MolMini base, chembl_test_macfonts
Failure breakdown
Attempted: 3000 | exact: 2665 (0.888)
| failure mode |
n |
share of all |
share of failures |
what it points at |
| unparseable output |
7 |
0.002 |
0.021 |
decoder emitted a string RDKit cannot read |
| stereochemistry only |
113 |
0.038 |
0.337 |
skeleton correct; wedge/hash reading wrong |
| right formula, wrong bonds |
8 |
0.003 |
0.024 |
atoms counted correctly, connectivity misread |
| other structural error |
207 |
0.069 |
0.618 |
a genuinely different molecule |
Accuracy by molecule size
| heavy atoms |
n |
exact |
valid |
| <=10 |
9 |
1.000 |
1.000 |
| <=15 |
84 |
0.810 |
0.988 |
| <=20 |
324 |
0.923 |
0.997 |
| <=25 |
741 |
0.926 |
0.997 |
| <=30 |
741 |
0.908 |
0.999 |
| <=35 |
576 |
0.884 |
0.998 |
| <=40 |
327 |
0.872 |
1.000 |
| >40 |
198 |
0.687 |
0.995 |
Accuracy by rendering style
| style |
n |
exact |
valid |
| clean |
1000 |
0.949 |
0.997 |
| degraded |
1000 |
0.794 |
0.997 |
| varied |
1000 |
0.922 |
0.999 |
Sequence length
- Reference SMILES length, all attempted: median 47 tokens
- Reference SMILES length, correct predictions: median 46 tokens
Longest failures
| reference |
predicted |
CC(=O)OC[C@H]1O[C@@H](O[C@H]2[C@H](OC(C)=O)[C@@H](OC(C)=O)[C@H](S(N)(= |
CC(=O)OC[C@H]1O[C@@H](S(N)(=O)=O)[C@H](OC(C)=O)[C@@H](OC(C)=O)[C@@H]1O |
CC(=O)OC[C@H]1O[C@@H](O[C@H]2[C@H](OC(C)=O)[C@@H](OC(C)=O)[C@H](S(N)(= |
CC(=O)OC[C@H]1O[C@@H](O[C@@H]2O[C@H](COC(C)=O)[C@@H](OC(C)=O)[C@H](OC( |
C[C@H]1[C@@H](c2cc(C(F)(F)F)cc(C(F)(F)F)c2)OC(=O)N1CC1=C(c2cc(-c3ccc(C |
C[C@H]1[C@@H](c2cc(C(F)(F)F)cc(C(F)(F)F)c2)OC(=O)N1CC1=C(c2ccc(C(=O)O) |
C[C@H]1[C@@H](c2cc(C(F)(F)F)cc(C(F)(F)F)c2)OC(=O)N1CC1=C(c2cc(-c3ccc(C |
C[C@H]1[C@@H](c2cc(C(F)(F)F)cc(C(F)(F)F)c2)OC(=O)N1CC1=C(c2ccc(C(=O)O) |
C=C(C)[C@H]1C[C@@H](C[C@@]23C[C@@H](/C=C/C(C)C)C(C)(C)[C@@](CC=C(C)C)( |
C=C(C)[C@H]1C[C@@H](C[C@]23C[C@@H](/C=C/C(C)C)C(C)(C)[C@](CC=C(C)C)(C2 |