metadata

base_model: BAAI/bge-base-en-v1.5
datasets:
  - bhlim/patentmatch_for_finetuning
language:
  - en
library_name: sentence-transformers
license: apache-2.0
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:10136
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      The UE sends the uplink signal including the identifier of the uplink
      serving node to the downlink serving node and in this case the downlink
      serving node learns the mapping relationship among the UE the uplink
      serving node and the downlink serving node.The UE sends the uplink signal
      including the identifier of the downlink serving node to the uplink
      serving node an in thiscase the uplink serving node learns the mapping
      relationship among the UE the uplink serving node and the downlink serving
      node.
    sentences:
      - >-
        A terminal for use in a wireless communication network comprising a
        plurality of base stations the terminal arranged to communicate with the
        network via at least two cells of a plurality of cells and to transmit a
        request for uplink resources wherein the terminal is arranged to select
        at least one cell from among said plurality of said cells for
        transmission of said request a resource for transmission of said request
        from among a plurality of resources provided by a cell and a
        characteristic of a signal used to transmit said request and to perform
        the selection in dependence on at least one of the reason for said
        request the characteristics of an uplink channel for transmission of
        said request and the preference of the network.
      - >-
        The electronic device of any of claims 15 wherein the processor is
        further configured to check whether the specific audio data is stored at
        the memory in response to a play request on the specific audio data.
      - >-
        The system of claim 1 or claim 2 comprising a plurality of said
        radiation emitting devices.
  - source_sentence: >-
      Further in the example of Fig.35 the sound adjusting circuit 210 controls
      the sound outputs of the first to fourth speakers 161 to 164 based on the
      sound data from the first to fifth detection sensors 420 to 428 so that
      the first sound corresponding to the second display image is localized in
      the first area 810 where the occupant of the driver seat 13 and the
      occupant of the rear seat 18 equipped with the headrest 25 are
      located.Likewise the sound adjusting circuit 210 controls the sound
      outputs of the first to fourth speakers 161 to 164 based on the sound data
      from the first to fifth detection sensors 420 to 428 so that the second
      sound corresponding to the first display image is localized in the second
      area 820 where the occupant of the assistant drivers seat 12 and the
      occupants of the rear seats 18 equipped with the headrests 26 and 27
      respectively are located.Accordingly the occupant of the rear seat 18
      equipped with the headrest 26 who is located in the crosstalk area 603 in
      Fig.33 can now hear the second sound clearly.
    sentences:
      - >-
        A gas turbine engine comprising a bladed rotor assembly 100200300400
        according to any one of Claims 1 to 9.
      - >-
        The method of claim 1 further comprising sensing a distance between the
        display and a user wherein applying the sound setting comprises applying
        the sound setting based on the sensed distance between the display and
        the user and the obtained curvature of the panel of the display.
      - >-
        A developer carrying member that is capable of carrying a developer on a
        surface thereof and that supplies the developer carried on the surface
        to a surface of an image bearing member when a voltage is applied
        thereto comprising an elastic layer and a surface layer that covers the
        elastic layer contains alumina and has a higher volume resistivity than
        the elastic layer.
  - source_sentence: >-
      In the example of fig.1 a user 107 who arrives in the underground area 109
      and who has not yet subscribed to the electronic ticket service may
      subscribe to the service by connecting his Bluetooth device 107a to a
      Bluetooth access point 104 of the service provider via a Bluetooth service
      device 104a.At the access point 104 the customer 104 may perform a payment
      transaction select a desired subscription and receive a link key.With the
      link key the users Bluetooth device 107a may subsequently establish secure
      Bluetooth connections with the Bluetooth transceivers 101 and 102af.
    sentences:
      - >-
        A wireless communications device 102 for setting up a local service
        session in a shortrange wireless communication network comprising means
        for sending 222 a request for preconfiguration information over a
        longrange network 104 to a remote destination 112 the preconfiguration
        information enabling establishment of the local service session with a
        proximate wireless communications device 110means for receiving 222 from
        the remote destination 112 the requested preconfiguration information
        wherein the requested preconfiguration information includes one or more
        security keys for performing an authentication process with the
        proximate wireless communications device 110 over shortrange wireless
        communication means for performing 220 an authentication process for
        establishing the local service session with the proximate wireless
        communications device 110 over the shortrange wireless communication
        using the received one or more security keys and means for establishing
        220 the local service session with the proximate wireless communications
        device 112 over the shortrange wireless communications after the
        authentication process.
      - >-
        The mobile terminal any one of claims 2 to 4 wherein the controller 180
        is further configured to differently process a color of the image
        corresponding to the trajectory of the second touch based on a position
        of the first touch.
      - >-
        A detergent box assembly for a washing machine comprising a detergent
        box a distributor box having a front plate a rear plate and a receiving
        chamber provided therebetween said receiving chamber configured to store
        a laundry treat agent the distributor box being movably disposed within
        the detergent box and adapted to move between an open position and a
        closed position a keypress being provided in the front plate and a
        driving subassembly disposed in at least one of the detergent box and
        the distributor box and configured to drive the distributor box to move
        from the closed position to the open position when the keypress is
        pressed.
  - source_sentence: >-
      The step of determining may comprisemeasuring a distance between each
      surrogate server and each subnetwork according to the subnetwork of the
      user selecting a surrogate server with the smallest distance.
    sentences:
      - >-
        The computer system of Claim 13 comprising a memory storing instructions
        which when implemented on the one or more processors configure the
        computer system to carry out the method of any one of Claims 1 to 10
      - >-
        A cooking oven 1 comprising a housing 2 a cooking cavity 3 formed in the
        housing 2 and closable by a door 5 heating means 6 6 placed in thermal
        exchange relationship with the cooking cavity 3 ventilating means placed
        in the housing 2 and having one or more electrical fans 7 8 7 8 adapted
        to ventilate on one or more thermally sensitive areas of the oven 1 a
        control system 10 connected to the heating means 6 6 and to the
        ventilating means and having a temperature detector 12 associated with
        the cooking cavity 3 wherein the control system is configured to
        activate and deactivate the heating means 6 6 depending on a temperature
        detected by the temperature detector 12 characterized in that the
        control system 10 activates and deactivates at least one of said one or
        more fans 7 8 7 8 automatically together with the respective activation
        and deactivation of the heating means 6 6.
      - >-
        The method of claim 12 wherein selecting the target control parameter
        further comprises for the respective selected control parameters
        comparing the initial turbine output with the predicted turbine output
        while operating the selected control parameter with the adjustment of
        the selected control parameter to determine an adjustment differential
        and selecting the target control parameter having the target adjustment
        by using the adjustment differential of the target control parameter.
  - source_sentence: >-
      Referring to FIG.32 a a sink device 3200 is designed to display thumbnail
      images in the metadata of contents received from source devices connected
      via an integrated wire interface.As mentioned in the foregoing description
      if a remote controller 3250 capable of outputting a pointing signal is
      situated within a region of a specific thumbnail image 3260 side
      information e.g.Amanda 1st album singer.Song etc.is displayed together.
    sentences:
      - >-
        The method of any one of claims 8 to 12 wherein the requesting for the
        broadcast channel information comprises transmitting to the server image
        data obtained by capturing the content being reproduced by the display
        apparatus or audio data obtained by recording the content for a certain
        time.
      - >-
        The electrode assembly of any one of the preceding claims wherein the
        first electrode comprises a substrate 113 wherein the first active
        material layer comprises active material layers 112 on both surfaces of
        the substrate and the ceramic layer comprises ceramic material layers 50
        on both surfaces of the substrate.
      - >-
        A method according to claim 1 wherein said topsheet assembly is a
        threeply laminate comprising an acquisition layer a nonwoven layer and a
        cuff assembly.
model-index:
  - name: BGE base PatentMatch Matryoshka
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.042620363062352014
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.10142067876874507
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.14483030781373324
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.23204419889502761
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.042620363062352014
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.03380689292291502
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.02896606156274665
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.023204419889502764
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.042620363062352014
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.10142067876874507
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.14483030781373324
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.23204419889502761
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.12169609468606697
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.08838588842535165
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.10140867877546615
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.04222573007103394
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.09352801894238358
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.14285714285714285
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.22454617205998423
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.04222573007103394
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.031176006314127862
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.028571428571428574
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.02245461720599842
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.04222573007103394
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.09352801894238358
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.14285714285714285
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.22454617205998423
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.11822400593872298
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.08611580912291245
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.09959411357742169
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.04025256511444357
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.09155485398579322
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.13970007892659828
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.21981057616416733
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.04025256511444357
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.03051828466193107
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.02794001578531966
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.021981057616416732
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.04025256511444357
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.09155485398579322
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.13970007892659828
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.21981057616416733
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.11513294301691931
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.08350856917352567
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.09631638060202527
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.037884767166535126
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.08602999210734018
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.13180741910023677
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.2079715864246251
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.037884767166535126
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.028676664035780054
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.02636148382004736
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.02079715864246251
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.037884767166535126
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.08602999210734018
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.13180741910023677
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.2079715864246251
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.10894233297304821
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.07907489883614581
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.09087791679720966
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.032754538279400155
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.07419100236779795
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.11444356748224152
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.18468823993685873
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.032754538279400155
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.024730334122599312
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.022888713496448304
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.018468823993685875
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.032754538279400155
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.07419100236779795
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.11444356748224152
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.18468823993685873
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.0959638876946607
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.06921471166735564
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.08022788346205763
            name: Cosine Map@100

BGE base PatentMatch Matryoshka

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5 on the bhlim/patentmatch_for_finetuning dataset. 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: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Training Dataset:
- bhlim/patentmatch_for_finetuning
Language: en
License: apache-2.0

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("bhlim/bge-base-patentmatch")
# Run inference
sentences = [
    'Referring to FIG.32 a a sink device 3200 is designed to display thumbnail images in the metadata of contents received from source devices connected via an integrated wire interface.As mentioned in the foregoing description if a remote controller 3250 capable of outputting a pointing signal is situated within a region of a specific thumbnail image 3260 side information e.g.Amanda 1st album singer.Song etc.is displayed together.',
    'The method of any one of claims 8 to 12 wherein the requesting for the broadcast channel information comprises transmitting to the server image data obtained by capturing the content being reproduced by the display apparatus or audio data obtained by recording the content for a certain time.',
    'The electrode assembly of any one of the preceding claims wherein the first electrode comprises a substrate 113 wherein the first active material layer comprises active material layers 112 on both surfaces of the substrate and the ceramic layer comprises ceramic material layers 50 on both surfaces of the substrate.',
]
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]

Evaluation

Metrics

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0426
cosine_accuracy@3	0.1014
cosine_accuracy@5	0.1448
cosine_accuracy@10	0.232
cosine_precision@1	0.0426
cosine_precision@3	0.0338
cosine_precision@5	0.029
cosine_precision@10	0.0232
cosine_recall@1	0.0426
cosine_recall@3	0.1014
cosine_recall@5	0.1448
cosine_recall@10	0.232
cosine_ndcg@10	0.1217
cosine_mrr@10	0.0884
cosine_map@100	0.1014

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0422
cosine_accuracy@3	0.0935
cosine_accuracy@5	0.1429
cosine_accuracy@10	0.2245
cosine_precision@1	0.0422
cosine_precision@3	0.0312
cosine_precision@5	0.0286
cosine_precision@10	0.0225
cosine_recall@1	0.0422
cosine_recall@3	0.0935
cosine_recall@5	0.1429
cosine_recall@10	0.2245
cosine_ndcg@10	0.1182
cosine_mrr@10	0.0861
cosine_map@100	0.0996

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0403
cosine_accuracy@3	0.0916
cosine_accuracy@5	0.1397
cosine_accuracy@10	0.2198
cosine_precision@1	0.0403
cosine_precision@3	0.0305
cosine_precision@5	0.0279
cosine_precision@10	0.022
cosine_recall@1	0.0403
cosine_recall@3	0.0916
cosine_recall@5	0.1397
cosine_recall@10	0.2198
cosine_ndcg@10	0.1151
cosine_mrr@10	0.0835
cosine_map@100	0.0963

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0379
cosine_accuracy@3	0.086
cosine_accuracy@5	0.1318
cosine_accuracy@10	0.208
cosine_precision@1	0.0379
cosine_precision@3	0.0287
cosine_precision@5	0.0264
cosine_precision@10	0.0208
cosine_recall@1	0.0379
cosine_recall@3	0.086
cosine_recall@5	0.1318
cosine_recall@10	0.208
cosine_ndcg@10	0.1089
cosine_mrr@10	0.0791
cosine_map@100	0.0909

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.0328
cosine_accuracy@3	0.0742
cosine_accuracy@5	0.1144
cosine_accuracy@10	0.1847
cosine_precision@1	0.0328
cosine_precision@3	0.0247
cosine_precision@5	0.0229
cosine_precision@10	0.0185
cosine_recall@1	0.0328
cosine_recall@3	0.0742
cosine_recall@5	0.1144
cosine_recall@10	0.1847
cosine_ndcg@10	0.096
cosine_mrr@10	0.0692
cosine_map@100	0.0802

Training Details

Training Dataset

bhlim/patentmatch_for_finetuning

Dataset: bhlim/patentmatch_for_finetuning at 8d60f21
Size: 10,136 training samples
Columns: positive and anchor
Approximate statistics based on the first 1000 samples:
positive anchor
type string string
details
min: 5 tokens
mean: 136.61 tokens
max: 512 tokens

min: 12 tokens
mean: 76.35 tokens
max: 512 tokens

	positive	anchor
type	string	string
details	min: 5 tokens mean: 136.61 tokens max: 512 tokens	min: 12 tokens mean: 76.35 tokens max: 512 tokens

Samples:

positive	anchor
Furthermore according to this liquid consuming apparatus if the decompression level acting on the liquid sensing chamber 21 of the liquid container 1 i.e.the pressure loss arising in the connecting passage between the liquid storage portion 7 and the liquid sensing chamber 21 due to the flow rate outflowing from the liquid storage portion 7 because of distension of the diaphragm pump through application of the external force when external force is applied in the direction of expansion of volume of the diaphragm pump 42 asdepicted in FIG.6 has been set to a low level if sufficient liquid is present in the liquid container 1 the liquid sensing chamber 21 will experience substantially no change in volume.	`The liquid cartridge according to any of claims 4 to 5 further comprising a ground terminal 175c 176c 177c positioned in the second line.`
`It is highly desirable for tires to have good wet skid resistance low rolling resistance and good wear characteristics.It has traditionally been very difficult to improve a tires wear characteristics without sacrificing its wet skid resistance and traction characteristics.These properties depend to a great extent on the dynamic viscoelastic properties of the rubbers utilized in making the tire.`	`The pneumatic tire of at least one of the previous claims wherein the rubber composition comprises from 5 to 20 phr of the oil and from 45 to 70 phr of the terpene phenol resin.`
Before setting the environment of the mobile communication terminal a user stores a multimedia message composed of different kinds of contents i.e.images sounds and texts.For example reference block 201 indicates a multimedia message composed of several images sounds and texts.The user can select an image A a sound A and a text A for environment setting elements of the mobile communication terminal from the contents of the multimedia message and construct a theme like in block 203 using the selected image A sound A and text A.The MPU 101 maps the contents of the theme to environment setting elements of the mobile communication terminal i.e.a background screen a ringtone and a user name like in block 205.The MPU 101 then sets the environment of the mobile communication terminal using the mapped elements like in block 207 thereby automatically and collectively changing the environment of the mobile communication terminal.Mapping information about mapping between the selected contents of the multimediamessage and the environment setting elements of the mobile communication terminal is stored in the flash RAM 107.	`A terminal for processing data comprising an output unit configured to output a chatting service window a receiving unit configured to receive a request for executing a chatting service and a first download request for downloading first data through the chatting service from a user and a controller configured to control to output the first data downloaded in response to the received first download request to a background screen of the chatting service window.`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 2e-05
num_train_epochs: 4
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: True
tf32: True
load_best_model_at_end: True
optim: adamw_torch_fused
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 16
eval_accumulation_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: 4
max_steps: -1
lr_scheduler_type: cosine
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: True
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: True
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: True
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
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: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
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
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_128_cosine_map@100	dim_256_cosine_map@100	dim_512_cosine_map@100	dim_64_cosine_map@100	dim_768_cosine_map@100
0.5047	10	10.0459	-	-	-	-	-
0.9590	19	-	0.0849	0.0915	0.0939	0.0778	0.0966
1.0095	20	7.1373	-	-	-	-	-
1.5142	30	5.9969	-	-	-	-	-
1.9685	39	-	0.0890	0.0965	0.1007	0.0795	0.1012
2.0189	40	5.2984	-	-	-	-	-
2.5237	50	4.884	-	-	-	-	-
2.9779	59	-	0.091	0.0967	0.099	0.0801	0.1013
3.0284	60	4.6633	-	-	-	-	-
3.5331	70	4.5226	-	-	-	-	-
3.8360	76	-	0.0909	0.0963	0.0996	0.0802	0.1014

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.1.2+cu121
Accelerate: 0.32.1
Datasets: 2.19.1
Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@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}
}