File size: 8,739 Bytes
e94bdab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
"""Compute SSE candidate scores for all entries based on the 5-gate screening system.

Populates the ssb_screening block with:
  - gates_passed: list of passed gate names
  - sse_candidate_score: composite score (0-100)
  - thermo_stable: bool (gate 1)
  - electronic_insulation: bool (gate 2)

Gates:
  1. thermo_stability: E_hull < 0.025 eV/atom (stable or near-stable)
  2. electronic_insulation: band_gap > 1.0 eV (not metallic)
  3. ionic_mobility: cavd_channel_dimensionality in ["2D", "3D"] (when available)
  4. electrochemical_window: window_width > 1.0 V (when available)
  5. mechanical: dendrite_suppression_flag (when available)

Score is transparent and compositional:
  - Gate 1 (thermo): 30 points
  - Gate 2 (electronic): 25 points
  - Gate 3 (mobility proxy): 20 points (partial credit for 1D channels)
  - Gate 4 (electrochemical): 15 points
  - Gate 5 (mechanical): 10 points

Usage:
    python scripts/compute_sse_candidate_score.py
    python scripts/compute_sse_candidate_score.py --subset battery
    python scripts/compute_sse_candidate_score.py --limit 10000 --dry-run
"""
import json, os, sys, time, argparse, warnings
from pathlib import Path
warnings.filterwarnings("ignore")

WIDTH = 60

# Gate thresholds
GATES = {
    "thermo_stability": {
        "weight": 30,
        "field": "thermo_stable",
        "description": "E_hull < 0.025 eV/atom",
        "check": lambda e: e.get("ssb_screening", {}).get("thermo_stable", False)
    },
    "electronic_insulation": {
        "weight": 25,
        "field": "electronic_insulation",
        "description": "band_gap > 1.0 eV",
        "check": lambda e: e.get("ssb_screening", {}).get("electronic_insulation", False)
    },
    "ionic_mobility": {
        "weight": 20,
        "field": "cavd_channel_dimensionality",
        "description": "2D/3D percolation channels",
        "check": lambda e: _check_mobility(e)
    },
    "electrochemical_window": {
        "weight": 15,
        "field": "stability_window_low_V",
        "description": "window_width > 1.0 V",
        "check": lambda e: _check_window(e)
    },
    "mechanical": {
        "weight": 10,
        "field": "dendrite_suppression_flag",
        "description": "shear_modulus > 6 GPa",
        "check": lambda e: e.get("ssb_screening", {}).get("dendrite_suppression_flag", False)
    }
}

def _check_mobility(e):
    ss = e.get("ssb_screening", {})
    dim = ss.get("cavd_channel_dimensionality")
    if dim in ("3D",):
        return True
    if dim in ("2D",):
        return True
    if dim in ("1D",):
        # Partial: mobile ions exist but channels are 1D
        return False
    return False

def _check_window(e):
    ss = e.get("ssb_screening", {})
    low = ss.get("stability_window_low_V")
    high = ss.get("stability_window_high_V")
    if low is not None and high is not None:
        return (high - low) >= 1.0
    return False

def _check_mechanical(e):
    return e.get("ssb_screening", {}).get("dendrite_suppression_flag", False)


def compute_gate_score(e, gate_name, gate_config):
    """Compute gate score. Gate passes = full weight, else 0."""
    try:
        passed = gate_config["check"](e)
        return gate_config["weight"] if passed else 0, passed
    except Exception:
        return 0, False


def main():
    parser = argparse.ArgumentParser(description="Compute SSE candidate scores")
    parser.add_argument("--subset", choices=["battery", "electrolyte", "gold", "full"], default="full")
    parser.add_argument("--limit", type=int, default=None)
    parser.add_argument("--dry-run", action="store_true")
    parser.add_argument("--output", type=str, default=None)
    args = parser.parse_args()
    
    if args.limit and not args.dry_run and args.output is None:
        print("ERROR: Refusing to save limited runs. Use --dry-run or --output.")
        sys.exit(1)
    
    BASE_DIR = Path(__file__).resolve().parent.parent
    DATASET_PATH = BASE_DIR / "dataset"
    
    print("=" * WIDTH)
    print("  SSE CANDIDATE SCORE — 5-GATE SCREENING SYSTEM")
    print("=" * WIDTH)
    print()
    print("  Gate weights:")
    for gate_name, config in GATES.items():
        print(f"    {config['weight']:2d} pts — {gate_name}: {config['description']}")
    print()
    
    print("Loading entries...")
    t0 = time.time()
    with open(DATASET_PATH / "entries_final_v3.json") as f:
        all_entries = json.load(f)
    print(f"  {len(all_entries):,} entries ({time.time()-t0:.1f}s)")
    
    # Select working subset
    if args.subset == "battery":
        with open(DATASET_PATH / "battery_candidate_subset_v1.json") as f:
            entries = json.load(f)
    elif args.subset == "electrolyte":
        with open(DATASET_PATH / "solid_electrolyte_candidate_subset_v1.json") as f:
            entries = json.load(f)
    elif args.subset == "gold":
        entries = [e for e in all_entries if e.get("tier") == "gold"]
    else:
        entries = all_entries
    
    if args.limit:
        entries = entries[:args.limit]
    
    print(f"  Working subset: {len(entries):,} entries")
    
    if not entries:
        print("No entries to process.")
        return
    
    # Score all entries
    print(f"\n{'─' * WIDTH}")
    print("  Computing scores...")
    print(f"{'─' * WIDTH}")
    
    score_dist = {}
    gate_counts = {g: {"pass": 0, "total": 0} for g in GATES}
    
    # Track entries that need to be synced back to all_entries
    updated_keys = set()
    
    for idx, e in enumerate(entries):
        if "ssb_screening" not in e:
            e["ssb_screening"] = {}
        
        ss = e["ssb_screening"]
        
        total_score = 0
        gates_passed = []
        
        for gate_name, config in GATES.items():
            score, passed = compute_gate_score(e, gate_name, config)
            total_score += score
            gate_counts[gate_name]["total"] += 1
            if passed:
                gates_passed.append(gate_name)
                gate_counts[gate_name]["pass"] += 1
        
        ss["sse_candidate_score"] = total_score
        ss["gates_passed"] = gates_passed
        
        # Record distribution
        bin_key = f"{(total_score // 10) * 10}-{(total_score // 10) * 10 + 9}"
        score_dist[bin_key] = score_dist.get(bin_key, 0) + 1
        
        # Keep track of which entries were updated
        source_id = e.get("source_id", "") + e.get("source", "")
        updated_keys.add(source_id)
    
    # Print results
    print(f"\n  Score distribution:")
    for key in sorted(score_dist.keys(), key=lambda x: int(x.split("-")[0])):
        count = score_dist[key]
        bar = "█" * min(count // 1000, 50)
        print(f"    {key:>6}: {count:>6,}  {bar}")
    
    print(f"\n  Per-gate pass rates:")
    for gate_name, counts in gate_counts.items():
        pct = counts["pass"] / max(counts["total"], 1) * 100
        print(f"    {gate_name:25s}: {counts['pass']:>6,}/{counts['total']:<6,} ({pct:.1f}%)")
    
    # Top scores
    all_sorted = sorted(entries, key=lambda e: e.get("ssb_screening", {}).get("sse_candidate_score", 0), reverse=True)
    print(f"\n  Top 10 candidates:")
    for e in all_sorted[:10]:
        ss = e.get("ssb_screening", {})
        print(f"    Score {ss.get('sse_candidate_score', 0):3d} | {e.get('structured_formula', e.get('formula','')):20s} | "
              f"{e.get('sse_family', '?'):15s} | Gates: {ss.get('gates_passed', [])}")
    
    # Sync back to all_entries
    if args.subset in ("full",):
        save_data = all_entries
    elif args.subset == "gold":
        save_data = all_entries
        entry_map = {}
        for e in entries:
            key = e.get("source_id", "") + e.get("source", "")
            entry_map[key] = e
        for e in save_data:
            key = e.get("source_id", "") + e.get("source", "")
            if key in entry_map:
                e["ssb_screening"] = entry_map[key].get("ssb_screening", {})
    else:
        save_data = entries
    
    # Save
    if args.subset == "battery":
        output_path = DATASET_PATH / "battery_candidate_subset_v1.json"
    elif args.subset == "electrolyte":
        output_path = DATASET_PATH / "solid_electrolyte_candidate_subset_v1.json"
    elif args.subset == "gold":
        output_path = DATASET_PATH / "entries_final_v3.json"
    else:
        output_path = DATASET_PATH / "entries_final_v3.json"
    
    if args.dry_run:
        print(f"\n  (dry-run — not saved)")
    else:
        print(f"\n  Writing to {output_path}...")
        t_write = time.time()
        with open(output_path, "w") as f:
            json.dump(save_data, f)
        print(f"  Done ({time.time()-t_write:.1f}s)")
    
    print("=" * WIDTH)


if __name__ == "__main__":
    main()