[3]))\n",
"\n",
" data_to_process.append((match[2], inp_path, outp_path))\n",
"\n",
"print('Files to process:')\n",
"for data_descr in sorted(data_to_process, key=lambda x: f'{x[0]}{x[1]}'):\n",
" print(f'{data_descr[0]}\\t{data_descr[1]}')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Obtain and prepare calibration constants"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"metadata = CalibrationMetadata(metadata_folder or out_folder)\n",
"# Constant paths & timestamps are saved under retrieved-constants in calibration_metadata.yml\n",
"const_yaml = metadata.setdefault(\"retrieved-constants\", {})"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"const_data = {}\n",
"const_load_mp = psh.ProcessContext(num_workers=24)\n",
"\n",
"if const_yaml: # Read constants from YAML file.\n",
" start = perf_counter()\n",
" for da, ccvs in const_yaml.items():\n",
"\n",
" for calibration_name, ccv in ccvs['constants'].items():\n",
"\n",
" dtype = np.uint32 if calibration_name.startswith('BadPixels') else np.float32\n",
"\n",
" const_data[(da, calibration_name)] = dict(\n",
" path=Path(ccv['path']),\n",
" dataset=ccv['dataset'],\n",
" data=const_load_mp.alloc(shape=(256, 256, mem_cells, 3), dtype=dtype)\n",
" )\n",
"else: # Retrieve constants from CALCAT.\n",
" print('Querying calibration database', end='', flush=True)\n",
" start = perf_counter()\n",
" for calibrations, condition in [\n",
" (dark_calibrations, dark_condition),\n",
" (illuminated_calibrations, illuminated_condition)\n",
" ]:\n",
" resp = CalibrationConstantVersion.get_closest_by_time_by_detector_conditions(\n",
" client, karabo_id, list(calibrations.keys()),\n",
" {'parameters_conditions_attributes': condition},\n",
" karabo_da='', event_at=creation_time.isoformat()\n",
" )\n",
"\n",
" if not resp['success']:\n",
" raise RuntimeError(resp)\n",
"\n",
" for ccv in resp['data']:\n",
" cc = ccv['calibration_constant']\n",
" da = ccv['physical_detector_unit']['karabo_da']\n",
" calibration_name = calibrations[cc['calibration_id']]\n",
" \n",
" dtype = np.uint32 if calibration_name.startswith('BadPixels') else np.float32\n",
" \n",
" const_data[(da, calibration_name)] = dict(\n",
" path=Path(ccv['path_to_file']) / ccv['file_name'],\n",
" dataset=ccv['data_set_name'],\n",
" data=const_load_mp.alloc(shape=(256, 256, mem_cells, 3), dtype=dtype)\n",
" )\n",
" print('.', end='', flush=True)\n",
" \n",
"total_time = perf_counter() - start\n",
"print(f'{total_time:.1f}s')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def load_constant_dataset(wid, index, const_descr):\n",
" ccv_entry = const_data[const_descr]\n",
" \n",
" with h5py.File(cal_db_root / ccv_entry['path'], 'r') as fp:\n",
" fp[ccv_entry['dataset'] + '/data'].read_direct(ccv_entry['data'])\n",
" \n",
" print('.', end='', flush=True)\n",
"\n",
"print('Loading calibration data', end='', flush=True)\n",
"start = perf_counter()\n",
"const_load_mp.map(load_constant_dataset, list(const_data.keys()))\n",
"total_time = perf_counter() - start\n",
"\n",
"print(f'{total_time:.1f}s')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# These are intended in order cell, X, Y, gain\n",
"ccv_offsets = {}\n",
"ccv_gains = {}\n",
"ccv_masks = {}\n",
"\n",
"ccv_shape = (mem_cells, 256, 256, 3)\n",
"\n",
"constant_order = {\n",
" 'Offset': (2, 1, 0, 3),\n",
" 'BadPixelsDark': (2, 1, 0, 3),\n",
" 'RelativeGain': (2, 1, 0, 3),\n",
" 'FFMap': (2, 0, 1, 3),\n",
" 'BadPixelsFF': (2, 0, 1, 3),\n",
" 'GainAmpMap': (2, 0, 1, 3),\n",
"}\n",
"\n",
"def prepare_constants(wid, index, aggregator):\n",
" consts = {calibration_name: entry['data']\n",
" for (aggregator_, calibration_name), entry\n",
" in const_data.items()\n",
" if aggregator == aggregator_}\n",
" \n",
" def _prepare_data(calibration_name, dtype):\n",
" return consts[calibration_name] \\\n",
" .transpose(constant_order[calibration_name]) \\\n",
" .astype(dtype, copy=True) # Make sure array is contiguous.\n",
" \n",
" if offset_corr and 'Offset' in consts:\n",
" ccv_offsets[aggregator] = _prepare_data('Offset', np.float32)\n",
" else:\n",
" ccv_offsets[aggregator] = np.zeros(ccv_shape, dtype=np.float32)\n",
" \n",
" ccv_gains[aggregator] = np.ones(ccv_shape, dtype=np.float32)\n",
" \n",
" if 'BadPixelsDark' in consts:\n",
" ccv_masks[aggregator] = _prepare_data('BadPixelsDark', np.uint32)\n",
" else:\n",
" ccv_masks[aggregator] = np.zeros(ccv_shape, dtype=np.uint32)\n",
" \n",
" if rel_gain and 'RelativeGain' in consts:\n",
" ccv_gains[aggregator] *= _prepare_data('RelativeGain', np.float32)\n",
" \n",
" if ff_map and 'FFMap' in consts:\n",
" ccv_gains[aggregator] *= _prepare_data('FFMap', np.float32)\n",
" \n",
" if 'BadPixelsFF' in consts:\n",
" np.bitwise_or(ccv_masks[aggregator], _prepare_data('BadPixelsFF', np.uint32),\n",
" out=ccv_masks[aggregator])\n",
" \n",
" if gain_amp_map and 'GainAmpMap' in consts:\n",
" ccv_gains[aggregator] *= _prepare_data('GainAmpMap', np.float32)\n",
" \n",
" print('.', end='', flush=True)\n",
" \n",
"\n",
"print('Preparing constants', end='', flush=True)\n",
"start = perf_counter()\n",
"psh.ThreadContext(num_workers=len(karabo_da)).map(prepare_constants, karabo_da)\n",
"total_time = perf_counter() - start\n",
"print(f'{total_time:.1f}s')\n",
"\n",
"const_data.clear() # Clear raw constants data now to save memory.\n",
"gc.collect();"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def correct_file(wid, index, work):\n",
" aggregator, inp_path, outp_path = work\n",
" module_index = int(aggregator[-2:])\n",
" \n",
" start = perf_counter()\n",
" dc = xd.H5File(inp_path, inc_suspect_trains=False).select('*', 'image.*', require_all=True)\n",
" inp_source = dc[input_source.format(karabo_id=karabo_id, module_index=module_index)]\n",
" open_time = perf_counter() - start\n",
" \n",
" # Load raw data for this file.\n",
" # Reshaping gets rid of the extra 1-len dimensions without\n",
" # mangling the frame axis for an actual frame count of 1.\n",
" start = perf_counter()\n",
" in_raw = inp_source['image.data'].ndarray().reshape(-1, 256, 256)\n",
" in_cell = inp_source['image.cellId'].ndarray().reshape(-1)\n",
" in_pulse = inp_source['image.pulseId'].ndarray().reshape(-1)\n",
" read_time = perf_counter() - start\n",
" \n",
" # Allocate output arrays.\n",
" out_data = np.zeros((in_raw.shape[0], 256, 256), dtype=np.float32)\n",
" out_gain = np.zeros((in_raw.shape[0], 256, 256), dtype=np.uint8)\n",
" out_mask = np.zeros((in_raw.shape[0], 256, 256), dtype=np.uint32)\n",
" \n",
" start = perf_counter()\n",
" correct_lpd_frames(in_raw, in_cell,\n",
" out_data, out_gain, out_mask,\n",
" ccv_offsets[aggregator], ccv_gains[aggregator], ccv_masks[aggregator],\n",
" num_threads=num_threads_per_worker)\n",
" correct_time = perf_counter() - start\n",
" \n",
" image_counts = inp_source['image.data'].data_counts(labelled=False)\n",
" \n",
" start = perf_counter()\n",
" if (not outp_path.exists() or overwrite) and image_counts.sum() > 0:\n",
" outp_source_name = output_source.format(karabo_id=karabo_id, module_index=module_index)\n",
"\n",
" with DataFile(outp_path, 'w') as outp_file: \n",
" outp_file.create_index(dc.train_ids, from_file=dc.files[0])\n",
" outp_file.create_metadata(like=dc, instrument_channels=(f'{outp_source_name}/image',))\n",
" \n",
" outp_source = outp_file.create_instrument_source(outp_source_name)\n",
" \n",
" outp_source.create_index(image=image_counts)\n",
" outp_source.create_key('image.cellId', data=in_cell,\n",
" chunks=(min(chunks_ids, in_cell.shape[0]),))\n",
" outp_source.create_key('image.pulseId', data=in_pulse,\n",
" chunks=(min(chunks_ids, in_pulse.shape[0]),))\n",
" outp_source.create_key('image.data', data=out_data,\n",
" chunks=(min(chunks_data, out_data.shape[0]), 256, 256))\n",
" outp_source.create_compressed_key('image.gain', data=out_gain)\n",
" outp_source.create_compressed_key('image.mask', data=out_mask)\n",
" write_time = perf_counter() - start\n",
" \n",
" total_time = open_time + read_time + correct_time + write_time\n",
" frame_rate = in_raw.shape[0] / total_time\n",
" \n",
" print('{}\\t{}\\t{:.3f}\\t{:.3f}\\t{:.3f}\\t{:.3f}\\t{:.3f}\\t{}\\t{:.1f}'.format(\n",
" wid, aggregator, open_time, read_time, correct_time, write_time, total_time,\n",
" in_raw.shape[0], frame_rate))\n",
" \n",
" in_raw = None\n",
" in_cell = None\n",
" in_pulse = None\n",
" out_data = None\n",
" out_gain = None\n",
" out_mask = None\n",
" gc.collect()\n",
"\n",
"print('worker\\tDA\\topen\\tread\\tcorrect\\twrite\\ttotal\\tframes\\trate')\n",
"start = perf_counter()\n",
"psh.ProcessContext(num_workers=num_workers).map(correct_file, data_to_process)\n",
"total_time = perf_counter() - start\n",
"print(f'Total time: {total_time:.1f}s')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Data preview for first train"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"geom = xg.LPD_1MGeometry.from_quad_positions(\n",
" [(11.4, 299), (-11.5, 8), (254.5, -16), (278.5, 275)])\n",
"\n",
"output_paths = [outp_path for _, _, outp_path in data_to_process if outp_path.exists()]\n",
"dc = xd.DataCollection.from_paths(output_paths).select_trains(np.s_[0])\n",
"\n",
"det = LPD1M(dc, detector_name=karabo_id)\n",
"data = det.get_array('image.data')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Intensity histogram across all cells"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"left_edge_ratio = 0.01\n",
"right_edge_ratio = 0.99\n",
"\n",
"fig, ax = plt.subplots(num=1, clear=True, figsize=(15, 6))\n",
"values, bins, _ = ax.hist(np.ravel(data.data), bins=2000, range=(-1500, 2000))\n",
"\n",
"def find_nearest_index(array, value):\n",
" return (np.abs(array - value)).argmin()\n",
"\n",
"cum_values = np.cumsum(values)\n",
"vmin = bins[find_nearest_index(cum_values, cum_values[-1]*left_edge_ratio)]\n",
"vmax = bins[find_nearest_index(cum_values, cum_values[-1]*right_edge_ratio)]\n",
"\n",
"max_value = values.max()\n",
"ax.vlines([vmin, vmax], 0, max_value, color='red', linewidth=5, alpha=0.2)\n",
"ax.text(vmin, max_value, f'{left_edge_ratio*100:.0f}%',\n",
" color='red', ha='center', va='bottom', size='large')\n",
"ax.text(vmax, max_value, f'{right_edge_ratio*100:.0f}%',\n",
" color='red', ha='center', va='bottom', size='large')\n",
"ax.text(vmax+(vmax-vmin)*0.01, max_value/2, 'Colormap interval',\n",
" color='red', rotation=90, ha='left', va='center', size='x-large')\n",
"\n",
"ax.set_xlim(vmin-(vmax-vmin)*0.1, vmax+(vmax-vmin)*0.1)\n",
"ax.set_ylim(0, max_value*1.1)\n",
"pass"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### First memory cell"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"fig, ax = plt.subplots(num=2, figsize=(15, 15), clear=True, nrows=1, ncols=1)\n",
"geom.plot_data_fast(data[:, 0, 0], ax=ax, vmin=vmin, vmax=vmax)\n",
"pass"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Train average"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2018-11-13T18:24:57.547563Z",
"start_time": "2018-11-13T18:24:56.995005Z"
}
},
"outputs": [],
"source": [
"fig, ax = plt.subplots(num=3, figsize=(15, 15), clear=True, nrows=1, ncols=1)\n",
"geom.plot_data_fast(data[:, 0].mean(axis=1), ax=ax, vmin=vmin, vmax=vmax)\n",
"pass"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Lowest gain stage per pixel"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"highest_gain_stage = det.get_array('image.gain', pulses=np.s_[:]).max(axis=(1, 2))\n",
"\n",
"fig, ax = plt.subplots(num=4, figsize=(15, 15), clear=True, nrows=1, ncols=1)\n",
"p = geom.plot_data_fast(highest_gain_stage, ax=ax, vmin=0, vmax=2);\n",
"\n",
"cb = ax.images[0].colorbar\n",
"cb.set_ticks([0, 1, 2])\n",
"cb.set_ticklabels(['High gain', 'Medium gain', 'Low gain'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Create virtual CXI file"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"if create_virtual_cxi_in:\n",
" vcxi_folder = Path(create_virtual_cxi_in.format(\n",
" run=run, proposal_folder=str(Path(in_folder).parent)))\n",
" vcxi_folder.mkdir(parents=True, exist_ok=True)\n",
" \n",
" def sort_files_by_seq(by_seq, outp_path):\n",
" by_seq.setdefault(int(outp_path.stem[-5:]), []).append(outp_path)\n",
" return by_seq\n",
" \n",
" from functools import reduce\n",
" reduce(sort_files_by_seq, output_paths, output_by_seq := {})\n",
" \n",
" for seq_number, seq_output_paths in output_by_seq.items():\n",
" # Create data collection and detector components only for this sequence.\n",
" try:\n",
" det = LPD1M(xd.DataCollection.from_paths(seq_output_paths), detector_name=karabo_id, min_modules=4)\n",
" except ValueError: # Couldn't find enough data for min_modules\n",
" continue\n",
" det.write_virtual_cxi(vcxi_folder / f'VCXI-LPD-R{run:04d}-S{seq_number:05d}.cxi')"
]
}
],
"metadata": {
"kernelspec": {
"display_name": ".cal2_venv",
"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": "3.8.11 (default, Jul 2 2021, 14:23:46) \n[GCC 4.8.5 20150623 (Red Hat 4.8.5-44)]"
},
"vscode": {
"interpreter": {
"hash": "8817b9fbc45d4ec68a62bc54fa52c1c5ac00d5619f7fe87d01def1798ea4889e"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}