metadata

base_model: BAAI/bge-small-en-v1.5
library_name: setfit
metrics:
  - accuracy
pipeline_tag: text-classification
tags:
  - setfit
  - absa
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
widget:
  - text: >-
      book:I was hooked!! Garth Nix is a awesome writer and though the book is a
      little babyish - its definetly worth a read! I thought the whole minute -
      hand - is - a - key part was a real good idea plus the names are so fun!
      The only thing I didn't like was that Arthur doesen't take his rightful
      place as "Monday"
  - text: >-
      book:The lawyer says he tracked Jack from his book and would like Jack to
      investigate the brutal murders of thirty-seven year old Gina Anderson and
      her son Joshua in their Seattle home; the house was trashed and the
      husband a lecturer at the nearby community college vanished
  - text: >-
      beings:Arthur with the lesser key to the lower kingdom of the House in
      hand, must travel into the House to find a cure for the mysterious plague
      that is striking the people of his town and his loved ones and find out
      why there are beings intent on getting the key from him, even if it means
      killing him
  - text: >-
      figures:But when a fight emerges between the two figures - Mister Monday
      and Sneezer - they both disappear without any further regard to Arthur
  - text: >-
      book:I could not put this book down if my life depended on it! I have
      never in my life read a book this fast
inference: false

SetFit Aspect Model with BAAI/bge-small-en-v1.5

This is a SetFit model that can be used for Aspect Based Sentiment Analysis (ABSA). This SetFit model uses BAAI/bge-small-en-v1.5 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification. In particular, this model is in charge of filtering aspect span candidates.

The model has been trained using an efficient few-shot learning technique that involves:

Fine-tuning a Sentence Transformer with contrastive learning.
Training a classification head with features from the fine-tuned Sentence Transformer.

This model was trained within the context of a larger system for ABSA, which looks like so:

Use a spaCy model to select possible aspect span candidates.
Use this SetFit model to filter these possible aspect span candidates.
Use a SetFit model to classify the filtered aspect span candidates.

Model Details

Model Description

Model Type: SetFit
Sentence Transformer body: BAAI/bge-small-en-v1.5
Classification head: a LogisticRegression instance
spaCy Model: en_core_web_lg
SetFitABSA Aspect Model: omymble/books-full-bge-aspect
SetFitABSA Polarity Model: omymble/books-full-bge-polarity
Maximum Sequence Length: 512 tokens
Number of Classes: 2 classes

Model Sources

Repository: SetFit on GitHub
Paper: Efficient Few-Shot Learning Without Prompts
Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

Model Labels

Label	Examples
aspect	"younger ones:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!" 'Nix:-enjoy the genre of fantasies, of a unknown world, as Nix weaves a wonderful tale of the things that will open your eyes to a different world' 'mystery:The mystery is secondary to the rest of the story and is only really approached in the remaining 30 pages of the book'
no aspect	"point:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!" "discussion:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!" "child:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!"

Label

Examples

aspect

"younger ones:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!"
'Nix:-enjoy the genre of fantasies, of a unknown world, as Nix weaves a wonderful tale of the things that will open your eyes to a different world'
'mystery:The mystery is secondary to the rest of the story and is only really approached in the remaining 30 pages of the book'

no aspect

"point:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!"
"discussion:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!"
"child:It might be an entertaining point of discussion with a child 12 or older, but it's too dark for younger ones, unless you're ready to talk about true evil, adult motivations, supernatural forces, and fratricide!"

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

from setfit import AbsaModel

# Download from the 🤗 Hub
model = AbsaModel.from_pretrained(
    "omymble/books-full-bge-aspect",
    "omymble/books-full-bge-polarity",
)
# Run inference
preds = model("The food was great, but the venue is just way too busy.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	2	25.9648	72

Label	Training Sample Count
no aspect	572
aspect	167

Training Hyperparameters

batch_size: (64, 64)
num_epochs: (5, 5)
max_steps: -1
sampling_strategy: oversampling
body_learning_rate: (2e-05, 1e-05)
head_learning_rate: 0.01
loss: CosineSimilarityLoss
distance_metric: cosine_distance
margin: 0.25
end_to_end: False
use_amp: True
warmup_proportion: 0.1
seed: 42
eval_max_steps: -1
load_best_model_at_end: True

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0002	1	0.2687	-
0.0090	50	0.2516	-
0.0180	100	0.2619	-
0.0270	150	0.2499	-
0.0360	200	0.2428	-
0.0450	250	0.2443	-
0.0540	300	0.246	-
0.0629	350	0.249	-
0.0719	400	0.2354	-
0.0809	450	0.2347	-
0.0899	500	0.2154	-
0.0989	550	0.2285	-
0.1079	600	0.1812	-
0.1169	650	0.1446	-
0.1259	700	0.165	-
0.1349	750	0.1125	-
0.1439	800	0.0971	-
0.1529	850	0.1059	-
0.1619	900	0.0866	-
0.1709	950	0.0492	-
0.1799	1000	0.0546	0.274
0.1888	1050	0.037	-
0.1978	1100	0.0189	-
0.2068	1150	0.0279	-
0.2158	1200	0.004	-
0.2248	1250	0.0309	-
0.2338	1300	0.0049	-
0.2428	1350	0.0286	-
0.2518	1400	0.0234	-
0.2608	1450	0.0158	-
0.2698	1500	0.0354	-
0.2788	1550	0.0062	-
0.2878	1600	0.0172	-
0.2968	1650	0.0389	-
0.3058	1700	0.0221	-
0.3147	1750	0.0065	-
0.3237	1800	0.0128	-
0.3327	1850	0.0225	-
0.3417	1900	0.0021	-
0.3507	1950	0.0102	-
0.3597	2000	0.012	0.3429
0.3687	2050	0.0249	-
0.3777	2100	0.0054	-
0.3867	2150	0.0014	-
0.3957	2200	0.0014	-
0.4047	2250	0.0143	-
0.4137	2300	0.0078	-
0.4227	2350	0.0195	-
0.4317	2400	0.0006	-
0.4406	2450	0.0014	-
0.4496	2500	0.0083	-
0.4586	2550	0.0141	-
0.4676	2600	0.0046	-
0.4766	2650	0.01	-
0.4856	2700	0.0268	-
0.4946	2750	0.0008	-
0.5036	2800	0.0076	-
0.5126	2850	0.0004	-
0.5216	2900	0.0037	-
0.5306	2950	0.0005	-
0.5396	3000	0.0065	0.3565
0.5486	3050	0.002	-
0.5576	3100	0.0072	-
0.5665	3150	0.0141	-
0.5755	3200	0.0004	-
0.5845	3250	0.0086	-
0.5935	3300	0.0098	-
0.6025	3350	0.0048	-
0.6115	3400	0.0013	-
0.6205	3450	0.007	-
0.6295	3500	0.0059	-
0.6385	3550	0.0174	-
0.6475	3600	0.0003	-
0.6565	3650	0.0004	-
0.6655	3700	0.0032	-
0.6745	3750	0.0004	-
0.6835	3800	0.0035	-
0.6924	3850	0.0019	-
0.7014	3900	0.015	-
0.7104	3950	0.0204	-
0.7194	4000	0.0016	0.3404
0.7284	4050	0.0003	-
0.7374	4100	0.0036	-
0.7464	4150	0.0016	-
0.7554	4200	0.0104	-
0.7644	4250	0.003	-
0.7734	4300	0.0159	-
0.7824	4350	0.0029	-
0.7914	4400	0.0068	-
0.8004	4450	0.0021	-
0.8094	4500	0.006	-
0.8183	4550	0.006	-
0.8273	4600	0.0038	-
0.8363	4650	0.008	-
0.8453	4700	0.0003	-
0.8543	4750	0.0126	-
0.8633	4800	0.0002	-
0.8723	4850	0.0041	-
0.8813	4900	0.0002	-
0.8903	4950	0.0137	-
0.8993	5000	0.0041	0.3363
0.9083	5050	0.0252	-
0.9173	5100	0.0023	-
0.9263	5150	0.0062	-
0.9353	5200	0.0152	-
0.9442	5250	0.0014	-
0.9532	5300	0.0224	-
0.9622	5350	0.0174	-
0.9712	5400	0.0066	-
0.9802	5450	0.0002	-
0.9892	5500	0.0136	-
0.9982	5550	0.0036	-
1.0072	5600	0.0102	-
1.0162	5650	0.011	-
1.0252	5700	0.0035	-
1.0342	5750	0.0002	-
1.0432	5800	0.0002	-
1.0522	5850	0.0044	-
1.0612	5900	0.0125	-
1.0701	5950	0.0061	-
1.0791	6000	0.0165	0.3591
1.0881	6050	0.006	-
1.0971	6100	0.0003	-
1.1061	6150	0.0074	-
1.1151	6200	0.0019	-
1.1241	6250	0.0002	-
1.1331	6300	0.0064	-
1.1421	6350	0.0127	-
1.1511	6400	0.0012	-
1.1601	6450	0.0003	-
1.1691	6500	0.0251	-
1.1781	6550	0.0002	-
1.1871	6600	0.0003	-
1.1960	6650	0.0002	-
1.2050	6700	0.0002	-
1.2140	6750	0.0123	-
1.2230	6800	0.0055	-
1.2320	6850	0.0098	-
1.2410	6900	0.0028	-
1.25	6950	0.0049	-
1.2590	7000	0.0021	0.3537
1.2680	7050	0.0147	-
1.2770	7100	0.003	-
1.2860	7150	0.0002	-
1.2950	7200	0.0049	-
1.3040	7250	0.0033	-
1.3129	7300	0.0002	-
1.3219	7350	0.0065	-
1.3309	7400	0.0043	-
1.3399	7450	0.0107	-
1.3489	7500	0.0184	-
1.3579	7550	0.0116	-
1.3669	7600	0.0041	-
1.3759	7650	0.0001	-
1.3849	7700	0.0001	-
1.3939	7750	0.0074	-
1.4029	7800	0.0002	-
1.4119	7850	0.0087	-
1.4209	7900	0.0014	-
1.4299	7950	0.0045	-
1.4388	8000	0.0018	0.3439
1.4478	8050	0.0039	-
1.4568	8100	0.007	-
1.4658	8150	0.0066	-
1.4748	8200	0.0101	-
1.4838	8250	0.0047	-
1.4928	8300	0.0021	-
1.5018	8350	0.0002	-
1.5108	8400	0.0116	-
1.5198	8450	0.0017	-
1.5288	8500	0.0032	-
1.5378	8550	0.0053	-
1.5468	8600	0.0038	-
1.5558	8650	0.0001	-
1.5647	8700	0.002	-
1.5737	8750	0.0065	-
1.5827	8800	0.0064	-
1.5917	8850	0.0001	-
1.6007	8900	0.0049	-
1.6097	8950	0.0002	-
1.6187	9000	0.0083	0.3486
1.6277	9050	0.0105	-
1.6367	9100	0.0019	-
1.6457	9150	0.0002	-
1.6547	9200	0.0049	-
1.6637	9250	0.0001	-
1.6727	9300	0.0097	-
1.6817	9350	0.0098	-
1.6906	9400	0.0022	-
1.6996	9450	0.0142	-
1.7086	9500	0.0025	-
1.7176	9550	0.0147	-
1.7266	9600	0.0086	-
1.7356	9650	0.0062	-
1.7446	9700	0.0002	-
1.7536	9750	0.0103	-
1.7626	9800	0.0186	-
1.7716	9850	0.0112	-
1.7806	9900	0.0042	-
1.7896	9950	0.0166	-
1.7986	10000	0.0002	0.3571
1.8076	10050	0.0029	-
1.8165	10100	0.0055	-
1.8255	10150	0.0057	-
1.8345	10200	0.0163	-
1.8435	10250	0.0093	-
1.8525	10300	0.0083	-
1.8615	10350	0.0073	-
1.8705	10400	0.0089	-
1.8795	10450	0.0068	-
1.8885	10500	0.0001	-
1.8975	10550	0.0232	-
1.9065	10600	0.0161	-
1.9155	10650	0.0088	-
1.9245	10700	0.0002	-
1.9335	10750	0.0093	-
1.9424	10800	0.0103	-
1.9514	10850	0.002	-
1.9604	10900	0.0113	-
1.9694	10950	0.0055	-
1.9784	11000	0.0148	0.3461
1.9874	11050	0.0001	-
1.9964	11100	0.0017	-
2.0054	11150	0.0001	-
2.0144	11200	0.0204	-
2.0234	11250	0.0032	-
2.0324	11300	0.0029	-
2.0414	11350	0.002	-
2.0504	11400	0.0001	-
2.0594	11450	0.005	-
2.0683	11500	0.0001	-
2.0773	11550	0.0051	-
2.0863	11600	0.0095	-
2.0953	11650	0.0093	-
2.1043	11700	0.0171	-
2.1133	11750	0.0059	-
2.1223	11800	0.0026	-
2.1313	11850	0.0092	-
2.1403	11900	0.0002	-
2.1493	11950	0.0069	-
2.1583	12000	0.006	0.3572
2.1673	12050	0.009	-
2.1763	12100	0.008	-
2.1853	12150	0.0001	-
2.1942	12200	0.0062	-
2.2032	12250	0.0086	-
2.2122	12300	0.0001	-
2.2212	12350	0.0001	-
2.2302	12400	0.0001	-
2.2392	12450	0.0001	-
2.2482	12500	0.0022	-
2.2572	12550	0.0014	-
2.2662	12600	0.0014	-
2.2752	12650	0.009	-
2.2842	12700	0.0001	-
2.2932	12750	0.0081	-
2.3022	12800	0.0127	-
2.3112	12850	0.0001	-
2.3201	12900	0.0028	-
2.3291	12950	0.0016	-
2.3381	13000	0.0051	0.3587
2.3471	13050	0.0044	-
2.3561	13100	0.0133	-
2.3651	13150	0.0043	-
2.3741	13200	0.0001	-
2.3831	13250	0.0017	-
2.3921	13300	0.0095	-
2.4011	13350	0.008	-
2.4101	13400	0.0074	-
2.4191	13450	0.0181	-
2.4281	13500	0.0141	-
2.4371	13550	0.0114	-
2.4460	13600	0.0046	-
2.4550	13650	0.0053	-
2.4640	13700	0.0001	-
2.4730	13750	0.0001	-
2.4820	13800	0.0114	-
2.4910	13850	0.0001	-
2.5	13900	0.0075	-
2.5090	13950	0.0016	-
2.5180	14000	0.0014	0.3376
2.5270	14050	0.0075	-
2.5360	14100	0.0001	-
2.5450	14150	0.0001	-
2.5540	14200	0.0013	-
2.5629	14250	0.0001	-
2.5719	14300	0.0082	-
2.5809	14350	0.0021	-
2.5899	14400	0.0001	-
2.5989	14450	0.0001	-
2.6079	14500	0.0016	-
2.6169	14550	0.0001	-
2.6259	14600	0.0001	-
2.6349	14650	0.0058	-
2.6439	14700	0.0223	-
2.6529	14750	0.0001	-
2.6619	14800	0.0001	-
2.6709	14850	0.0249	-
2.6799	14900	0.008	-
2.6888	14950	0.0071	-
2.6978	15000	0.0237	0.3769
2.7068	15050	0.0001	-
2.7158	15100	0.0016	-
2.7248	15150	0.0031	-
2.7338	15200	0.0063	-
2.7428	15250	0.0001	-
2.7518	15300	0.0127	-
2.7608	15350	0.0001	-
2.7698	15400	0.0114	-
2.7788	15450	0.0106	-
2.7878	15500	0.0086	-
2.7968	15550	0.0083	-
2.8058	15600	0.0001	-
2.8147	15650	0.0001	-
2.8237	15700	0.0035	-
2.8327	15750	0.0095	-
2.8417	15800	0.0041	-
2.8507	15850	0.0001	-
2.8597	15900	0.0001	-
2.8687	15950	0.0001	-
2.8777	16000	0.0001	0.3509
2.8867	16050	0.0001	-
2.8957	16100	0.0124	-
2.9047	16150	0.0083	-
2.9137	16200	0.0017	-
2.9227	16250	0.0001	-
2.9317	16300	0.0042	-
2.9406	16350	0.0058	-
2.9496	16400	0.0001	-
2.9586	16450	0.0001	-
2.9676	16500	0.0021	-
2.9766	16550	0.0025	-
2.9856	16600	0.0068	-
2.9946	16650	0.0099	-
3.0036	16700	0.0015	-
3.0126	16750	0.0086	-
3.0216	16800	0.0162	-
3.0306	16850	0.0001	-
3.0396	16900	0.0181	-
3.0486	16950	0.0083	-
3.0576	17000	0.0045	0.346
3.0665	17050	0.0072	-
3.0755	17100	0.0045	-
3.0845	17150	0.005	-
3.0935	17200	0.003	-
3.1025	17250	0.0069	-
3.1115	17300	0.0001	-
3.1205	17350	0.003	-
3.1295	17400	0.0077	-
3.1385	17450	0.0001	-
3.1475	17500	0.0001	-
3.1565	17550	0.0166	-
3.1655	17600	0.0001	-
3.1745	17650	0.0001	-
3.1835	17700	0.0084	-
3.1924	17750	0.0106	-
3.2014	17800	0.0027	-
3.2104	17850	0.0092	-
3.2194	17900	0.0001	-
3.2284	17950	0.0001	-
3.2374	18000	0.0066	0.3501
3.2464	18050	0.0037	-
3.2554	18100	0.0035	-
3.2644	18150	0.0029	-
3.2734	18200	0.0017	-
3.2824	18250	0.0001	-
3.2914	18300	0.0034	-
3.3004	18350	0.0121	-
3.3094	18400	0.0051	-
3.3183	18450	0.0024	-
3.3273	18500	0.0019	-
3.3363	18550	0.0014	-
3.3453	18600	0.0167	-
3.3543	18650	0.0097	-
3.3633	18700	0.0025	-
3.3723	18750	0.0065	-
3.3813	18800	0.011	-
3.3903	18850	0.0001	-
3.3993	18900	0.0001	-
3.4083	18950	0.0072	-
3.4173	19000	0.0132	0.3511
3.4263	19050	0.0084	-
3.4353	19100	0.0015	-
3.4442	19150	0.0014	-
3.4532	19200	0.011	-
3.4622	19250	0.0083	-
3.4712	19300	0.0073	-
3.4802	19350	0.0024	-
3.4892	19400	0.002	-
3.4982	19450	0.0155	-
3.5072	19500	0.0042	-
3.5162	19550	0.0001	-
3.5252	19600	0.0043	-
3.5342	19650	0.0026	-
3.5432	19700	0.0022	-
3.5522	19750	0.002	-
3.5612	19800	0.0018	-
3.5701	19850	0.0001	-
3.5791	19900	0.0012	-
3.5881	19950	0.002	-
3.5971	20000	0.0089	0.3516
3.6061	20050	0.003	-
3.6151	20100	0.0036	-
3.6241	20150	0.0001	-
3.6331	20200	0.0001	-
3.6421	20250	0.0156	-
3.6511	20300	0.0001	-
3.6601	20350	0.0174	-
3.6691	20400	0.0001	-
3.6781	20450	0.011	-
3.6871	20500	0.0001	-
3.6960	20550	0.0047	-
3.7050	20600	0.0132	-
3.7140	20650	0.007	-
3.7230	20700	0.0001	-
3.7320	20750	0.0025	-
3.7410	20800	0.0049	-
3.75	20850	0.0074	-
3.7590	20900	0.002	-
3.7680	20950	0.0112	-
3.7770	21000	0.0001	0.3483
3.7860	21050	0.0001	-
3.7950	21100	0.0064	-
3.8040	21150	0.0133	-
3.8129	21200	0.0001	-
3.8219	21250	0.0112	-
3.8309	21300	0.0001	-
3.8399	21350	0.0001	-
3.8489	21400	0.0001	-
3.8579	21450	0.0025	-
3.8669	21500	0.0047	-
3.8759	21550	0.0001	-
3.8849	21600	0.0062	-
3.8939	21650	0.0001	-
3.9029	21700	0.0315	-
3.9119	21750	0.002	-
3.9209	21800	0.0034	-
3.9299	21850	0.004	-
3.9388	21900	0.0046	-
3.9478	21950	0.008	-
3.9568	22000	0.0103	0.3474
3.9658	22050	0.0142	-
3.9748	22100	0.0207	-
3.9838	22150	0.0105	-
3.9928	22200	0.0114	-
4.0018	22250	0.002	-
4.0108	22300	0.0121	-
4.0198	22350	0.0001	-
4.0288	22400	0.0058	-
4.0378	22450	0.0045	-
4.0468	22500	0.0001	-
4.0558	22550	0.0086	-
4.0647	22600	0.0121	-
4.0737	22650	0.0045	-
4.0827	22700	0.0001	-
4.0917	22750	0.0046	-
4.1007	22800	0.0076	-
4.1097	22850	0.0001	-
4.1187	22900	0.0154	-
4.1277	22950	0.0108	-
4.1367	23000	0.0058	0.3575
4.1457	23050	0.0088	-
4.1547	23100	0.0019	-
4.1637	23150	0.0055	-
4.1727	23200	0.0299	-
4.1817	23250	0.0085	-
4.1906	23300	0.0016	-
4.1996	23350	0.0001	-
4.2086	23400	0.0001	-
4.2176	23450	0.0072	-
4.2266	23500	0.0092	-
4.2356	23550	0.0001	-
4.2446	23600	0.0064	-
4.2536	23650	0.0065	-
4.2626	23700	0.0001	-
4.2716	23750	0.0017	-
4.2806	23800	0.0083	-
4.2896	23850	0.0001	-
4.2986	23900	0.0039	-
4.3076	23950	0.002	-
4.3165	24000	0.0037	0.357
4.3255	24050	0.0095	-
4.3345	24100	0.002	-
4.3435	24150	0.017	-
4.3525	24200	0.0086	-
4.3615	24250	0.007	-
4.3705	24300	0.0023	-
4.3795	24350	0.0122	-
4.3885	24400	0.0097	-
4.3975	24450	0.0027	-
4.4065	24500	0.0081	-
4.4155	24550	0.0043	-
4.4245	24600	0.0055	-
4.4335	24650	0.0001	-
4.4424	24700	0.0014	-
4.4514	24750	0.0001	-
4.4604	24800	0.0091	-
4.4694	24850	0.0087	-
4.4784	24900	0.0101	-
4.4874	24950	0.0001	-
4.4964	25000	0.013	0.3566
4.5054	25050	0.013	-
4.5144	25100	0.0082	-
4.5234	25150	0.0063	-
4.5324	25200	0.0046	-
4.5414	25250	0.0087	-
4.5504	25300	0.0063	-
4.5594	25350	0.0019	-
4.5683	25400	0.0061	-
4.5773	25450	0.004	-
4.5863	25500	0.0001	-
4.5953	25550	0.0001	-
4.6043	25600	0.0088	-
4.6133	25650	0.0191	-
4.6223	25700	0.0124	-
4.6313	25750	0.0001	-
4.6403	25800	0.0023	-
4.6493	25850	0.0001	-
4.6583	25900	0.0068	-
4.6673	25950	0.0001	-
4.6763	26000	0.0034	0.3563
4.6853	26050	0.0138	-
4.6942	26100	0.0001	-
4.7032	26150	0.0068	-
4.7122	26200	0.0091	-
4.7212	26250	0.0001	-
4.7302	26300	0.0152	-
4.7392	26350	0.0064	-
4.7482	26400	0.0021	-
4.7572	26450	0.0088	-
4.7662	26500	0.0001	-
4.7752	26550	0.0042	-
4.7842	26600	0.0022	-
4.7932	26650	0.0065	-
4.8022	26700	0.0039	-
4.8112	26750	0.0039	-
4.8201	26800	0.0001	-
4.8291	26850	0.0155	-
4.8381	26900	0.0021	-
4.8471	26950	0.0039	-
4.8561	27000	0.002	0.3555
4.8651	27050	0.0092	-
4.8741	27100	0.0001	-
4.8831	27150	0.0081	-
4.8921	27200	0.0081	-
4.9011	27250	0.0037	-
4.9101	27300	0.0104	-
4.9191	27350	0.0022	-
4.9281	27400	0.004	-
4.9371	27450	0.0076	-
4.9460	27500	0.0043	-
4.9550	27550	0.0142	-
4.9640	27600	0.0126	-
4.9730	27650	0.0038	-
4.9820	27700	0.0107	-
4.9910	27750	0.0019	-
5.0	27800	0.0104	-

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
SetFit: 1.0.3
Sentence Transformers: 3.0.1
spaCy: 3.7.4
Transformers: 4.39.0
PyTorch: 2.3.1+cu121
Datasets: 2.20.0
Tokenizers: 0.15.2

Citation

BibTeX

@article{https://doi.org/10.48550/arxiv.2209.11055,
    doi = {10.48550/ARXIV.2209.11055},
    url = {https://arxiv.org/abs/2209.11055},
    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
    title = {Efficient Few-Shot Learning Without Prompts},
    publisher = {arXiv},
    year = {2022},
    copyright = {Creative Commons Attribution 4.0 International}
}