Add SetFit ABSA model

README.md

@@ -6,23 +6,19 @@ tags:
 - text-classification
 - generated_from_setfit_trainer
 widget:
-- text: Filet Mignon:Please try the Filet Mignon, its just the most tender piece ever.
+- text: bench:Get your food to go, find a bench, and kick back with a plate of dumplings.
 - text: comparison:Frankly, when you compare what you can have here for lunch, versus
     McDs or so many other sandwich shops in the city, there is no comparison.
 - text: ton:We had crawfish boiled and despite making a mess, it was a ton of fun
     and quite tasty as well.
-- text: street:It is set far from the small street it's on, and there is no traffic
-    noise.
-- text: service:The only thing more wonderful than the food (which is exceptional)
-    is the service.
+- text: traffic noise:It is set far from the small street it's on, and there is no
+    traffic noise.
+- text: food:The only thing more wonderful than the food (which is exceptional) is
+    the service.
 metrics:
-- accuracy
 - f1_micro
 - f1_macro
-- f1_weighted
-- precision_micro
 - precision_macro
-- recall_micro
 - recall_macro
 pipeline_tag: text-classification
 library_name: setfit
@@ -39,29 +35,17 @@ model-index:
       type: unknown
       split: test
     metrics:
-    - type: accuracy
-      value: 0.85
-      name: Accuracy
     - type: f1_micro
-      value: 0.85
+      value: 0.8516772438803264
       name: F1_Micro
     - type: f1_macro
-      value: 0.8429732868757259
+      value: 0.8441110611976916
       name: F1_Macro
-    - type: f1_weighted
-      value: 0.8499330789226259
-      name: F1_Weighted
-    - type: precision_micro
-      value: 0.85
-      name: Precision_Micro
     - type: precision_macro
-      value: 0.8432831750154752
+      value: 0.8482610861593047
       name: Precision_Macro
-    - type: recall_micro
-      value: 0.85
-      name: Recall_Micro
     - type: recall_macro
-      value: 0.84267110841914
+      value: 0.8409649439480325
       name: Recall_Macro
 ---
 
@@ -86,7 +70,7 @@ This model was trained within the context of a larger system for ABSA, which loo
 - **Model Type:** SetFit
 - **Sentence Transformer body:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2)
 - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
-- **spaCy Model:** en_core_web_trf
+- **spaCy Model:** en_core_web_lg
 - **SetFitABSA Aspect Model:** [ronalhung/setfit-absa-restaurants-aspect](https://huggingface.co/ronalhung/setfit-absa-restaurants-aspect)
 - **SetFitABSA Polarity Model:** [ronalhung/setfit-absa-restaurants-polarity](https://huggingface.co/ronalhung/setfit-absa-restaurants-polarity)
 - **Maximum Sequence Length:** 256 tokens
@@ -110,9 +94,9 @@ This model was trained within the context of a larger system for ABSA, which loo
 ## Evaluation
 
 ### Metrics
-| Label   | Accuracy | F1_Micro | F1_Macro | F1_Weighted | Precision_Micro | Precision_Macro | Recall_Micro | Recall_Macro |
-|:--------|:---------|:---------|:---------|:------------|:----------------|:----------------|:-------------|:-------------|
-| **all** | 0.85     | 0.85     | 0.8430   | 0.8499      | 0.85            | 0.8433          | 0.85         | 0.8427       |
+| Label   | F1_Micro | F1_Macro | Precision_Macro | Recall_Macro |
+|:--------|:---------|:---------|:----------------|:-------------|
+| **all** | 0.8517   | 0.8441   | 0.8483          | 0.8410       |
 
 ## Uses
 
@@ -167,11 +151,11 @@ preds = model("The food was great, but the venue is just way too busy.")
 ### Training Set Metrics
 | Training set | Min | Median  | Max |
 |:-------------|:----|:--------|:----|
-| Word count   | 4   | 19.4282 | 45  |
+| Word count   | 4   | 19.4181 | 45  |
 
 | Label     | Training Sample Count |
 |:----------|:----------------------|
-| no aspect | 164                   |
+| no aspect | 167                   |
 | aspect    | 254                   |
 
 ### Training Hyperparameters
@@ -195,19 +179,19 @@ preds = model("The food was great, but the venue is just way too busy.")
 ### Training Results
 | Epoch  | Step | Training Loss | Validation Loss |
 |:------:|:----:|:-------------:|:---------------:|
-| 0.0007 | 1    | 0.3248        | -               |
-| 0.0348 | 50   | 0.321         | 0.3073          |
-| 0.0697 | 100  | 0.2806        | 0.2588          |
-| 0.1045 | 150  | 0.2523        | 0.2490          |
-| 0.1394 | 200  | 0.2406        | 0.2377          |
-| 0.1742 | 250  | 0.2215        | 0.2164          |
-| 0.2091 | 300  | 0.1754        | 0.1884          |
-| 0.2439 | 350  | 0.0979        | 0.1852          |
-| 0.2787 | 400  | 0.042         | 0.1994          |
-| 0.3136 | 450  | 0.0155        | 0.1929          |
-| 0.3484 | 500  | 0.006         | 0.1928          |
-| 0.3833 | 550  | 0.0029        | 0.1978          |
-| 0.4181 | 600  | 0.0025        | 0.1918          |
+| 0.0007 | 1    | 0.3998        | -               |
+| 0.0345 | 50   | 0.3187        | 0.3072          |
+| 0.0689 | 100  | 0.2744        | 0.2600          |
+| 0.1034 | 150  | 0.2494        | 0.2504          |
+| 0.1378 | 200  | 0.2459        | 0.2408          |
+| 0.1723 | 250  | 0.2242        | 0.2210          |
+| 0.2068 | 300  | 0.1802        | 0.1815          |
+| 0.2412 | 350  | 0.1085        | 0.1787          |
+| 0.2757 | 400  | 0.0435        | 0.1918          |
+| 0.3101 | 450  | 0.0143        | 0.1832          |
+| 0.3446 | 500  | 0.0063        | 0.1971          |
+| 0.3790 | 550  | 0.004         | 0.1945          |
+| 0.4135 | 600  | 0.002         | 0.2005          |
 
 ### Framework Versions
 - Python: 3.11.13

config_setfit.json

@@ -1,9 +1,9 @@
 {
-  "spacy_model": "en_core_web_trf",
-  "span_context": 0,
-  "normalize_embeddings": false,
   "labels": [
     "no aspect",
     "aspect"
-  ]
+  ],
+  "normalize_embeddings": false,
+  "spacy_model": "en_core_web_lg",
+  "span_context": 0
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7982fccd2bed366bcfa239c170d3bb119bc78bb6e7db0d79529f74d83251a4e0
+oid sha256:7dba18907c4154fbe59eb8021d311243efdaeece1eda4861e3fa03d57c17e663
 size 90864192

model_head.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b206f4cd840216880af7cab55a698c06e209513567eec53c6c1b1681730cff7d
+oid sha256:65a867cfda07fb8c763af0c400c0384eb89e526113dcdc0fe90e87e33f099f75
 size 3919