Skip to content

EPICS2023 Database Preparation

This section covers the extraction and optimization of atomic data from the EPICS2023 (EADL, EEDL, EPDL) datasets provided by IAEA.

EPICS2023 Database Preparation

This section covers the extraction and optimization of atomic data from the EPICS2023 (EADL, EEDL, EPDL) datasets.

Source Databases Download

The raw ENDF ASCII data files for Germanium (Z=32) must be downloaded directly from the official IAEA Nuclear Data Section data paths. You can fetch the exact element files using the following direct links:

After downloading, save these files as EADL.dat, EEDL.dat, and EPDL.dat respectively in the project root directory before running the database builder script.

Core Extraction Functions

The following functions parse raw position-based ENDL tables, convert them to fast NumPy structures, and enforce energy monotonicity for cubic spline interpolations.

process_db.EPICS2023(list_of_files, tables)

Parses ENDL-formatted data files and extracts tables based on active descriptors.

Parameters:

Name Type Description Default
list_of_files list of str

File paths for EADL, EEDL, and EPDL database sheets.

 required
tables dict

A combined translation dictionary containing parsing lambdas or 'skip' flags.

 required

Returns:

Type Description
tuple(int, float, dict)

Z (Atomic Number), M (Atomic Mass), and the raw nested data hierarchy.
Source code in process_db.py 
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
def EPICS2023(list_of_files, tables):
    """
    Parses ENDL-formatted data files and extracts tables based on active descriptors.

    Parameters
    ----------
    list_of_files : list of str
        File paths for EADL, EEDL, and EPDL database sheets.
    tables : dict
        A combined translation dictionary containing parsing lambdas or 'skip' flags.

    Returns
    -------
    tuple (int, float, dict)
        Z (Atomic Number), M (Atomic Mass), and the raw nested data hierarchy.
    """
    DATA = {}
    for f in list_of_files:
        with open(f, 'r') as fp:
            print(f"{10*'+-'} Reading Data from {f} {10*'-+'}")
            H = fp.readline(), fp.readline()
            Z, M = int(H[0][0:3]), float(H[0][13:24])
            while H[0]:
                Yi, Yo = int(H[0][7:9]), int(H[0][10:12])
                C, I_prop = int(H[1][0:2]), int(H[1][2:5])
                S, X = int(H[1][5:8]), int(float(H[1][21:32]))
                S = 0 if S == 0 else X
                print(f'Yi = {Yi:2d} Yo={Yo:2d} C={C:2d} I={I_prop:3d} S={S:2d}', end='')
                key = (Yi, Yo, C, I_prop)
                if tables[key] != 'skip':
                    if Yi not in DATA.keys():
                        DATA[Yi] = {}
                    if Yo not in DATA[Yi].keys():
                        DATA[Yi][Yo] = {}
                    if C not in DATA[Yi][Yo].keys():
                        DATA[Yi][Yo][C] = {}
                    if I_prop not in DATA[Yi][Yo][C].keys():
                        DATA[Yi][Yo][C][I_prop] = {}
                    if S not in DATA[Yi][Yo][C][I_prop].keys():
                        DATA[Yi][Yo][C][I_prop][S] = {}
                else:
                    print(' (skipped)')

                D = fp.readline().rstrip('\r\n')
                while D != blank_marker:
                    if tables[key] != 'skip':
                        D = D.replace('D', 'e')
                        row = [float(D[i:16+i]) for i in range(0, len(D), 16)]
                        col = tables[key](row).items()
                        if not DATA[Yi][Yo][C][I_prop][S]:
                            DATA[Yi][Yo][C][I_prop][S] = {k: [v] for k, v in col}
                        else:
                            for k, v in col:
                                DATA[Yi][Yo][C][I_prop][S][k].append(v)
                    D = fp.readline().rstrip('\r\n')
                H = fp.readline(), fp.readline()
                if tables[key] != 'skip':
                    In = DATA[Yi][Yo][C][I_prop][S]
                    Tu = tuple(In.keys())
                    Ou = f"  {Tu}: [{len(In[Tu[0]])}] "
                    print(Ou)
    return Z, M, DATA

process_db.Data_to_Arrays(D)

Recursively converts standard Python lists into fast NumPy arrays.

Walks down the 5-layer nested dictionary and performs in-place modification of all data terminal tables.

Parameters:

Name Type Description Default
D dict

The nested data dictionary to transform.

 required
Source code in process_db.py 
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
def Data_to_Arrays(D):
    """
    Recursively converts standard Python lists into fast NumPy arrays.

    Walks down the 5-layer nested dictionary and performs in-place modification
    of all data terminal tables.

    Parameters
    ----------
    D : dict
        The nested data dictionary to transform.
    """
    for Yi in D:
        for Yo in D[Yi]:
            for C in D[Yi][Yo]:
                for I_prop in D[Yi][Yo][C]:
                    for S in D[Yi][Yo][C][I_prop]:
                        for k, v in D[Yi][Yo][C][I_prop][S].items():
                            D[Yi][Yo][C][I_prop][S][k] = np.array(v)

process_db.Shift_E(D)

Forces strict monotonic increase of energy thresholds for spline safety.

Modifies adjacent identical energy boundaries near atomic shell transitions by a small fractional value to ensure compatibility with cubic splines.

Parameters:

Name Type Description Default
D dict

The global multi-level interaction dataset.

 required
Source code in process_db.py 
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
def Shift_E(D):
    """
    Forces strict monotonic increase of energy thresholds for spline safety.

    Modifies adjacent identical energy boundaries near atomic shell transitions
    by a small fractional value to ensure compatibility with cubic splines.

    Parameters
    ----------
    D : dict
        The global multi-level interaction dataset.
    """
    E = D[7][0][73][0][0]['Eγ']
    for i in range(len(E)-1):
        if E[i] == E[i+1]:
            D[7][0][73][0][0]['Eγ'][i] = 0.9999 * E[i]
            D[7][0][73][0][0]['Eγ'][i+1] = 1.0001 * E[i+1]

process_db.Split_Arrays(T)

Splits multi-dimensional distribution tables over discrete incident energy values.

Parameters:

Name Type Description Default
T dict

A table structure with 3 equivalent-length data lists.

 required

Returns:

Type Description
dict

A reassigned layout mapped by unique isolated baseline variables.
Source code in process_db.py 
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
def Split_Arrays(T):
    """
    Splits multi-dimensional distribution tables over discrete incident energy values.

    Parameters
    ----------
    T : dict
        A table structure with 3 equivalent-length data lists.

    Returns
    -------
    dict
        A reassigned layout mapped by unique isolated baseline variables.
    """
    D = {}
    K = list(T.keys())
    U = sorted(list(set(T[K[0]])))
    A = np.swapaxes(np.array([T[K[0]], T[K[1]], T[K[2]]]), 0, 1)
    for u in U:
        D[u] = {K[1]: A[A[:, 0] == u, 1], K[2]: A[A[:, 0] == u, 2]}
    return D