Add new SentenceTransformer model.

fbf1aae verified over 1 year ago

69.3 kB

base_model: BAAI/bge-base-en-v1.5
datasets: []
language: []
library_name: sentence-transformers
metrics:
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@5
  - cosine_ndcg@10
  - cosine_mrr@5
  - cosine_mrr@10
  - cosine_map@5
  - cosine_map@10
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - loss:CoSENTLoss
  - dataset_size:102962
  - loss:WeightedDenoisingAutoEncoderLoss
  - loss:WeightedMultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      , antenna, or other sensor to attain mission performance levels that

      currently cannot be achieved by a monolithic satellite. Most aspects of
      this concept have been widely studied, but

      the first implementation has yet to be realized, with the exception of a
      few initial experiments.

      A distributed satellite system taxonomy is shown in Fig. 1 with a
      discussion of current and planned systems to

      follow. At the end of this section, a candidate distributed space mission
      is presented as a common reference for

      Table 1 presents a selection of current distributed satellite systems,
      grouped in the four typical mission

      categories
    sentences:
      - >+
        What are the peaks that appear on waterfall plots but not on zero speed
        curves?

      - >+
        What is the main challenge in implementing a distributed satellite
        system?

      - >+
        What are the remaining challenges that need to be addressed for the
        successful implementation of optical links?

  - source_sentence: >-
      :250,000 scale for regional context) . Near-term efforts should focus on
      high-priority locations .

      [16] Terrain hazard (e .g ., slope, surface roughness), line-of-sight (i
      .e ., viewshed), and time-dependent

      illumination maps at appropriate scales (e .g ., best-available supported
      by the data) are high-priority derived products essential in mission
      planning, and they should be made available as soon as possible .

      [17] South polar data products could be initially controlled to coarser
      data and known surface reference points to support early Artemis missions
      and other surface activities, but establishment of a local control network
      applied to all necessary data layers would facilitate interoperability and
      provide more precision for specific sites .

      Higher-order data products are tied to controlled foundational data and
      are derived from source data, such as measurements of elemental abundance,
      temperature or reflectance at multiple wavelengths, observations of solar
      illumination, and output from space weather models . Higher-order data
      products derived from these source data will play an essential role in
      planning and executing south polar missions . Planning the science
      activities to be carried out on the lunar surface will be based on these
      higher-order data products, and, in turn, the science returned by those
      activities will be used to update those same products . For example,
      geologic maps based on remotely sensed data prior to early Artemis
      landings will be a likely outcome of site assessments and will form the
      critical basis for traverse plans and planning of science tasks . The
      observations, samples, and measurements made during Artemis surface
      activities will feed back into updating the geologic maps, to the benefit
      of future crewed or robotic missions to the same area . Similarly,
      resource maps will drive the selection of landing sites for missions
      focused on resource discovery, characterization, and utilization, and the
      findings of those missions will be used to iteratively update the resource
      maps . In these cases, and others
    sentences:
      - >+
        Who are the authors of the NASA document "Space Radiation Cancer Risk
        Projections for Explorative Missions: Uncertainty Reduction and
        Mitigation"?

      - >+
        What is the aggregate data rate of the outputs of the 7-band CCD-in-CMOS
        TDI sensor?

      - >+
        What are the essential derived products in mission planning, and why are
        they crucial for south polar missions?

  - source_sentence: As LisR as a demonstrator mission under development
    sentences:
      - ' As LisR only serves as a technology demonstrator, the follow-up mission HiVE is already under development'
      - ' Schematic of the 3 grid extraction system in an ion gridded thruster showing one ion beamlet and the corresponding axial potential proﬁle (not to scale)'
      - |2-
         The diagram of PUC is shown as follows:
        The propellent type is Sulfur Dioxide (SO2)
  - source_sentence: >-
      Conclusion This provides formation processes application the impact on,
      small moon of the Didymos binary
    sentences:
      - >2-
         The mission network model,
        parameters, commodity demand and supply used in this case study are
        presented in Fig
      - >2-
         Conclusion
        This paper provides an overview of ejecta formation and evolution
        processes with speciﬁc application to the hypervelo- city impact of the
        DART spacecraft on Dimorphos, the small moon of the Didymos binary
        asteroid system
      - >-

        vantage points from NASA and non-NASA sources, including in orbit,
        airborne and even in-situ sensors to create a more dynamic and complete
        picture of a natural physical process
  - source_sentence: Table of
    sentences:
      - |-

        [29] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun
      - ' Nevertheless, both formulations used in this work allow enough flexibility to adapt them to the most common mission requirements, while still being able to reduce the searching space for the optimization process'
      - ' Table 4 oﬀers a description of the selected FoM'
model-index:
  - name: SentenceTransformer based on BAAI/bge-base-en-v1.5
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@5
            value: 0.8196517412935324
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8606965174129353
            name: Cosine Accuracy@10
          - type: cosine_precision@5
            value: 0.16393034825870648
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08606965174129352
            name: Cosine Precision@10
          - type: cosine_recall@5
            value: 0.8196517412935324
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8606965174129353
            name: Cosine Recall@10
          - type: cosine_ndcg@5
            value: 0.7116804364524553
            name: Cosine Ndcg@5
          - type: cosine_ndcg@10
            value: 0.7249564355877233
            name: Cosine Ndcg@10
          - type: cosine_mrr@5
            value: 0.6753316749585401
            name: Cosine Mrr@5
          - type: cosine_mrr@10
            value: 0.6808181315643997
            name: Cosine Mrr@10
          - type: cosine_map@5
            value: 0.6753316749585406
            name: Cosine Map@5
          - type: cosine_map@10
            value: 0.6808181315644002
            name: Cosine Map@10
          - type: cosine_accuracy@5
            value: 0.8685897435897436
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.9134615384615384
            name: Cosine Accuracy@10
          - type: cosine_precision@5
            value: 0.1737179487179487
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09134615384615384
            name: Cosine Precision@10
          - type: cosine_recall@5
            value: 0.8685897435897436
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9134615384615384
            name: Cosine Recall@10
          - type: cosine_ndcg@5
            value: 0.7286906150877144
            name: Cosine Ndcg@5
          - type: cosine_ndcg@10
            value: 0.7431205481598633
            name: Cosine Ndcg@10
          - type: cosine_mrr@5
            value: 0.6814636752136752
            name: Cosine Mrr@5
          - type: cosine_mrr@10
            value: 0.6873728123728122
            name: Cosine Mrr@10
          - type: cosine_map@5
            value: 0.6814636752136752
            name: Cosine Map@5
          - type: cosine_map@10
            value: 0.6873728123728124
            name: Cosine Map@10
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@5
            value: 0.8034825870646766
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8470149253731343
            name: Cosine Accuracy@10
          - type: cosine_precision@5
            value: 0.1606965174129353
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08470149253731342
            name: Cosine Precision@10
          - type: cosine_recall@5
            value: 0.8034825870646766
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8470149253731343
            name: Cosine Recall@10
          - type: cosine_ndcg@5
            value: 0.6967322721990336
            name: Cosine Ndcg@5
          - type: cosine_ndcg@10
            value: 0.7108049353049597
            name: Cosine Ndcg@10
          - type: cosine_mrr@5
            value: 0.6608001658374787
            name: Cosine Mrr@5
          - type: cosine_mrr@10
            value: 0.6666044776119396
            name: Cosine Mrr@10
          - type: cosine_map@5
            value: 0.6608001658374792
            name: Cosine Map@5
          - type: cosine_map@10
            value: 0.6666044776119403
            name: Cosine Map@10
          - type: cosine_accuracy@5
            value: 0.8557692307692307
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.907051282051282
            name: Cosine Accuracy@10
          - type: cosine_precision@5
            value: 0.17115384615384616
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0907051282051282
            name: Cosine Precision@10
          - type: cosine_recall@5
            value: 0.8557692307692307
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.907051282051282
            name: Cosine Recall@10
          - type: cosine_ndcg@5
            value: 0.7241884927554256
            name: Cosine Ndcg@5
          - type: cosine_ndcg@10
            value: 0.7406789864779515
            name: Cosine Ndcg@10
          - type: cosine_mrr@5
            value: 0.6800747863247864
            name: Cosine Mrr@5
          - type: cosine_mrr@10
            value: 0.6868208180708181
            name: Cosine Mrr@10
          - type: cosine_map@5
            value: 0.6800747863247864
            name: Cosine Map@5
          - type: cosine_map@10
            value: 0.6868208180708181
            name: Cosine Map@10
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@5
            value: 0.777363184079602
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8258706467661692
            name: Cosine Accuracy@10
          - type: cosine_precision@5
            value: 0.15547263681592036
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08258706467661689
            name: Cosine Precision@10
          - type: cosine_recall@5
            value: 0.777363184079602
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8258706467661692
            name: Cosine Recall@10
          - type: cosine_ndcg@5
            value: 0.6732366988651133
            name: Cosine Ndcg@5
          - type: cosine_ndcg@10
            value: 0.6890994908635195
            name: Cosine Ndcg@10
          - type: cosine_mrr@5
            value: 0.638246268656716
            name: Cosine Mrr@5
          - type: cosine_mrr@10
            value: 0.6448970425649527
            name: Cosine Mrr@10
          - type: cosine_map@5
            value: 0.6382462686567164
            name: Cosine Map@5
          - type: cosine_map@10
            value: 0.644897042564953
            name: Cosine Map@10
          - type: cosine_accuracy@5
            value: 0.842948717948718
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8814102564102564
            name: Cosine Accuracy@10
          - type: cosine_precision@5
            value: 0.16858974358974357
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08814102564102565
            name: Cosine Precision@10
          - type: cosine_recall@5
            value: 0.842948717948718
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8814102564102564
            name: Cosine Recall@10
          - type: cosine_ndcg@5
            value: 0.7221379927354293
            name: Cosine Ndcg@5
          - type: cosine_ndcg@10
            value: 0.7350654713813302
            name: Cosine Ndcg@10
          - type: cosine_mrr@5
            value: 0.6817307692307693
            name: Cosine Mrr@5
          - type: cosine_mrr@10
            value: 0.6873613654863654
            name: Cosine Mrr@10
          - type: cosine_map@5
            value: 0.6817307692307691
            name: Cosine Map@5
          - type: cosine_map@10
            value: 0.6873613654863655
            name: Cosine Map@10

SentenceTransformer based on BAAI/bge-base-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5 on the tsdae and sup datasets. 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 Datasets:
- tsdae
- sup

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("federicovolponi/BAAI-bge-base-en-v1.5-space-multitask-tsdae")
# Run inference
sentences = [
    'Table of',
    ' Table 4 oﬀers a description of the selected FoM',
    '\n[29] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun',
]
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@5	0.8197
cosine_accuracy@10	0.8607
cosine_precision@5	0.1639
cosine_precision@10	0.0861
cosine_recall@5	0.8197
cosine_recall@10	0.8607
cosine_ndcg@5	0.7117
cosine_ndcg@10	0.725
cosine_mrr@5	0.6753
cosine_mrr@10	0.6808
cosine_map@5	0.6753
cosine_map@10	0.6808

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@5	0.8035
cosine_accuracy@10	0.847
cosine_precision@5	0.1607
cosine_precision@10	0.0847
cosine_recall@5	0.8035
cosine_recall@10	0.847
cosine_ndcg@5	0.6967
cosine_ndcg@10	0.7108
cosine_mrr@5	0.6608
cosine_mrr@10	0.6666
cosine_map@5	0.6608
cosine_map@10	0.6666

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@5	0.7774
cosine_accuracy@10	0.8259
cosine_precision@5	0.1555
cosine_precision@10	0.0826
cosine_recall@5	0.7774
cosine_recall@10	0.8259
cosine_ndcg@5	0.6732
cosine_ndcg@10	0.6891
cosine_mrr@5	0.6382
cosine_mrr@10	0.6449
cosine_map@5	0.6382
cosine_map@10	0.6449

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@5	0.8686
cosine_accuracy@10	0.9135
cosine_precision@5	0.1737
cosine_precision@10	0.0913
cosine_recall@5	0.8686
cosine_recall@10	0.9135
cosine_ndcg@5	0.7287
cosine_ndcg@10	0.7431
cosine_mrr@5	0.6815
cosine_mrr@10	0.6874
cosine_map@5	0.6815
cosine_map@10	0.6874

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@5	0.8558
cosine_accuracy@10	0.9071
cosine_precision@5	0.1712
cosine_precision@10	0.0907
cosine_recall@5	0.8558
cosine_recall@10	0.9071
cosine_ndcg@5	0.7242
cosine_ndcg@10	0.7407
cosine_mrr@5	0.6801
cosine_mrr@10	0.6868
cosine_map@5	0.6801
cosine_map@10	0.6868

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@5	0.8429
cosine_accuracy@10	0.8814
cosine_precision@5	0.1686
cosine_precision@10	0.0881
cosine_recall@5	0.8429
cosine_recall@10	0.8814
cosine_ndcg@5	0.7221
cosine_ndcg@10	0.7351
cosine_mrr@5	0.6817
cosine_mrr@10	0.6874
cosine_map@5	0.6817
cosine_map@10	0.6874

Training Details

Training Datasets

tsdae

Dataset: tsdae
Size: 95,730 training samples
Columns: damaged_sentence and orginal_sentence
Approximate statistics based on the first 1000 samples:
damaged_sentence orginal_sentence
type string string
details
min: 3 tokens
mean: 13.28 tokens
max: 174 tokens

min: 6 tokens
mean: 30.02 tokens
max: 374 tokens

	damaged_sentence	orginal_sentence
type	string	string
details	min: 3 tokens mean: 13.28 tokens max: 174 tokens	min: 6 tokens mean: 30.02 tokens max: 374 tokens

Samples:

damaged_sentence	orginal_sentence
`, the described above allows continue this`	`However, the modularization into functional units described above allows to continue this idea and form a well-defined functional hierarchy`
`Solar scientific military and the stage for Change mission technology improvements—continued advances in will mass/volume`	`Solar sails can perform unique scientific, commercial, and military missions, and the stage is set for near-term UPGRADE/REPLACE PAYLOADS • Change of mission • Take advantage of technology improvements—continued advances in electronics will cause payload components to shrink in mass/volume, while capabilities increase`
`4mm Hexcell 5052 aluminum honeycomb with 1`	`4mm thick Hexcell 5052 alloy hexagonal aluminum honeycomb with 1`

Loss: losses.WeightedDenoisingAutoEncoderLoss

sup

Dataset: sup
Size: 7,232 training samples
Columns: positive and anchor
Approximate statistics based on the first 1000 samples:
positive anchor
type string string
details
min: 5 tokens
mean: 354.69 tokens
max: 512 tokens

min: 9 tokens
mean: 19.21 tokens
max: 40 tokens

	positive	anchor
type	string	string
details	min: 5 tokens mean: 354.69 tokens max: 512 tokens	min: 9 tokens mean: 19.21 tokens max: 40 tokens

Samples:

positive	anchor
, using diverse software or hardware designs may double design and veriﬁcation costs due to having to build two different components for the same functionality. Hence, although DCLS execution also halves performance efﬁciency (the corresponding functionality is executed twice), it allows reusing the same design (e.g. the same core design) for the primary and the redundant paths (e.g. with staggered execution), thus containing design and veriﬁcation costs. Redundancy can be applied at different granularities accord- ing to the sphere of replication (SoR). Choosing the right SoR depends on several tradeoffs like area overheads, re- design costs, fault detection time, and overall system costs. In the context of DCLS, the SoR is placed at the level of the CPU (core), as done for the AURIX processors. This requires including two replicas of the same core and compare their memory transactions, which requires roughly duplicating com- putational resources in the chip and being able to ensure that replicas can provide independent behavior. On the other hand, storage (memories, caches) and communication means (buses, crossbars) do not need to be fully replicated and can build upon Error Correction Codes (ECC) and Cyclic Redundancy Check (CRC) as a form of lightweight redundancy with diversity. HPC ASIL-D capable platforms typically combine a low- performance microcontroller amenable for the automotive do- main (i.e. ASIL-D capable) and an HPC accelerator deliv- ering high computation throughput, but whose adherence to ISO26262 requirements is unknown, so its appropriate use for ASIL-C/D systems needs to be investigated. Without loss of generality, we consider an NVIDIA GPU accelerator, thus analogous to those in NVIDIA Drive and Xavier families for the automotive domain. However, the ﬁndings in this paper can easily be extrapolated to other products. Software faults and some hardware faults are regarded as systematic, and it must be proven that their failure risk is residual. However, random hardware faults cannot be avoided, and means are required to prevent them from causing hazards. Those faults can be caused by, for example, voltage droops	`What are the advantages of using the same design for the primary and redundant paths in DCLS execution?`
: First, the TT&C spectrum requirements of the new satellites shall be assessed. Second, the utilization of existing TT&C frequency allocations and their potential to incorporate the future number of satellites is studied. Only for the case that this study results in the need for new spectrum, the study groups were asked to investigate new potential TT&C frequency allocations in the frequency ranges 150.05-174 MHz and 400.15-420 MHz. The studies shall be completed for WRC-19. This paper presents the intermediate results of the study groups. A study of the spectrum requirements of small satellites has been completed. The required spectrum for TT&C is expected to be less than 2.5 MHz for downlink and less than 1 MHz for uplink. Consequently, the study groups conducted sharing studies in various bands which will be summarized and evaluated from a satellite developer’s perspective. After the Cubesat design standard was introduced in 1999 and first satellites of this new class have been launched in the subsequent years, small satellites have become increasingly popular in the past five years. Today not only universities use small satellite platforms for education and technology demonstration, but also commercial operators started to develop and deploy satellites with masses of typically less than 50 kg and reasonably short development times. Currently more than hundred new satellites are currently launched into space per year. The increase of launches was recognized by the International Telecommunication Union (ITU) which is responsible for the coordination of the shared use of frequencies. As the first Cubesats were mainly launched by new entrants into the space sector, mandatory regulatory procedures like frequency coordination were omitted or underestimated by the developers. Additionally, the new developers complaint that the existing regulatory procedures are too complicated and time-consuming for satellites with short development times. The ITU therefore decided at the WRC-12 to study the characteristics of picosatellites and nanosatellites and their current practice in filing satellites to the ITU. The studies were concluded in 2015 with two reports on the characteristics [1] and current filing practice [2]. In these reports it was identified that the characteristics that define small satellites (low mass, small dimensions, low power, …) are not relevant from a frequency coordination perspective and that the short development times are still long enough to properly file the systems to the ITU. As a result	`What are the spectrum requirements for TT&C of small satellites?`
:287–299, Dec 2019. [20] Tam´as Vink´o and Dario Izzo. Global optimi- sation heuristics and test problems for prelimi- nary spacecraft trajectory design. Technical re- port, 2008. [21] Matej Petkovic, Luke Lucas, Dragi Kocev, Saˇso Dˇzeroski, Redouane Boumghar, and Nikola Simidjievski. Quantifying the effects of gyro- less flying of the mars express spacecraft with machine learning. In 2019 IEEE International [22] Janhavi H. Borse, Dipti D. Patil, Vinod Kumar, and Sudhir Kumar. Soft landing parameter measurements for candidate navigation trajec- tories using deep learning and ai-enabled plan- etary descent. Mathematical Problems in Engi- neering, 2022	`What are some of the research topics and methods explored in the provided references?`

Loss: losses.WeightedMultipleNegativesRankingLoss with these parameters:
```
{
    "scale": 20,
    "similarity_fct": "cos_sim"
}
```

Evaluation Datasets

tsdae

Dataset: tsdae
Size: 10,637 evaluation samples
Columns: damaged_sentence and orginal_sentence
Approximate statistics based on the first 1000 samples:
damaged_sentence orginal_sentence
type string string
details
min: 3 tokens
mean: 13.52 tokens
max: 182 tokens

min: 5 tokens
mean: 30.74 tokens
max: 452 tokens

	damaged_sentence	orginal_sentence
type	string	string
details	min: 3 tokens mean: 13.52 tokens max: 182 tokens	min: 5 tokens mean: 30.74 tokens max: 452 tokens

Samples:

damaged_sentence	orginal_sentence
`from providing student licenses the OirthirSAT team`	`The authors thank Michael Doherty from Ansys for providing student licenses for STK to the OirthirSAT team`
`at 205`	`4 as observed by TROPICS Pathfinder at 205 GHz`
`this reason of chemistry needed to radiative heating`	`For this reason, careful reexaminations of the chemistry models are needed to reduce the uncertainties in the radiative heating`

Loss: losses.WeightedDenoisingAutoEncoderLoss

sup

Dataset: sup
Size: 804 evaluation samples
Columns: positive and anchor
Approximate statistics based on the first 1000 samples:
positive anchor
type string string
details
min: 4 tokens
mean: 351.15 tokens
max: 512 tokens

min: 8 tokens
mean: 19.36 tokens
max: 45 tokens

	positive	anchor
type	string	string
details	min: 4 tokens mean: 351.15 tokens max: 512 tokens	min: 8 tokens mean: 19.36 tokens max: 45 tokens

Samples:

positive	anchor
, the total number of test thermocouples has been rationalized taking into account redundancy needs, accommodation constraints and hardware passivation needs for flight. The test is subdivided into 19 phases (see Figure 12) with two phases before and after the test for the health check functional tests under room conditions. Functional tests demonstrate anomalies such as the PCDU Reset and operational malfunctions of the RAX instrument at its high temperatures. The PCDU Reset anomaly was solved during the test by a software patch and validated during the final hot and cold plateaus. To address the RAX anomaly at hot, various test configurations were simulated using the thermal numerical model during the test to actually perform RAX functional test at an intermediate plateau facilitating mission operational constraints for flight. Data collected from hot and cold thermal balance test phases, as well as the rover OFF transition from hot to cold, are the inputs for correlation activities conducted post-TV/TB test. The thermal numerical model updates mainly focus on conductive couplings	`What was the solution to the PCDU Reset anomaly during the test?`
, where +Z axis orients to the earth, and sun pointing attitude mode during day time orienting -Z axis to the sun. Therefore, attitude control subsystem is required to maneuver the satellite attitude twice per revolution around its pitch axis. Figure 6 shows concept of the attitude maneuverer. Another attitude maneuverer is necessary to perform SAR observation and SAR data download to a to ground station, because X-band transmit antenna is oriented to +Z, so the satellite has to offset its attitude to orient the X-band transmit antenna toward the ground station. 3.4 High pointing accuracy Disturbance torque and system momentum profiles during few revolutions were estimated as shown in Figure 7 and 8. Four micro reaction wheels, which can respond to these profiles were selected which enable attitude maneuvers within a short period of time. In order to perform a pitch attitude maneuver quickly, two wheels are located on pitch axis while one wheel was located on each of the remaining roll and yaw axes. Figure 9 shows the satellite attitudes during SAR observation. There are three kinds of attitude, strip map mode, sliding spot light mode, and spotlight mode. Large change of momentum is required for pitch axis when the satellite is in spotlight mode. However, two pitch reaction wheels do not generate enough momentum to execute spotlight mode. So, sliding spotlight mode was selected for high resolution SAR observation mode instead of spotlight mode, in order to relax the torque and momentum requirements to the pitch wheels. In addition, two pitch Figure 7. Disturbance torque profile Figure 8. System momentum profile reaction wheels are accelerated to plus direction or minus direction by using magnet torque before observation. In order to obtain a high resolution SAR data, high attitude control accuracy is required for spotlight mode observation. To achieve high pointing accuracy against a defined ground target point, the attitude control loop applied feed forward compensation with estimated attitude angle and rate. Figure 10 shows an example of dynamic error during a spotlight mode observation maneuver.[4] Equipment for SAR mission consumes total large power more than 1300W, therefore PCDU has a risk of causing electrical and RF influence to the bus power and signal line. In order to research the system, electrical interface check was performed using bread board model of PCDU, battery	`What is the reason for selecting sliding spotlight mode instead of spotlight mode for high resolution SAR observation?`
, body shape and motion assumptions. Then, ORSAT uses DCA to determine the reentry risk posed to the Earth’s population based on the year of reentry and orbit inclination. It also predicts impact kinetic energy (impact velocity and impact mass) of objects that survive reentry[18]. ORSAT has been in use for the last decade and currently in its 6.0 version. However, unlike DAS, OR- SAT is not readily available. Only personnel at the Johnson Space Center, Orbital Debris Program Oﬃce run ORSAT. ORSAT is limited to ballistic reentry, only tumbling motions or stable orientations of objects are allowed which produce no lift. Partial melting of objects is considered by a demise factor and almost all materials in the database are temperature de- pendent. Heating by oxidation is also considered [20]. Therefore, ORSAT determines when and if a reentry object demises by using integrated trajectory, atmospheric, aerodynamic, aero-thermodynamic, and thermal models as outlined in section 3.1 [17, 18, 20]. Reentry demisability analysis using DAS requires the spacecraft to be deﬁned to the level of each individual hardware part constituting the spacecraft. This step facilitates population of the DAS Spacecraft Deﬁnition Module . Section 3.2.1 illustrates a generic spacecraft subdivision approach that can be followed to itemize the individual parts spacecraft parts. Subsequently, non-demisable parts are identiﬁed before or by the actual reentry analysis as explained in section 3.2.2. Itemization of the demisable spacecraft basic parts can be best approached by decompos- ing the spacecraft according to the Hierarchical System Terminology deﬁned in the NASA Systems Engineering Handbook [14]. Tables 3.2, 3.3 and 3.4 illustrate a generic approach to decompose a spacecraft into basic parts [29, 30, 9] excluding the payload. Description of the speciﬁc product for the basic part identiﬁed completes the process. Though slight vari- ations are likely to occur in the decomposition of diﬀerent missions, the Generic Spacecraft Subsystems Hierarchical Subdivision approach is robust, hence	`What is the limitation of ORSAT in terms of object motion?`

Loss: losses.WeightedMultipleNegativesRankingLoss with these parameters:
```
{
    "scale": 20,
    "similarity_fct": "cos_sim"
}
```

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
learning_rate: 3e-06
weight_decay: 0.001
num_train_epochs: 6
bf16: True
tf32: False
load_best_model_at_end: 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: 32
per_device_eval_batch_size: 32
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
learning_rate: 3e-06
weight_decay: 0.001
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 6
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
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: False
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
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	sup loss	tsdae loss	dim_256_cosine_map@10	dim_512_cosine_map@10	dim_768_cosine_map@10
0.0311	100	0.1372	-	-	-	-	-
0.0622	200	0.1061	-	-	-	-	-
0.0932	300	0.1161	-	-	-	-	-
0.1243	400	0.0881	-	-	-	-	-
0.1554	500	0.0878	0.2867	0.0724	0.6238	0.6501	0.6502
0.1865	600	0.0929	-	-	-	-	-
0.2175	700	0.0979	-	-	-	-	-
0.2486	800	0.0902	-	-	-	-	-
0.2797	900	0.0755	-	-	-	-	-
0.3108	1000	0.0885	0.2262	0.0714	0.6380	0.6669	0.6639
0.3418	1100	0.0854	-	-	-	-	-
0.3729	1200	0.0975	-	-	-	-	-
0.4040	1300	0.1104	-	-	-	-	-
0.4351	1400	0.0829	-	-	-	-	-
0.4661	1500	0.0846	0.1949	0.0710	0.6529	0.6803	0.6765
0.4972	1600	0.0821	-	-	-	-	-
0.5283	1700	0.0892	-	-	-	-	-
0.5594	1800	0.0859	-	-	-	-	-
0.5904	1900	0.0936	-	-	-	-	-
0.6215	2000	0.0829	0.1703	0.0706	0.6579	0.6837	0.6851
0.6526	2100	0.0972	-	-	-	-	-
0.6837	2200	0.0797	-	-	-	-	-
0.7147	2300	0.0868	-	-	-	-	-
0.7458	2400	0.0781	-	-	-	-	-
0.7769	2500	0.0837	0.1588	0.0704	0.6633	0.7016	0.6915
0.8080	2600	0.0778	-	-	-	-	-
0.8390	2700	0.0873	-	-	-	-	-
0.8701	2800	0.086	-	-	-	-	-
0.9012	2900	0.0832	-	-	-	-	-
0.9323	3000	0.0931	0.1502	0.0697	0.6733	0.6951	0.6927
0.9633	3100	0.0891	-	-	-	-	-
0.9944	3200	0.0787	-	-	-	-	-
1.0255	3300	0.0843	-	-	-	-	-
1.0566	3400	0.0705	-	-	-	-	-
1.0876	3500	0.0808	0.1484	0.0686	0.6782	0.6880	0.6824
1.1187	3600	0.0754	-	-	-	-	-
1.1498	3700	0.0714	-	-	-	-	-
1.1809	3800	0.0734	-	-	-	-	-
1.2119	3900	0.0732	-	-	-	-	-
1.2430	4000	0.0702	0.1508	0.0679	0.6674	0.6803	0.6770
1.2741	4100	0.0712	-	-	-	-	-
1.3052	4200	0.0719	-	-	-	-	-
1.3362	4300	0.0744	-	-	-	-	-
1.3673	4400	0.0796	-	-	-	-	-
1.3984	4500	0.0823	0.1377	0.0673	0.6677	0.6872	0.6835
1.4295	4600	0.0693	-	-	-	-	-
1.4605	4700	0.0718	-	-	-	-	-
1.4916	4800	0.0726	-	-	-	-	-
1.5227	4900	0.0739	-	-	-	-	-
1.5538	5000	0.0746	0.1366	0.0669	0.6671	0.6900	0.6846
1.5848	5100	0.0757	-	-	-	-	-
1.6159	5200	0.0747	-	-	-	-	-
1.6470	5300	0.0729	-	-	-	-	-
1.6781	5400	0.0747	-	-	-	-	-
1.7091	5500	0.0726	0.1357	0.0666	0.6598	0.6806	0.6904
1.7402	5600	0.0735	-	-	-	-	-
1.7713	5700	0.0709	-	-	-	-	-
1.8024	5800	0.0698	-	-	-	-	-
1.8334	5900	0.0714	-	-	-	-	-
1.8645	6000	0.0732	0.1348	0.0662	0.6729	0.6908	0.6923
1.8956	6100	0.0752	-	-	-	-	-
1.9267	6200	0.0744	-	-	-	-	-
1.9577	6300	0.0775	-	-	-	-	-
1.9888	6400	0.0702	-	-	-	-	-
2.0199	6500	0.0713	0.1311	0.0660	0.6874	0.6868	0.6874

Framework Versions

Python: 3.12.0
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.1+cu118
Accelerate: 0.31.0
Datasets: 2.20.0
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",
}

WeightedDenoisingAutoEncoderLoss

@inproceedings{wang-2021-TSDAE,
    title = "TSDAE: Using Transformer-based Sequential Denoising Auto-Encoderfor Unsupervised Sentence Embedding Learning",
    author = "Wang, Kexin and Reimers, Nils and Gurevych, Iryna", 
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
    month = nov,
    year = "2021",
    address = "Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    pages = "671--688",
    url = "https://arxiv.org/abs/2104.06979",
}

WeightedMultipleNegativesRankingLoss

@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}
}