dataset build 2026-08-24 16:04

.gitattributes

@@ -58,3 +58,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+train.jsonl filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,60 @@
+---
+license: mit
+task_categories:
+- text2text-generation
+tags:
+- scheduler
+- text-to-json
+- tagalog
+- taglish
+- synthetic
+- from-scratch
+size_categories:
+- 100K<n<1M
+---
+
+# Schedula dataset
+
+Fully synthetic, seeded, reproducible corpus for training **Schedula** - a tiny
+from-scratch model that turns messy natural language (English / Taglish /
+Filipino) into structured schedule JSON.
+
+## Modes
+| mode | input | output |
+|---|---|---|
+| `event_extract` | sentences, bullet dumps, decorated announcements (unicode stars/stylized headers), edit statements; embedded schedule JSON must be ignored | `{"events":[{title, category, date, start_time, duration_min, location, priority}]}` |
+| `duplicate_detect` | CURRENT LIST json (+ optional `"time_created": "... ##ignore"` metadata) + numbered NEW ITEMS (1-8) | `{"verdicts":[{duplicate, match_id, confidence, reason}]}` |
+
+## Conventions
+- `<today>` anchor date resolves every relative expression ("next week tue",
+  "bukas", "sa loob ng 2 linggo").
+- Fields suffixed ` ##ignore` are app-owned metadata: present in inputs,
+  never copied to outputs.
+- Items under "daily reminders"-style headers are undated (`null`) by design.
+- Category labels come from the per-example schema hint (default enum plus
+  alternate English/Tagalog sets) so the model learns hint-driven mapping.
+- Duplicate boundaries include numbered schoolwork (`quiz #2` vs `quiz #3`
+  never match) and EN<->Filipino translation pairs (do match).
+
+## Splits
+{
+  "train": {
+    "duplicate_detect": 63671,
+    "event_extract": 136329
+  },
+  "val": {
+    "duplicate_detect": 959,
+    "event_extract": 2041
+  },
+  "test": {
+    "duplicate_detect": 974,
+    "event_extract": 2026
+  }
+}
+
+Robustness noise (typos/filler/casing) appears only in val/test (18%).
+
+## Reproducibility
+Built by `python -m schedula.data.build --config configs/default.yaml`
+(seed 1337, created 2026-08-24). The generator lives at
+https://github.com/maxie-12321/schedula

manifest.json

@@ -0,0 +1,26 @@
+{
+  "seed": 1337,
+  "config_data": {
+    "out_dir": "data",
+    "n_train": 200000,
+    "n_val": 3000,
+    "n_test": 3000,
+    "robustness_frac": 0.2,
+    "max_input_chars": 2400,
+    "lang_mix": {
+      "en": 0.45,
+      "taglish": 0.35,
+      "fil": 0.2
+    },
+    "mix": {
+      "event_extract": 0.68,
+      "duplicate_detect": 0.32
+    }
+  },
+  "counts": {
+    "train": 200000,
+    "val": 3000,
+    "test": 3000
+  },
+  "created": "2026-08-24"
+}

train.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ae87c554ad1c79bbfdf0a37f1f4e466efca43b820a2eefd72df6f86e4ae4cec
+size 201240631