Release 722975 product configurations with attributes and menu options
Browse filesFour verified city catalog shards; anonymous location joins for 8568 requests; stable user/request IDs; normalized attributes and source menu options; checksum manifests. No raw responses, coordinates, credentials or evaluation answers.
- README.md +70 -30
- catalog_beijing.parquet +3 -0
- catalog_guangzhou.parquet +3 -0
- catalog_manifest.json +109 -0
- catalog_shanghai.parquet +3 -0
- catalog_shenzhen.parquet +3 -0
- locations.jsonl +20 -0
- locations.parquet +3 -0
- preview_manifest.json +60 -33
- requests.jsonl +0 -0
- requests.parquet +3 -0
- users.parquet +3 -0
README.md
CHANGED
|
@@ -3,7 +3,7 @@ pretty_name: DeliveryBench
|
|
| 3 |
language:
|
| 4 |
- zh
|
| 5 |
size_categories:
|
| 6 |
-
-
|
| 7 |
tags:
|
| 8 |
- agents
|
| 9 |
- personalization
|
|
@@ -11,73 +11,113 @@ tags:
|
|
| 11 |
- food-ordering
|
| 12 |
- longitudinal
|
| 13 |
configs:
|
| 14 |
-
- config_name:
|
| 15 |
default: true
|
| 16 |
data_files:
|
| 17 |
- split: full
|
| 18 |
-
path:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
- config_name: requests
|
| 20 |
data_files:
|
| 21 |
- split: full
|
| 22 |
-
path: requests.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
---
|
| 24 |
|
| 25 |
# DeliveryBench
|
| 26 |
|
| 27 |
DeliveryBench supports personalized food-ordering agents through real, location-specific menus, constructed user profiles and longitudinal natural-language requests. It covers Guangzhou, Shenzhen, Shanghai and Beijing.
|
| 28 |
|
| 29 |
-
**
|
| 30 |
|
| 31 |
## Dataset scale
|
| 32 |
|
| 33 |
-
|
|
| 34 |
-
| --- | ---: | ---: |
|
| 35 |
-
|
|
| 36 |
-
|
|
| 37 |
-
|
|
| 38 |
-
|
|
| 39 |
-
|
|
| 40 |
-
| Audited menu configurations | 722,975 | To be released |
|
| 41 |
|
| 42 |
-
The
|
| 43 |
|
| 44 |
-
|
|
| 45 |
| --- | ---: | ---: |
|
| 46 |
| Training | 700 | 5,938 |
|
| 47 |
| Validation | 150 | 1,239 |
|
| 48 |
| Test | 150 | 1,391 |
|
| 49 |
-
| Total | 1,000 | 8,568 |
|
| 50 |
|
| 51 |
-
|
| 52 |
|
| 53 |
-
## Files and
|
| 54 |
|
| 55 |
-
|
|
| 56 |
| --- | --- | ---: |
|
| 57 |
-
| `users.jsonl` |
|
| 58 |
-
| `requests.jsonl` |
|
| 59 |
-
| `
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
Profiles describe constructed preferences rather than identified people. Real ordering traces inform activity, timing and expenditure context; semantic food-family and attribute preferences are separately constructed and grounded in menu supply. Natural-language requests are synthetic ordering inputs, not real customer conversations.
|
| 62 |
|
| 63 |
-
|
| 64 |
|
| 65 |
-
|
| 66 |
|
| 67 |
## Loading
|
| 68 |
|
| 69 |
```python
|
| 70 |
from datasets import load_dataset
|
| 71 |
|
|
|
|
| 72 |
users = load_dataset("fstxwd/DeliveryBench", "users", split="full")
|
| 73 |
requests = load_dataset("fstxwd/DeliveryBench", "requests", split="full")
|
|
|
|
| 74 |
|
| 75 |
-
|
| 76 |
-
train_requests = requests.filter(lambda row: row["split"] == "train")
|
| 77 |
-
|
| 78 |
assert len(users) == 1000 and len(requests) == 8568
|
| 79 |
-
|
| 80 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
```
|
| 82 |
|
| 83 |
-
The catalog
|
|
|
|
| 3 |
language:
|
| 4 |
- zh
|
| 5 |
size_categories:
|
| 6 |
+
- 100K<n<1M
|
| 7 |
tags:
|
| 8 |
- agents
|
| 9 |
- personalization
|
|
|
|
| 11 |
- food-ordering
|
| 12 |
- longitudinal
|
| 13 |
configs:
|
| 14 |
+
- config_name: catalogs
|
| 15 |
default: true
|
| 16 |
data_files:
|
| 17 |
- split: full
|
| 18 |
+
path: catalog_*.parquet
|
| 19 |
+
- config_name: users
|
| 20 |
+
data_files:
|
| 21 |
+
- split: full
|
| 22 |
+
path: users.parquet
|
| 23 |
- config_name: requests
|
| 24 |
data_files:
|
| 25 |
- split: full
|
| 26 |
+
path: requests.parquet
|
| 27 |
+
- config_name: locations
|
| 28 |
+
data_files:
|
| 29 |
+
- split: full
|
| 30 |
+
path: locations.parquet
|
| 31 |
---
|
| 32 |
|
| 33 |
# DeliveryBench
|
| 34 |
|
| 35 |
DeliveryBench supports personalized food-ordering agents through real, location-specific menus, constructed user profiles and longitudinal natural-language requests. It covers Guangzhou, Shenzhen, Shanghai and Beijing.
|
| 36 |
|
| 37 |
+
**Available data: 1,000 synthetic user profiles, 8,568 eligible ordering requests and 722,975 menu configurations with product attributes and selected specification options, covering 20 episode-linked delivery locations.**
|
| 38 |
|
| 39 |
## Dataset scale
|
| 40 |
|
| 41 |
+
| City | Users | Eligible requests | Menu configurations | Catalog locations |
|
| 42 |
+
| --- | ---: | ---: | ---: | ---: |
|
| 43 |
+
| Guangzhou | 250 | 2,305 | 139,724 | 5 |
|
| 44 |
+
| Shenzhen | 250 | 1,916 | 223,439 | 5 |
|
| 45 |
+
| Shanghai | 250 | 2,110 | 232,427 | 5 |
|
| 46 |
+
| Beijing | 250 | 2,237 | 127,385 | 5 |
|
| 47 |
+
| Total | 1,000 | 8,568 | 722,975 | 20 |
|
|
|
|
| 48 |
|
| 49 |
+
The original constructed corpus contains 8,873 episodes; 8,568 passed the menu-feasibility audit and appear in this release. Each city contributes 250 user profiles. The `split` column retains the original user-disjoint assignment.
|
| 50 |
|
| 51 |
+
| Partition | Users | Eligible requests |
|
| 52 |
| --- | ---: | ---: |
|
| 53 |
| Training | 700 | 5,938 |
|
| 54 |
| Validation | 150 | 1,239 |
|
| 55 |
| Test | 150 | 1,391 |
|
|
|
|
| 56 |
|
| 57 |
+
The broader collection targets 200 locations, with 198 completed, and contains 64,095 complete menu snapshots, 4,926,925 product observations and 35,052,171 specification-option observations. Those broader collection counts are not the downloadable catalog's size. This release contains the audited 722,975-configuration catalog at the 20 episode-linked locations. Configurations and location-specific observations are not counts of unique dishes or unique businesses.
|
| 58 |
|
| 59 |
+
## Files and joins
|
| 60 |
|
| 61 |
+
| Files | Contents | Rows |
|
| 62 |
| --- | --- | ---: |
|
| 63 |
+
| `users.parquet`; `users.jsonl` | User ID, city, partition and constructed food/drink preferences | 1,000 each |
|
| 64 |
+
| `requests.parquet`; `requests.jsonl` | Natural-language requests, sequence, time context, partition and catalog location | 8,568 each |
|
| 65 |
+
| `locations.parquet`; `locations.jsonl` | Anonymous location ID, city, catalog count, request count and collection status | 20 each |
|
| 66 |
+
| `catalog_guangzhou.parquet`, `catalog_shenzhen.parquet`, `catalog_shanghai.parquet`, `catalog_beijing.parquet` | Location-specific menu configurations, product information, attributes and selected options | 722,975 combined |
|
| 67 |
+
| `catalog_manifest.json` | Per-file checksums, row counts, attribute statistics, missingness and integrity checks | — |
|
| 68 |
+
| `preview_manifest.json` | Versioned release scope and file checksums | — |
|
| 69 |
+
|
| 70 |
+
Parquet and JSONL copies of the small input tables contain the same records. Requests join to profiles through `user_id` and to the catalog through `location_id`. Anonymous location IDs preserve these joins without publishing addresses or coordinates. `sequence_index` preserves request order within each user. Existing user/request IDs are retained from the input-data release.
|
| 71 |
+
|
| 72 |
+
## Product configuration fields
|
| 73 |
+
|
| 74 |
+
Each catalog row represents one enumerated configuration from a location-specific menu snapshot.
|
| 75 |
+
|
| 76 |
+
| Field | Meaning |
|
| 77 |
+
| --- | --- |
|
| 78 |
+
| `configuration_id` | Unique opaque configuration identifier |
|
| 79 |
+
| `location_id`, `city` | Anonymous catalog location and city |
|
| 80 |
+
| `shop_id`, `item_id`, `sku_id` | Opaque, location-scoped merchant, item and SKU identifiers |
|
| 81 |
+
| `name`, `shop_name`, `brand` | Product name, merchant name and recorded brand |
|
| 82 |
+
| `family`, `domain` | Processed food-family and food/drink classification |
|
| 83 |
+
| `sku_name` | Source SKU label, such as size or portion |
|
| 84 |
+
| `attributes` | Normalized menu-derived `{name, value}` pairs, including available size, sweetness, ice level or spiciness fields |
|
| 85 |
+
| `options` | Selected source-menu `{group, value, option_id}` entries, preserving labels such as temperature, sweetness and add-ins |
|
| 86 |
+
| `price_cents`, `price_scope` | Recorded base/SKU price in integer cents and its scope |
|
| 87 |
+
| `rating`, `sales`, `eta`, `distance` | Available provider service-quality fields; missing evidence remains null |
|
| 88 |
+
| `checkout_verified` | Whether the source configuration was verified at checkout |
|
| 89 |
+
|
| 90 |
+
Menu labels and normalized attributes describe the available menu evidence. They are not independent ingredient, allergy or preparation guarantees. `price_scope` distinguishes base/SKU prices from checkout totals; selected extras and final prices may remain unverified. Provider quality fields retain their source values and missingness; no common cross-provider scale or unit conversion is asserted.
|
| 91 |
+
|
| 92 |
+
The catalog preserves the source enumeration bounds. `bounded_complete` in the location table denotes completion within recorded collection/enumeration bounds, not a guarantee that every platform merchant or possible option combination was obtained. Opaque IDs support snapshot-local joins, not global merchant identity resolution.
|
| 93 |
+
|
| 94 |
+
## Users and requests
|
| 95 |
|
| 96 |
Profiles describe constructed preferences rather than identified people. Real ordering traces inform activity, timing and expenditure context; semantic food-family and attribute preferences are separately constructed and grounded in menu supply. Natural-language requests are synthetic ordering inputs, not real customer conversations.
|
| 97 |
|
| 98 |
+
Profile preferences contain domain/family entries and personal attribute name/value pairs. `hour` is 0–23; `weekday` retains the source 0–6 encoding, and `period` retains the source categorical code. `task_type` preserves the original request category.
|
| 99 |
|
| 100 |
+
Raw API responses, credentials, precise addresses/coordinates, source-user linkage hashes and per-episode evaluation answers are excluded. Contact-like strings in menu text are removed. The catalog and input tables support data inspection and retrieval/ranking integration; the version-matched evaluation labels and evaluation resources will be available soon. Profile fields supplied for inspection should not be treated as hidden-goal evaluation observations.
|
| 101 |
|
| 102 |
## Loading
|
| 103 |
|
| 104 |
```python
|
| 105 |
from datasets import load_dataset
|
| 106 |
|
| 107 |
+
catalogs = load_dataset("fstxwd/DeliveryBench", "catalogs", split="full")
|
| 108 |
users = load_dataset("fstxwd/DeliveryBench", "users", split="full")
|
| 109 |
requests = load_dataset("fstxwd/DeliveryBench", "requests", split="full")
|
| 110 |
+
locations = load_dataset("fstxwd/DeliveryBench", "locations", split="full")
|
| 111 |
|
| 112 |
+
assert len(catalogs) == 722975
|
|
|
|
|
|
|
| 113 |
assert len(users) == 1000 and len(requests) == 8568
|
| 114 |
+
|
| 115 |
+
request = requests[0]
|
| 116 |
+
local_candidates = catalogs.filter(
|
| 117 |
+
lambda row: row["location_id"] == request["location_id"]
|
| 118 |
+
)
|
| 119 |
+
print(request["message"])
|
| 120 |
+
print(local_candidates[0])
|
| 121 |
```
|
| 122 |
|
| 123 |
+
For a specific partition, filter the input tables by their `split` column. The catalog is partitioned by city for storage; `location_id` identifies the candidate pool for a request. No purchase submission or payment is needed to inspect these files.
|
catalog_beijing.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9827939b98ea97e5f42cdd6dffd676a71111a2151648e43d6055ff7923cb796a
|
| 3 |
+
size 6872799
|
catalog_guangzhou.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff9643a395a5bcc99bc5e6694cdf27684e066eb60c131b02c1111e358dff54ed
|
| 3 |
+
size 7036812
|
catalog_manifest.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dataset": "DeliveryBench",
|
| 3 |
+
"version": "0.3.0-catalog",
|
| 4 |
+
"configurations": 722975,
|
| 5 |
+
"locations": 20,
|
| 6 |
+
"cities": 4,
|
| 7 |
+
"users": 1000,
|
| 8 |
+
"requests": 8568,
|
| 9 |
+
"files": {
|
| 10 |
+
"catalog_beijing.parquet": {
|
| 11 |
+
"rows": 127385,
|
| 12 |
+
"bytes": 6872799,
|
| 13 |
+
"sha256": "9827939b98ea97e5f42cdd6dffd676a71111a2151648e43d6055ff7923cb796a"
|
| 14 |
+
},
|
| 15 |
+
"catalog_guangzhou.parquet": {
|
| 16 |
+
"rows": 139724,
|
| 17 |
+
"bytes": 7036812,
|
| 18 |
+
"sha256": "ff9643a395a5bcc99bc5e6694cdf27684e066eb60c131b02c1111e358dff54ed"
|
| 19 |
+
},
|
| 20 |
+
"catalog_shanghai.parquet": {
|
| 21 |
+
"rows": 232427,
|
| 22 |
+
"bytes": 12066182,
|
| 23 |
+
"sha256": "563d14b3b2ab8dcaf6851576f85323d2a570538dfd05f5b05ab5167850d3a65b"
|
| 24 |
+
},
|
| 25 |
+
"catalog_shenzhen.parquet": {
|
| 26 |
+
"rows": 223439,
|
| 27 |
+
"bytes": 11157363,
|
| 28 |
+
"sha256": "7f2331c0e43662172cfc7998714227e72d08bbfab731968415282211d0fd9e37"
|
| 29 |
+
},
|
| 30 |
+
"requests.jsonl": {
|
| 31 |
+
"rows": 8568,
|
| 32 |
+
"bytes": 2184886,
|
| 33 |
+
"sha256": "c1feaa6d79cdf0ec0acf4806129a3cff1b682f73f6398b672c298a0442e93d71"
|
| 34 |
+
},
|
| 35 |
+
"locations.jsonl": {
|
| 36 |
+
"rows": 20,
|
| 37 |
+
"bytes": 2598,
|
| 38 |
+
"sha256": "17313ff3a92824b54ae25c7b04c0eb473a331420a5f964c53cd39bf04004c2c6"
|
| 39 |
+
},
|
| 40 |
+
"users.parquet": {
|
| 41 |
+
"rows": 1000,
|
| 42 |
+
"bytes": 12640,
|
| 43 |
+
"sha256": "a4b87afd008a17e521c67de0323c551087a1515d9c4adde8dbe59afe68f92f57"
|
| 44 |
+
},
|
| 45 |
+
"users.jsonl": {
|
| 46 |
+
"rows": 1000,
|
| 47 |
+
"bytes": 281782,
|
| 48 |
+
"sha256": "91d355e14afe47b36dffa1a60d6d1de2ec4572ae3e2b00bc7305d6647292c20e"
|
| 49 |
+
},
|
| 50 |
+
"requests.parquet": {
|
| 51 |
+
"rows": 8568,
|
| 52 |
+
"bytes": 49664,
|
| 53 |
+
"sha256": "7cf8a18253a962be0bbde67c6211fd7e4fd04230783f2d0eaab557638a48a793"
|
| 54 |
+
},
|
| 55 |
+
"locations.parquet": {
|
| 56 |
+
"rows": 20,
|
| 57 |
+
"bytes": 2379,
|
| 58 |
+
"sha256": "5cca8f7513dd749b429bfc97001be7c53cafdb98faf5bffd37ee10a1e5257e29"
|
| 59 |
+
}
|
| 60 |
+
},
|
| 61 |
+
"per_city_configurations": {
|
| 62 |
+
"广州市": 139724,
|
| 63 |
+
"深圳市": 223439,
|
| 64 |
+
"上海市": 232427,
|
| 65 |
+
"北京市": 127385
|
| 66 |
+
},
|
| 67 |
+
"configuration_statistics": {
|
| 68 |
+
"with_options": 669976,
|
| 69 |
+
"with_attributes": 664330,
|
| 70 |
+
"option_selections": 2418987,
|
| 71 |
+
"checkout_verified": 0
|
| 72 |
+
},
|
| 73 |
+
"attribute_occurrences": {
|
| 74 |
+
"size": 472464,
|
| 75 |
+
"sweetness": 367148,
|
| 76 |
+
"ice_level": 236908,
|
| 77 |
+
"spiciness": 20934,
|
| 78 |
+
"flavor": 306005,
|
| 79 |
+
"temperature": 340444
|
| 80 |
+
},
|
| 81 |
+
"null_quality_fields": {
|
| 82 |
+
"rating": 1236,
|
| 83 |
+
"sales": 221462,
|
| 84 |
+
"eta": 0,
|
| 85 |
+
"distance": 0,
|
| 86 |
+
"price_cents": 0
|
| 87 |
+
},
|
| 88 |
+
"text_redactions": {
|
| 89 |
+
"phone": 0,
|
| 90 |
+
"email": 0,
|
| 91 |
+
"url": 0
|
| 92 |
+
},
|
| 93 |
+
"checks": {
|
| 94 |
+
"frozen_source_hashes": true,
|
| 95 |
+
"unique_configuration_ids": true,
|
| 96 |
+
"request_catalog_location_join": true,
|
| 97 |
+
"private_reference_location_resolution": true,
|
| 98 |
+
"precise_coordinates_exported": false,
|
| 99 |
+
"source_linkage_hashes_exported": false,
|
| 100 |
+
"raw_api_responses_exported": false,
|
| 101 |
+
"evaluation_answers_exported": false,
|
| 102 |
+
"input_parquet_round_trip": true
|
| 103 |
+
},
|
| 104 |
+
"price_note": "price_cents is a base/SKU price according to price_scope; final totals and extras may be unverified.",
|
| 105 |
+
"quality_note": "Provider rating/sales/ETA/distance values retain missingness; no common cross-provider scale is asserted.",
|
| 106 |
+
"attribute_note": "attributes are normalized menu-derived fields; options preserve selected source menu group/value labels.",
|
| 107 |
+
"coverage_note": "The 20 episode-linked catalogs have explicit collection bounds. They do not enumerate all platform supply.",
|
| 108 |
+
"identifier_note": "Opaque configuration/shop/item/SKU/option IDs and remapped location IDs; no coordinates or addresses."
|
| 109 |
+
}
|
catalog_shanghai.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:563d14b3b2ab8dcaf6851576f85323d2a570538dfd05f5b05ab5167850d3a65b
|
| 3 |
+
size 12066182
|
catalog_shenzhen.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f2331c0e43662172cfc7998714227e72d08bbfab731968415282211d0fd9e37
|
| 3 |
+
size 11157363
|
locations.jsonl
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"location_id": "DB_L001", "city": "广州市", "configurations": 6289, "collection_status": "complete", "requests": 462}
|
| 2 |
+
{"location_id": "DB_L002", "city": "广州市", "configurations": 74341, "collection_status": "bounded_complete", "requests": 461}
|
| 3 |
+
{"location_id": "DB_L003", "city": "广州市", "configurations": 2435, "collection_status": "bounded_complete", "requests": 470}
|
| 4 |
+
{"location_id": "DB_L004", "city": "广州市", "configurations": 13726, "collection_status": "complete", "requests": 455}
|
| 5 |
+
{"location_id": "DB_L005", "city": "广州市", "configurations": 42933, "collection_status": "bounded_complete", "requests": 457}
|
| 6 |
+
{"location_id": "DB_L006", "city": "深圳市", "configurations": 33862, "collection_status": "bounded_complete", "requests": 362}
|
| 7 |
+
{"location_id": "DB_L007", "city": "深圳市", "configurations": 60222, "collection_status": "bounded_complete", "requests": 399}
|
| 8 |
+
{"location_id": "DB_L008", "city": "深圳市", "configurations": 3514, "collection_status": "bounded_complete", "requests": 407}
|
| 9 |
+
{"location_id": "DB_L009", "city": "深圳市", "configurations": 60875, "collection_status": "bounded_complete", "requests": 361}
|
| 10 |
+
{"location_id": "DB_L010", "city": "深圳市", "configurations": 64966, "collection_status": "bounded_complete", "requests": 387}
|
| 11 |
+
{"location_id": "DB_L011", "city": "上海市", "configurations": 50143, "collection_status": "bounded_complete", "requests": 418}
|
| 12 |
+
{"location_id": "DB_L012", "city": "上海市", "configurations": 894, "collection_status": "bounded_complete", "requests": 417}
|
| 13 |
+
{"location_id": "DB_L013", "city": "上海市", "configurations": 73572, "collection_status": "bounded_complete", "requests": 427}
|
| 14 |
+
{"location_id": "DB_L014", "city": "上海市", "configurations": 37930, "collection_status": "bounded_complete", "requests": 425}
|
| 15 |
+
{"location_id": "DB_L015", "city": "上海市", "configurations": 69888, "collection_status": "bounded_complete", "requests": 423}
|
| 16 |
+
{"location_id": "DB_L016", "city": "北京市", "configurations": 35447, "collection_status": "bounded_complete", "requests": 442}
|
| 17 |
+
{"location_id": "DB_L017", "city": "北京市", "configurations": 40337, "collection_status": "bounded_complete", "requests": 452}
|
| 18 |
+
{"location_id": "DB_L018", "city": "北京市", "configurations": 28838, "collection_status": "bounded_complete", "requests": 451}
|
| 19 |
+
{"location_id": "DB_L019", "city": "北京市", "configurations": 1170, "collection_status": "bounded_complete", "requests": 459}
|
| 20 |
+
{"location_id": "DB_L020", "city": "北京市", "configurations": 21593, "collection_status": "bounded_complete", "requests": 433}
|
locations.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5cca8f7513dd749b429bfc97001be7c53cafdb98faf5bffd37ee10a1e5257e29
|
| 3 |
+
size 2379
|
preview_manifest.json
CHANGED
|
@@ -1,48 +1,75 @@
|
|
| 1 |
{
|
| 2 |
"dataset": "DeliveryBench",
|
| 3 |
-
"version": "0.
|
| 4 |
-
"release_scope": "
|
| 5 |
"counts": {
|
| 6 |
"users": 1000,
|
| 7 |
"requests": 8568,
|
| 8 |
-
"
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
"test": 150
|
| 12 |
-
},
|
| 13 |
-
"request_splits": {
|
| 14 |
-
"train": 5938,
|
| 15 |
-
"validation": 1239,
|
| 16 |
-
"test": 1391
|
| 17 |
-
},
|
| 18 |
-
"users_per_city": {
|
| 19 |
-
"上海市": 250,
|
| 20 |
-
"北京市": 250,
|
| 21 |
-
"广州市": 250,
|
| 22 |
-
"深圳市": 250
|
| 23 |
-
},
|
| 24 |
-
"requests_per_city": {
|
| 25 |
-
"上海市": 2110,
|
| 26 |
-
"北京市": 2237,
|
| 27 |
-
"广州市": 2305,
|
| 28 |
-
"深圳市": 1916
|
| 29 |
-
},
|
| 30 |
-
"audited_menu_configurations": 722975,
|
| 31 |
-
"episode_linked_locations": 20,
|
| 32 |
-
"original_episodes": 8873
|
| 33 |
},
|
| 34 |
"files": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
"users.jsonl": {
|
| 36 |
"rows": 1000,
|
|
|
|
| 37 |
"sha256": "91d355e14afe47b36dffa1a60d6d1de2ec4572ae3e2b00bc7305d6647292c20e"
|
| 38 |
},
|
| 39 |
-
"requests.
|
| 40 |
"rows": 8568,
|
| 41 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
}
|
| 43 |
},
|
| 44 |
-
"identifier_policy": "
|
| 45 |
-
"
|
| 46 |
-
"
|
| 47 |
-
"profile_scope": "Constructed food/drink preferences are included as inspectable profile inputs. Per-episode answer labels are excluded."
|
| 48 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"dataset": "DeliveryBench",
|
| 3 |
+
"version": "0.3.0-catalog",
|
| 4 |
+
"release_scope": "Synthetic user/request inputs and menu-grounded product configurations with attributes and selected options",
|
| 5 |
"counts": {
|
| 6 |
"users": 1000,
|
| 7 |
"requests": 8568,
|
| 8 |
+
"configurations": 722975,
|
| 9 |
+
"catalog_locations": 20,
|
| 10 |
+
"cities": 4
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
},
|
| 12 |
"files": {
|
| 13 |
+
"catalog_beijing.parquet": {
|
| 14 |
+
"rows": 127385,
|
| 15 |
+
"bytes": 6872799,
|
| 16 |
+
"sha256": "9827939b98ea97e5f42cdd6dffd676a71111a2151648e43d6055ff7923cb796a"
|
| 17 |
+
},
|
| 18 |
+
"catalog_guangzhou.parquet": {
|
| 19 |
+
"rows": 139724,
|
| 20 |
+
"bytes": 7036812,
|
| 21 |
+
"sha256": "ff9643a395a5bcc99bc5e6694cdf27684e066eb60c131b02c1111e358dff54ed"
|
| 22 |
+
},
|
| 23 |
+
"catalog_shanghai.parquet": {
|
| 24 |
+
"rows": 232427,
|
| 25 |
+
"bytes": 12066182,
|
| 26 |
+
"sha256": "563d14b3b2ab8dcaf6851576f85323d2a570538dfd05f5b05ab5167850d3a65b"
|
| 27 |
+
},
|
| 28 |
+
"catalog_shenzhen.parquet": {
|
| 29 |
+
"rows": 223439,
|
| 30 |
+
"bytes": 11157363,
|
| 31 |
+
"sha256": "7f2331c0e43662172cfc7998714227e72d08bbfab731968415282211d0fd9e37"
|
| 32 |
+
},
|
| 33 |
+
"requests.jsonl": {
|
| 34 |
+
"rows": 8568,
|
| 35 |
+
"bytes": 2184886,
|
| 36 |
+
"sha256": "c1feaa6d79cdf0ec0acf4806129a3cff1b682f73f6398b672c298a0442e93d71"
|
| 37 |
+
},
|
| 38 |
+
"locations.jsonl": {
|
| 39 |
+
"rows": 20,
|
| 40 |
+
"bytes": 2598,
|
| 41 |
+
"sha256": "17313ff3a92824b54ae25c7b04c0eb473a331420a5f964c53cd39bf04004c2c6"
|
| 42 |
+
},
|
| 43 |
+
"users.parquet": {
|
| 44 |
+
"rows": 1000,
|
| 45 |
+
"bytes": 12640,
|
| 46 |
+
"sha256": "a4b87afd008a17e521c67de0323c551087a1515d9c4adde8dbe59afe68f92f57"
|
| 47 |
+
},
|
| 48 |
"users.jsonl": {
|
| 49 |
"rows": 1000,
|
| 50 |
+
"bytes": 281782,
|
| 51 |
"sha256": "91d355e14afe47b36dffa1a60d6d1de2ec4572ae3e2b00bc7305d6647292c20e"
|
| 52 |
},
|
| 53 |
+
"requests.parquet": {
|
| 54 |
"rows": 8568,
|
| 55 |
+
"bytes": 49664,
|
| 56 |
+
"sha256": "7cf8a18253a962be0bbde67c6211fd7e4fd04230783f2d0eaab557638a48a793"
|
| 57 |
+
},
|
| 58 |
+
"locations.parquet": {
|
| 59 |
+
"rows": 20,
|
| 60 |
+
"bytes": 2379,
|
| 61 |
+
"sha256": "5cca8f7513dd749b429bfc97001be7c53cafdb98faf5bffd37ee10a1e5257e29"
|
| 62 |
+
},
|
| 63 |
+
"catalog_manifest.json": {
|
| 64 |
+
"bytes": 3487,
|
| 65 |
+
"sha256": "61f70ccb279dd91bc1cb4e91488add8f3180b33eeffc7e6e3e3749655fe933ad"
|
| 66 |
+
},
|
| 67 |
+
"README.md": {
|
| 68 |
+
"bytes": 7266,
|
| 69 |
+
"sha256": "0900e2ac4b260308f57e26c2f60b4e8752e0ed01c251eb940089411c4e09f43b"
|
| 70 |
}
|
| 71 |
},
|
| 72 |
+
"identifier_policy": "Existing user/request IDs retained; anonymous location joins added; catalog IDs opaque.",
|
| 73 |
+
"evaluation_answers_included": false,
|
| 74 |
+
"raw_api_responses_included": false
|
|
|
|
| 75 |
}
|
requests.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
requests.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cf8a18253a962be0bbde67c6211fd7e4fd04230783f2d0eaab557638a48a793
|
| 3 |
+
size 49664
|
users.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a4b87afd008a17e521c67de0323c551087a1515d9c4adde8dbe59afe68f92f57
|
| 3 |
+
size 12640
|