Add SetFit model

README.md

@@ -5,13 +5,15 @@ tags:
 - text-classification
 - generated_from_setfit_trainer
 widget:
-- text: caption summarizes main takeaway
 - text: the caption is long so can take a little time to understand the context of
     the chart
-- text: Informative caption but
-- text: The chart's caption is long but includes an overview of the data
+- text: The caption is informative, it also has some unnecessary information that
+    might not be needed to interpret the charts
+- text: The caption is detailed and
 - text: Although the caption is long, it expands on each of the studies making it
     easier to understand the main takeaway
+- text: The different bars in the chart are not easy to comprehend without reading
+    the captions
 metrics:
 - accuracy
 pipeline_tag: text-classification
@@ -48,10 +50,10 @@ The model has been trained using an efficient few-shot learning technique that i
 - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
 
 ### Model Labels
-| Label | Examples                                                                                                                                                                                                                                                                                                                                                               |
-|:------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| 0     | <ul><li>'The caption provides thorough information about the anomalies, but it can take a while to understand the context about the takeaway'</li><li>'The caption is detailed, but does not provide the clear takeaway'</li><li>'The caption provided is extensive, which is informative but it adds to the complexity'</li></ul>                                     |
-| 1     | <ul><li>'The different bars in the chart are not easy to comprehend without reading the captions'</li><li>'Although the caption is long, it expands on each of the studies making it easier to understand the main takeaway'</li><li>'The caption is informative, it also has some unnecessary information that might not be needed to interpret the charts'</li></ul> |
+| Label | Examples                                                                                                                                                                                                                                                                                   |
+|:------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| 0     | <ul><li>'the caption is long so can take a little time to understand the context of the chart'</li><li>'The caption is detailed, but does not provide the clear takeaway'</li><li>'The caption is long and hard to interpret'</li></ul>                                                    |
+| 1     | <ul><li>'The caption summarizes the key trend'</li><li>'The caption is informative, it also has some unnecessary information that might not be needed to interpret the charts'</li><li>'The different bars in the chart are not easy to comprehend without reading the captions'</li></ul> |
 
 ## Uses
 
@@ -71,7 +73,7 @@ from setfit import SetFitModel
 # Download from the 🤗 Hub
 model = SetFitModel.from_pretrained("an778/sentence-transformers")
 # Run inference
-preds = model("Informative caption but")
+preds = model("The caption is detailed and")
 ```
 
 <!--
@@ -101,9 +103,9 @@ preds = model("Informative caption but")
 ## Training Details
 
 ### Training Set Metrics
-| Training set | Min | Median  | Max |
-|:-------------|:----|:--------|:----|
-| Word count   | 3   | 11.8125 | 21  |
+| Training set | Min | Median | Max |
+|:-------------|:----|:-------|:----|
+| Word count   | 3   | 11.375 | 21  |
 
 | Label | Training Sample Count |
 |:------|:----------------------|
@@ -131,11 +133,11 @@ preds = model("Informative caption but")
 ### Training Results
 | Epoch  | Step | Training Loss | Validation Loss |
 |:------:|:----:|:-------------:|:---------------:|
-| 0.1111 | 1    | 0.2351        | -               |
-| 1.0    | 9    | -             | 0.2198          |
-| 2.0    | 18   | -             | 0.2101          |
-| 3.0    | 27   | -             | 0.1996          |
-| 4.0    | 36   | -             | 0.1938          |
+| 0.1111 | 1    | 0.2791        | -               |
+| 1.0    | 9    | -             | 0.2136          |
+| 2.0    | 18   | -             | 0.2055          |
+| 3.0    | 27   | -             | 0.1965          |
+| 4.0    | 36   | -             | 0.1903          |
 
 ### Framework Versions
 - Python: 3.11.11

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab77a24647ddd258dd29a0454016c323e09523feb1be759dd72c33463b5e6571
+oid sha256:5745f07bd3a8217762862016906cf2c9aaec5c2839babf66664f682c8bb9e84f
 size 90864192

model_head.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d145f91bfa7d1b4cc5cc470af35cb7378777e4d9746a8484da63acfb656b64a6
+oid sha256:737e6c8bc539d02329266772cbb4664ed615e4d72338bb045da5d6c55e9a640b
 size 3935