Eval_02_Final_5 / README.md

Mr-FineTuner

Fine-tuned paraphrase-MiniLM-L6-v2 for CEFR classification

6702f4c verified 9 months ago

preview code

raw

history blame contribute delete

15.3 kB

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:14356
  - loss:MultipleNegativesRankingLoss
base_model: sentence-transformers/paraphrase-MiniLM-L6-v2
widget:
  - source_sentence: >-
      Pear trees are usually productive for 50 to 75 years though some still
      produce fruit after 100 years .
    sentences:
      - In the late 1950s , he studied cinema in France .
      - >-
        Pear trees are usually productive for 50 to 75 years though some still
        produce fruit after 100 years .
      - >-
        A recording medium is a physical material that holds data expressed in
        any of the existing recording formats .
  - source_sentence: On poor , dry soils there are tropical heathlands .
    sentences:
      - On poor , dry soils there are tropical heathlands .
      - There are plans to build a new library at my school .
      - >-
        These are forest birds that tend to feed on insects at or near the
        ground .
  - source_sentence: >-
      According to Statistics Canada , the county has a total area of 2004.44
      km2 .
    sentences:
      - >-
        In 2018 , there are eleven senators holding ministerial positions and
        the head of state , the First mayor .
      - >-
        According to Statistics Canada , the county has a total area of 2004.44
        km2 .
      - >-
        There are some common ways used to stretch piercings , of different
        origins and useful for different people .
  - source_sentence: Oll , who was married , fell into severe depressions after he divorced .
    sentences:
      - >-
        Tide pools are a home for hardy organisms such as sea stars , mussels
        and clams .
      - >-
        Endgames can be studied according to the types of pieces that remain on
        board .
      - Oll , who was married , fell into severe depressions after he divorced .
  - source_sentence: She often shared her boots with her sister .
    sentences:
      - She often shared her boots with her sister .
      - >-
        Many of the Greek city -states also had a god or goddess associated with
        that city .
      - Until 1 April 2010 the Departments were as follows .
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: SentenceTransformer based on sentence-transformers/paraphrase-MiniLM-L6-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: dev
          type: dev
        metrics:
          - type: pearson_cosine
            value: 0.17226242076011888
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.15567680488974325
            name: Spearman Cosine

SentenceTransformer based on sentence-transformers/paraphrase-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-MiniLM-L6-v2. It maps sentences & paragraphs to a 384-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: sentence-transformers/paraphrase-MiniLM-L6-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 384 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': 128, '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})
)

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("Mr-FineTuner/Eval_02_Final_5")
# Run inference
sentences = [
    'She often shared her boots with her sister .',
    'She often shared her boots with her sister .',
    'Until 1 April 2010 the Departments were as follows .',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

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

Evaluation

Metrics

Semantic Similarity

Dataset: dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.1723
spearman_cosine	0.1557

Training Details

Training Dataset

Unnamed Dataset

Size: 14,356 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: 7 tokens
mean: 18.64 tokens
max: 42 tokens

min: 7 tokens
mean: 18.64 tokens
max: 42 tokens

min: 1.0
mean: 3.32
max: 6.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 7 tokens mean: 18.64 tokens max: 42 tokens	min: 7 tokens mean: 18.64 tokens max: 42 tokens	min: 1.0 mean: 3.32 max: 6.0

Samples:

sentence_0	sentence_1	label
`Construction of the temple complex started in approximately 1264 BC and lasted for about 20 years , until 1244 BC .`	`Construction of the temple complex started in approximately 1264 BC and lasted for about 20 years , until 1244 BC .`	`3.0`
`He knew which bag to buy for his older sister 's birthday .`	`He knew which bag to buy for his older sister 's birthday .`	`3.0`
`The precise origin of absinthe is unclear .`	`The precise origin of absinthe is unclear .`	`4.0`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
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: 16
per_device_eval_batch_size: 16
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: 3
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}
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: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	dev cosine similarity loss	dev_spearman_cosine
0.5568	500	0.0017	0.4277	0.1245
1.0	898	-	0.4394	0.1433
1.1136	1000	0.001	0.4388	0.1568
1.6704	1500	0.0006	0.4518	0.1503
2.0	1796	-	0.4394	0.1487
2.2272	2000	0.0008	0.4454	0.1471
2.7840	2500	0.0009	0.4532	0.1542
3.0	2694	-	0.4534	0.1557

Framework Versions

Python: 3.11.11
Sentence Transformers: 3.4.1
Transformers: 4.48.1
PyTorch: 2.5.1+cu124
Accelerate: 1.3.0
Datasets: 3.2.0
Tokenizers: 0.21.0

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}
}