From ddcce7225cf63710a8560fe7d060cac3c08b46f8 Mon Sep 17 00:00:00 2001
From: Leon Teichroeb <leon.teichroeb@cern.ch>
Date: Thu, 12 Dec 2024 18:06:46 +0100
Subject: [PATCH] Cleaned up generate_ref_data.py
---
utils/generate_ref_data.py | 214 ++++++++++---------------------------
1 file changed, 58 insertions(+), 156 deletions(-)
diff --git a/utils/generate_ref_data.py b/utils/generate_ref_data.py
index d64b45f..f5db0a7 100644
--- a/utils/generate_ref_data.py
+++ b/utils/generate_ref_data.py
@@ -1,164 +1,66 @@
-import itertools
-import yaml
-import numpy as np
-from steammaterials.STEAM_materials import STEAM_materials
-from datetime import datetime
-from pathlib import Path
-import os
-import pandas as pd
+#!/usr/bin/env python3
+import itertools
+import yaml
+import numpy as np
+from steammaterials.STEAM_materials import STEAM_materials
+from datetime import datetime
+from pathlib import Path
+import os
+import sys
+import pandas as pd
-################ FILL THIS IN #################
-csv_file = pd.read_csv(Path('../all_material_functions.csv'))
+def create_all_ref_files(output_path=None):
+ repository_root = Path(os.path.dirname(__file__)).parent
+ if output_path is None:
+ output_path = repository_root / 'tests' / 'ref_data'
+ csv_file = pd.read_csv(repository_root / 'all_material_functions.csv')
+
+ errors = 0
+ for i, row in csv_file.iterrows():
+ if row['make_ref']:
+ try:
+ create_ref_file(row, output_path)
+ except Exception as e:
+ print(f'Failed to generate reference file for {row["ref_name"]}')
+ print(f'Failed with error: {e}')
+ errors += 1
+ if errors:
+ print(f'!!! During generation, {errors} errors occurred !!!')
-for i in range(len(csv_file)):
- if csv_file.iloc[i]['make_ref']:
- input_parameters = int(csv_file.iloc[i]['input_parameters'])
- gen_by = csv_file.iloc[i]['ref_generated_by']
- ref_name = csv_file.iloc[i]['ref_name']
+def create_ref_file(row, output_path):
+ """
+ Creates a file to act as reference for the expected output of the given material function.
+ Expects a dictionary-like `row` containing the number of 'input_parameters', the material name
+ 'ref_name', author 'ref_generated_by' and inputs with 'input{i}'.
+ """
+ parameter_count = int(row['input_parameters'])
+ ref_name = row['ref_name']
+ if parameter_count < 1:
+ raise ValueError('Incorrect number of input parameters for {ref_name}')
- # you = 'Tim Mulder'
- # # material_function = 'CFUN_rhoHast_v2'
- # # input_parameters = 1
- # material_function = 'CFUN_rhoCu_NIST_v2'
- # input_parameters = 3
- ###############################################
+ inputs = [eval(row[f'input{i+1}']) for i in range(parameter_count)]
+ input_mat = np.array(list(itertools.product(*inputs))).T
+ out = STEAM_materials(ref_name, input_mat.shape[0], input_mat.shape[1]).evaluate(input_mat)
- if input_parameters == 1:
- input1 = eval(csv_file.iloc[i]['input1'])
- numpy2d = input1.reshape((1, len(input1)))
- input_list = np.array([[i] for i in input1])
- elif input_parameters == 2:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input_list = np.array([i for i in itertools.product(input1, input2)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1]))
- elif input_parameters == 3:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2]))
- elif input_parameters == 4:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3]))
- elif input_parameters == 5:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4]))
- elif input_parameters == 6:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5]))
- elif input_parameters == 7:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input7 = eval(csv_file.iloc[i]['input7'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6, input7)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5], input_list[:, 6]))
- elif input_parameters == 8:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input7 = eval(csv_file.iloc[i]['input7'])
- input8 = eval(csv_file.iloc[i]['input8'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6, input7, input8)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5], input_list[:, 6], input_list[:, 7]))
- elif input_parameters == 9:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input7 = eval(csv_file.iloc[i]['input7'])
- input8 = eval(csv_file.iloc[i]['input8'])
- input9 = eval(csv_file.iloc[i]['input9'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6, input7, input8, input9)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5], input_list[:, 6], input_list[:, 7], input_list[:, 8]))
- elif input_parameters == 10:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input7 = eval(csv_file.iloc[i]['input7'])
- input8 = eval(csv_file.iloc[i]['input8'])
- input9 = eval(csv_file.iloc[i]['input9'])
- input10 = eval(csv_file.iloc[i]['input10'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6, input7, input8, input9, input10)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5], input_list[:, 6], input_list[:, 7], input_list[:, 8], input_list[:, 9]))
- elif input_parameters == 11:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input7 = eval(csv_file.iloc[i]['input7'])
- input8 = eval(csv_file.iloc[i]['input8'])
- input9 = eval(csv_file.iloc[i]['input9'])
- input10 = eval(csv_file.iloc[i]['input10'])
- input11 = eval(csv_file.iloc[i]['input11'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6, input7, input8, input9, input10, input11)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5], input_list[:, 6], input_list[:, 7], input_list[:, 8], input_list[:, 9], input_list[:, 10]))
- elif input_parameters == 12:
- input1 = eval(csv_file.iloc[i]['input1'])
- input2 = eval(csv_file.iloc[i]['input2'])
- input3 = eval(csv_file.iloc[i]['input3'])
- input4 = eval(csv_file.iloc[i]['input4'])
- input5 = eval(csv_file.iloc[i]['input5'])
- input6 = eval(csv_file.iloc[i]['input6'])
- input7 = eval(csv_file.iloc[i]['input7'])
- input8 = eval(csv_file.iloc[i]['input8'])
- input9 = eval(csv_file.iloc[i]['input9'])
- input10 = eval(csv_file.iloc[i]['input10'])
- input11 = eval(csv_file.iloc[i]['input11'])
- input12 = eval(csv_file.iloc[i]['input12'])
- input_list = np.array([i for i in itertools.product(input1, input2, input3, input4, input5, input6, input7, input8, input9, input10, input11, input12)])
- numpy2d = np.vstack((input_list[:, 0], input_list[:, 1], input_list[:, 2], input_list[:, 3], input_list[:, 4], input_list[:, 5], input_list[:, 6], input_list[:, 7], input_list[:, 8], input_list[:, 9], input_list[:, 10], input_list[:, 11]))
- else:
- print('Wrong amount of input parameters')
+ dictionary_output = {
+ 'author': row['ref_generated_by'],
+ 'date': datetime.today().strftime('%d.%m.%Y'),
+ 'material_function': ref_name,
+ 'input': input_mat.T.tolist(),
+ 'output': [i.item() for i in out],
+ 'input_parameters': parameter_count,
+ }
+ try:
+ with (output_path / f'{ref_name}.yaml').open("x") as f:
+ yaml.dump(dictionary_output, f, default_flow_style=False)
+ except FileExistsError:
+ print(f'Reference file for {ref_name} already exists. You cannot overwrite it.')
- ###############################################
- if input_parameters >= 1:
- out = STEAM_materials(ref_name, numpy2d.shape[0], numpy2d.shape[1]).evaluate(numpy2d)
- else:
- print('Wrong amount of input parameters')
- ############### Making the yaml ################
- dictionary_output = {
- 'author': gen_by,
- 'date': datetime.today().strftime('%d.%m.%Y'),
- 'material_function': ref_name,
- 'input': [[i.item() for i in k] for k in input_list],
- 'output': [i.item() for i in out],
- 'input_parameters': input_parameters,
- }
- try:
- with Path(os.path.dirname(os.path.dirname(__file__)) + os.sep + 'tests' + os.sep + 'ref_data' + os.sep + ref_name+".yaml").open("x") as f:
- yaml.dump(dictionary_output, f, default_flow_style=False)
- except:
- print('Reference file for ' + ref_name + ' already excists. You cannot overwrite it.')
+if __name__ == '__main__':
+ if len(sys.argv) > 1:
+ create_all_ref_files(Path(sys.argv[1]))
+ else:
+ create_all_ref_files()
--
GitLab