@mailwoman/neural-weights-fr-fr 4.0.0 → 4.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/model-card.json +10 -12
- package/model.onnx +0 -0
- package/package.json +1 -1
package/model-card.json
CHANGED
|
@@ -1,18 +1,16 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "neural-weights-fr-fr",
|
|
3
|
-
"version": "4.
|
|
4
|
-
"model_lineage": "shares the en-us
|
|
5
|
-
"phase": "Stage
|
|
3
|
+
"version": "4.2.0",
|
|
4
|
+
"model_lineage": "shares the en-us v1.0.2-consolidation-runB multi-locale model (step 20000) — shipped as the unified 4.2.0 release version; FR postcode 99.6 / house_number 94.6 (best ever) / region 27.6 (open gap #330)",
|
|
5
|
+
"phase": "Stage 3 — multi-locale (FR via the shared model)",
|
|
6
6
|
"license": "AGPL-3.0-only",
|
|
7
7
|
"locale": "fr-fr",
|
|
8
8
|
"training": {
|
|
9
|
-
"corpus_version": "0.
|
|
10
|
-
"tokenizer_version": "0.
|
|
11
|
-
"steps":
|
|
12
|
-
"hardware": "
|
|
13
|
-
"
|
|
14
|
-
"started_at": null,
|
|
15
|
-
"completed_at": "2026-05-23T06:21:51.190078Z"
|
|
9
|
+
"corpus_version": "0.4.5-unit-v2",
|
|
10
|
+
"tokenizer_version": "0.6.0-a0",
|
|
11
|
+
"steps": 20000,
|
|
12
|
+
"hardware": "NVIDIA A100-SXM4-40GB (Modal cloud)",
|
|
13
|
+
"note": "Identical artifact to neural-weights-en-us 4.1.0 (one multi-locale model serves both). See that card's training/recipe for full provenance."
|
|
16
14
|
},
|
|
17
15
|
"components_supported": [
|
|
18
16
|
"country",
|
|
@@ -183,7 +181,7 @@
|
|
|
183
181
|
"particle-honorific kryptonite (e.g. FR 'Saint-Just-Saint-Rambert') if not in synth set",
|
|
184
182
|
"non-Latin scripts (CJK, Cyrillic) fall through to byte-fallback tokens; F1 unknown"
|
|
185
183
|
],
|
|
186
|
-
"notes": "v0.4.0
|
|
184
|
+
"notes": "v0.4.0 — issue #116. Same encoder geometry as v0.3.0 (8.87M params, 6L/256H/4-heads, 21 BIO labels, linear-chain CRF). Issue proposed (1) per-token CRF NLL normalization + (3) class-weighted CE biased toward coarse labels + (4) source-weight rebalance. Empirical iteration found that §1 and §3 destabilize sustained training at every LR tested (5e-4, 3e-4, 1.5e-4 — even the v0.3.0-safe LR), and on the golden v0.1.2 eval (4535 entries) they slightly REGRESS country/postcode F1 vs v0.3.0. SHIPPED recipe is the §4-only ablation (v0.3.0 dual-loss + v0.4.0 source-weight rebalance) at lr=1.5e-4, step 2200. Modest fine-label gains (street +0.03, house_number +0.01), modest coarse-F1 regression (country -0.07, postcode -0.07). §1/§3 deferred to v0.4.1 corpus-side investigation per issue's '2K divergence' clause. Full ablation matrix retrospective in LOG.md.",
|
|
187
185
|
"format": {
|
|
188
186
|
"model": "ONNX int8 dynamic",
|
|
189
187
|
"tokenizer": "SentencePiece unigram, byte_fallback=true, vocab_size=16000",
|
|
@@ -196,4 +194,4 @@
|
|
|
196
194
|
"model_card": "model-card.json"
|
|
197
195
|
},
|
|
198
196
|
"base_relpath": "/data/models/checkpoints/v0_4_0-stableLR-source-only/step-002200"
|
|
199
|
-
}
|
|
197
|
+
}
|
package/model.onnx
CHANGED
|
Binary file
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mailwoman/neural-weights-fr-fr",
|
|
3
|
-
"version": "4.
|
|
3
|
+
"version": "4.2.0",
|
|
4
4
|
"description": "Mailwoman neural-classifier weights for locale 'fr-fr'. Data-only package — loaded by @mailwoman/neural at runtime.",
|
|
5
5
|
"license": "AGPL-3.0-only",
|
|
6
6
|
"repository": {
|