diff --git "a/README.md" "b/README.md"
--- "a/README.md"
+++ "b/README.md"
@@ -1,16 +1,36 @@
---
+base_model: sentence-transformers/paraphrase-mpnet-base-v2
library_name: setfit
+metrics:
+- accuracy
+pipeline_tag: text-classification
tags:
- setfit
- sentence-transformers
- text-classification
- generated_from_setfit_trainer
-base_model: sentence-transformers/paraphrase-mpnet-base-v2
-metrics:
-- accuracy
-widget: []
-pipeline_tag: text-classification
+widget:
+- text: We will get back to you.
+- text: I think so.
+- text: So I wish those were quick calling.
+- text: ' If you''re about any Medicare benefits go website, I''m more than happy
+ to assist you. We''ll let you know all the benefits and saving options.'
+- text: i'll have to decline sorry
inference: true
+model-index:
+- name: SetFit with sentence-transformers/paraphrase-mpnet-base-v2
+ results:
+ - task:
+ type: text-classification
+ name: Text Classification
+ dataset:
+ name: Unknown
+ type: unknown
+ split: test
+ metrics:
+ - type: accuracy
+ value: 0.8268669169795578
+ name: Accuracy
---
# SetFit with sentence-transformers/paraphrase-mpnet-base-v2
@@ -29,7 +49,7 @@ The model has been trained using an efficient few-shot learning technique that i
- **Sentence Transformer body:** [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2)
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
- **Maximum Sequence Length:** 512 tokens
-- **Number of Classes:** 29 classes
+- **Number of Classes:** 30 classes
@@ -40,6 +60,47 @@ The model has been trained using an efficient few-shot learning technique that i
- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
+### Model Labels
+| Label | Examples |
+|:---------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| GreetBack | - "hey how's your energy flow today all good"
- 'hello how are you'
- "what's the word how you feeling"
|
+| answering_machine | - 'The new number for Supercenter is 520-912-2943.'
- " Alright now, so leave me a message and I'll get back with you."
- 'We have your name, your number, and the nature of your call.'
|
+| affirmation | - " Hello? Yes, ma'am."
- 'Yes, I do. All right, so great.'
- 'Yes, I only have Medicare.'
|
+| calling_about | - "what's the rationale behind this call"
- "what's the purpose behind this call today"
- 'why did you select this time to reach out to me'
|
+| decline | - 'No, no, no, no, no, no.'
- "I won't."
- "I don't know. My white panel is all right."
|
+| BUSY | - 'Can you call me in maybe 15 minutes? Because I am attending, my passion now is very good. All right.'
- "Oh no, honey, I don't have a minute, but thank you for calling anyway."
- 'If you can call back this afternoon.'
|
+| who_are_you | - 'start by telling me who you are'
- 'Hello. Who are you?'
- "what's the name you're known by"
|
+| hold_a_sec | - "i'll be right back don't disconnect"
- 'I have a minute.'
- 'Please call the right back.'
|
+| where_are_you_calling_from | - "can you confirm if your company's origin is india"
- 'are you operating from india'
- "Leave a message. I'm in Korea."
|
+| other | - 'God will do it all the time.'
- " Hi, I'm Doug."
- 'Anybody on the phone? Anybody on the phone? Hello? Hello?'
|
+| interested | - "I'm just connecting you to right now."
- "I'll try."
- 'Oh, Stacy right now, honey.'
|
+| weather | - 'describe the weather at this moment'
- 'tell me about the weather situation right now'
- 'describe the weather for today'
|
+| Not_Interested | - "I don't know. I thought I was all set."
- "i'm not enthusiastic about it"
- 'I should hang up.'
|
+| provide_age | - "I'm 19. Okay, you're not."
- "I'm 82."
- 'What is your age? 92.'
|
+| DNC | - 'Please, take my number off your list.'
- 'The patient is my name from your calling list.'
- "We've been through this before. I guess if we get some more and then stop."
|
+| sorry_greeting | - "i'm not feeling very optimistic today"
- "I'm struggling."
- "I'm under the weather."
|
+| where_get_number | - "i'd like to know my number's source"
- "How'd you get my number?"
- "I'm going to go through number right now."
|
+| abusive | - 'Big balls.'
- 'crikey'
- 'what the fuck you calling me for'
|
+| can_you_email | - 'can you email me the service agreement'
- 'Send me something in writing.'
- 'can you send an email confirmation of the reservation'
|
+| already | - 'i have already reviewed the document thoroughly'
- "Well, I'm already enrolled in Medicare. All right."
- 'I did already from somebody else.'
|
+| transfer_request | - 'please transfer my call to your supervisor'
- "i'm requesting to speak with your superior asap"
- 'i want to discuss this with your supervisor'
|
+| DNQ | - 'not meeting qualifications'
- "i'm not a suitable candidate for this position"
- 'unfortunately i cannot proceed'
|
+| greetings | - 'Thanks so much and have a wonderful day.'
- 'Have a good day and all that stuff.'
- ' Well, good afternoon.'
|
+| language_barrier | - "Oh, I'm sorry, my English."
- 'espaol explain'
- "i'm sorry to speak in spanish"
|
+| are_you_bot | - 'You can talk to me. Are you talking or through a computer? All right.'
- "can you confirm if you're not a human"
- "can you tell me if you're automated"
|
+| complain_calls | - "but y'all just keep calling and calling and calling"
- 'these calls are ruining my day'
- "No, you've called me, two other people's called me twice in the last half hour."
|
+| not_decision_maker | - 'No, I have to change my Medicare plan.'
- "That's my mother. You have to call back and speak to her. Okay? She's not here right now. Okay."
- "I'm not the owner of the house or whatnot."
|
+| scam | - 'scammers are persistent how do you protect against fraudulent activity'
- "No, I'm not interested and you should start scamming people because you know it's against the law. Well, I apologize."
- "Oh, I know. Nothing but a scammer that's so quick calling this moment. Hmm. So, I apologize."
|
+| say_again | - 'Do you have?'
- 'Could you go back to that card and show the pieces?'
- "I'm sorry, I can't hear you."
|
+| callback | - ' me, give me a call back or text me. Thank you.'
- "I've already made my decision until October. You can call me back then. All right."
- 'Give me a brief call back. I will return your call. Barbara Hayes.'
|
+
+## Evaluation
+
+### Metrics
+| Label | Accuracy |
+|:--------|:---------|
+| **all** | 0.8269 |
+
## Uses
### Direct Use for Inference
@@ -56,9 +117,9 @@ Then you can load this model and run inference.
from setfit import SetFitModel
# Download from the 🤗 Hub
-model = SetFitModel.from_pretrained("setfit_model_id")
+model = SetFitModel.from_pretrained("m-aliabbas1/medicare_idrak")
# Run inference
-preds = model("I loved the spiderman movie!")
+preds = model("I think so.")
```