SentenceTransformer based on distilbert/distilroberta-base

This is a sentence-transformers model finetuned from distilbert/distilroberta-base on the qqp_triplets 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: distilbert/distilroberta-base
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- qqp_triplets
Language: en

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: RobertaModel 
  (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("ravi259/distilroberta-base-sentence-transformer_finetuned")
# Run inference
sentences = [
    'A dog is in the water.',
    'Wet brown dog swims towards camera.',
    'The dog is rolling around in the grass.',
]
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

Triplet

Datasets: all-nli-dev and all-nli-test
Evaluated with TripletEvaluator

Metric	all-nli-dev	all-nli-test
cosine_accuracy	0.9557	0.9048

Training Details

Training Dataset

qqp_triplets

Dataset: qqp_triplets at f475d9c
Size: 101,762 training samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 7 tokens mean: 10.38 tokens max: 45 tokens	min: 6 tokens mean: 12.8 tokens max: 39 tokens	min: 6 tokens mean: 13.4 tokens max: 50 tokens

Samples:

anchor	positive	negative
`A person on a horse jumps over a broken down airplane.`	`A person is outdoors, on a horse.`	`A person is at a diner, ordering an omelette.`
`Children smiling and waving at camera`	`There are children present`	`The kids are frowning`
`A boy is jumping on skateboard in the middle of a red bridge.`	`The boy does a skateboarding trick.`	`The boy skates down the sidewalk.`

Loss: TripletLoss with these parameters:

{
    "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
    "triplet_margin": 5
}

Evaluation Dataset

all-nli

Dataset: all-nli at d482672
Size: 6,584 evaluation samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 6 tokens mean: 18.02 tokens max: 66 tokens	min: 5 tokens mean: 9.81 tokens max: 29 tokens	min: 5 tokens mean: 10.37 tokens max: 29 tokens

Samples:

anchor	positive	negative
`Two women are embracing while holding to go packages.`	`Two woman are holding packages.`	`The men are fighting outside a deli.`
`Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.`	`Two kids in numbered jerseys wash their hands.`	`Two kids in jackets walk to school.`
`A man selling donuts to a customer during a world exhibition event held in the city of Angeles`	`A man selling donuts to a customer.`	`A woman drinks her coffee in a small cafe.`

Loss: TripletLoss with these parameters:

{
    "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
    "triplet_margin": 5
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 2e-05
num_train_epochs: 1
warmup_ratio: 0.1
fp16: 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: 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: 2e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
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: True
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: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	all-nli-dev_cosine_accuracy	all-nli-test_cosine_accuracy
0.0029	100	4.8503	-	-	-
0.0057	200	4.8595	-	-	-
0.0086	300	4.8404	-	-	-
0.0115	400	4.7216	-	-	-
0.0143	500	4.5376	-	-	-
0.0172	600	4.2746	-	-	-
0.0201	700	3.8115	-	-	-
0.0229	800	3.4799	-	-	-
0.0258	900	3.3466	-	-	-
0.0287	1000	3.1246	-	-	-
0.0315	1100	3.0024	-	-	-
0.0344	1200	2.6864	-	-	-
0.0373	1300	2.6236	-	-	-
0.0402	1400	2.312	-	-	-
0.0430	1500	2.2408	-	-	-
0.0459	1600	2.0932	-	-	-
0.0488	1700	1.9609	-	-	-
0.0516	1800	1.9065	-	-	-
0.0545	1900	2.0037	-	-	-
0.0574	2000	1.8347	-	-	-
0.0602	2100	1.7955	-	-	-
-1	-1	-	-	0.8873	0.9048
0.0029	100	1.6861	1.5452	0.8873	-
0.0057	200	1.7841	1.5419	0.8873	-
0.0086	300	1.7054	1.5365	0.8878	-
0.0115	400	1.7229	1.5309	0.8885	-
0.0143	500	1.569	1.5230	0.8885	-
0.0172	600	1.7132	1.5100	0.8902	-
0.0201	700	1.6673	-	-	-
0.0143	500	1.6086	1.4758	0.8917	-
0.0287	1000	1.562	1.4129	0.8958	-
0.0430	1500	1.4604	1.3243	0.8990	-
0.0574	2000	1.2731	1.2694	0.9014	-
0.0717	2500	1.3045	1.1753	0.9058	-
0.0860	3000	1.261	1.0945	0.9139	-
0.1004	3500	1.2035	1.0355	0.9201	-
0.1147	4000	1.1648	0.9687	0.9242	-
0.1291	4500	1.1255	0.9390	0.9256	-
0.1434	5000	1.0097	0.9202	0.9228	-
0.1577	5500	0.997	0.8762	0.9297	-
0.1721	6000	0.9698	0.8487	0.9338	-
0.1864	6500	0.8949	0.8460	0.9339	-
0.2008	7000	0.9007	0.8203	0.9345	-
0.2151	7500	0.8834	0.8189	0.9353	-
0.2294	8000	0.8699	0.8025	0.9359	-
0.2438	8500	0.8574	0.7930	0.9371	-
0.2868	10000	0.5934	-	-	-
0.5736	20000	0.6866	-	-	-
0.0029	100	0.5108	0.6478	0.9474	-
0.0057	200	0.6647	0.6448	0.9479	-
0.0086	300	0.6496	0.6370	0.9481	-
0.0115	400	0.558	0.6321	0.9490	-
0.0143	500	0.5273	0.6300	0.9484	-
0.0172	600	0.5374	0.6234	0.9491	-
0.0201	700	0.5382	0.6216	0.9473	-
0.0229	800	0.5819	0.6190	0.9484	-
0.0258	900	0.5833	0.6116	0.9490	-
0.0287	1000	0.4391	0.6126	0.9476	-
0.0315	1100	0.6385	0.6041	0.9484	-
0.0344	1200	0.5859	0.5952	0.9519	-
0.0373	1300	0.5665	0.5914	0.9505	-
0.0402	1400	0.5657	0.5939	0.9485	-
0.0430	1500	0.5685	0.6102	0.9481	-
0.0459	1600	0.5033	0.5939	0.9491	-
0.0488	1700	0.4487	0.5958	0.9485	-
0.0516	1800	0.4801	0.6010	0.9461	-
0.0545	1900	0.5075	0.6012	0.9484	-
0.0574	2000	0.518	0.5895	0.9496	-
0.0602	2100	0.4062	0.5886	0.9484	-
0.0631	2200	0.4278	0.6018	0.9467	-
0.0660	2300	0.4356	0.6112	0.9473	-
0.0688	2400	0.4446	0.6068	0.9503	-
0.0717	2500	0.4782	0.6000	0.9481	-
0.0746	2600	0.4841	0.6058	0.9511	-
0.0774	2700	0.4566	0.5983	0.9508	-
0.0803	2800	0.4021	0.6063	0.9491	-
0.0832	2900	0.5318	0.6231	0.9499	-
0.0860	3000	0.4938	0.6310	0.9465	-
0.0889	3100	0.5207	0.6022	0.9520	-
0.0918	3200	0.4585	0.6072	0.9517	-
0.0946	3300	0.4797	0.6051	0.9509	-
0.0975	3400	0.4313	0.6169	0.9494	-
0.1004	3500	0.5005	0.6386	0.9497	-
0.1033	3600	0.4712	0.6337	0.9440	-
0.1061	3700	0.4868	0.6309	0.9491	-
0.1090	3800	0.5115	0.6558	0.9476	-
0.1119	3900	0.4655	0.6351	0.9482	-
0.1147	4000	0.4614	0.6397	0.9470	-
0.1176	4100	0.5194	0.6409	0.9449	-
0.1205	4200	0.4946	0.6423	0.9453	-
0.1233	4300	0.5083	0.6323	0.9474	-
0.1262	4400	0.4596	0.6240	0.9481	-
0.1291	4500	0.4472	0.6323	0.9487	-
0.1319	4600	0.4135	0.6158	0.9525	-
0.1348	4700	0.4167	0.6240	0.9482	-
0.1377	4800	0.4163	0.6278	0.9484	-
0.1405	4900	0.3373	0.6335	0.9468	-
0.1434	5000	0.4124	0.6374	0.9443	-
0.1463	5100	0.4058	0.6522	0.9456	-
0.1491	5200	0.4267	0.6638	0.9435	-
0.1520	5300	0.4091	0.6874	0.9399	-
0.1549	5400	0.3793	0.6383	0.9482	-
0.1577	5500	0.3941	0.6700	0.9435	-
0.1606	5600	0.4212	0.6721	0.9468	-
0.1635	5700	0.3591	0.6548	0.9453	-
0.1664	5800	0.3368	0.6837	0.9433	-
0.1692	5900	0.4446	0.6728	0.9438	-
0.1721	6000	0.3587	0.6567	0.9482	-
0.1750	6100	0.2734	0.6608	0.9471	-
0.1778	6200	0.3344	0.6621	0.9452	-
0.1807	6300	0.3618	0.6798	0.9408	-
0.1836	6400	0.3687	0.6727	0.9412	-
0.1864	6500	0.3313	0.6448	0.9449	-
0.1893	6600	0.3492	0.6545	0.9440	-
0.1922	6700	0.3462	0.6598	0.9446	-
0.1950	6800	0.3201	0.6751	0.9424	-
0.1979	6900	0.3084	0.6400	0.9430	-
0.2008	7000	0.3034	0.6589	0.9453	-
0.2036	7100	0.3229	0.6881	0.9427	-
0.2065	7200	0.2896	0.6497	0.9435	-
0.2094	7300	0.3159	0.6333	0.9471	-
0.2122	7400	0.2931	0.6273	0.9470	-
0.2151	7500	0.2994	0.6412	0.9464	-
0.2180	7600	0.2507	0.6144	0.9478	-
0.2208	7700	0.266	0.6523	0.9421	-
0.2237	7800	0.2872	0.6294	0.9464	-
0.2266	7900	0.2913	0.6471	0.9461	-
0.2294	8000	0.2095	0.6418	0.9443	-
0.2323	8100	0.2603	0.6900	0.9383	-
0.2352	8200	0.2553	0.6363	0.9437	-
0.2381	8300	0.309	0.6238	0.9452	-
0.2409	8400	0.2574	0.6488	0.9414	-
0.2438	8500	0.2083	0.6528	0.9444	-
0.2467	8600	0.2371	0.6723	0.9400	-
0.2495	8700	0.2962	0.6354	0.9426	-
0.2524	8800	0.4068	0.6418	0.9420	-
0.2553	8900	0.2896	0.6188	0.9447	-
0.2581	9000	0.3449	0.6375	0.9424	-
0.2610	9100	0.308	0.6172	0.9470	-
0.2639	9200	0.4177	0.6231	0.9458	-
0.2667	9300	0.3373	0.6260	0.9488	-
0.2696	9400	0.3533	0.6268	0.9496	-
0.2725	9500	0.3793	0.6539	0.9455	-
0.2753	9600	0.3553	0.6253	0.9443	-
0.2782	9700	0.3545	0.6050	0.9471	-
0.2811	9800	0.3108	0.6228	0.9459	-
0.2839	9900	0.414	0.6265	0.9467	-
0.2868	10000	0.3686	0.6263	0.9459	-
0.2897	10100	0.366	0.6178	0.9453	-
0.2925	10200	0.3388	0.6428	0.9429	-
0.2954	10300	0.3805	0.6244	0.9473	-
0.2983	10400	0.3446	0.6195	0.9458	-
0.3012	10500	0.2602	0.6242	0.9456	-
0.3040	10600	0.328	0.6323	0.9438	-
0.3069	10700	0.3151	0.6179	0.9485	-
0.3098	10800	0.2682	0.6179	0.9465	-
0.3126	10900	0.3493	0.6365	0.9435	-
0.3155	11000	0.3194	0.6297	0.9449	-
0.3184	11100	0.2754	0.6478	0.9397	-
0.3212	11200	0.3181	0.6163	0.9450	-
0.3241	11300	0.2817	0.6100	0.9441	-
0.3270	11400	0.3091	0.5994	0.9465	-
0.3298	11500	0.2963	0.6135	0.9468	-
0.3327	11600	0.2824	0.6086	0.9455	-
0.3356	11700	0.2495	0.6214	0.9473	-
0.3384	11800	0.3144	0.6338	0.9424	-
0.3413	11900	0.2904	0.6220	0.9459	-
0.3442	12000	0.2964	0.6120	0.9478	-
0.3470	12100	0.2887	0.6104	0.9471	-
0.3499	12200	0.2619	0.6152	0.9476	-
0.3528	12300	0.3758	0.6147	0.9485	-
0.3556	12400	0.2787	0.6149	0.9465	-
0.3585	12500	0.2811	0.6044	0.9496	-
0.3614	12600	0.2409	0.6021	0.9487	-
0.3643	12700	0.2835	0.6113	0.9474	-
0.3671	12800	0.3025	0.6181	0.9467	-
0.3700	12900	0.2741	0.6016	0.9500	-
0.3729	13000	0.2868	0.5940	0.9505	-
0.3757	13100	0.2739	0.6104	0.9478	-
0.3786	13200	0.2912	0.6229	0.9456	-
0.3815	13300	0.3091	0.6329	0.9443	-
0.3843	13400	0.2513	0.6309	0.9414	-
0.3872	13500	0.2921	0.6273	0.9437	-
0.3901	13600	0.272	0.6149	0.9444	-
0.3929	13700	0.2553	0.6398	0.9421	-
0.3958	13800	0.2647	0.6282	0.9465	-
0.3987	13900	0.2125	0.6196	0.9467	-
0.4015	14000	0.2639	0.6072	0.9423	-
0.4044	14100	0.2206	0.6136	0.9446	-
0.4073	14200	0.2165	0.6117	0.9459	-
0.4101	14300	0.1993	0.6256	0.9406	-
0.4130	14400	0.2288	0.6332	0.9447	-
0.4159	14500	0.2434	0.6109	0.9435	-
0.4187	14600	0.2274	0.6106	0.9458	-
0.4216	14700	0.2088	0.5842	0.9478	-
0.4245	14800	0.2133	0.6151	0.9458	-
0.4274	14900	0.2033	0.5970	0.9464	-
0.4302	15000	0.2469	0.6124	0.9461	-
0.4331	15100	0.2415	0.6001	0.9478	-
0.4360	15200	0.1785	0.6013	0.9465	-
0.4388	15300	0.2257	0.6015	0.9465	-
0.4417	15400	0.1985	0.5986	0.9491	-
0.4446	15500	0.2581	0.6048	0.9502	-
0.4474	15600	0.2702	0.6154	0.9474	-
0.4503	15700	0.2028	0.6003	0.9493	-
0.4532	15800	0.1722	0.6275	0.9450	-
0.4560	15900	0.1977	0.6310	0.9432	-
0.4589	16000	0.2191	0.6123	0.9482	-
0.4618	16100	0.2124	0.6260	0.9458	-
0.4646	16200	0.2143	0.6099	0.9474	-
0.4675	16300	0.2018	0.5997	0.9462	-
0.4704	16400	0.1887	0.6123	0.9449	-
0.4732	16500	0.2036	0.6009	0.9505	-
0.4761	16600	0.1788	0.5969	0.9508	-
0.4790	16700	0.2213	0.6137	0.9464	-
0.4818	16800	0.2031	0.5984	0.9488	-
0.4847	16900	0.1904	0.6026	0.9488	-
0.4876	17000	0.173	0.6161	0.9488	-
0.4904	17100	0.2011	0.6114	0.9456	-
0.4933	17200	0.2513	0.5981	0.9482	-
0.4962	17300	0.2176	0.5912	0.9503	-
0.4991	17400	0.1753	0.5911	0.9506	-
0.5019	17500	0.227	0.6049	0.9467	-
0.5048	17600	0.2112	0.5998	0.9473	-
0.5077	17700	0.2064	0.5989	0.9478	-
0.5105	17800	0.1722	0.6224	0.9458	-
0.5134	17900	0.1682	0.6007	0.9468	-
0.5163	18000	0.1685	0.5990	0.9476	-
0.5191	18100	0.2159	0.6031	0.9465	-
0.5220	18200	0.1772	0.6014	0.9482	-
0.5249	18300	0.1915	0.6040	0.9485	-
0.5277	18400	0.1838	0.6020	0.9485	-
0.5306	18500	0.1922	0.6198	0.9455	-
0.5335	18600	0.2625	0.6013	0.9471	-
0.5363	18700	0.1749	0.6006	0.9478	-
0.5392	18800	0.154	0.6084	0.9478	-
0.5421	18900	0.1681	0.6162	0.9479	-
0.5449	19000	0.2006	0.5959	0.9481	-
0.5478	19100	0.162	0.5910	0.9476	-
0.5507	19200	0.1558	0.5829	0.9500	-
0.5535	19300	0.1847	0.5866	0.9478	-
0.5564	19400	0.1702	0.5864	0.9494	-
0.5593	19500	0.1791	0.6086	0.9474	-
0.5622	19600	0.1601	0.5851	0.9490	-
0.5650	19700	0.1999	0.5939	0.9479	-
0.5679	19800	0.179	0.5996	0.9490	-
0.5708	19900	0.1723	0.6054	0.9456	-
0.5736	20000	0.2368	0.6067	0.9464	-
0.5765	20100	0.1903	0.5984	0.9473	-
0.5794	20200	0.1705	0.5928	0.9473	-
0.5822	20300	0.1571	0.5949	0.9464	-
0.5851	20400	0.1701	0.6009	0.9433	-
0.5880	20500	0.1319	0.5947	0.9473	-
0.5908	20600	0.1597	0.6022	0.9462	-
0.5937	20700	0.1543	0.6083	0.9461	-
0.5966	20800	0.1665	0.5959	0.9471	-
0.5994	20900	0.1956	0.5885	0.9468	-
0.6023	21000	0.146	0.5836	0.9471	-
0.6052	21100	0.153	0.5982	0.9438	-
0.6080	21200	0.1282	0.5898	0.9468	-
0.6109	21300	0.177	0.5761	0.9494	-
0.6138	21400	0.4914	0.5753	0.9496	-
0.6166	21500	0.4644	0.5833	0.9496	-
0.6195	21600	0.5082	0.5849	0.9502	-
0.6224	21700	0.5107	0.5743	0.9509	-
0.6253	21800	0.5229	0.5728	0.9499	-
0.6281	21900	0.4995	0.5659	0.9506	-
0.6310	22000	0.4214	0.5685	0.9508	-
0.6339	22100	0.5147	0.5692	0.9490	-
0.6367	22200	0.4704	0.5564	0.9515	-
0.6396	22300	0.5525	0.5605	0.9506	-
0.6425	22400	0.476	0.5599	0.9526	-
0.6453	22500	0.4799	0.5550	0.9517	-
0.6482	22600	0.4673	0.5651	0.9500	-
0.6511	22700	0.473	0.5556	0.9512	-
0.6539	22800	0.4194	0.5522	0.9506	-
0.6568	22900	0.487	0.5441	0.9532	-
0.6597	23000	0.4595	0.5492	0.9526	-
0.6625	23100	0.4526	0.5527	0.9519	-
0.6654	23200	0.4496	0.5467	0.9522	-
0.6683	23300	0.4539	0.5513	0.9525	-
0.6711	23400	0.4223	0.5473	0.9522	-
0.6740	23500	0.4955	0.5548	0.9506	-
0.6769	23600	0.4903	0.5498	0.9526	-
0.6797	23700	0.4565	0.5393	0.9552	-
0.6826	23800	0.4346	0.5436	0.9543	-
0.6855	23900	0.4921	0.5393	0.9544	-
0.6883	24000	0.4091	0.5386	0.9549	-
0.6912	24100	0.4693	0.5379	0.9561	-
0.6941	24200	0.4479	0.5384	0.9557	-
0.6970	24300	0.4646	0.5333	0.9558	-
0.6998	24400	0.3953	0.5451	0.9534	-
0.7027	24500	0.4193	0.5381	0.9557	-
0.7056	24600	0.4397	0.5298	0.9547	-
0.7084	24700	0.5095	0.5351	0.9553	-
0.7113	24800	0.4872	0.5316	0.9553	-
0.7142	24900	0.4571	0.5276	0.9550	-
0.7170	25000	0.5045	0.5335	0.9537	-
0.7199	25100	0.4686	0.5426	0.9509	-
0.7228	25200	0.4558	0.5369	0.9526	-
0.7256	25300	0.5359	0.5304	0.9532	-
0.7285	25400	0.4151	0.5360	0.9535	-
0.7314	25500	0.4559	0.5414	0.9512	-
0.7342	25600	0.4454	0.5321	0.9534	-
0.7371	25700	0.4951	0.5325	0.9543	-
0.7400	25800	0.4476	0.5281	0.9550	-
0.7428	25900	0.4266	0.5280	0.9563	-
0.7457	26000	0.4356	0.5307	0.9561	-
0.7486	26100	0.466	0.5261	0.9560	-
0.7514	26200	0.4873	0.5256	0.9555	-
0.7543	26300	0.4595	0.5178	0.9553	-
0.7572	26400	0.367	0.5218	0.9561	-
0.7601	26500	0.4913	0.5174	0.9560	-
0.7629	26600	0.4464	0.5138	0.9564	-
0.7658	26700	0.4651	0.5167	0.9552	-
0.7687	26800	0.3585	0.5182	0.9543	-
0.7715	26900	0.3361	0.5213	0.9561	-
0.7744	27000	0.4138	0.5211	0.9549	-
0.7773	27100	0.3869	0.5251	0.9535	-
0.7801	27200	0.4386	0.5257	0.9538	-
0.7830	27300	0.4725	0.5288	0.9525	-
0.7859	27400	0.4149	0.5338	0.9519	-
0.7887	27500	0.4375	0.5342	0.9520	-
0.7916	27600	0.3823	0.5314	0.9535	-
0.7945	27700	0.3709	0.5305	0.9528	-
0.7973	27800	0.4664	0.5307	0.9540	-
0.8002	27900	0.392	0.5292	0.9537	-
0.8031	28000	0.4618	0.5298	0.9526	-
0.8059	28100	0.3979	0.5287	0.9523	-
0.8088	28200	0.4698	0.5212	0.9540	-
0.8117	28300	0.3464	0.5210	0.9541	-
0.8145	28400	0.456	0.5210	0.9543	-
0.8174	28500	0.4301	0.5210	0.9550	-
0.8203	28600	0.4716	0.5183	0.9561	-
0.8232	28700	0.3684	0.5215	0.9550	-
0.8260	28800	0.4643	0.5176	0.9557	-
0.8289	28900	0.391	0.5223	0.9543	-
0.8318	29000	0.3916	0.5202	0.9529	-
0.8346	29100	0.3636	0.5223	0.9543	-
0.8375	29200	0.4098	0.5208	0.9538	-
0.8404	29300	0.4074	0.5216	0.9538	-
0.8432	29400	0.4149	0.5177	0.9535	-
0.8461	29500	0.3674	0.5167	0.9549	-
0.8490	29600	0.4507	0.5159	0.9557	-
0.8518	29700	0.4066	0.5141	0.9569	-
0.8547	29800	0.3429	0.5111	0.9563	-
0.8576	29900	0.4127	0.5152	0.9560	-
0.8604	30000	0.402	0.5132	0.9555	-
0.8633	30100	0.4486	0.5112	0.9553	-
0.8662	30200	0.4669	0.5103	0.9558	-
0.8690	30300	0.3961	0.5135	0.9549	-
0.8719	30400	0.4236	0.5130	0.9549	-
0.8748	30500	0.4198	0.5130	0.9550	-
0.8776	30600	0.3946	0.5099	0.9561	-
0.8805	30700	0.4098	0.5087	0.9552	-
0.8834	30800	0.3808	0.5095	0.9546	-
0.8863	30900	0.4149	0.5106	0.9544	-
0.8891	31000	0.4102	0.5106	0.9560	-
0.8920	31100	0.4009	0.5121	0.9553	-
0.8949	31200	0.3867	0.5090	0.9550	-
0.8977	31300	0.4151	0.5087	0.9553	-
0.9006	31400	0.4549	0.5074	0.9557	-
0.9035	31500	0.3757	0.5080	0.9558	-
0.9063	31600	0.3685	0.5088	0.9550	-
0.9092	31700	0.3433	0.5102	0.9549	-
0.9121	31800	0.3207	0.5082	0.9555	-
0.9149	31900	0.5076	0.5073	0.9553	-
0.9178	32000	0.4057	0.5056	0.9557	-
0.9207	32100	0.426	0.5071	0.9557	-
0.9235	32200	0.3753	0.5068	0.9561	-
0.9264	32300	0.3639	0.5062	0.9566	-
0.9293	32400	0.4551	0.5060	0.9564	-
0.9321	32500	0.3991	0.5042	0.9567	-
0.9350	32600	0.3713	0.5040	0.9572	-
0.9379	32700	0.4199	0.5042	0.9572	-
0.9407	32800	0.4355	0.5041	0.9567	-
0.9436	32900	0.4045	0.5050	0.9561	-
0.9465	33000	0.4686	0.5063	0.9560	-
0.9493	33100	0.4204	0.5067	0.9560	-
0.9522	33200	0.3856	0.5055	0.9557	-
0.9551	33300	0.371	0.5054	0.9553	-
0.9580	33400	0.3439	0.5055	0.9550	-
0.9608	33500	0.3519	0.5044	0.9550	-
0.9637	33600	0.4555	0.5041	0.9553	-
0.9666	33700	0.4009	0.5035	0.9555	-
0.9694	33800	0.4453	0.5035	0.9557	-
0.9723	33900	0.3833	0.5036	0.9552	-
0.9752	34000	0.4048	0.5027	0.9553	-
0.9780	34100	0.4663	0.5034	0.9547	-
0.9809	34200	0.3675	0.5031	0.9547	-
0.9838	34300	0.4779	0.5032	0.9552	-
0.9866	34400	0.504	0.5030	0.9553	-
0.9895	34500	0.4018	0.5028	0.9555	-
0.9924	34600	0.4258	0.5026	0.9557	-
0.9952	34700	0.4134	0.5024	0.9557	-
0.9981	34800	0.3956	0.5025	0.9557	-

Framework Versions

Python: 3.11.11
Sentence Transformers: 3.4.1
Transformers: 4.48.3
PyTorch: 2.5.1+cu124
Accelerate: 1.3.0
Datasets: 3.3.2
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",
}

TripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}