Team Ai
Modelpublic

meandyou200175/pB-query-sql

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes14downloads
Model Card

SentenceTransformer based on vinai/phobert-base-v2

This is a sentence-transformers model finetuned from vinai/phobert-base-v2. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

  • —Model Type: Sentence Transformer
  • —Base model: vinai/phobert-base-v2 <!-- at revision e2375d266bdf39c6e8e9a87af16a5da3190b0cc8 -->
  • —Maximum Sequence Length: 256 tokens
  • —Output Dimensionality: 768 dimensions
  • —Similarity Function: Cosine Similarity <!-- - Training Dataset: Unknown --> <!-- - Language: Unknown --> <!-- - License: Unknown -->

Model Sources

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: RobertaModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

bash
pip install -U sentence-transformers

Then you can load this model and run inference.

python
from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("meandyou200175/pB-query-sql")
# Run inference
sentences = [
    'mình cần robot hút bụi pin lớn hơn 3120mAh',
    'Robot hút bụi Ecovacs Deebot T10, Pin 5200mAh, Hút 5000Pa, Giá: 12.900.000',
    'Robot hút bụi Ecovacs Deebot T10, Pin 2076mAh, Hút 5000Pa, Giá: 12.900.000',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

<!--

Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details> -->

<!--

Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details> -->

<!--

Out-of-Scope Use

List how the model may foreseeably be misused and address what users ought not to do with the model. -->

Evaluation

Metrics

Information Retrieval
MetricValue
cosine_accuracy@10.4825
cosine_accuracy@20.7364
cosine_accuracy@50.9538
cosine_accuracy@100.9888
cosine_accuracy@1000.9985
cosine_precision@10.4825
cosine_precision@20.3682
cosine_precision@50.1908
cosine_precision@100.0989
cosine_precision@1000.01
cosine_recall@10.4825
cosine_recall@20.7364
cosine_recall@50.9538
cosine_recall@100.9888
cosine_recall@1000.9985
cosine_ndcg@100.755
cosine_mrr@10.4825
cosine_mrr@20.6095
cosine_mrr@50.6724
cosine_mrr@100.6773
cosine_mrr@1000.6778
cosine_map@1000.6778

<!--

Bias, Risks and Limitations

What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model. -->

<!--

Recommendations

What are recommendations with respect to the foreseeable issues? For example, filtering explicit content. -->

Training Details

Training Dataset

Unnamed Dataset
  • —Size: 12,085 training samples
  • —Columns: <code>query</code>, <code>positive</code>, <code>negative1</code>, <code>negative2</code>, <code>negative3</code>, <code>negative4</code>, <code>negative5</code>, <code>negative6</code>, <code>negative7</code>, <code>negative8</code>, <code>negative9</code>, <code>negative10</code>, <code>negative11</code>, <code>negative12</code>, <code>negative13</code>, <code>negative14</code>, and <code>negative_15</code>
  • —Approximate statistics based on the first 1000 samples: | | query | positive | negative1 | negative2 | negative3 | negative4 | negative5 | negative6 | negative7 | negative8 | negative9 | negative10 | negative11 | negative12 | negative13 | negative14 | negative_15 | |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------| | type | string | string | string | string | string | string | string | string | string | string | string | string | string | string | string | string | string | | details | <ul><li>min: 6 tokens</li><li>mean: 14.32 tokens</li><li>max: 28 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 27.93 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.05 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.21 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.28 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.36 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 28.3 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.31 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.42 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 28.57 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.36 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.3 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 27.7 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 27.71 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 27.84 tokens</li><li>max: 100 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 27.67 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 27.88 tokens</li><li>max: 97 tokens</li></ul> |
  • —Samples: | query | positive | negative1 | negative2 | negative3 | negative4 | negative5 | negative6 | negative7 | negative8 | negative9 | negative10 | negative11 | negative12 | negative13 | negative14 | negative_15 | |:--------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------| | <code>tôi cần máy lọc không khí giá dưới 4 triệu</code> | <code>Máy lọc không khí Xiaomi Mi Air Purifier 4, công suất 50W, lọc bụi PM2.5, Giá: 3.200.000</code> | <code>Máy lọc không khí Xiaomi Mi Air Purifier 6, công suất 50W, lọc bụi PM2.5, Giá: 3.200.000</code> | <code>Máy lọc không khí Xiaomi Mi Air Purifier 5, công suất 50W, lọc bụi PM2.5, Giá: 3.200.000</code> | <code>Bếp gas Rinnai RV-377GN, Công suất 4.2kW, 2 bếp, Giá: 2.300.000</code> | <code>Máy in HP LaserJet Pro M404dn, Tốc độ 30 trang/phút, Kết nối LAN, Giá: 6.800.000</code> | <code>Máy lọc không khí Sharp FP-J30E-A, Diện tích lọc: 23m2, Bộ lọc HEPA, Plasmacluster ion, Giá: 2,190,000</code> | <code>Bàn ủi Philips HI114, công suất 1500W, mặt đế chống dính, Giá: 950.000</code> | <code>Máy rửa bát Bosch SMS46NI05E, 12 bộ, 6 chương trình rửa, Tiết kiệm nước, Giá: 16.500.000</code> | <code>Bếp từ Sunhouse SHD6158, công suất 2300W, hẹn giờ 3h, Giá 1100000</code> | <code>Tai nghe Sony WH-1000XM5, Chống ồn chủ động, Pin 30h, Bluetooth 5.2, Giá: 9.500.000</code> | <code>Công tắc Tuya Smart WiFi, chịu tải 16A, kết nối 2.4GHz, Giá: 420.000</code> | <code>Màn hình Samsung Odyssey Neo G9, 49 inch, 240Hz, 2K, Giá: 42.000.000</code> | <code>Router TP-Link Archer C6, băng tần 2.4/5GHz, tốc độ 1200Mbps, Giá 950000</code> | <code>Máy pha cà phê DeLonghi ECAM22.110B, Bình 2.2L, Tự động, Giá: 14.000.000</code> | <code>Máy lọc không khí Philips AC1214, diện tích 30 mét vuông, công suất 35W, Giá 1350000</code> | <code>Tủ lạnh Sharp Inverter SJ-X196E, Cao 1m65, Dung tích 196L, Giá: 5.500.000</code> | | <code>Máy sấy dung tích lớn hơn 3kg</code> | <code>Máy sấy Electrolux EDV7052, dung tích 7kg, công suất 1600W, Giá 5250000</code> | <code>Máy sấy Electrolux EDV7052, dung tích 3kg, công suất 1600W, Giá 5250000</code> | <code>Máy sấy Electrolux EDV7052, dung tích 2kg, công suất 1600W, Giá 5250000</code> | <code>Máy lọc không khí Coway AP-1009CH, Diện tích 50m2, Giá: 6.800.000</code> | <code>Laptop Asus Vivobook 15, màn hình 15.6 inch (~39.6cm), trọng lượng 2.4kg (2400g), RAM 16GB, SSD 512GB, Giá: 18.500.000</code> | <code>Loa JBL Charge 5, Phát nhạc 20h, Chống nước IP67, Giá: 3.900.000</code> | <code>Máy tính xách tay ASUS Zenbook 14 OLED UX3402, Trọng lượng: 1.4kg, CPU: i5-1240P, RAM 16GB, SSD 512GB, Màn OLED 2.8K, Giá: 23,990,000</code> | <code>Máy pha cà phê Delonghi EC685, Công suất 1450W, Áp suất 15 bar, Giá: 9.800.000</code> | <code>Xe côn tay Yamaha R15 V4, Động cơ 155cc, Công nghệ VVA, Giá: 74.000.000</code> | <code>Đèn bàn LED Xiaomi Mi, công suất 12W, điều chỉnh độ sáng + nhiệt màu, Giá: 580.000</code> | <code>Tablet iPad 9th Gen, Màn 10.2 inch Retina, Chip A13, Giá: 10.900.000</code> | <code>Ghế sofa chữ L, Dài 2m7, Da PU, 5 chỗ ngồi, Giá: 12.000.000</code> | <code>Máy ảnh Canon EOS R10, Nặng 429g, Cảm biến APS-C 24MP, Giá: 24.500.000</code> | <code>Máy giặt sấy LG Inverter 10kg, AI DD, TurboWash, Giá: 19.500.000</code> | <code>Bàn là hơi nước Philips GC4880, công suất 1500W, Giá 700000</code> | <code>Máy hút ẩm FujiE HM-920EC, Công suất 20L/ngày, Giá: 5.900.000</code> | | <code>tôi cần điện thoại giá nhỏ hơn 8 triệu</code> | <code>Smartphone Xiaomi Redmi Note 12, Màn AMOLED 6.67", Pin 5000mAh, Giá: 5.890.000</code> | <code>Smartphone Xiaomi Redmi Note 9, Màn AMOLED 6.67", Pin 5000mAh, Giá: 5.890.000</code> | <code>Smartphone Xiaomi Redmi Note 11, Màn AMOLED 6.67", Pin 5000mAh, Giá: 5.890.000</code> | <code>Smartphone Xiaomi Redmi Note 10, Màn AMOLED 6.67", Pin 5000mAh, Giá: 5.890.000</code> | <code>Máy quét phim Plustek OpticFilm 8200, 16MP, Giá: 2.650.000</code> | <code>Máy xay Philips HR2115, công suất 500W, cối 1.5L, Giá: 1.200.000</code> | <code>Bàn phím Logitech K380, Kết nối Bluetooth, Nhỏ gọn, Giá: 1.200.000</code> | <code>Máy xay cà phê cầm tay Hario, dung tích 50g, tay quay, Giá: 480.000</code> | <code>Ghế xoay Ergohuman Plus, Tải trọng 120kg, Lưng lưới, Giá: 8.900.000</code> | <code>Điện thoại Xiaomi Redmi A1, RAM 2GB, Pin 5000mAh, Giá: 2.100.000</code> | <code>Quạt điều hòa Sunhouse SHD7725, Bình 45L, Công suất 200W, Giá: 3.600.000</code> | <code>Smartphone iPhone 13 Mini, Trọng lượng: 140g, Màn hình 5.4 inch Super Retina XDR, Chip A15 Bionic, RAM 4GB, ROM 128GB, Giá: 17,990,000</code> | <code>Xe máy điện VinFast Theon S, Tốc độ tối đa 90km/h, Pin 3500W, Giá: 63.000.000</code> | <code>Tivi Samsung Neo QLED 65", 4K UHD, HDR10+, Giá: 25.500.000</code> | <code>Tủ đông Aqua AQF-C300, Công suất 280W, Dung tích 295L, Giá: 8.600.000</code> | <code>Máy lọc nước Karofi 12000ml, 8 lõi lọc, công suất 95W, Giá: 7.000.000</code> |
  • —Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:
json
  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }

Evaluation Dataset

Unnamed Dataset
  • —Size: 1,343 evaluation samples
  • —Columns: <code>query</code>, <code>positive</code>, <code>negative1</code>, <code>negative2</code>, <code>negative3</code>, <code>negative4</code>, <code>negative5</code>, <code>negative6</code>, <code>negative7</code>, <code>negative8</code>, <code>negative9</code>, <code>negative10</code>, <code>negative11</code>, <code>negative12</code>, <code>negative13</code>, <code>negative14</code>, and <code>negative_15</code>
  • —Approximate statistics based on the first 1000 samples: | | query | positive | negative1 | negative2 | negative3 | negative4 | negative5 | negative6 | negative7 | negative8 | negative9 | negative10 | negative11 | negative12 | negative13 | negative14 | negative_15 | |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------| | type | string | string | string | string | string | string | string | string | string | string | string | string | string | string | string | string | string | | details | <ul><li>min: 6 tokens</li><li>mean: 14.19 tokens</li><li>max: 27 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 27.94 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.02 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 28.1 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.23 tokens</li><li>max: 83 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.1 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.26 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.35 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 28.33 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.28 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 28.14 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 28.31 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 27.77 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 27.87 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 27.67 tokens</li><li>max: 112 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 27.52 tokens</li><li>max: 97 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 28.0 tokens</li><li>max: 112 tokens</li></ul> |
  • —Samples: | query | positive | negative1 | negative2 | negative3 | negative4 | negative5 | negative6 | negative7 | negative8 | negative9 | negative10 | negative11 | negative12 | negative13 | negative14 | negative_15 | |:---------------------------------------------------------------|:---------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------| | <code>mình cần nồi áp suất điện công suất lớn hơn 1040W</code> | <code>Nồi áp suất Philips HD2137, Công suất 1300W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 700W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 751W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 653W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 770W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 958W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 699W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 797W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 679W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 893W, 14 chức năng, Giá: 2.900.000</code> | <code>Nồi áp suất Philips HD2137, Công suất 897W, 14 chức năng, Giá: 2.900.000</code> | <code>Công tắc Tuya Smart, điều khiển qua App, chịu tải 16A, Giá: 420.000</code> | <code>Ghế văn phòng Noble WB-204, xoay 360°, chịu lực 160kg, Giá: 2.450.000</code> | <code>Máy in Canon LBP6230DN, In 2 mặt, Tốc độ 28 trang/phút, Giá: 4.600.000</code> | <code>Tablet Lenovo Tab M10 Gen 3, màn 10.1", RAM 4GB, ROM 64GB, Giá: 5.900.000</code> | <code>Bình đun Kangaroo 1800W, dung tích 1.7L, tự ngắt khi sôi, Giá: 500.000</code> | | <code>tôi muốn mua loa công suất nhỏ hơn 240W</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 150W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 281W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 253W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 282W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 352W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 319W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 293W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 295W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 301W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 320W, Chống nước, Giá: 3.200.000</code> | <code>Loa Bluetooth Sony SRS-XB33, Công suất 356W, Chống nước, Giá: 3.200.000</code> | <code>Smart Tivi LG QNED 75, Màn hình 75 inch 4K, Dolby Vision, Giá: 29.000.000</code> | <code>Quạt đứng Asia D16018, 3 tốc độ, chiều cao 1.2m, Giá: 580.000</code> | <code>Loa kéo JBL PartyBox 1000, công suất 1000W, trọng lượng 15kg, Bluetooth, Giá: 22.500.000</code> | <code>Vali du lịch Sakos, Size 22 inch, Khóa TSA, Nhựa ABS, Giá: 1.850.000</code> | <code>Bình nóng lạnh Ariston Andris2 20L, công suất 2500W, chống giật ELCB, Giá 2390000</code> | | <code>Tai nghe không dây chống ồn ANC và pin trên 20h</code> | <code>Tai nghe Bose QuietComfort 45, ANC, pin 24h, Bluetooth 5.1, Giá 7490000</code> | <code>Máy pha cà phê Delonghi EC685, Công suất 1350W, Áp suất 15 bar, Giá: 6.200.000</code> | <code>Máy lọc nước Karofi KAQ-U95, Công suất 25L/h, 10 lõi lọc, Giá: 9.500.000</code> | <code>Máy lọc Kangaroo KG111, công suất 25L/h, 9 lõi lọc, vòi nhựa ABS, Giá 7550000</code> | <code>Đèn năng lượng mặt trời Sunhouse 20W, pin lithium, chiếu sáng 10h, Giá: 460.000</code> | <code>Ghế massage Daikiosan DK-150, công suất 120W, nhiều chế độ, Giá: 3.850.000</code> | <code>Máy hút bụi Hitachi CV-SE230V, Công suất 2300W, Lọc HEPA, Giá: 4.500.000</code> | <code>Tủ giày gỗ công nghiệp, Rộng 70cm, 3 tầng, Giá: 1.200.000</code> | <code>Thiết bị lọc Sharp FP-J30E-A, Diện tích lọc: 23m2, Bộ lọc HEPA, Plasmacluster ion, Giá: 2,190,000</code> | <code>Bàn ủi Philips GC2990, công suất 1800W, đế chống dính, Giá: 680.000</code> | <code>Bàn là hơi nước Philips GC4880, công suất 1500W, Giá 700000</code> | <code>Máy lọc không khí Xiaomi Mi Air Purifier 4, công suất 50W, lọc bụi PM2.5, Giá: 3.200.000</code> | <code>Smartphone Xiaomi Redmi Note 12, Màn AMOLED 6.67", Pin 5000mAh, Giá: 5.890.000</code> | <code>Tủ lạnh Hitachi R-WB640VGV0, Dung tích 569L, Inverter, Giá: 25.000.000</code> | <code>Máy cưa bàn Makita MLT100, Công suất 1650W, Bàn cắt 690mm, Giá: 9.200.000</code> | <code>Loa Bluetooth Anker Soundcore, Pin 12h, Công suất 10W, Giá: 1.200.000</code> |
  • —Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:
json
  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }

Training Hyperparameters

Non-Default Hyperparameters
  • —eval_strategy: steps
  • —per_device_train_batch_size: 4
  • —per_device_eval_batch_size: 4
  • —learning_rate: 2e-05
  • —num_train_epochs: 10
  • —warmup_ratio: 0.1
  • —fp16: True
  • —batch_sampler: no_duplicates
All Hyperparameters

<details><summary>Click to expand</summary>

  • —overwrite_output_dir: False
  • —do_predict: False
  • —eval_strategy: steps
  • —prediction_loss_only: True
  • —per_device_train_batch_size: 4
  • —per_device_eval_batch_size: 4
  • —per_gpu_train_batch_size: None
  • —per_gpu_eval_batch_size: None
  • —gradient_accumulation_steps: 1
  • —eval_accumulation_steps: None
  • —torch_empty_cache_steps: None
  • —learning_rate: 2e-05
  • —weight_decay: 0.0
  • —adam_beta1: 0.9
  • —adam_beta2: 0.999
  • —adam_epsilon: 1e-08
  • —max_grad_norm: 1.0
  • —num_train_epochs: 10
  • —max_steps: -1
  • —lr_scheduler_type: linear
  • —lr_scheduler_kwargs: {}
  • —warmup_ratio: 0.1
  • —warmup_steps: 0
  • —log_level: passive
  • —log_level_replica: warning
  • —log_on_each_node: True
  • —logging_nan_inf_filter: True
  • —save_safetensors: True
  • —save_on_each_node: False
  • —save_only_model: False
  • —restore_callback_states_from_checkpoint: False
  • —no_cuda: False
  • —use_cpu: False
  • —use_mps_device: False
  • —seed: 42
  • —data_seed: None
  • —jit_mode_eval: False
  • —use_ipex: False
  • —bf16: False
  • —fp16: True
  • —fp16_opt_level: O1
  • —half_precision_backend: auto
  • —bf16_full_eval: False
  • —fp16_full_eval: False
  • —tf32: None
  • —local_rank: 0
  • —ddp_backend: None
  • —tpu_num_cores: None
  • —tpu_metrics_debug: False
  • —debug: []
  • —dataloader_drop_last: False
  • —dataloader_num_workers: 0
  • —dataloader_prefetch_factor: None
  • —past_index: -1
  • —disable_tqdm: False
  • —remove_unused_columns: True
  • —label_names: None
  • —load_best_model_at_end: False
  • —ignore_data_skip: False
  • —fsdp: []
  • —fsdp_min_num_params: 0
  • —fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
  • —fsdp_transformer_layer_cls_to_wrap: None
  • —accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
  • —deepspeed: None
  • —label_smoothing_factor: 0.0
  • —optim: adamw_torch
  • —optim_args: None
  • —adafactor: False
  • —group_by_length: False
  • —length_column_name: length
  • —ddp_find_unused_parameters: None
  • —ddp_bucket_cap_mb: None
  • —ddp_broadcast_buffers: False
  • —dataloader_pin_memory: True
  • —dataloader_persistent_workers: False
  • —skip_memory_metrics: True
  • —use_legacy_prediction_loop: False
  • —push_to_hub: False
  • —resume_from_checkpoint: None
  • —hub_model_id: None
  • —hub_strategy: every_save
  • —hub_private_repo: None
  • —hub_always_push: False
  • —gradient_checkpointing: False
  • —gradient_checkpointing_kwargs: None
  • —include_inputs_for_metrics: False
  • —include_for_metrics: []
  • —eval_do_concat_batches: True
  • —fp16_backend: auto
  • —push_to_hub_model_id: None
  • —push_to_hub_organization: None
  • —mp_parameters:
  • —auto_find_batch_size: False
  • —full_determinism: False
  • —torchdynamo: None
  • —ray_scope: last
  • —ddp_timeout: 1800
  • —torch_compile: False
  • —torch_compile_backend: None
  • —torch_compile_mode: None
  • —include_tokens_per_second: False
  • —include_num_input_tokens_seen: False
  • —neftune_noise_alpha: None
  • —optim_target_modules: None
  • —batch_eval_metrics: False
  • —eval_on_start: False
  • —use_liger_kernel: False
  • —eval_use_gather_object: False
  • —average_tokens_across_devices: False
  • —prompts: None
  • —batch_sampler: no_duplicates
  • —multi_dataset_batch_sampler: proportional

</details>

Training Logs

<details><summary>Click to expand</summary>

EpochStepTraining LossValidation Losscosine_ndcg@10
-1-1--0.1714
0.03311003.1377--
0.06622002.503--
0.09933002.2899--
0.13244002.2639--
0.16555002.2148--
0.19856002.1753--
0.23167002.066--
0.26478001.9329--
0.29789001.8195--
0.330910001.76381.60240.4015
0.364011001.6775--
0.397112001.5961--
0.430213001.4757--
0.463314001.5213--
0.496415001.3906--
0.529516001.4209--
0.562517001.3032--
0.595618001.2807--
0.628719001.2214--
0.661820001.13690.98190.4349
0.694921001.0557--
0.728022001.0554--
0.761123000.9938--
0.794224000.885--
0.827325000.9139--
0.860426000.9002--
0.893427000.8333--
0.926528000.7317--
0.959629000.7898--
0.992730000.74670.53640.5163
1.025831000.59--
1.058932000.6862--
1.092033000.6216--
1.125134000.5714--
1.158235000.5407--
1.191336000.4424--
1.224437000.4588--
1.257438000.4185--
1.290539000.3664--
1.323640000.40790.32220.5620
1.356741000.4488--
1.389842000.3677--
1.422943000.3706--
1.456044000.3379--
1.489145000.2752--
1.522246000.3173--
1.555347000.3361--
1.588448000.3541--
1.621449000.3089--
1.654550000.2880.23240.5978
1.687651000.3473--
1.720752000.2599--
1.753853000.2065--
1.786954000.3111--
1.820055000.2551--
1.853156000.2377--
1.886257000.2837--
1.919358000.2271--
1.952359000.1589--
1.985460000.31150.17890.6191
2.018561000.1572--
2.051662000.1874--
2.084763000.1606--
2.117864000.1562--
2.150965000.2054--
2.184066000.1909--
2.217167000.2015--
2.250268000.2208--
2.283369000.1983--
2.316370000.16340.15030.6417
2.349471000.1692--
2.382572000.1868--
2.415673000.1422--
2.448774000.1455--
2.481875000.2187--
2.514976000.142--
2.548077000.1148--
2.581178000.1341--
2.614279000.1284--
2.647380000.15550.12020.6534
2.680381000.1207--
2.713482000.1153--
2.746583000.1114--
2.779684000.1158--
2.812785000.1241--
2.845886000.1345--
2.878987000.132--
2.912088000.113--
2.945189000.1101--
2.978290000.11770.10140.6733
3.011391000.1358--
3.044392000.1148--
3.077493000.097--
3.110594000.1303--
3.143695000.1077--
3.176796000.116--
3.209897000.0878--
3.242998000.0984--
3.276099000.1017--
3.3091100000.14520.09500.6854
3.3422101000.0905--
3.3752102000.0811--
3.4083103000.0802--
3.4414104000.1408--
3.4745105000.0622--
3.5076106000.0779--
3.5407107000.0754--
3.5738108000.0645--
3.6069109000.0642--
3.6400110000.05860.09980.6864
3.6731111000.0556--
3.7062112000.0948--
3.7392113000.0791--
3.7723114000.0587--
3.8054115000.1135--
3.8385116000.0886--
3.8716117000.0807--
3.9047118000.0656--
3.9378119000.0506--
3.9709120000.10590.07830.7082
4.0040121000.0551--
4.0371122000.0421--
4.0702123000.0334--
4.1032124000.0651--
4.1363125000.0363--
4.1694126000.0525--
4.2025127000.0428--
4.2356128000.0646--
4.2687129000.0647--
4.3018130000.05280.07570.7215
4.3349131000.0812--
4.3680132000.051--
4.4011133000.0401--
4.4341134000.037--
4.4672135000.0283--
4.5003136000.0483--
4.5334137000.0616--
4.5665138000.0622--
4.5996139000.0552--
4.6327140000.06330.08130.7209
4.6658141000.0811--
4.6989142000.0586--
4.7320143000.0458--
4.7651144000.0418--
4.7981145000.07--
4.8312146000.0498--
4.8643147000.0864--
4.8974148000.0442--
4.9305149000.0481--
4.9636150000.05360.07110.7243
4.9967151000.1027--
5.0298152000.0291--
5.0629153000.0437--
5.0960154000.0541--
5.1291155000.0217--
5.1621156000.0315--
5.1952157000.0417--
5.2283158000.0429--
5.2614159000.0176--
5.2945160000.03580.07590.7176
5.3276161000.0374--
5.3607162000.0509--
5.3938163000.0473--
5.4269164000.0367--
5.4600165000.0479--
5.4931166000.0338--
5.5261167000.0557--
5.5592168000.0556--
5.5923169000.0443--
5.6254170000.0730.07510.7414
5.6585171000.0892--
5.6916172000.0262--
5.7247173000.0306--
5.7578174000.0345--
5.7909175000.0222--
5.8240176000.0586--
5.8570177000.0326--
5.8901178000.0255--
5.9232179000.0593--
5.9563180000.03740.06770.7365
5.9894181000.0318--
6.0225182000.0659--
6.0556183000.0206--
6.0887184000.0452--
6.1218185000.0347--
6.1549186000.0236--
6.1880187000.0385--
6.2210188000.0425--
6.2541189000.015--
6.2872190000.0260.06420.7403
6.3203191000.0279--
6.3534192000.0163--
6.3865193000.0256--
6.4196194000.031--
6.4527195000.0435--
6.4858196000.0298--
6.5189197000.0346--
6.5520198000.0155--
6.5850199000.0431--
6.6181200000.03580.06240.7382
6.6512201000.0224--
6.6843202000.0451--
6.7174203000.0437--
6.7505204000.0832--
6.7836205000.0542--
6.8167206000.0243--
6.8498207000.0225--
6.8829208000.0384--
6.9159209000.0214--
6.9490210000.02960.06200.7423
6.9821211000.0244--
7.0152212000.0136--
7.0483213000.0145--
7.0814214000.0378--
7.1145215000.0215--
7.1476216000.0214--
7.1807217000.0269--
7.2138218000.015--
7.2469219000.0463--
7.2799220000.02730.06050.7452
7.3130221000.0276--
7.3461222000.022--
7.3792223000.0443--
7.4123224000.0106--
7.4454225000.0169--
7.4785226000.024--
7.5116227000.0356--
7.5447228000.0167--
7.5778229000.019--
7.6109230000.02330.05800.7459
7.6439231000.0158--
7.6770232000.02--
7.7101233000.013--
7.7432234000.0378--
7.7763235000.0186--
7.8094236000.0143--
7.8425237000.0364--
7.8756238000.022--
7.9087239000.0178--
7.9418240000.0380.05250.7471
7.9749241000.022--
8.0079242000.0279--
8.0410243000.0368--
8.0741244000.0319--
8.1072245000.0114--
8.1403246000.0041--
8.1734247000.0337--
8.2065248000.0094--
8.2396249000.0171--
8.2727250000.02640.05790.7519
8.3058251000.0269--
8.3388252000.0308--
8.3719253000.0208--
8.4050254000.0062--
8.4381255000.016--
8.4712256000.0165--
8.5043257000.0198--
8.5374258000.0211--
8.5705259000.0355--
8.6036260000.03150.05530.7574
8.6367261000.016--
8.6698262000.0144--
8.7028263000.0208--
8.7359264000.0205--
8.7690265000.0155--
8.8021266000.0284--
8.8352267000.0204--
8.8683268000.0163--
8.9014269000.0382--
8.9345270000.02670.05890.7558
8.9676271000.0204--
9.0007272000.0193--
9.0338273000.0079--
9.0668274000.0156--
9.0999275000.0234--
9.1330276000.0113--
9.1661277000.0216--
9.1992278000.0217--
9.2323279000.0028--
9.2654280000.01550.06270.7521
9.2985281000.0346--
9.3316282000.0165--
9.3647283000.0214--
9.3977284000.0224--
9.4308285000.0095--
9.4639286000.0081--
9.4970287000.0219--
9.5301288000.0272--
9.5632289000.0468--
9.5963290000.00330.05860.7548
9.6294291000.0161--
9.6625292000.0263--
9.6956293000.0156--
9.7287294000.0114--
9.7617295000.0184--
9.7948296000.0098--
9.8279297000.0453--
9.8610298000.0117--
9.8941299000.0142--
9.9272300000.03180.05650.7550
9.9603301000.0192--
9.9934302000.0187--

</details>

Framework Versions

  • —Python: 3.11.13
  • —Sentence Transformers: 4.1.0
  • —Transformers: 4.52.4
  • —PyTorch: 2.6.0+cu124
  • —Accelerate: 1.8.1
  • —Datasets: 3.6.0
  • —Tokenizers: 0.21.2

Citation

BibTeX

Sentence Transformers
bibtex
@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}
MultipleNegativesRankingLoss
bibtex
@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

<!--

Glossary

Clearly define terms in order to be accessible across audiences. -->

<!--

Model Card Authors

Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction. -->

<!--

Model Card Contact

Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors. -->