pq_cache_4 / checkpoint-2240 /README.md

1shoomun

Updated Weights

9748d38 verified 8 months ago

preview code

raw

history blame contribute delete

18.8 kB

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:3570
  - loss:MultipleNegativesRankingLoss
  - loss:CosineSimilarityLoss
base_model: jinaai/jina-embedding-b-en-v1
widget:
  - source_sentence: How do I change my stocks to mutual funds?
    sentences:
      - How can I swap my stocks for mutual funds?
      - Show my stocks
      - What are the profits I have gained in my portfolio
  - source_sentence: What percentage of my investments are in large cap?
    sentences:
      - Show some of my best performing holdings
      - Suggest recommendations for me
      - Can you show what percentage of my portfolio consists of large cap
  - source_sentence: How do I change my risk profile?
    sentences:
      - What can I do to bring down the volatility in my portfolio?
      - I want to change my risk profile
      - What is the total value of my portfolio
  - source_sentence: >-
      Is now a good time to buy energy stocks considering the war in the Middle
      East and rising fuel prices?
    sentences:
      - Am I investing in the small cap market more?
      - >-
        I saw in the news that there is a war going on in the Middle East and
        fuel will be more costly now, should I buy energy sector stocks?
      - Are my ETFs giving better returns compare to my mutual funds?
  - source_sentence: Look for funds that fit my stock holdings
    sentences:
      - Can you tell me if my investments will grow well in the long run?
      - Do I have any stocks in my portfolio?
      - Explore funds that match my stock portfolio
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
model-index:
  - name: SentenceTransformer based on jinaai/jina-embedding-b-en-v1
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: test eval
          type: test-eval
        metrics:
          - type: cosine_accuracy@1
            value: 0.8659217877094972
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9916201117318436
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.9972067039106145
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.8659217877094972
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.33054003724394787
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.1994413407821229
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09999999999999999
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.8659217877094972
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9916201117318436
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.9972067039106145
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9460695277624867
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9273743016759775
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9273743016759777
            name: Cosine Map@100

SentenceTransformer based on jinaai/jina-embedding-b-en-v1

This is a sentence-transformers model finetuned from jinaai/jina-embedding-b-en-v1. 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: jinaai/jina-embedding-b-en-v1
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity

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': False}) with Transformer model: T5EncoderModel 
  (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:

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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Look for funds that fit my stock holdings',
    'Explore funds that match my stock portfolio',
    'Can you tell me if my investments will grow well in the long run?',
]
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: test-eval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8659
cosine_accuracy@3	0.9916
cosine_accuracy@5	0.9972
cosine_accuracy@10	1.0
cosine_precision@1	0.8659
cosine_precision@3	0.3305
cosine_precision@5	0.1994
cosine_precision@10	0.1
cosine_recall@1	0.8659
cosine_recall@3	0.9916
cosine_recall@5	0.9972
cosine_recall@10	1.0
cosine_ndcg@10	0.9461
cosine_mrr@10	0.9274
cosine_map@100	0.9274

Training Details

Training Datasets

Unnamed Dataset

Size: 1,785 training samples
Columns: sentence_0, sentence_1, and label
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 label
type string string float
details
min: 4 tokens
mean: 11.4 tokens
max: 26 tokens

min: 4 tokens
mean: 10.11 tokens
max: 33 tokens

min: 1.0
mean: 1.0
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 4 tokens mean: 11.4 tokens max: 26 tokens	min: 4 tokens mean: 10.11 tokens max: 33 tokens	min: 1.0 mean: 1.0 max: 1.0

Samples:

sentence_0	sentence_1	label
`How can I lower the risk in my investments?`	`How to reduce my risk`	`1.0`
`How is my asset allocation divided?`	`What is my asset allocation breakdown?`	`1.0`
`Any specific swap recommendations for better returns?`	`What are the specific swap suggestions to improve my returns?`	`1.0`

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Unnamed Dataset

Size: 1,785 training samples
Columns: sentence_0, sentence_1, and label
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 label
type string string float
details
min: 4 tokens
mean: 11.28 tokens
max: 26 tokens

min: 4 tokens
mean: 9.98 tokens
max: 33 tokens

min: 1.0
mean: 1.0
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 4 tokens mean: 11.28 tokens max: 26 tokens	min: 4 tokens mean: 9.98 tokens max: 33 tokens	min: 1.0 mean: 1.0 max: 1.0

Samples:

sentence_0	sentence_1	label
`What should I do to improve my investment returns?`	`How can I improve my returns?`	`1.0`
`Can you give me an overview of my portfolio?`	`Do you have any insights on my portfolio`	`1.0`
`Reveal my stock assets`	`Show my stocks`	`1.0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
num_train_epochs: 20
multi_dataset_batch_sampler: round_robin

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
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1
num_train_epochs: 20
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: False
fp16: False
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}
tp_size: 0
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
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: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	test-eval_cosine_ndcg@10
1.0	112	-	0.9013
2.0	224	-	0.9112
3.0	336	-	0.9250
4.0	448	-	0.9307
4.4643	500	0.1949	0.9337
5.0	560	-	0.9342
6.0	672	-	0.9381
7.0	784	-	0.9423
8.0	896	-	0.9426
8.9286	1000	0.1347	0.9452
9.0	1008	-	0.9442
10.0	1120	-	0.9461
11.0	1232	-	0.9461
12.0	1344	-	0.9461
13.0	1456	-	0.9461
13.3929	1500	0.1193	0.9461
14.0	1568	-	0.9461
15.0	1680	-	0.9461
16.0	1792	-	0.9461
17.0	1904	-	0.9461
17.8571	2000	0.117	0.9461
18.0	2016	-	0.9461
19.0	2128	-	0.9461

Framework Versions

Python: 3.10.16
Sentence Transformers: 4.1.0
Transformers: 4.51.3
PyTorch: 2.7.0
Accelerate: 1.6.0
Datasets: 3.5.0
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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}