README.md

metadata

language:
  - en
license: apache-2.0
tags:
  - sentence-transformers
  - sparse-encoder
  - sparse
  - splade
  - generated_from_trainer
  - dataset_size:99000
  - loss:SpladeLoss
widget:
  - source_sentence: who are the dancers in the limp bizkit rollin video
    sentences:
      - >-
        Voting age Before the Second World War, the voting age in almost all
        countries was 21 years or higher. Czechoslovakia was the first to reduce
        the voting age to 20 years in 1946, and by 1968 a total of 17 countries
        had lowered their voting age.[1] Many countries, particularly in Western
        Europe, reduced their voting ages to 18 years during the 1970s, starting
        with the United Kingdom (1969),[2] with the United States (26th
        Amendment) (1971), Canada, West Germany (1972), Australia (1974), France
        (1974), and others following soon afterwards. By the end of the 20th
        century, 18 had become by far the most common voting age. However, a few
        countries maintain a voting age of 20 years or higher. It was argued
        that young men could be drafted to go to war at 18, and many people felt
        they should be able to vote at the age of 18.[3]
      - >-
        Rollin' (Limp Bizkit song) The music video was filmed atop the South
        Tower of the former World Trade Center in New York City. The
        introduction features Ben Stiller and Stephen Dorff mistaking Fred Durst
        for the valet and giving him the keys to their Bentley Azure. Also
        making a cameo is break dancer Mr. Wiggles. The rest of the video has
        several cuts to Durst and his bandmates hanging out of the Bentley as
        they drive about Manhattan. The song Ben Stiller is playing at the
        beginning is "My Generation" from the same album. The video also
        features scenes of Fred Durst with five girls dancing in a room. The
        video was filmed around the same time as the film Zoolander, which
        explains Stiller and Dorff's appearance. Fred Durst has a small cameo in
        that film.
      - >-
        Eobard Thawne When Thawne reappears, he murders the revived Johnny
        Quick,[9] before proceeding to trap Barry and the revived Max Mercury
        inside the negative Speed Force. Thawne then attempts to kill Wally
        West's children through their connection to the Speed Force in front of
        Linda Park-West, only to be stopped by Jay Garrick and Bart Allen.
        Thawne defeats Jay and prepares to kill Bart, but Barry, Max, Wally,
        Jesse Quick, and Impulse arrive to prevent the villain from doing
        so.[8][10] In the ensuing fight, Thawne reveals that he is responsible
        for every tragedy that has occurred in Barry's life, including the death
        of his mother. Thawne then decides to destroy everything the Flash holds
        dear by killing Barry's wife, Iris, before they even met.[10]
  - source_sentence: who wins season 14 of hell's kitchen
    sentences:
      - >-
        Hell's Kitchen (U.S. season 14) Season 14 of the American competitive
        reality television series Hell's Kitchen premiered on March 3, 2015 on
        Fox. The prize is a head chef position at Gordon Ramsay Pub & Grill in
        Caesars Atlantic City.[1] Gordon Ramsay returned as head chef with Andi
        Van Willigan and James Avery returning as sous-chefs for both their
        respective kitchens as well as Marino Monferrato as the maître d'.
        Executive chef Meghan Gill from Roanoke, Virginia, won the competition,
        thus becoming the fourteenth winner of Hell's Kitchen.
      - >-
        Maze Runner: The Death Cure On April 22, 2017, the studio delayed the
        release date once again, to February 9, 2018, in order to allow more
        time for post-production; months later, on August 25, the studio moved
        the release forward two weeks.[17] The film will premiere on January 26,
        2018 in 3D, IMAX and IMAX 3D.[18][19]
      - >-
        North American Plate On its western edge, the Farallon Plate has been
        subducting under the North American Plate since the Jurassic Period. The
        Farallon Plate has almost completely subducted beneath the western
        portion of the North American Plate leaving that part of the North
        American Plate in contact with the Pacific Plate as the San Andreas
        Fault. The Juan de Fuca, Explorer, Gorda, Rivera, Cocos and Nazca plates
        are remnants of the Farallon Plate.
  - source_sentence: who played the dj in the movie the warriors
    sentences:
      - "List of Arrow episodes As of May\_17, 2018,[update] 138 episodes of Arrow\_have aired, concluding the\_sixth season. On April 2, 2018, the CW renewed the series for a seventh season.[1]"
      - >-
        Lynne Thigpen Cherlynne Theresa "Lynne" Thigpen (December 22, 1948 –
        March 12, 2003) was an American actress, best known for her role as "The
        Chief" of ACME in the various Carmen Sandiego television series and
        computer games from 1991 to 1997. For her varied television work,
        Thigpen was nominated for six Daytime Emmy Awards; she won a Tony Award
        in 1997 for portraying Dr. Judith Kaufman in An American Daughter.
      - >-
        The Washington Post The Washington Post is an American daily newspaper.
        It is the most widely circulated newspaper published in Washington,
        D.C., and was founded on December 6, 1877,[7] making it the area's
        oldest extant newspaper. In February 2017, amid a barrage of criticism
        from President Donald Trump over the paper's coverage of his campaign
        and early presidency as well as concerns among the American press about
        Trump's criticism and threats against journalists who provide coverage
        he deems unfavorable, the Post adopted the slogan "Democracy Dies in
        Darkness".[8]
  - source_sentence: how old was messi when he started his career
    sentences:
      - >-
        Lionel Messi Born and raised in central Argentina, Messi was diagnosed
        with a growth hormone deficiency as a child. At age 13, he relocated to
        Spain to join Barcelona, who agreed to pay for his medical treatment.
        After a fast progression through Barcelona's youth academy, Messi made
        his competitive debut aged 17 in October 2004. Despite being
        injury-prone during his early career, he established himself as an
        integral player for the club within the next three years, finishing 2007
        as a finalist for both the Ballon d'Or and FIFA World Player of the Year
        award, a feat he repeated the following year. His first uninterrupted
        campaign came in the 2008–09 season, during which he helped Barcelona
        achieve the first treble in Spanish football. At 22 years old, Messi won
        the Ballon d'Or and FIFA World Player of the Year award by record voting
        margins.
      - >-
        We Are Marshall Filming of We Are Marshall commenced on April 3, 2006,
        in Huntington, West Virginia, and was completed in Atlanta, Georgia. The
        premiere for the film was held at the Keith Albee Theater on December
        12, 2006, in Huntington; other special screenings were held at Pullman
        Square. The movie was released nationwide on December 22, 2006.
      - >-
        One Fish, Two Fish, Red Fish, Blue Fish One Fish, Two Fish, Red Fish,
        Blue Fish is a 1960 children's book by Dr. Seuss. It is a simple rhyming
        book for beginning readers, with a freewheeling plot about a boy and a
        girl named Jay and Kay and the many amazing creatures they have for
        friends and pets. Interspersed are some rather surreal and unrelated
        skits, such as a man named Ned whose feet stick out from his bed, and a
        creature who has a bird in his ear. As of 2001, over 6 million copies of
        the book had been sold, placing it 13th on a list of "All-Time
        Bestselling Children's Books" from Publishers Weekly.[1] Based on a 2007
        online poll, the United States' National Education Association labor
        union named the book one of its "Teachers' Top 100 Books for
        Children."[2]
  - source_sentence: is send in the clowns from a musical
    sentences:
      - >-
        Money in the Bank ladder match The first match was contested in 2005 at
        WrestleMania 21, after being invented (in kayfabe) by Chris Jericho.[1]
        At the time, it was exclusive to wrestlers of the Raw brand, and Edge
        won the inaugural match.[1] From then until 2010, the Money in the Bank
        ladder match, now open to all WWE brands, became a WrestleMania
        mainstay. 2010 saw a second and third Money in the Bank ladder match
        when the Money in the Bank pay-per-view debuted in July. Unlike the
        matches at WrestleMania, this new event featured two such ladder matches
        – one each for a contract for the WWE Championship and World
        Heavyweight Championship, respectively.
      - >-
        The Suite Life on Deck The Suite Life on Deck is an American sitcom that
        aired on Disney Channel from September 26, 2008 to May 6, 2011. It is a
        sequel/spin-off of the Disney Channel Original Series The Suite Life of
        Zack & Cody. The series follows twin brothers Zack and Cody Martin and
        hotel heiress London Tipton in a new setting, the SS Tipton, where they
        attend classes at "Seven Seas High School" and meet Bailey Pickett while
        Mr. Moseby manages the ship. The ship travels around the world to
        nations such as Italy, France, Greece, India, Sweden and the United
        Kingdom where the characters experience different cultures, adventures,
        and situations.[1]
      - >-
        Send In the Clowns "Send In the Clowns" is a song written by Stephen
        Sondheim for the 1973 musical A Little Night Music, an adaptation of
        Ingmar Bergman's film Smiles of a Summer Night. It is a ballad from Act
        Two, in which the character Desirée reflects on the ironies and
        disappointments of her life. Among other things, she looks back on an
        affair years earlier with the lawyer Fredrik, who was deeply in love
        with her but whose marriage proposals she had rejected. Meeting him
        after so long, she realizes she is in love with him and finally ready to
        marry him, but now it is he who rejects her: he is in an unconsummated
        marriage with a much younger woman. Desirée proposes marriage to rescue
        him from this situation, but he declines, citing his dedication to his
        bride. Reacting to his rejection, Desirée sings this song. The song is
        later reprised as a coda after Fredrik's young wife runs away with his
        son, and Fredrik is finally free to accept Desirée's offer.[1]
datasets:
  - sentence-transformers/natural-questions
pipeline_tag: feature-extraction
library_name: sentence-transformers
metrics:
  - dot_accuracy@1
  - dot_accuracy@3
  - dot_accuracy@5
  - dot_accuracy@10
  - dot_precision@1
  - dot_precision@3
  - dot_precision@5
  - dot_precision@10
  - dot_recall@1
  - dot_recall@3
  - dot_recall@5
  - dot_recall@10
  - dot_ndcg@10
  - dot_mrr@10
  - dot_map@100
co2_eq_emissions:
  emissions: 10.656630177765601
  energy_consumed: 0.027415938631047954
  source: codecarbon
  training_type: fine-tuning
  on_cloud: false
  cpu_model: 13th Gen Intel(R) Core(TM) i7-13700K
  ram_total_size: 31.777088165283203
  hours_used: 0.082
  hardware_used: 1 x NVIDIA GeForce RTX 3090
model-index:
  - name: SPLADE BERT-tiny trained on Natural-Questions tuples
    results:
      - task:
          type: sparse-information-retrieval
          name: Sparse Information Retrieval
        dataset:
          name: NanoMSMARCO
          type: NanoMSMARCO
        metrics:
          - type: dot_accuracy@1
            value: 0.22
            name: Dot Accuracy@1
          - type: dot_accuracy@3
            value: 0.36
            name: Dot Accuracy@3
          - type: dot_accuracy@5
            value: 0.4
            name: Dot Accuracy@5
          - type: dot_accuracy@10
            value: 0.54
            name: Dot Accuracy@10
          - type: dot_precision@1
            value: 0.22
            name: Dot Precision@1
          - type: dot_precision@3
            value: 0.11999999999999998
            name: Dot Precision@3
          - type: dot_precision@5
            value: 0.08000000000000002
            name: Dot Precision@5
          - type: dot_precision@10
            value: 0.054000000000000006
            name: Dot Precision@10
          - type: dot_recall@1
            value: 0.22
            name: Dot Recall@1
          - type: dot_recall@3
            value: 0.36
            name: Dot Recall@3
          - type: dot_recall@5
            value: 0.4
            name: Dot Recall@5
          - type: dot_recall@10
            value: 0.54
            name: Dot Recall@10
          - type: dot_ndcg@10
            value: 0.3571008976876618
            name: Dot Ndcg@10
          - type: dot_mrr@10
            value: 0.30113492063492064
            name: Dot Mrr@10
          - type: dot_map@100
            value: 0.31600753362153616
            name: Dot Map@100
      - task:
          type: sparse-information-retrieval
          name: Sparse Information Retrieval
        dataset:
          name: NanoNFCorpus
          type: NanoNFCorpus
        metrics:
          - type: dot_accuracy@1
            value: 0.24
            name: Dot Accuracy@1
          - type: dot_accuracy@3
            value: 0.38
            name: Dot Accuracy@3
          - type: dot_accuracy@5
            value: 0.48
            name: Dot Accuracy@5
          - type: dot_accuracy@10
            value: 0.58
            name: Dot Accuracy@10
          - type: dot_precision@1
            value: 0.24
            name: Dot Precision@1
          - type: dot_precision@3
            value: 0.21333333333333332
            name: Dot Precision@3
          - type: dot_precision@5
            value: 0.184
            name: Dot Precision@5
          - type: dot_precision@10
            value: 0.16
            name: Dot Precision@10
          - type: dot_recall@1
            value: 0.01910619386686893
            name: Dot Recall@1
          - type: dot_recall@3
            value: 0.03647891009411463
            name: Dot Recall@3
          - type: dot_recall@5
            value: 0.043286562520389434
            name: Dot Recall@5
          - type: dot_recall@10
            value: 0.0624423217616165
            name: Dot Recall@10
          - type: dot_ndcg@10
            value: 0.1824659316003306
            name: Dot Ndcg@10
          - type: dot_mrr@10
            value: 0.3309444444444445
            name: Dot Mrr@10
          - type: dot_map@100
            value: 0.0640015611933746
            name: Dot Map@100
      - task:
          type: sparse-information-retrieval
          name: Sparse Information Retrieval
        dataset:
          name: NanoNQ
          type: NanoNQ
        metrics:
          - type: dot_accuracy@1
            value: 0.1
            name: Dot Accuracy@1
          - type: dot_accuracy@3
            value: 0.24
            name: Dot Accuracy@3
          - type: dot_accuracy@5
            value: 0.34
            name: Dot Accuracy@5
          - type: dot_accuracy@10
            value: 0.44
            name: Dot Accuracy@10
          - type: dot_precision@1
            value: 0.1
            name: Dot Precision@1
          - type: dot_precision@3
            value: 0.08
            name: Dot Precision@3
          - type: dot_precision@5
            value: 0.068
            name: Dot Precision@5
          - type: dot_precision@10
            value: 0.046000000000000006
            name: Dot Precision@10
          - type: dot_recall@1
            value: 0.09
            name: Dot Recall@1
          - type: dot_recall@3
            value: 0.22
            name: Dot Recall@3
          - type: dot_recall@5
            value: 0.32
            name: Dot Recall@5
          - type: dot_recall@10
            value: 0.41
            name: Dot Recall@10
          - type: dot_ndcg@10
            value: 0.24844109892252747
            name: Dot Ndcg@10
          - type: dot_mrr@10
            value: 0.20507142857142857
            name: Dot Mrr@10
          - type: dot_map@100
            value: 0.208667797146501
            name: Dot Map@100
      - task:
          type: sparse-nano-beir
          name: Sparse Nano BEIR
        dataset:
          name: NanoBEIR mean
          type: NanoBEIR_mean
        metrics:
          - type: dot_accuracy@1
            value: 0.18666666666666665
            name: Dot Accuracy@1
          - type: dot_accuracy@3
            value: 0.32666666666666666
            name: Dot Accuracy@3
          - type: dot_accuracy@5
            value: 0.4066666666666667
            name: Dot Accuracy@5
          - type: dot_accuracy@10
            value: 0.52
            name: Dot Accuracy@10
          - type: dot_precision@1
            value: 0.18666666666666665
            name: Dot Precision@1
          - type: dot_precision@3
            value: 0.13777777777777778
            name: Dot Precision@3
          - type: dot_precision@5
            value: 0.11066666666666668
            name: Dot Precision@5
          - type: dot_precision@10
            value: 0.08666666666666667
            name: Dot Precision@10
          - type: dot_recall@1
            value: 0.10970206462228964
            name: Dot Recall@1
          - type: dot_recall@3
            value: 0.20549297003137154
            name: Dot Recall@3
          - type: dot_recall@5
            value: 0.25442885417346317
            name: Dot Recall@5
          - type: dot_recall@10
            value: 0.3374807739205388
            name: Dot Recall@10
          - type: dot_ndcg@10
            value: 0.26266930940350663
            name: Dot Ndcg@10
          - type: dot_mrr@10
            value: 0.27905026455026455
            name: Dot Mrr@10
          - type: dot_map@100
            value: 0.1962256306538039
            name: Dot Map@100

SPLADE BERT-tiny trained on Natural-Questions tuples

This is a SPLADE Sparse Encoder model trained on the natural-questions dataset using the sentence-transformers library. It maps sentences & paragraphs to a 30522-dimensional sparse vector space and can be used for semantic search and sparse retrieval.

This model was trained using train_script.py.

Model Details

Model Description

• Model Type: SPLADE Sparse Encoder
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 30522 dimensions
• Similarity Function: Dot Product
• Training Dataset:
  • natural-questions
• Language: en
• License: apache-2.0

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Sparse Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sparse Encoders on Hugging Face

Full Model Architecture

SparseEncoder(
  (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
  (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
)

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 SparseEncoder

# Download from the 🤗 Hub
model = SparseEncoder("sparse-encoder-testing/splade-bert-tiny-nq")
# Run inference
sentences = [
    'is send in the clowns from a musical',
    'Send In the Clowns "Send In the Clowns" is a song written by Stephen Sondheim for the 1973 musical A Little Night Music, an adaptation of Ingmar Bergman\'s film Smiles of a Summer Night. It is a ballad from Act Two, in which the character Desirée reflects on the ironies and disappointments of her life. Among other things, she looks back on an affair years earlier with the lawyer Fredrik, who was deeply in love with her but whose marriage proposals she had rejected. Meeting him after so long, she realizes she is in love with him and finally ready to marry him, but now it is he who rejects her: he is in an unconsummated marriage with a much younger woman. Desirée proposes marriage to rescue him from this situation, but he declines, citing his dedication to his bride. Reacting to his rejection, Desirée sings this song. The song is later reprised as a coda after Fredrik\'s young wife runs away with his son, and Fredrik is finally free to accept Desirée\'s offer.[1]',
    'The Suite Life on Deck The Suite Life on Deck is an American sitcom that aired on Disney Channel from September 26, 2008 to May 6, 2011. It is a sequel/spin-off of the Disney Channel Original Series The Suite Life of Zack & Cody. The series follows twin brothers Zack and Cody Martin and hotel heiress London Tipton in a new setting, the SS Tipton, where they attend classes at "Seven Seas High School" and meet Bailey Pickett while Mr. Moseby manages the ship. The ship travels around the world to nations such as Italy, France, Greece, India, Sweden and the United Kingdom where the characters experience different cultures, adventures, and situations.[1]',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# (3, 30522)

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

Evaluation

Metrics

Sparse Information Retrieval

• Datasets: NanoMSMARCO, NanoNFCorpus and NanoNQ
• Evaluated with SparseInformationRetrievalEvaluator

Metric	NanoMSMARCO	NanoNFCorpus	NanoNQ
dot_accuracy@1	0.22	0.24	0.1
dot_accuracy@3	0.36	0.38	0.24
dot_accuracy@5	0.4	0.48	0.34
dot_accuracy@10	0.54	0.58	0.44
dot_precision@1	0.22	0.24	0.1
dot_precision@3	0.12	0.2133	0.08
dot_precision@5	0.08	0.184	0.068
dot_precision@10	0.054	0.16	0.046
dot_recall@1	0.22	0.0191	0.09
dot_recall@3	0.36	0.0365	0.22
dot_recall@5	0.4	0.0433	0.32
dot_recall@10	0.54	0.0624	0.41
dot_ndcg@10	0.3571	0.1825	0.2484
dot_mrr@10	0.3011	0.3309	0.2051
dot_map@100	0.316	0.064	0.2087

Sparse Nano BEIR

• Dataset: NanoBEIR_mean
• Evaluated with SparseNanoBEIREvaluator with these parameters:

  {
      "dataset_names": [
          "msmarco",
          "nfcorpus",
          "nq"
      ]
  }
  

Metric	Value
dot_accuracy@1	0.1867
dot_accuracy@3	0.3267
dot_accuracy@5	0.4067
dot_accuracy@10	0.52
dot_precision@1	0.1867
dot_precision@3	0.1378
dot_precision@5	0.1107
dot_precision@10	0.0867
dot_recall@1	0.1097
dot_recall@3	0.2055
dot_recall@5	0.2544
dot_recall@10	0.3375
dot_ndcg@10	0.2627
dot_mrr@10	0.2791
dot_map@100	0.1962

Training Details

Training Dataset

natural-questions

• Dataset: natural-questions at f9e894e
• Size: 99,000 training samples
• Columns: query and answer
• Approximate statistics based on the first 1000 samples:

  	query	answer
  type	string	string
  details	min: 29 characters mean: 46.96 characters max: 93 characters	min: 10 characters mean: 582.13 characters max: 2141 characters

• Samples:

  query	answer
  who played the father in papa don't preach	Alex McArthur Alex McArthur (born March 6, 1957) is an American actor.
  where was the location of the battle of hastings	Battle of Hastings The Battle of Hastings[a] was fought on 14 October 1066 between the Norman-French army of William, the Duke of Normandy, and an English army under the Anglo-Saxon King Harold Godwinson, beginning the Norman conquest of England. It took place approximately 7 miles (11 kilometres) northwest of Hastings, close to the present-day town of Battle, East Sussex, and was a decisive Norman victory.
  how many puppies can a dog give birth to	Canine reproduction The largest litter size to date was set by a Neapolitan Mastiff in Manea, Cambridgeshire, UK on November 29, 2004; the litter was 24 puppies.[22]

• Loss: SpladeLoss with these parameters:

  {'loss': SparseMultipleNegativesRankingLoss(
    (model): SparseEncoder(
      (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
      (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
    )
    (cross_entropy_loss): CrossEntropyLoss()
  ), 'lambda_corpus': 3e-05, 'lambda_query': 5e-05, 'corpus_regularizer': FlopsLoss(
    (model): SparseEncoder(
      (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
      (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
    )
  ), 'query_regularizer': FlopsLoss(
    (model): SparseEncoder(
      (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
      (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
    )
  )}
  

Evaluation Dataset

natural-questions

• Dataset: natural-questions at f9e894e
• Size: 1,000 evaluation samples
• Columns: query and answer
• Approximate statistics based on the first 1000 samples:

  	query	answer
  type	string	string
  details	min: 30 characters mean: 47.2 characters max: 96 characters	min: 58 characters mean: 598.96 characters max: 2480 characters

• Samples:

  query	answer
  where is the tiber river located in italy	Tiber The Tiber (/ˈtaɪbər/, Latin: Tiberis,[1] Italian: Tevere [ˈteːvere])[2] is the third-longest river in Italy, rising in the Apennine Mountains in Emilia-Romagna and flowing 406 kilometres (252 mi) through Tuscany, Umbria and Lazio, where it is joined by the river Aniene, to the Tyrrhenian Sea, between Ostia and Fiumicino.[3] It drains a basin estimated at 17,375 square kilometres (6,709 sq mi). The river has achieved lasting fame as the main watercourse of the city of Rome, founded on its eastern banks.
  what kind of car does jay gatsby drive	Jay Gatsby At the Buchanan home, Jordan Baker, Nick, Jay, and the Buchanans decide to visit New York City. Tom borrows Gatsby's yellow Rolls Royce to drive up to the city. On the way to New York City, Tom makes a detour at a gas station in "the Valley of Ashes", a run-down part of Long Island. The owner, George Wilson, shares his concern that his wife, Myrtle, may be having an affair. This unnerves Tom, who has been having an affair with Myrtle, and he leaves in a hurry.
  who sings if i can dream about you	I Can Dream About You "I Can Dream About You" is a song performed by American singer Dan Hartman on the soundtrack album of the film Streets of Fire. Released in 1984 as a single from the soundtrack, and included on Hartman's album I Can Dream About You, it reached number 6 on the Billboard Hot 100.[1]

• Loss: SpladeLoss with these parameters:

  {'loss': SparseMultipleNegativesRankingLoss(
    (model): SparseEncoder(
      (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
      (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
    )
    (cross_entropy_loss): CrossEntropyLoss()
  ), 'lambda_corpus': 3e-05, 'lambda_query': 5e-05, 'corpus_regularizer': FlopsLoss(
    (model): SparseEncoder(
      (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
      (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
    )
  ), 'query_regularizer': FlopsLoss(
    (model): SparseEncoder(
      (0): MLMTransformer({'max_seq_length': 512, 'do_lower_case': False}) with MLMTransformer model: BertForMaskedLM 
      (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
    )
  )}
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• 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: 1
• 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: {'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
• 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
• 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
• 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

Training Logs

Epoch	Step	Training Loss	Validation Loss	NanoMSMARCO_dot_ndcg@10	NanoNFCorpus_dot_ndcg@10	NanoNQ_dot_ndcg@10	NanoBEIR_mean_dot_ndcg@10
0.0129	20	1688.0217	-	-	-	-	-
0.0259	40	1557.9103	-	-	-	-	-
0.0388	60	1205.4178	-	-	-	-	-
0.0517	80	692.3048	-	-	-	-	-
0.0646	100	297.4244	-	-	-	-	-
0.0776	120	144.2392	-	-	-	-	-
0.0905	140	75.7438	-	-	-	-	-
0.1034	160	35.3506	-	-	-	-	-
0.1164	180	20.7095	-	-	-	-	-
0.1293	200	12.5446	6.9048	0.1524	0.0784	0.0574	0.0961
0.1422	220	8.1351	-	-	-	-	-
0.1551	240	6.1495	-	-	-	-	-
0.1681	260	4.4986	-	-	-	-	-
0.1810	280	3.5353	-	-	-	-	-
0.1939	300	3.0714	-	-	-	-	-
0.2069	320	2.4237	-	-	-	-	-
0.2198	340	1.9325	-	-	-	-	-
0.2327	360	1.8585	-	-	-	-	-
0.2456	380	1.491	-	-	-	-	-
0.2586	400	1.4503	0.8541	0.2248	0.1322	0.1045	0.1538
0.2715	420	1.3789	-	-	-	-	-
0.2844	440	1.3195	-	-	-	-	-
0.2973	460	1.198	-	-	-	-	-
0.3103	480	1.1532	-	-	-	-	-
0.3232	500	1.1931	-	-	-	-	-
0.3361	520	1.1989	-	-	-	-	-
0.3491	540	1.008	-	-	-	-	-
0.3620	560	0.9798	-	-	-	-	-
0.3749	580	0.9551	-	-	-	-	-
0.3878	600	0.9687	0.4356	0.2709	0.1438	0.1519	0.1888
0.4008	620	0.8331	-	-	-	-	-
0.4137	640	0.6947	-	-	-	-	-
0.4266	660	0.7768	-	-	-	-	-
0.4396	680	0.7101	-	-	-	-	-
0.4525	700	0.6902	-	-	-	-	-
0.4654	720	0.6766	-	-	-	-	-
0.4783	740	0.6001	-	-	-	-	-
0.4913	760	0.6231	-	-	-	-	-
0.5042	780	0.5953	-	-	-	-	-
0.5171	800	0.6846	0.3068	0.2958	0.1543	0.2071	0.2190
0.5301	820	0.5851	-	-	-	-	-
0.5430	840	0.579	-	-	-	-	-
0.5559	860	0.5659	-	-	-	-	-
0.5688	880	0.553	-	-	-	-	-
0.5818	900	0.4812	-	-	-	-	-
0.5947	920	0.5389	-	-	-	-	-
0.6076	940	0.4658	-	-	-	-	-
0.6206	960	0.5309	-	-	-	-	-
0.6335	980	0.484	-	-	-	-	-
0.6464	1000	0.4655	0.2527	0.3131	0.1660	0.2294	0.2362
0.6593	1020	0.5617	-	-	-	-	-
0.6723	1040	0.4786	-	-	-	-	-
0.6852	1060	0.5561	-	-	-	-	-
0.6981	1080	0.4869	-	-	-	-	-
0.7111	1100	0.5134	-	-	-	-	-
0.7240	1120	0.4702	-	-	-	-	-
0.7369	1140	0.4481	-	-	-	-	-
0.7498	1160	0.4758	-	-	-	-	-
0.7628	1180	0.4625	-	-	-	-	-
0.7757	1200	0.4733	0.2330	0.3498	0.1748	0.2357	0.2534
0.7886	1220	0.4527	-	-	-	-	-
0.8016	1240	0.4735	-	-	-	-	-
0.8145	1260	0.3818	-	-	-	-	-
0.8274	1280	0.4546	-	-	-	-	-
0.8403	1300	0.4724	-	-	-	-	-
0.8533	1320	0.4194	-	-	-	-	-
0.8662	1340	0.4352	-	-	-	-	-
0.8791	1360	0.3926	-	-	-	-	-
0.8920	1380	0.397	-	-	-	-	-
0.9050	1400	0.4157	0.2206	0.3558	0.1785	0.2495	0.2613
0.9179	1420	0.4426	-	-	-	-	-
0.9308	1440	0.4077	-	-	-	-	-
0.9438	1460	0.4227	-	-	-	-	-
0.9567	1480	0.4184	-	-	-	-	-
0.9696	1500	0.4838	-	-	-	-	-
0.9825	1520	0.4991	-	-	-	-	-
0.9955	1540	0.3889	-	-	-	-	-
-1	-1	-	-	0.3571	0.1825	0.2484	0.2627

Environmental Impact

Carbon emissions were measured using CodeCarbon.

• Energy Consumed: 0.027 kWh
• Carbon Emitted: 0.011 kg of CO2
• Hours Used: 0.082 hours

Training Hardware

• On Cloud: No
• GPU Model: 1 x NVIDIA GeForce RTX 3090
• CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
• RAM Size: 31.78 GB

Framework Versions

• Python: 3.11.6
• Sentence Transformers: 4.2.0.dev0
• Transformers: 4.49.0
• PyTorch: 2.6.0+cu124
• Accelerate: 1.5.1
• Datasets: 2.21.0
• Tokenizers: 0.21.1

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",
}

SpladeLoss

@misc{formal2022distillationhardnegativesampling,
      title={From Distillation to Hard Negative Sampling: Making Sparse Neural IR Models More Effective},
      author={Thibault Formal and Carlos Lassance and Benjamin Piwowarski and Stéphane Clinchant},
      year={2022},
      eprint={2205.04733},
      archivePrefix={arXiv},
      primaryClass={cs.IR},
      url={https://arxiv.org/abs/2205.04733},
}