---
language:
- en
license: apache-2.0
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:50000
- loss:CachedGISTEmbedLoss
base_model: microsoft/mpnet-base
widget:
- source_sentence: what does the accounts receivable turnover measure?
sentences:
- The accounts receivable turnover ratio is an accounting measure used to quantify
a company's effectiveness in collecting its receivables or money owed by clients.
The ratio shows how well a company uses and manages the credit it extends to customers
and how quickly that short-term debt is collected or is paid.
- Capital budgeting, and investment appraisal, is the planning process used to determine
whether an organization's long term investments such as new machinery, replacement
of machinery, new plants, new products, and research development projects are
worth the funding of cash through the firm's capitalization structure ( ...
- The accounts receivable turnover ratio is an accounting measure used to quantify
a company's effectiveness in collecting its receivables or money owed by clients.
The ratio shows how well a company uses and manages the credit it extends to customers
and how quickly that short-term debt is collected or is paid.
- source_sentence: does gabapentin cause liver problems?
sentences:
- Gabapentin has no appreciable liver metabolism, yet, suspected cases of gabapentin-induced
hepatotoxicity have been reported. Per literature review, two cases of possible
gabapentin-induced liver injury have been reported.
- Strongholds are a type of story mission which only unlocks after enough progression
through the game. There are three Stronghold's during the first section of progression
through The Division 2. You'll need to complete the first two and have reached
level 30 before being able to unlock the final Stronghold.
- The most-common side effects attributed to Gabapentin include mild sedation, ataxia,
and occasional diarrhea. Sedation can be minimized by tapering from a smaller
starting dose to the desired dose. When treating seizures, it is ideal to wean
off the drug to reduce the risk of withdrawal seizures.
- source_sentence: how long should you wait to give blood after eating?
sentences:
- Until the bleeding has stopped it is natural to taste blood or to see traces of
blood in your saliva. You may stop using gauze after the flow stops – usually
around 8 hours after surgery.
- Before donation The first and most important rule—never donate blood on an empty
stomach. “Eat a wholesome meal about 2-3 hours before donating to keep your blood
sugar stable," says Dr Chaturvedi. The timing of the meal is important too. You
need to allow the food to be digested properly before the blood is drawn.
- While grid computing involves virtualizing computing resources to store massive
amounts of data, whereas cloud computing is where an application doesn't access
resources directly, rather it accesses them through a service over the internet.
...
- source_sentence: what is the difference between chicken francese and chicken marsala?
sentences:
- Chicken is the species name, equivalent to our “human.” Rooster is an adult male,
equivalent to “man.” Hen is an adult female, equivalent to “woman.” Cockerel is
a juvenile male, equivalent to “boy/young man.”
- What is 99 kg in pounds? - 99 kg is equal to 218.26 pounds.
- The difference between the two is for Francese, the chicken breast is first dipped
in flour, then into a beaten egg mixture, before being cooked. For piccata, the
chicken is first dipped in egg and then in flour. Both are then simmered in a
lemony butter sauce, but the piccata sauce includes capers.”
- source_sentence: what energy is released when coal is burned?
sentences:
- When coal is burned, it reacts with the oxygen in the air. This chemical reaction
converts the stored solar energy into thermal energy, which is released as heat.
But it also produces carbon dioxide and methane.
- When coal is burned it releases a number of airborne toxins and pollutants. They
include mercury, lead, sulfur dioxide, nitrogen oxides, particulates, and various
other heavy metals.
- Squad Building Challenges allow you to exchange sets of players for coins, packs,
and special items in FUT 20. Each of these challenges come with specific requirements,
such as including players from certain teams. ... Live SBCs are time-limited challenges
which often give out unique, high-rated versions of players.
datasets:
- tomaarsen/gooaq-hard-negatives
pipeline_tag: sentence-similarity
library_name: sentence-transformers
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
co2_eq_emissions:
emissions: 40.414352043491995
energy_consumed: 0.10397258579449552
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: 0.297
hardware_used: 1 x NVIDIA GeForce RTX 3090
model-index:
- name: MPNet base trained on Natural Questions pairs
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoClimateFEVER
type: NanoClimateFEVER
metrics:
- type: cosine_accuracy@1
value: 0.22
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.44
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.5
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.72
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.22
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.16666666666666663
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.11599999999999999
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.092
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.09333333333333332
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.195
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.22666666666666666
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.36733333333333335
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.27399441682859144
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.36331746031746026
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.20383864617106084
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoDBPedia
type: NanoDBPedia
metrics:
- type: cosine_accuracy@1
value: 0.46
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.64
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.76
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.84
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.46
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.4
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.3760000000000001
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.34
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.03065300183409328
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.07824513873584021
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.12190077086725051
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.21649668807903738
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.38922276974007985
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.5768571428571428
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.27830317958127815
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoFEVER
type: NanoFEVER
metrics:
- type: cosine_accuracy@1
value: 0.38
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.54
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.58
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.38
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.18
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.12
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.07200000000000001
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.37
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.52
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.57
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.68
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.5226736410648857
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.4790238095238095
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.4780826341570998
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoFiQA2018
type: NanoFiQA2018
metrics:
- type: cosine_accuracy@1
value: 0.28
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.52
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.58
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.28
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.22
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.16
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.09599999999999997
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.1371904761904762
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.32535714285714284
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.3653571428571428
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.42940476190476184
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.34407947120145826
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.3881666666666667
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.2878001762783797
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoHotpotQA
type: NanoHotpotQA
metrics:
- type: cosine_accuracy@1
value: 0.34
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.58
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.64
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.74
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.34
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.21333333333333332
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.14800000000000002
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.094
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.17
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.32
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.37
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.47
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.3814617400581295
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.4695793650793651
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.30635937490171045
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoMSMARCO
type: NanoMSMARCO
metrics:
- type: cosine_accuracy@1
value: 0.12
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.3
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.56
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.66
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.12
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.1
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.11200000000000002
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.066
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.12
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.3
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.56
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.66
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.3660111210554949
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.27407936507936503
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.2870484235732714
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoNFCorpus
type: NanoNFCorpus
metrics:
- type: cosine_accuracy@1
value: 0.32
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.4
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.44
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.5
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.32
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.21333333333333332
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.196
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.144
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.012173283062756207
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.02038195250132044
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.028711609969173105
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.04001132454412933
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.17348600988460589
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.37288095238095237
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.04905096401443591
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoNQ
type: NanoNQ
metrics:
- type: cosine_accuracy@1
value: 0.16
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.38
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.46
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.58
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.16
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.12666666666666665
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.09200000000000001
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.06000000000000001
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.15
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.36
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.43
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.54
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.34266003995975836
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.2898015873015873
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.29298840458552056
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoQuoraRetrieval
type: NanoQuoraRetrieval
metrics:
- type: cosine_accuracy@1
value: 0.8
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.9
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.92
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.96
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.8
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.3666666666666666
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.244
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.13399999999999998
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.7006666666666667
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.8553333333333333
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.8993333333333333
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.9566666666666666
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.8760206475896655
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.8545238095238095
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.8464432234432234
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoSCIDOCS
type: NanoSCIDOCS
metrics:
- type: cosine_accuracy@1
value: 0.36
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.48
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.54
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.68
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.36
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.25999999999999995
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.21600000000000003
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.15200000000000002
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.07566666666666669
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.16166666666666665
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.22266666666666668
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.31466666666666665
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.2989790025477086
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.44196825396825395
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.23644335943955802
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoArguAna
type: NanoArguAna
metrics:
- type: cosine_accuracy@1
value: 0.18
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.56
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.64
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.84
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.18
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.18666666666666668
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.128
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.08399999999999999
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.18
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.56
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.64
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.84
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.5000949127836057
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.392515873015873
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.4000896669795642
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoSciFact
type: NanoSciFact
metrics:
- type: cosine_accuracy@1
value: 0.36
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.46
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.48
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.62
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.36
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.16666666666666663
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.10400000000000001
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.068
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.325
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.44
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.46
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.605
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.4619142281323308
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.43007936507936506
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.42744465636932816
name: Cosine Map@100
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: NanoTouche2020
type: NanoTouche2020
metrics:
- type: cosine_accuracy@1
value: 0.5306122448979592
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.7142857142857143
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.8571428571428571
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.9591836734693877
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.5306122448979592
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.44217687074829926
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.4122448979591837
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.3530612244897959
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.03881638827876476
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.09899647775241191
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.14503016807403868
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.23921899286976872
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.40436873020254516
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.6520651117589893
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.3151658515102787
name: Cosine Map@100
- task:
type: nano-beir
name: Nano BEIR
dataset:
name: NanoBEIR mean
type: NanoBEIR_mean
metrics:
- type: cosine_accuracy@1
value: 0.3469701726844584
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 0.5303296703296704
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 0.6074725274725276
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 0.7214756671899529
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.3469701726844584
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.23401360544217684
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.18648037676609108
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.13500470957613817
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.18488460123328906
name: Cosine Recall@1
- type: cosine_recall@3
value: 0.32576774706513195
name: Cosine Recall@3
- type: cosine_recall@5
value: 0.3876666429564824
name: Cosine Recall@5
- type: cosine_recall@10
value: 0.48913834108187415
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.4103820562345277
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.4603737509655877
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.3391583508465161
name: Cosine Map@100
---
# MPNet base trained on Natural Questions pairs
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [microsoft/mpnet-base](https://huggingface.co/microsoft/mpnet-base) on the [gooaq-hard-negatives](https://huggingface.co/datasets/tomaarsen/gooaq-hard-negatives) 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:** [microsoft/mpnet-base](https://huggingface.co/microsoft/mpnet-base)
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
- **Training Dataset:**
- [gooaq-hard-negatives](https://huggingface.co/datasets/tomaarsen/gooaq-hard-negatives)
- **Language:** en
- **License:** apache-2.0
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: MPNetModel
(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("tomaarsen/mpnet-base-nq-cgist-triplet-gt")
# Run inference
sentences = [
'what energy is released when coal is burned?',
'When coal is burned, it reacts with the oxygen in the air. This chemical reaction converts the stored solar energy into thermal energy, which is released as heat. But it also produces carbon dioxide and methane.',
'When coal is burned it releases a number of airborne toxins and pollutants. They include mercury, lead, sulfur dioxide, nitrogen oxides, particulates, and various other heavy metals.',
]
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
* Datasets: `NanoClimateFEVER`, `NanoDBPedia`, `NanoFEVER`, `NanoFiQA2018`, `NanoHotpotQA`, `NanoMSMARCO`, `NanoNFCorpus`, `NanoNQ`, `NanoQuoraRetrieval`, `NanoSCIDOCS`, `NanoArguAna`, `NanoSciFact` and `NanoTouche2020`
* Evaluated with [InformationRetrievalEvaluator
](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | NanoClimateFEVER | NanoDBPedia | NanoFEVER | NanoFiQA2018 | NanoHotpotQA | NanoMSMARCO | NanoNFCorpus | NanoNQ | NanoQuoraRetrieval | NanoSCIDOCS | NanoArguAna | NanoSciFact | NanoTouche2020 |
|:--------------------|:-----------------|:------------|:-----------|:-------------|:-------------|:------------|:-------------|:-----------|:-------------------|:------------|:------------|:------------|:---------------|
| cosine_accuracy@1 | 0.22 | 0.46 | 0.38 | 0.28 | 0.34 | 0.12 | 0.32 | 0.16 | 0.8 | 0.36 | 0.18 | 0.36 | 0.5306 |
| cosine_accuracy@3 | 0.44 | 0.64 | 0.54 | 0.5 | 0.58 | 0.3 | 0.4 | 0.38 | 0.9 | 0.48 | 0.56 | 0.46 | 0.7143 |
| cosine_accuracy@5 | 0.5 | 0.76 | 0.58 | 0.52 | 0.64 | 0.56 | 0.44 | 0.46 | 0.92 | 0.54 | 0.64 | 0.48 | 0.8571 |
| cosine_accuracy@10 | 0.72 | 0.84 | 0.7 | 0.58 | 0.74 | 0.66 | 0.5 | 0.58 | 0.96 | 0.68 | 0.84 | 0.62 | 0.9592 |
| cosine_precision@1 | 0.22 | 0.46 | 0.38 | 0.28 | 0.34 | 0.12 | 0.32 | 0.16 | 0.8 | 0.36 | 0.18 | 0.36 | 0.5306 |
| cosine_precision@3 | 0.1667 | 0.4 | 0.18 | 0.22 | 0.2133 | 0.1 | 0.2133 | 0.1267 | 0.3667 | 0.26 | 0.1867 | 0.1667 | 0.4422 |
| cosine_precision@5 | 0.116 | 0.376 | 0.12 | 0.16 | 0.148 | 0.112 | 0.196 | 0.092 | 0.244 | 0.216 | 0.128 | 0.104 | 0.4122 |
| cosine_precision@10 | 0.092 | 0.34 | 0.072 | 0.096 | 0.094 | 0.066 | 0.144 | 0.06 | 0.134 | 0.152 | 0.084 | 0.068 | 0.3531 |
| cosine_recall@1 | 0.0933 | 0.0307 | 0.37 | 0.1372 | 0.17 | 0.12 | 0.0122 | 0.15 | 0.7007 | 0.0757 | 0.18 | 0.325 | 0.0388 |
| cosine_recall@3 | 0.195 | 0.0782 | 0.52 | 0.3254 | 0.32 | 0.3 | 0.0204 | 0.36 | 0.8553 | 0.1617 | 0.56 | 0.44 | 0.099 |
| cosine_recall@5 | 0.2267 | 0.1219 | 0.57 | 0.3654 | 0.37 | 0.56 | 0.0287 | 0.43 | 0.8993 | 0.2227 | 0.64 | 0.46 | 0.145 |
| cosine_recall@10 | 0.3673 | 0.2165 | 0.68 | 0.4294 | 0.47 | 0.66 | 0.04 | 0.54 | 0.9567 | 0.3147 | 0.84 | 0.605 | 0.2392 |
| **cosine_ndcg@10** | **0.274** | **0.3892** | **0.5227** | **0.3441** | **0.3815** | **0.366** | **0.1735** | **0.3427** | **0.876** | **0.299** | **0.5001** | **0.4619** | **0.4044** |
| cosine_mrr@10 | 0.3633 | 0.5769 | 0.479 | 0.3882 | 0.4696 | 0.2741 | 0.3729 | 0.2898 | 0.8545 | 0.442 | 0.3925 | 0.4301 | 0.6521 |
| cosine_map@100 | 0.2038 | 0.2783 | 0.4781 | 0.2878 | 0.3064 | 0.287 | 0.0491 | 0.293 | 0.8464 | 0.2364 | 0.4001 | 0.4274 | 0.3152 |
#### Nano BEIR
* Dataset: `NanoBEIR_mean`
* Evaluated with [NanoBEIREvaluator
](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.NanoBEIREvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.347 |
| cosine_accuracy@3 | 0.5303 |
| cosine_accuracy@5 | 0.6075 |
| cosine_accuracy@10 | 0.7215 |
| cosine_precision@1 | 0.347 |
| cosine_precision@3 | 0.234 |
| cosine_precision@5 | 0.1865 |
| cosine_precision@10 | 0.135 |
| cosine_recall@1 | 0.1849 |
| cosine_recall@3 | 0.3258 |
| cosine_recall@5 | 0.3877 |
| cosine_recall@10 | 0.4891 |
| **cosine_ndcg@10** | **0.4104** |
| cosine_mrr@10 | 0.4604 |
| cosine_map@100 | 0.3392 |
## Training Details
### Training Dataset
#### gooaq-hard-negatives
* Dataset: [gooaq-hard-negatives](https://huggingface.co/datasets/tomaarsen/gooaq-hard-negatives) at [87594a1](https://huggingface.co/datasets/tomaarsen/gooaq-hard-negatives/tree/87594a1e6c58e88b5843afa9da3a97ffd75d01c2)
* Size: 50,000 training samples
* Columns: question
, answer
, and negative
* Approximate statistics based on the first 1000 samples:
| | question | answer | negative |
|:--------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
| type | string | string | string |
| details |
what is the difference between calories from fat and total fat?
| Fat has more than twice as many calories per gram as carbohydrates and proteins. A gram of fat has about 9 calories, while a gram of carbohydrate or protein has about 4 calories. In other words, you could eat twice as much carbohydrates or proteins as fat for the same amount of calories.
| Fat has more than twice as many calories per gram as carbohydrates and proteins. A gram of fat has about 9 calories, while a gram of carbohydrate or protein has about 4 calories. In other words, you could eat twice as much carbohydrates or proteins as fat for the same amount of calories.
|
| what is the difference between return transcript and account transcript?
| A tax return transcript usually meets the needs of lending institutions offering mortgages and student loans. ... Tax Account Transcript - shows basic data such as return type, marital status, adjusted gross income, taxable income and all payment types. It also shows changes made after you filed your original return.
| Trial balance is not a financial statement whereas a balance sheet is a financial statement. Trial balance is solely used for internal purposes whereas a balance sheet is used for purposes other than internal i.e. external. In a trial balance, each and every account is divided into debit (dr.) and credit (cr.)
|
| how long does my dog need to fast before sedation?
| Now, guidelines are aimed towards 6-8 hours before surgery. This pre-op fasting time is much more beneficial for your pets because you have enough food in there to neutralize the stomach acid, preventing it from coming up the esophagus that causes regurgitation under anesthetic.
| Try not to let your pooch rapidly wolf down his/her food! Do not let the dog play or exercise (e.g. go for a walk) for at least two hours after having a meal. Ensure continuous fresh water is available to avoid your pet gulping down a large amount after eating.
|
* Loss: [CachedGISTEmbedLoss
](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
```json
{'guide': SentenceTransformer(
(0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, '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})
(2): Normalize()
), 'temperature': 0.01}
```
### Evaluation Dataset
#### gooaq-hard-negatives
* Dataset: [gooaq-hard-negatives](https://huggingface.co/datasets/tomaarsen/gooaq-hard-negatives) at [87594a1](https://huggingface.co/datasets/tomaarsen/gooaq-hard-negatives/tree/87594a1e6c58e88b5843afa9da3a97ffd75d01c2)
* Size: 10,048,700 evaluation samples
* Columns: question
, answer
, and negative
* Approximate statistics based on the first 1000 samples:
| | question | answer | negative |
|:--------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
| type | string | string | string |
| details | how is height width and length written?
| The Graphics' industry standard is width by height (width x height). Meaning that when you write your measurements, you write them from your point of view, beginning with the width.
| The Graphics' industry standard is width by height (width x height). Meaning that when you write your measurements, you write them from your point of view, beginning with the width. That's important.
|
| what is the difference between pork shoulder and loin?
| All the recipes I've found for pulled pork recommends a shoulder/butt. Shoulders take longer to cook than a loin, because they're tougher. Loins are lean, while shoulders have marbled fat inside.
| They are extracted from the loin, which runs from the hip to the shoulder, and it has a small strip of meat called the tenderloin. Unlike other pork, this pork chop is cut from four major sections, which are the shoulder, also known as the blade chops, ribs chops, loin chops, and the last, which is the sirloin chops.
|
| is the yin yang symbol religious?
| The ubiquitous yin-yang symbol holds its roots in Taoism/Daoism, a Chinese religion and philosophy. The yin, the dark swirl, is associated with shadows, femininity, and the trough of a wave; the yang, the light swirl, represents brightness, passion and growth.
| Yin energy is in the calm colors around you, in the soft music, in the soothing sound of a water fountain, or the relaxing images of water. Yang (active energy) is the feng shui energy expressed in strong, vibrant sounds and colors, bright lights, upward moving energy, tall plants, etc.
|
* Loss: [CachedGISTEmbedLoss
](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
```json
{'guide': SentenceTransformer(
(0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, '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})
(2): Normalize()
), 'temperature': 0.01}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 2048
- `per_device_eval_batch_size`: 2048
- `learning_rate`: 2e-05
- `num_train_epochs`: 1
- `warmup_ratio`: 0.1
- `seed`: 12
- `bf16`: True
#### All Hyperparameters