Add official PerSHOP release with dialogue-level splits

LICENSE-DATA

@@ -0,0 +1,24 @@
+PerSHOP dataset
+Copyright (c) 2023-2026 Keyvan Mahmoudi and Heshaam Faili, University of Tehran
+
+The PerSHOP dataset — the dialogues, their annotations and the product
+database (PerSHOP.xlsx, Products_DataBASE.xlsx, and the files under data/ and
+splits/) — is licensed under the Creative Commons Attribution-NonCommercial 4.0
+International License (CC BY-NC 4.0).
+
+Full legal code: https://creativecommons.org/licenses/by-nc/4.0/legalcode
+Summary:         https://creativecommons.org/licenses/by-nc/4.0/
+
+Attribution: cite
+  K. Mahmoudi and H. Faili, "PerSHOP: A Multi-Domain Persian Dataset for
+  Shopping Dialogue Systems," 2026 International Interdisciplinary Conference
+  on Artificial Intelligence: Engineering, Health, Finance and Humanities
+  (IICAI), Tehran, Iran, 2026, pp. 1-6, doi: 10.1109/IICAI70155.2026.11620935.
+
+Commercial use: any commercial use, including training, fine-tuning or
+deploying machine learning models for commercial products or services, is not
+permitted under this license and requires a separate written agreement with
+the authors.
+
+This license applies to every copy of the dataset, including copies
+redistributed by third parties.

data/products/slots.jsonl

@@ -0,0 +1,35 @@
+{"slot": "product", "description": "The general name of the product, such as milk."}
+{"slot": "category", "description": "The general category of the product such as mutton in the meat product."}
+{"slot": "type", "description": "The product type like honey type in cream product."}
+{"slot": "fat", "description": "Amount of the fat content of a food product such as milk or yogurt."}
+{"slot": "fresh", "description": "Freshness or shelf life of an edible product such as milk."}
+{"slot": "size", "description": "Descriptive size of a product, for example, large-size mineral water."}
+{"slot": "brand", "description": "The brand of a product."}
+{"slot": "sub-brand", "description": "Sub-brand of a famous brand. for example, Pompina is a sub-brand of the Kaleh brand."}
+{"slot": "volume", "description": "The volume of a product such as milk."}
+{"slot": "cheese-shape", "description": "Different shapes of cheeses."}
+{"slot": "weight", "description": "The weight of a product."}
+{"slot": "flavor", "description": "The taste of an edible product such as the taste of a juice."}
+{"slot": "carbonate", "description": "Carbonated or non-carbonated products such as soft drinks."}
+{"slot": "vegetables-of-buttermilk", "description": "Buttermilk with vegetables or without vegetables."}
+{"slot": "family-drink", "description": "Large size for beverage products such as soda or mineral water."}
+{"slot": "package", "description": "The type of packaging of a product such as bottled soda."}
+{"slot": "bag", "description": "The Brewable or urgency of products such as tea or coffee."}
+{"slot": "bag-number", "description": "Number of bags for products such as tea bags."}
+{"slot": "size-number", "description": "The numerical size of some products. Like a shoe in size 36."}
+{"slot": "color", "description": "Descriptive color of a product. For example, a blue T-shirt."}
+{"slot": "color-number", "description": "The numerical color of a product, for example, hair color."}
+{"slot": "country", "description": "Country of manufacture of a product."}
+{"slot": "percent", "description": "The percentage of something in a product, such as the percentage of meat in a hamburger."}
+{"slot": "height", "description": "The height of a product."}
+{"slot": "width", "description": "The width of a product."}
+{"slot": "length", "description": "The length of a product."}
+{"slot": "diag", "description": "The Diameter of a product of a product"}
+{"slot": "material", "description": "Material used in a product for example silk in clothing."}
+{"slot": "pack-number", "description": "The number in the packaging of products that are sold in bulk. For example, a pack involved 8 paper towels."}
+{"slot": "skin-type", "description": "Types of skin for which the product is suitable, such as face cream."}
+{"slot": "hair-type", "description": "Types of hair for which the product is suitable, such as shampoo."}
+{"slot": "gender", "description": "The gender of the person who consumes the product, such as women's shoes."}
+{"slot": "energy-rank", "description": "The energy consumption of an electrical product such as a refrigerator."}
+{"slot": "capacity", "description": "The capacity of a product. for example the capacity of a washing machine."}
+{"slot": "price", "description": "The price of a product."}

scripts/corrections.json

@@ -0,0 +1,25 @@
+{
+  "description": "Corrections applied to the original PerSHOP.xlsx by make_splits.py. Each markup fix only repairs the [span](slot) syntax so the annotation can be parsed; the annotated span and slot are kept as the annotator intended, except where noted.",
+  "markup_fixes": [
+    {"utterance_id": 327, "before": "[1000 میلی لیتری] [دلستر]", "after": "[1000 میلی لیتری](volume) [دلستر]", "reason": "slot label missing"},
+    {"utterance_id": 359, "before": "(volume)[200 میلی لیتر]", "after": "[200 میلی لیتر](volume)", "reason": "span and slot in reversed order"},
+    {"utterance_id": 1063, "before": "[400 گرمی] اگه", "after": "[400 گرمی](weight) اگه", "reason": "slot label missing"},
+    {"utterance_id": 3103, "before": "(volume)[1.5]", "after": "[1.5](volume)", "reason": "span and slot in reversed order"},
+    {"utterance_id": 5190, "before": "[۱.۵ لیتر]ی", "after": "[۱.۵ لیتر](volume)ی", "reason": "slot label missing"},
+    {"utterance_id": 7835, "before": "[کافی میکس](bag_number)[24] (product) عددی", "after": "[کافی میکس](product) [24](bag_number) عددی", "reason": "slot labels attached to the wrong spans (product and bag_number swapped)"},
+    {"utterance_id": 7879, "before": "[کاپوچینو](bag_number)[20] (product) عددی", "after": "[کاپوچینو](product) [20](bag_number) عددی", "reason": "slot labels attached to the wrong spans (product and bag_number swapped)"},
+    {"utterance_id": 10153, "before": "]زیتون](category)", "after": "[زیتون](category)", "reason": "wrong opening bracket"},
+    {"utterance_id": 11155, "before": "]مایع](category)", "after": "[مایع](category)", "reason": "wrong opening bracket"},
+    {"utterance_id": 13081, "before": "[120  گرم] (weight)", "after": "[120  گرم](weight)", "reason": "space between span and slot"},
+    {"utterance_id": 14845, "before": "(weight)[250]", "after": "[250](weight)", "reason": "span and slot in reversed order"},
+    {"utterance_id": 16581, "before": "]بدون لاکتوز](type)", "after": "[بدون لاکتوز](type)", "reason": "wrong opening bracket"},
+    {"utterance_id": 20202, "before": "وانیلی](type)", "after": "[وانیلی](type)", "reason": "opening bracket missing"}
+  ],
+  "removed_utterances": [
+    {"utterance_id": 12536, "reason": "empty user utterance"},
+    {"utterance_id": 13069, "reason": "empty user utterance"}
+  ],
+  "renamed_slots": {
+    "super-product": "super_product"
+  }
+}

scripts/make_splits.py

@@ -0,0 +1,212 @@
+"""Build the official PerSHOP release from the original spreadsheets.
+
+Usage:
+    python scripts/make_splits.py --dialogues PerSHOP.xlsx --products Products_DataBASE.xlsx --out .
+
+Outputs (relative to --out):
+    data/dialogues/{train,validation,test}.jsonl   one row per utterance
+    data/products/products.jsonl                   product database (750 items)
+    data/products/slots.jsonl                      slot names and descriptions
+    splits/{train,validation,test}_dialogue_ids.txt
+    splits/split_stats.json
+
+Splitting protocol
+------------------
+* The unit of splitting is the dialogue: all utterances of a dialogue go to the
+  same split, so no dialogue context leaks between train and evaluation data.
+* Ratio 80/10/10, stratified by each dialogue's primary domain (its most
+  frequent non-"Chats" domain), so every product domain appears in all splits
+  wherever it has enough dialogues.
+* Fixed seed (42). Re-running this script on the same input reproduces the
+  same split exactly.
+
+Corrections
+-----------
+The documented corrections in scripts/corrections.json are applied to the
+original annotations: 13 markup-syntax fixes, removal of 2 empty utterances,
+and renaming the slot "super-product" to "super_product". Each fix is checked
+against the original text, so the script fails if the input has changed.
+`raw_text` is kept verbatim. `text` and `entities` are derived from the
+corrected `annotated_text` by parsing the inline markup "[span](slot)".
+"""
+
+import argparse
+import json
+import random
+import re
+from collections import Counter
+from pathlib import Path
+
+import pandas as pd
+
+SEED = 42
+RATIOS = {"train": 0.8, "validation": 0.1, "test": 0.1}
+
+DOMAIN_EN = {
+    "Chats": "chit-chat",
+    "لبنیات": "dairy",
+    "نوشیدنی": "beverages",
+    "خوار و بار": "groceries",
+    "خانه و سبک زندگی": "home and lifestyle",
+    "پروتئین": "protein",
+    "کودک": "baby and child",
+    "کنسروجات": "canned food",
+    "آرایشی و بهداشتی": "cosmetics and personal care",
+    "خشکبار، دسر و شیرینی": "nuts, desserts and sweets",
+    "میوه و سبزیجات": "fruits and vegetables",
+    "مد و پوشاک": "fashion and clothing",
+    "دستمال و شوینده": "tissues and detergents",
+    "دیجیتال": "digital",
+    "خودرو، موتور، دوچرخه": "car, motorcycle and bicycle",
+}
+
+MARKUP = re.compile(r"\[([^\[\]]+?)\]\(([^()]+?)\)")
+
+
+def clean(value):
+    """Convert pandas missing values to None and strip strings."""
+    if pd.isna(value):
+        return None
+    return str(value).strip() if isinstance(value, str) else value
+
+
+def parse_markup(annotated):
+    """Return plain text and entity spans (character offsets on the plain text)."""
+    if annotated is None:
+        return None, []
+    text_parts, entities, cursor, out_len = [], [], 0, 0
+    for match in MARKUP.finditer(annotated):
+        before = annotated[cursor:match.start()]
+        text_parts.append(before)
+        out_len += len(before)
+        span, slot = match.group(1), match.group(2).strip()
+        entities.append({"value": span, "slot": slot, "start": out_len, "end": out_len + len(span)})
+        text_parts.append(span)
+        out_len += len(span)
+        cursor = match.end()
+    text_parts.append(annotated[cursor:])
+    return "".join(text_parts), entities
+
+
+def apply_corrections(df, corrections_path):
+    """Apply the documented corrections, checking each against the original text."""
+    corrections = json.loads(corrections_path.read_text(encoding="utf-8"))
+    df = df.copy()
+    for fix in corrections["markup_fixes"]:
+        mask = df.id == fix["utterance_id"]
+        original = df.loc[mask, "user_text"].iloc[0]
+        if original.count(fix["before"]) != 1:
+            raise ValueError(f"Correction does not match utterance {fix['utterance_id']}")
+        df.loc[mask, "user_text"] = original.replace(fix["before"], fix["after"])
+    removed = [r["utterance_id"] for r in corrections["removed_utterances"]]
+    if not df.loc[df.id.isin(removed), "user_text"].isna().all():
+        raise ValueError("An utterance listed for removal is not empty")
+    df = df[~df.id.isin(removed)]
+    for old, new in corrections["renamed_slots"].items():
+        df["user_text"] = df.user_text.str.replace(f"]({old})", f"]({new})", regex=False)
+    return df
+
+
+def split_dialogues(primary_domain):
+    """Stratified dialogue-level split with a fixed seed."""
+    rng = random.Random(SEED)
+    assignment = {}
+    for domain in sorted(primary_domain.unique()):
+        ids = sorted(primary_domain[primary_domain == domain].index.tolist())
+        rng.shuffle(ids)
+        n = len(ids)
+        n_val = round(n * RATIOS["validation"])
+        n_test = round(n * RATIOS["test"])
+        for i, dialogue_id in enumerate(ids):
+            if i < n_test:
+                assignment[dialogue_id] = "test"
+            elif i < n_test + n_val:
+                assignment[dialogue_id] = "validation"
+            else:
+                assignment[dialogue_id] = "train"
+    return assignment
+
+
+def build_rows(df):
+    rows = []
+    df = df.sort_values(["dialogue_number", "id"])
+    for dialogue_id, group in df.groupby("dialogue_number", sort=True):
+        for turn_index, record in enumerate(group.itertuples(index=False)):
+            annotated = clean(record.user_text)
+            text, entities = parse_markup(annotated)
+            domain = clean(record.domain)
+            rows.append({
+                "utterance_id": int(record.id),
+                "dialogue_id": int(dialogue_id),
+                "turn_index": turn_index,
+                "speaker": clean(record.Speecher).lower(),
+                "domain": domain,
+                "domain_en": DOMAIN_EN.get(domain, domain),
+                "intent": clean(record.intent),
+                "super_product": clean(record.SuperProduct),
+                "text": text,
+                "annotated_text": annotated,
+                "raw_text": clean(record.RawText),
+                "entities": entities,
+            })
+    return rows
+
+
+def write_jsonl(path, records):
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with path.open("w", encoding="utf-8") as f:
+        for record in records:
+            f.write(json.dumps(record, ensure_ascii=False) + "\n")
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--dialogues", required=True)
+    parser.add_argument("--products", required=True)
+    parser.add_argument("--out", default=".")
+    args = parser.parse_args()
+    out = Path(args.out)
+
+    df = pd.read_excel(args.dialogues, sheet_name="Chats").rename(columns={"Intents & Actions": "intent"})
+    df = apply_corrections(df, Path(__file__).with_name("corrections.json"))
+
+    product_turns = df[df.domain != "Chats"]
+    primary = product_turns.groupby("dialogue_number").domain.agg(lambda s: s.value_counts().index[0])
+    primary = primary.reindex(sorted(df.dialogue_number.unique())).fillna("Chats")
+    assignment = split_dialogues(primary)
+
+    rows = build_rows(df)
+    stats = {"seed": SEED, "ratios": RATIOS, "splits": {}}
+    for split in RATIOS:
+        split_rows = [r for r in rows if assignment[r["dialogue_id"]] == split]
+        write_jsonl(out / "data" / "dialogues" / f"{split}.jsonl", split_rows)
+        dialogue_ids = sorted({r["dialogue_id"] for r in split_rows})
+        ids_path = out / "splits" / f"{split}_dialogue_ids.txt"
+        ids_path.parent.mkdir(parents=True, exist_ok=True)
+        ids_path.write_text("\n".join(map(str, dialogue_ids)) + "\n", encoding="utf-8")
+        stats["splits"][split] = {
+            "dialogues": len(dialogue_ids),
+            "utterances": len(split_rows),
+            "primary_domain_dialogues": dict(Counter(DOMAIN_EN[primary[i]] for i in dialogue_ids).most_common()),
+        }
+
+    products = pd.read_excel(args.products, sheet_name="woz_product")
+    product_records = [
+        {col.replace("-", "_"): clean(val) for col, val in rec.items() if clean(val) is not None}
+        for rec in products.to_dict(orient="records")
+    ]
+    for rec in product_records:
+        rec["domain_en"] = DOMAIN_EN.get(rec.get("domain"), rec.get("domain"))
+    write_jsonl(out / "data" / "products" / "products.jsonl", product_records)
+
+    slots = pd.read_excel(args.products, sheet_name="Slots")
+    write_jsonl(out / "data" / "products" / "slots.jsonl",
+                [{"slot": clean(s), "description": clean(d)} for s, d in zip(slots.iloc[:, 0], slots.iloc[:, 1])])
+
+    stats["products"] = len(product_records)
+    (out / "splits" / "split_stats.json").write_text(json.dumps(stats, ensure_ascii=False, indent=2), encoding="utf-8")
+    print(json.dumps(stats, ensure_ascii=False, indent=2))
+
+
+if __name__ == "__main__":
+    main()

splits/split_stats.json

@@ -0,0 +1,70 @@
+{
+  "seed": 42,
+  "ratios": {
+    "train": 0.8,
+    "validation": 0.1,
+    "test": 0.1
+  },
+  "splits": {
+    "train": {
+      "dialogues": 849,
+      "utterances": 17482,
+      "primary_domain_dialogues": {
+        "dairy": 258,
+        "beverages": 240,
+        "groceries": 103,
+        "home and lifestyle": 43,
+        "baby and child": 28,
+        "canned food": 27,
+        "cosmetics and personal care": 25,
+        "nuts, desserts and sweets": 24,
+        "fruits and vegetables": 23,
+        "protein": 20,
+        "digital": 19,
+        "fashion and clothing": 19,
+        "tissues and detergents": 17,
+        "car, motorcycle and bicycle": 2,
+        "chit-chat": 1
+      }
+    },
+    "validation": {
+      "dialogues": 106,
+      "utterances": 2242,
+      "primary_domain_dialogues": {
+        "dairy": 32,
+        "beverages": 30,
+        "groceries": 13,
+        "home and lifestyle": 6,
+        "canned food": 4,
+        "protein": 3,
+        "fruits and vegetables": 3,
+        "cosmetics and personal care": 3,
+        "baby and child": 3,
+        "nuts, desserts and sweets": 3,
+        "fashion and clothing": 2,
+        "digital": 2,
+        "tissues and detergents": 2
+      }
+    },
+    "test": {
+      "dialogues": 106,
+      "utterances": 2199,
+      "primary_domain_dialogues": {
+        "dairy": 32,
+        "beverages": 30,
+        "groceries": 13,
+        "home and lifestyle": 6,
+        "canned food": 4,
+        "nuts, desserts and sweets": 3,
+        "fruits and vegetables": 3,
+        "baby and child": 3,
+        "cosmetics and personal care": 3,
+        "protein": 3,
+        "digital": 2,
+        "tissues and detergents": 2,
+        "fashion and clothing": 2
+      }
+    }
+  },
+  "products": 750
+}

splits/test_dialogue_ids.txt

@@ -0,0 +1,106 @@
+1
+8
+23
+39
+97
+100
+105
+110
+119
+150
+160
+180
+182
+185
+199
+200
+210
+214
+217
+223
+228
+243
+246
+259
+277
+297
+298
+306
+338
+343
+349
+358
+360
+366
+369
+381
+387
+388
+432
+448
+450
+472
+488
+490
+500
+517
+524
+525
+529
+549
+556
+565
+591
+592
+618
+621
+629
+632
+648
+653
+654
+663
+675
+678
+680
+682
+694
+698
+706
+715
+716
+727
+729
+731
+746
+752
+772
+781
+789
+800
+802
+842
+867
+874
+893
+897
+907
+915
+920
+923
+945
+947
+967
+970
+985
+991
+995
+1015
+1016
+1022
+1029
+1031
+1039
+1055
+1056
+1058

splits/train_dialogue_ids.txt

@@ -0,0 +1,849 @@
+2
+3
+4
+5
+6
+7
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+24
+26
+27
+28
+29
+30
+31
+32
+33
+34
+36
+37
+38
+40
+42
+43
+44
+45
+46
+47
+48
+49
+50
+51
+52
+53
+54
+55
+56
+57
+58
+59
+60
+61
+62
+63
+64
+66
+67
+68
+69
+70
+71
+72
+73
+75
+76
+77
+78
+79
+81
+82
+83
+84
+85
+86
+87
+89
+90
+91
+92
+93
+94
+96
+98
+99
+101
+102
+103
+104
+106
+107
+108
+109
+111
+112
+114
+115
+116
+118
+120
+121
+122
+123
+124
+125
+126
+127
+129
+131
+132
+133
+136
+137
+138
+139
+140
+141
+142
+143
+144
+145
+146
+147
+148
+149
+151
+152
+153
+154
+155
+156
+157
+158
+159
+161
+162
+163
+164
+166
+167
+168
+171
+172
+173
+175
+176
+177
+178
+179
+181
+183
+184
+186
+187
+188
+190
+191
+192
+193
+194
+195
+197
+198
+201
+202
+203
+204
+205
+206
+207
+208
+209
+211
+212
+213
+215
+218
+219
+220
+221
+226
+227
+230
+231
+232
+233
+234
+235
+236
+237
+239
+240
+241
+242
+244
+245
+247
+248
+249
+250
+251
+252
+253
+254
+255
+256
+257
+258
+260
+261
+262
+263
+265
+266
+267
+268
+269
+270
+271
+272
+273
+274
+275
+276
+278
+279
+280
+282
+283
+285
+286
+287
+288
+289
+290
+291
+292
+293
+295
+296
+299
+300
+301
+302
+303
+304
+305
+307
+308
+309
+310
+311
+312
+313
+314
+315
+316
+317
+318
+319
+320
+321
+322
+323
+324
+325
+326
+327
+328
+329
+331
+332
+333
+334
+335
+336
+337
+339
+340
+341
+344
+345
+347
+348
+350
+351
+352
+353
+354
+355
+356
+357
+359
+361
+362
+363
+364
+365
+367
+368
+370
+372
+373
+374
+375
+376
+377
+378
+379
+380
+382
+383
+384
+385
+386
+389
+390
+391
+393
+394
+395
+396
+397
+398
+399
+400
+401
+402
+403
+404
+405
+406
+407
+408
+409
+410
+412
+413
+414
+415
+416
+417
+418
+419
+420
+421
+422
+424
+425
+426
+427
+428
+429
+430
+431
+433
+434
+435
+436
+437
+438
+440
+441
+442
+443
+444
+445
+446
+447
+449
+451
+452
+453
+454
+456
+457
+458
+460
+462
+464
+465
+466
+467
+468
+469
+470
+471
+473
+474
+475
+476
+477
+478
+480
+481
+482
+483
+484
+485
+486
+487
+489
+491
+494
+495
+496
+497
+498
+499
+501
+502
+503
+504
+507
+508
+509
+510
+511
+512
+513
+514
+515
+516
+518
+519
+520
+521
+523
+527
+528
+530
+531
+532
+533
+534
+535
+536
+537
+538
+539
+540
+541
+542
+543
+544
+545
+546
+547
+548
+550
+552
+553
+554
+555
+558
+559
+560
+561
+562
+563
+564
+566
+568
+569
+570
+571
+572
+573
+574
+576
+577
+579
+580
+581
+582
+583
+584
+585
+586
+587
+588
+589
+590
+593
+594
+595
+596
+597
+599
+600
+601
+602
+603
+604
+605
+606
+607
+608
+610
+611
+612
+613
+614
+615
+616
+617
+619
+620
+622
+623
+624
+625
+626
+627
+628
+630
+631
+633
+634
+636
+637
+638
+639
+640
+641
+642
+643
+644
+645
+646
+647
+649
+650
+651
+652
+655
+656
+658
+659
+660
+661
+662
+664
+665
+666
+667
+668
+669
+670
+671
+672
+673
+674
+676
+677
+679
+681
+683
+684
+685
+686
+687
+688
+689
+690
+691
+692
+693
+695
+696
+697
+699
+700
+701
+702
+703
+704
+705
+707
+708
+709
+710
+712
+714
+717
+718
+719
+720
+721
+722
+723
+724
+725
+726
+728
+730
+732
+733
+734
+735
+736
+737
+738
+740
+741
+742
+743
+744
+745
+747
+748
+749
+750
+751
+753
+754
+755
+756
+757
+758
+759
+760
+761
+763
+764
+765
+767
+768
+770
+771
+773
+774
+776
+777
+778
+779
+780
+782
+783
+784
+785
+786
+787
+788
+790
+791
+792
+793
+794
+795
+796
+797
+798
+799
+801
+803
+804
+805
+806
+807
+808
+809
+810
+811
+812
+814
+815
+816
+817
+818
+819
+820
+821
+822
+823
+825
+826
+828
+829
+830
+832
+833
+834
+835
+836
+837
+839
+840
+841
+843
+845
+846
+847
+848
+849
+851
+852
+853
+854
+855
+856
+857
+858
+859
+860
+861
+863
+864
+865
+866
+869
+870
+871
+872
+873
+875
+876
+877
+878
+880
+882
+883
+884
+885
+886
+887
+888
+889
+890
+891
+894
+895
+896
+898
+899
+900
+901
+902
+905
+906
+908
+909
+912
+913
+914
+916
+917
+918
+919
+921
+924
+925
+926
+927
+929
+930
+931
+932
+933
+934
+935
+936
+937
+939
+940
+941
+944
+948
+949
+950
+951
+952
+953
+954
+955
+956
+958
+960
+961
+962
+963
+964
+965
+968
+971
+972
+973
+974
+975
+976
+978
+979
+980
+981
+982
+984
+986
+987
+988
+989
+992
+994
+996
+998
+1000
+1001
+1002
+1003
+1004
+1005
+1006
+1007
+1008
+1010
+1011
+1012
+1013
+1014
+1017
+1018
+1019
+1020
+1021
+1024
+1026
+1027
+1028
+1030
+1033
+1034
+1035
+1036
+1037
+1038
+1041
+1042
+1044
+1046
+1048
+1049
+1050
+1051
+1053
+1054
+1057
+1059
+1060
+1061

splits/validation_dialogue_ids.txt

@@ -0,0 +1,106 @@
+25
+35
+41
+65
+74
+80
+88
+95
+113
+117
+128
+130
+134
+135
+165
+169
+170
+174
+189
+196
+216
+222
+224
+225
+229
+238
+264
+281
+284
+294
+330
+342
+346
+371
+392
+411
+423
+439
+455
+459
+461
+463
+479
+492
+493
+505
+506
+522
+526
+551
+557
+567
+575
+578
+598
+609
+635
+657
+711
+713
+739
+762
+766
+769
+775
+813
+824
+827
+831
+838
+844
+850
+862
+868
+879
+881
+892
+903
+904
+910
+911
+922
+928
+938
+942
+943
+946
+957
+959
+966
+969
+977
+983
+990
+993
+997
+999
+1009
+1023
+1025
+1032
+1040
+1043
+1045
+1047
+1052