sigma-cls / README.md

paramitopia

Update SetFit classifier with new categories

81c4e40 verified 7 months ago

preview code

raw

history blame contribute delete

63.6 kB

metadata

tags:
  - setfit
  - sentence-transformers
  - text-classification
  - generated_from_setfit_trainer
widget:
  - text: >-
      What is an SRE? Use only Korean in your response and provide a title
      wrapped in double angular brackets, such as <<SRE>>. Use the keywords
      'indicator', 'objective' and 'management'.
  - text: >-
      Who is known as The Invincibles in English football?


      In football, "The Invincibles" is a nickname used to refer to the Preston
      North End team of the 1888–89 season, managed by William Sudell, and the
      Arsenal team of the 2003–04 season managed by Arsène Wenger.  Preston
      North End earned the nickname after completing an entire season undefeated
      in league and cup competition (27 games), while Arsenal were undefeated in
      the league (38 games) in a run that stretched to a record 49 games. The
      actual nickname of the Preston team was the "Old Invincibles" but both
      versions have been in use.
  - text: >-
      Security guard is in the vicinity of at mall. Mountain peak is in the
      vicinity of at apartment. Napkin holder is in the vicinity of at mall.
      Store is in the vicinity of at mall. Legal pad is not in the vicinity of
      desk. Napkin holder is not in the vicinity of at apartment. Room is in the
      vicinity of store. Legal pad is not in the vicinity of tsunami. Legal pad
      has property yellow. Desk is in the vicinity of room. Food is in the
      vicinity of store. Legal pad is not in the vicinity of haystack.


      Store is in the vicinity of at mall. What do you think about that
      statement?
  - text: >-
      How much PVC produced each year?


      Polyvinyl chloride (alternatively: poly(vinyl chloride), colloquial:
      polyvinyl, or simply vinyl; abbreviated: PVC) is the world's third-most
      widely produced synthetic polymer of plastic (after polyethylene and
      polypropylene). About 40 million tons of PVC are produced each year.
  - text: >-
      What is Bubble tea?


      Bubble tea (also known as pearl milk tea, bubble milk tea, tapioca milk
      tea, boba tea, or boba; Chinese: 珍珠奶茶; pinyin: zhēnzhū nǎichá, 波霸奶茶; bōbà
      nǎichá) is a tea-based drink that originated in Taiwan in the early 1980s.
      Taiwanese immigrants brought it to the United States in the 1990s,
      initially in California through regions like Los Angeles County, but the
      drink has also spread to other countries where there is a large East Asian
      diaspora population.


      Bubble tea most commonly consists of tea accompanied by chewy tapioca
      balls ("boba" or "pearls"), but it can be made with other toppings as
      well, such as grass jelly, aloe vera, red bean, or popping boba. It has
      many varieties and flavors, but the two most popular varieties are pearl
      black milk tea and pearl green milk tea ("pearl" signifies the tapioca
      balls at the bottom).
metrics:
  - accuracy
pipeline_tag: text-classification
library_name: setfit
inference: true
base_model: NovaSearch/stella_en_400M_v5

SetFit with NovaSearch/stella_en_400M_v5

This is a SetFit model that can be used for Text Classification. This SetFit model uses NovaSearch/stella_en_400M_v5 as the Sentence Transformer embedding model. A SetFitHead instance is used for classification.

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.

Model Details

Model Description

Model Type: SetFit
Sentence Transformer body: NovaSearch/stella_en_400M_v5
Classification head: a SetFitHead instance
Maximum Sequence Length: 512 tokens
Number of Classes: 7 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
Extraction	"Please convert the following to a list of unique names of persons mentioned in the article.\n\nHere's the article:\nBut a hotly contested primary is likely to drag the eventual nominee to the right, even on issues that could otherwise favor his party. Mr. DeSantis, widely seen as Mr. Trump’s most serious challenger, signed a ban on abortion in his state after six weeks, a threshold before many women know they are pregnant.\n\nAnd at some point, Republicans’ drive against transgender people and their fixation on social issues may appear to be bullying — or simply far afield from real issues in the lives of swing voters, said Ms. Caprara, the chief of staff for the Illinois governor.\n\n“There’s this toxic soup between abortion, guns, gay rights, library books, African American history,” she said. “It just comes across to people as, ‘Who are these people?’”\n\nThe biggest issue, however, may be the storm cloud on the horizon that may or may not burst — the economy. In 2020, Mr. Biden became one of the few presidential candidates in modern history to have triumphed over the candidate who was more trusted on the economy in polls." "What did the writer say they were probably going to upgrade to?\n\nI've owned almost every nexus and pixel device from google. I still have my nexus 6p lying around somewhere.\nI'm actually pretty happy with my pixel 6p. The battery life is good, the performance is excellent, no issues with the fp sensor, and contrary to other opinions, I quite like android 13.\nI've only had the phone overheat once and that was when it was super hot and humid outside and I was using walking directions with the gps on.\nThe modem isn't the best, but i've never been left without a signal and most of the time I'm close to wifi so its not a huge issue for me.\nI'll probably upgrade to the pixel 7 pro if it has some meaningful improvements to the 6p." "What is the name of the boys' dog? \n\nMike Brady (Robert Reed), a widowed architect with three sons—Greg (Barry Williams), Peter (Christopher Knight), and Bobby (Mike Lookinland)—marries Carol Martin (Florence Henderson), who herself has three daughters: Marcia (Maureen McCormick), Jan (Eve Plumb), and Cindy (Susan Olsen). Carol and her daughters take the Brady surname. Included in the blended family are Mike's live-in housekeeper, Alice Nelson (Ann B. Davis), and the boys' dog, Tiger. (In the pilot episode, the girls also have a pet: a cat named Fluffy. Fluffy never appears in any other episodes.) The setting is a large two-story house designed by Mike, located in a Los Angeles suburb.[4] The show never addressed what happened to Carol's first husband.[5]"
Math	'Given that the four-digit number $N$ is a perfect square, and each digit of $N$ is less than 7. If each digit is increased by 3, the resulting four-digit number is still a perfect square. Find $N$ .' 'Given $f(x) = ax^5 + bx^3 + cx + 8$, and $f(-2) = 10$, then $f(2) = \boxed{ }$\n\nA: $-2$\n\nB: $-6$\n\nC: $6$\n\nD: $8$' 'Find the seventh term of the geometric sequence with first term $3$ and second term $\frac{2}{5}$.'
Brainstorming	'Is this quote inspiring? Give me the words and phrases that make you think it is or is not inspiring. \n\nQuote: “The best way to not feel hopeless is to get up and do something. Don’t wait for good things to happen to you. If you go out and make some good things happen, you will fill the world with hope, you will fill yourself with hope.”' 'How can I manage my time effectively while balancing school and work?' 'Why did the Hood Canal Bridge sink in Washington in 1979?'
Factual QA	'Who was the first host of Top Chef?' 'In the 2016 World Rugby Nations Cup, what was the final score of the match between Namibia and Emerging Italy?' "What is the full name of Vic Clapham's great-grandson who completed the Comrades Marathon from 2012 to 2015?"
Generation	'Imagine you are a highly experienced spaceship engineer, and someone is asking your opinion on the most efficient design for a newly proposed interstellar spacecraft. Provide a brief overview of essential features for such a vessel.' 'Create a dialogue between two people who have completely opposite political views, but must find a way to work together.' 'Your response should contain at least 5 sentences. Include keywords [love, happiness, joy] in the response. In your response, the word "joy" should appear at least 3 times.\n\nHow can I express my feelings of joy and happiness to someone I love?'
Coding	'Your friend Claire has dragged you along to a speedcubing event that is happening in Eindhoven. These events are all about solving the Rubik’s cube and similar twisty puzzles as quickly as possible. The attendants of the event can enter into various competitions based on the type and size of the puzzle, and there are even special competitions where the puzzles need to be solved one-handed or blindfolded. \n\nClaire is competing in the most popular competition: speedsolving the $3\times 3\times 3$ Rubik’s cube, pictured on the right. Each contestant needs to solve the cube five times, each time with a different random scramble. After all solves are completed, the best and the worst times are discarded and the final score is the average of the remaining three times. The contestant with the smallest final score wins.\n\nClaire has done well in the competition so far and is among the contenders for the overall victory. All the other contestants have already finished their five solves, but Claire has one solve remaining. By looking at the final scores of the other contestants, she has deduced her own target final score. As long as her final score is less than or equal to this target score, she will be declared the overall winner. Is it possible for her to win the competition, and if so, what is the worst time she can have on her last solve in order to do so?\n\n-----Input-----\nThe input consists of:\n - One line with four real numbers $t_1$, $t_2$, $t_3$ and $t_4$, the times Claire got on her first four solves.\n - One line with a real number $t$, Claire’s target final score, the worst final score she can have in order to be declared the overall winner.\n\nEach number is between $1$ and $20$, inclusive, and is given with exactly two decimal places.\n\n-----Output-----\nIf it is not possible for Claire to win the event, output “impossible”. If she will win regardless of the time she gets on her last solve, output “infinite”. Otherwise, output the worst time she can have on her last solve in order to be declared the overall winner. Output the number to exactly two decimal places.\n\n-----Examples-----\nSample Input 1:\n6.38 7.20 6.95 8.11\n7.53\nSample Output 1:\ninfinite\n\nSample Input 2:\n6.38 7.20 6.95 8.11\n6.99\nSample Output 2:\n6.82' "Problem :\n\nBajirao is on a date with his girlfriend Avni. It is a romantic night and they are\nplaying a game of words. \n\nThe rule of this game is that if Bajirao says a word such that no adjacent letters occurring in the word are same then he gets a kiss from her otherwise he gets a slap.\n\nInput :\n\nThe first line consists of T the number of test cases. The next T lines are such that each line consists of a single word spoken by Bajirao.\n\nOutput\n\nFor every test case, on a new line print 'KISS' if Bajirao gets a kiss and 'SLAP' if Bajirao gets a slap.\n\nConstraints :\n\n1 ≤ T ≤ 100\n\n2 ≤ Length of Word spoken by Bajirao ≤ 100\n\nThe input word will comprise only of lower case English alphabets (a-z).\n\nProblem Setter : Shreyans\n\nProblem Tester : Sandeep\n\nProblem Statement : Ravi\n\n(By IIT Kgp HackerEarth Programming Club)\n\nSAMPLE INPUT\n2\nremember\noccurring\n\nSAMPLE OUTPUT\nKISS\nSLAP" 'Using the Arduino IDE, create an .ino sketch that creates an MQTT broker for the MKR WIFI 1010'
Reasoning	'Silver dollar pancakes are a variety that is smaller than traditional pancakes. Silver dollars are a type of American coin.\n\nAre some types of pancakes named after coins?' 'Most successful companies are related to good management. Enterprise management generally includes two aspects, namely management and management, of which management is more important. To do a good job of management, you need a variety of Managing talent also requires leaders to make the most of their role.\n\nIt follows from this:.\n\n1. A well-managed company will succeed.\n2. With good management talent, good management is guaranteed.\n3. Poorly managed companies will eventually fail in market competition.\n4. Leaders should pay attention to the role of subordinates.' 'The survey shows that the biggest difficulty in youth entrepreneurship is funding.64.2% of people believe that lack of sufficient funds is the main difficulty. Many people are unwilling to borrow or raise funds despite lack of funds, which reflects that many entrepreneurs are starting a business. There is a conservative mentality in the process. Another prominent difficulty is the excessive competition of peers, accounting for 26.9%. During the survey, it was found that the field of youth entrepreneurship is more concentrated, such as the college student group is more inclined to e-commerce, computer technology support, etc. Young farmers are more willing to engage in the planting and breeding industries that they are more familiar with. This kind of homogeneous entrepreneurship will inevitably lead to excessive competition while forming a scale effect.\n\nThe following statement is consistent with the original?\n\n1. Insufficient funding is a major factor in the failure of youth entrepreneurship.\n2. Inadequate financial services support for young entrepreneurs.\n3. Homogeneous entrepreneurship reflects the conservative mindset of entrepreneurs.\n4. The field of youth entrepreneurship is concentrated in certain fixed industries.'

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 SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("setfit_model_id")
# Run inference
preds = model("What is an SRE? Use only Korean in your response and provide a title wrapped in double angular brackets, such as <<SRE>>. Use the keywords 'indicator', 'objective' and 'management'.")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	110.8976	8430

Label	Training Sample Count
Brainstorming	250
Coding	253
Extraction	250
Factual QA	255
Generation	250
Math	250
Reasoning	250

Training Hyperparameters

batch_size: (16, 2)
num_epochs: (1, 15)
max_steps: 500
sampling_strategy: oversampling
body_learning_rate: (2e-05, 1e-05)
head_learning_rate: 0.0001
loss: CosineSimilarityLoss
distance_metric: cosine_distance
margin: 0.25
end_to_end: True
use_amp: False
warmup_proportion: 0.1
l2_weight: 0.01
seed: 42
run_name: stella_en_400M_v5
evaluation_strategy: no
eval_max_steps: -1
load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.002	1	0.2869	-
0.004	2	0.1469	-
0.006	3	0.2431	-
0.008	4	0.3568	-
0.01	5	0.2769	-
0.012	6	0.2425	-
0.014	7	0.2001	-
0.016	8	0.2825	-
0.018	9	0.2433	-
0.02	10	0.3096	-
0.022	11	0.2856	-
0.024	12	0.265	-
0.026	13	0.2476	-
0.028	14	0.1764	-
0.03	15	0.1491	-
0.032	16	0.3051	-
0.034	17	0.2445	-
0.036	18	0.249	-
0.038	19	0.1981	-
0.04	20	0.1892	-
0.042	21	0.1933	-
0.044	22	0.2331	-
0.046	23	0.2145	-
0.048	24	0.1708	-
0.05	25	0.2272	-
0.052	26	0.1714	-
0.054	27	0.2138	-
0.056	28	0.2178	-
0.058	29	0.1346	-
0.06	30	0.1939	-
0.062	31	0.1632	-
0.064	32	0.1934	-
0.066	33	0.1897	-
0.068	34	0.1558	-
0.07	35	0.1568	-
0.072	36	0.1116	-
0.074	37	0.1609	-
0.076	38	0.1294	-
0.078	39	0.1511	-
0.08	40	0.1654	-
0.082	41	0.1542	-
0.084	42	0.0887	-
0.086	43	0.0811	-
0.088	44	0.0991	-
0.09	45	0.0845	-
0.092	46	0.0875	-
0.094	47	0.0338	-
0.096	48	0.0945	-
0.098	49	0.0477	-
0.1	50	0.0696	-
0.102	51	0.136	-
0.104	52	0.099	-
0.106	53	0.0371	-
0.108	54	0.0513	-
0.11	55	0.0484	-
0.112	56	0.0194	-
0.114	57	0.0601	-
0.116	58	0.1149	-
0.118	59	0.0836	-
0.12	60	0.0865	-
0.122	61	0.0659	-
0.124	62	0.0849	-
0.126	63	0.0963	-
0.128	64	0.07	-
0.13	65	0.0233	-
0.132	66	0.1248	-
0.134	67	0.0561	-
0.136	68	0.0851	-
0.138	69	0.0638	-
0.14	70	0.0498	-
0.142	71	0.0311	-
0.144	72	0.1374	-
0.146	73	0.0502	-
0.148	74	0.0605	-
0.15	75	0.0137	-
0.152	76	0.065	-
0.154	77	0.0846	-
0.156	78	0.0347	-
0.158	79	0.0517	-
0.16	80	0.1447	-
0.162	81	0.0609	-
0.164	82	0.1423	-
0.166	83	0.0917	-
0.168	84	0.226	-
0.17	85	0.0595	-
0.172	86	0.0588	-
0.174	87	0.0228	-
0.176	88	0.0925	-
0.178	89	0.0595	-
0.18	90	0.044	-
0.182	91	0.0244	-
0.184	92	0.0939	-
0.186	93	0.0794	-
0.188	94	0.0501	-
0.19	95	0.1363	-
0.192	96	0.0502	-
0.194	97	0.0498	-
0.196	98	0.0562	-
0.198	99	0.0657	-
0.2	100	0.0397	-
0.202	101	0.0305	-
0.204	102	0.0559	-
0.206	103	0.0871	-
0.208	104	0.063	-
0.21	105	0.0143	-
0.212	106	0.0706	-
0.214	107	0.0627	-
0.216	108	0.1047	-
0.218	109	0.0487	-
0.22	110	0.0086	-
0.222	111	0.0562	-
0.224	112	0.0101	-
0.226	113	0.0235	-
0.228	114	0.0511	-
0.23	115	0.0295	-
0.232	116	0.0549	-
0.234	117	0.0554	-
0.236	118	0.0301	-
0.238	119	0.0152	-
0.24	120	0.0234	-
0.242	121	0.01	-
0.244	122	0.0372	-
0.246	123	0.0085	-
0.248	124	0.0205	-
0.25	125	0.0117	-
0.252	126	0.0039	-
0.254	127	0.0178	-
0.256	128	0.0276	-
0.258	129	0.0592	-
0.26	130	0.0143	-
0.262	131	0.0667	-
0.264	132	0.0059	-
0.266	133	0.0767	-
0.268	134	0.0088	-
0.27	135	0.0034	-
0.272	136	0.0031	-
0.274	137	0.0151	-
0.276	138	0.0072	-
0.278	139	0.0033	-
0.28	140	0.0188	-
0.282	141	0.0069	-
0.284	142	0.1552	-
0.286	143	0.0618	-
0.288	144	0.0043	-
0.29	145	0.0209	-
0.292	146	0.0094	-
0.294	147	0.0191	-
0.296	148	0.0119	-
0.298	149	0.0012	-
0.3	150	0.0014	-
0.302	151	0.0121	-
0.304	152	0.0018	-
0.306	153	0.0792	-
0.308	154	0.0027	-
0.31	155	0.0035	-
0.312	156	0.0009	-
0.314	157	0.0014	-
0.316	158	0.0068	-
0.318	159	0.0025	-
0.32	160	0.003	-
0.322	161	0.0116	-
0.324	162	0.0009	-
0.326	163	0.0404	-
0.328	164	0.0022	-
0.33	165	0.0011	-
0.332	166	0.0122	-
0.334	167	0.0006	-
0.336	168	0.0138	-
0.338	169	0.0101	-
0.34	170	0.0019	-
0.342	171	0.0033	-
0.344	172	0.0035	-
0.346	173	0.007	-
0.348	174	0.0008	-
0.35	175	0.002	-
0.352	176	0.0006	-
0.354	177	0.001	-
0.356	178	0.0011	-
0.358	179	0.0057	-
0.36	180	0.0003	-
0.362	181	0.001	-
0.364	182	0.0007	-
0.366	183	0.0016	-
0.368	184	0.0018	-
0.37	185	0.001	-
0.372	186	0.0009	-
0.374	187	0.0057	-
0.376	188	0.0008	-
0.378	189	0.0182	-
0.38	190	0.0005	-
0.382	191	0.053	-
0.384	192	0.0012	-
0.386	193	0.0158	-
0.388	194	0.0043	-
0.39	195	0.0074	-
0.392	196	0.0013	-
0.394	197	0.0016	-
0.396	198	0.0021	-
0.398	199	0.0007	-
0.4	200	0.002	-
0.402	201	0.0004	-
0.404	202	0.0008	-
0.406	203	0.0002	-
0.408	204	0.0026	-
0.41	205	0.0012	-
0.412	206	0.0004	-
0.414	207	0.0017	-
0.416	208	0.0038	-
0.418	209	0.0008	-
0.42	210	0.0008	-
0.422	211	0.0007	-
0.424	212	0.0577	-
0.426	213	0.0013	-
0.428	214	0.0005	-
0.43	215	0.0015	-
0.432	216	0.0006	-
0.434	217	0.0005	-
0.436	218	0.0017	-
0.438	219	0.001	-
0.44	220	0.0002	-
0.442	221	0.0005	-
0.444	222	0.003	-
0.446	223	0.0007	-
0.448	224	0.0002	-
0.45	225	0.001	-
0.452	226	0.0006	-
0.454	227	0.001	-
0.456	228	0.0506	-
0.458	229	0.0005	-
0.46	230	0.0009	-
0.462	231	0.0015	-
0.464	232	0.0003	-
0.466	233	0.0004	-
0.468	234	0.001	-
0.47	235	0.0004	-
0.472	236	0.0007	-
0.474	237	0.0014	-
0.476	238	0.0003	-
0.478	239	0.0004	-
0.48	240	0.0007	-
0.482	241	0.0002	-
0.484	242	0.0006	-
0.486	243	0.0003	-
0.488	244	0.0004	-
0.49	245	0.0587	-
0.492	246	0.0003	-
0.494	247	0.0007	-
0.496	248	0.0013	-
0.498	249	0.0507	-
0.5	250	0.0002	-
0.502	251	0.0004	-
0.504	252	0.0003	-
0.506	253	0.0004	-
0.508	254	0.0002	-
0.51	255	0.0003	-
0.512	256	0.0096	-
0.514	257	0.0002	-
0.516	258	0.0003	-
0.518	259	0.0003	-
0.52	260	0.0013	-
0.522	261	0.0004	-
0.524	262	0.0004	-
0.526	263	0.0007	-
0.528	264	0.0006	-
0.53	265	0.0003	-
0.532	266	0.0023	-
0.534	267	0.0008	-
0.536	268	0.0002	-
0.538	269	0.0018	-
0.54	270	0.0002	-
0.542	271	0.0007	-
0.544	272	0.0001	-
0.546	273	0.0004	-
0.548	274	0.0618	-
0.55	275	0.0192	-
0.552	276	0.0009	-
0.554	277	0.0142	-
0.556	278	0.0014	-
0.558	279	0.0006	-
0.56	280	0.0565	-
0.562	281	0.0006	-
0.564	282	0.0233	-
0.566	283	0.0004	-
0.568	284	0.0116	-
0.57	285	0.0002	-
0.572	286	0.0032	-
0.574	287	0.0001	-
0.576	288	0.0003	-
0.578	289	0.0004	-
0.58	290	0.0003	-
0.582	291	0.0003	-
0.584	292	0.0003	-
0.586	293	0.0012	-
0.588	294	0.0021	-
0.59	295	0.0002	-
0.592	296	0.0003	-
0.594	297	0.0022	-
0.596	298	0.0005	-
0.598	299	0.0005	-
0.6	300	0.0024	-
0.602	301	0.0008	-
0.604	302	0.0003	-
0.606	303	0.0022	-
0.608	304	0.0069	-
0.61	305	0.0009	-
0.612	306	0.0144	-
0.614	307	0.0004	-
0.616	308	0.0006	-
0.618	309	0.0006	-
0.62	310	0.0261	-
0.622	311	0.0002	-
0.624	312	0.0003	-
0.626	313	0.0003	-
0.628	314	0.0007	-
0.63	315	0.0603	-
0.632	316	0.0002	-
0.634	317	0.0003	-
0.636	318	0.0007	-
0.638	319	0.0006	-
0.64	320	0.0002	-
0.642	321	0.0016	-
0.644	322	0.0003	-
0.646	323	0.0003	-
0.648	324	0.0002	-
0.65	325	0.0006	-
0.652	326	0.0006	-
0.654	327	0.0006	-
0.656	328	0.0002	-
0.658	329	0.0004	-
0.66	330	0.0002	-
0.662	331	0.0002	-
0.664	332	0.0001	-
0.666	333	0.0466	-
0.668	334	0.0002	-
0.67	335	0.0003	-
0.672	336	0.0005	-
0.674	337	0.0013	-
0.676	338	0.0002	-
0.678	339	0.0004	-
0.68	340	0.0573	-
0.682	341	0.0001	-
0.684	342	0.0002	-
0.686	343	0.0002	-
0.688	344	0.0009	-
0.69	345	0.024	-
0.692	346	0.0003	-
0.694	347	0.0011	-
0.696	348	0.0002	-
0.698	349	0.0191	-
0.7	350	0.0001	-
0.702	351	0.0002	-
0.704	352	0.0009	-
0.706	353	0.0004	-
0.708	354	0.0001	-
0.71	355	0.0	-
0.712	356	0.0002	-
0.714	357	0.0002	-
0.716	358	0.0009	-
0.718	359	0.0005	-
0.72	360	0.0013	-
0.722	361	0.0046	-
0.724	362	0.0001	-
0.726	363	0.0005	-
0.728	364	0.0002	-
0.73	365	0.0017	-
0.732	366	0.0332	-
0.734	367	0.0004	-
0.736	368	0.0203	-
0.738	369	0.0003	-
0.74	370	0.0001	-
0.742	371	0.0003	-
0.744	372	0.0004	-
0.746	373	0.0133	-
0.748	374	0.0009	-
0.75	375	0.0017	-
0.752	376	0.0016	-
0.754	377	0.0022	-
0.756	378	0.0015	-
0.758	379	0.0004	-
0.76	380	0.0002	-
0.762	381	0.0001	-
0.764	382	0.0004	-
0.766	383	0.0001	-
0.768	384	0.0012	-
0.77	385	0.0005	-
0.772	386	0.0018	-
0.774	387	0.032	-
0.776	388	0.0002	-
0.778	389	0.0001	-
0.78	390	0.0019	-
0.782	391	0.001	-
0.784	392	0.0003	-
0.786	393	0.0001	-
0.788	394	0.0005	-
0.79	395	0.0016	-
0.792	396	0.0005	-
0.794	397	0.0018	-
0.796	398	0.0007	-
0.798	399	0.0002	-
0.8	400	0.0004	-
0.802	401	0.0002	-
0.804	402	0.001	-
0.806	403	0.0001	-
0.808	404	0.0002	-
0.81	405	0.0002	-
0.812	406	0.0004	-
0.814	407	0.0003	-
0.816	408	0.0001	-
0.818	409	0.0004	-
0.82	410	0.001	-
0.822	411	0.0005	-
0.824	412	0.0001	-
0.826	413	0.0002	-
0.828	414	0.0001	-
0.83	415	0.0004	-
0.832	416	0.0002	-
0.834	417	0.0002	-
0.836	418	0.0001	-
0.838	419	0.0002	-
0.84	420	0.0011	-
0.842	421	0.0002	-
0.844	422	0.0003	-
0.846	423	0.0002	-
0.848	424	0.0004	-
0.85	425	0.0002	-
0.852	426	0.0002	-
0.854	427	0.0501	-
0.856	428	0.0001	-
0.858	429	0.0002	-
0.86	430	0.0004	-
0.862	431	0.0003	-
0.864	432	0.0001	-
0.866	433	0.0001	-
0.868	434	0.0001	-
0.87	435	0.0002	-
0.872	436	0.0008	-
0.874	437	0.0001	-
0.876	438	0.0002	-
0.878	439	0.0002	-
0.88	440	0.0004	-
0.882	441	0.0002	-
0.884	442	0.0002	-
0.886	443	0.0001	-
0.888	444	0.0006	-
0.89	445	0.0002	-
0.892	446	0.0003	-
0.894	447	0.0002	-
0.896	448	0.0011	-
0.898	449	0.0002	-
0.9	450	0.0004	-
0.902	451	0.0001	-
0.904	452	0.0009	-
0.906	453	0.0001	-
0.908	454	0.0003	-
0.91	455	0.0006	-
0.912	456	0.0028	-
0.914	457	0.0002	-
0.916	458	0.0001	-
0.918	459	0.0002	-
0.92	460	0.0002	-
0.922	461	0.0004	-
0.924	462	0.0001	-
0.926	463	0.0001	-
0.928	464	0.0001	-
0.93	465	0.002	-
0.932	466	0.0003	-
0.934	467	0.0006	-
0.936	468	0.0001	-
0.938	469	0.0002	-
0.94	470	0.0002	-
0.942	471	0.0001	-
0.944	472	0.0002	-
0.946	473	0.0003	-
0.948	474	0.0003	-
0.95	475	0.001	-
0.952	476	0.0002	-
0.954	477	0.0001	-
0.956	478	0.0003	-
0.958	479	0.0002	-
0.96	480	0.0487	-
0.962	481	0.0002	-
0.964	482	0.0004	-
0.966	483	0.0002	-
0.968	484	0.0001	-
0.97	485	0.0003	-
0.972	486	0.0002	-
0.974	487	0.0003	-
0.976	488	0.0088	-
0.978	489	0.0003	-
0.98	490	0.0011	-
0.982	491	0.0003	-
0.984	492	0.0001	-
0.986	493	0.0001	-
0.988	494	0.0003	-
0.99	495	0.0002	-
0.992	496	0.0004	-
0.994	497	0.0003	-
0.996	498	0.0001	-
0.998	499	0.0002	-
1.0	500	0.0002	-

Framework Versions

Python: 3.11.3
SetFit: 1.2.0.dev0
Sentence Transformers: 3.4.1
Transformers: 4.49.0
PyTorch: 2.6.0+cu124
Datasets: 3.3.2
Tokenizers: 0.21.0

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}
}