metadata
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- mteb
datasets:
- ms_marco
model-index:
- name: E:\HuggingFaceDataDownloader\results\finetuned_models\2000\2000_finetune
results:
- task:
type: Classification
dataset:
type: DDSC/angry-tweets
name: MTEB AngryTweetsClassification
config: default
split: test
revision: 20b0e6081892e78179356fada741b7afa381443d
metrics:
- type: accuracy
value: 56.084049665711554
- type: f1
value: 55.198013156852625
- task:
type: BitextMining
dataset:
type: strombergnlp/bornholmsk_parallel
name: MTEB BornholmBitextMining
config: default
split: test
revision: 3bc5cfb4ec514264fe2db5615fac9016f7251552
metrics:
- type: accuracy
value: 47
- type: f1
value: 37.97365079365079
- type: precision
value: 34.48333333333334
- type: recall
value: 47
- task:
type: Classification
dataset:
type: danish_political_comments
name: MTEB DanishPoliticalCommentsClassification
config: default
split: train
revision: edbb03726c04a0efab14fc8c3b8b79e4d420e5a1
metrics:
- type: accuracy
value: 40.88398556758257
- type: f1
value: 37.604524785367076
- task:
type: Classification
dataset:
type: DDSC/lcc
name: MTEB LccSentimentClassification
config: default
split: test
revision: de7ba3406ee55ea2cc52a0a41408fa6aede6d3c6
metrics:
- type: accuracy
value: 59.599999999999994
- type: f1
value: 59.0619246469949
- task:
type: Classification
dataset:
type: strombergnlp/nordic_langid
name: MTEB NordicLangClassification
config: default
split: test
revision: e254179d18ab0165fdb6dbef91178266222bee2a
metrics:
- type: accuracy
value: 61.00333333333333
- type: f1
value: 60.45633325804296
- task:
type: Classification
dataset:
type: ScandEval/scala-da
name: MTEB ScalaDaClassification
config: default
split: test
revision: 1de08520a7b361e92ffa2a2201ebd41942c54675
metrics:
- type: accuracy
value: 50.43457031250001
- type: ap
value: 50.22017546538257
- type: f1
value: 50.03426509926491
e5-dansk-test
This is a sentence-transformers model: It maps sentences & paragraphs to a 1024 dimensional dense vector space and can be used for tasks like clustering or semantic search.
The model was trained by MS-MARCO english dataset machine translated into the danish language to test whether Machine translation high quality datasets to a foreign language produces good results
Usage (Sentence-Transformers)
Using this model becomes easy when you have sentence-transformers installed:
pip install -U sentence-transformers
Then you can use the model like this:
from sentence_transformers import SentenceTransformer
sentences = ["Dette er en dansk sætning", "Dette er en også en dansk sætning"]
model = SentenceTransformer('Jechto/e5-dansk-test-0.1')
embeddings = model.encode(sentences)
print(embeddings)
Training
The model was trained with the parameters:
DataLoader:
sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader of length 10327 with parameters:
{'batch_size': 16}
Loss:
sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss with parameters:
{'scale': 20.0, 'similarity_fct': 'cos_sim'}
Parameters of the fit()-Method:
{
"epochs": 1,
"evaluation_steps": 2000,
"evaluator": "sentence_transformers.evaluation.BinaryClassificationEvaluator.BinaryClassificationEvaluator",
"max_grad_norm": 1,
"optimizer_class": "<class 'torch.optim.adam.Adam'>",
"optimizer_params": {
"lr": 1e-05
},
"scheduler": "warmupconstant",
"steps_per_epoch": null,
"warmup_steps": 10000,
"weight_decay": 0.01
}
Full Model Architecture
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(1): Pooling({'word_embedding_dimension': 1024, '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})
(2): Normalize()
)