Update README.md

README.md

@@ -1,6 +1,9 @@
 ---
 library_name: setfit
 tags:
+- mtg
+- multilabel
+- magic
 - setfit
 - sentence-transformers
 - text-classification
@@ -8,34 +11,37 @@ tags:
 metrics:
 - accuracy
 widget:
-- text: 'Stoic Farmer
+- text: >-
+    Stoic Farmer
 
 
-    When Stoic Farmer enters the battlefield, search your library for a basic Plains
-    card and reveal it. If an opponent controls more lands than you, put it onto the
-    battlefield tapped. Otherwise put it into your hand. Then shuffle.
+    When Stoic Farmer enters the battlefield, search your library for a basic
+    Plains card and reveal it. If an opponent controls more lands than you, put
+    it onto the battlefield tapped. Otherwise put it into your hand. Then
+    shuffle.
 
-    Foretell {1}{W} (During your turn, you may pay {2} and exile this card from your
-    hand face down. Cast it on a later turn for its foretell cost.)'
-- text: 'Hibernation Sliver
+    Foretell {1}{W} (During your turn, you may pay {2} and exile this card from
+    your hand face down. Cast it on a later turn for its foretell cost.)
+- text: |-
+    Hibernation Sliver
 
+    All Slivers have "Pay 2 life: Return this permanent to its owner's hand."
+- text: |-
+    Scattershot Archer
 
-    All Slivers have "Pay 2 life: Return this permanent to its owner''s hand."'
-- text: 'Scattershot Archer
+    {T}: Scattershot Archer deals 1 damage to each creature with flying.
+- text: |-
+    Seize the Initiative
 
-
-    {T}: Scattershot Archer deals 1 damage to each creature with flying.'
-- text: 'Seize the Initiative
-
-
-    Target creature gets +1/+1 and gains first strike until end of turn.'
-- text: 'Voldaren Duelist
+    Target creature gets +1/+1 and gains first strike until end of turn.
+- text: >-
+    Voldaren Duelist
 
 
     Haste
 
-    When Voldaren Duelist enters the battlefield, target creature can''t block this
-    turn.'
+    When Voldaren Duelist enters the battlefield, target creature can't block
+    this turn.
 pipeline_tag: text-classification
 inference: false
 base_model: sentence-transformers/paraphrase-mpnet-base-v2
@@ -46,15 +52,32 @@ model-index:
       type: text-classification
       name: Text Classification
     dataset:
-      name: Unknown
-      type: unknown
+      name: mtg-coloridentity-multilabel-classification
+      type: joshuasundance/mtg-coloridentity-multilabel-classification
       split: test
     metrics:
     - type: accuracy
       value: 0.7145687016027372
       name: Accuracy
+license: mit
+datasets:
+- joshuasundance/mtg-coloridentity-multilabel-classification
+language:
+- en
 ---
 
+This is a proof-of-concept model trained on `datasets/joshuasundance/mtg-coloridentity-multilabel-classification`. 
+
+It takes card name + text as a single str as input and outputs color identity as an encoding:
+
+```python
+colors = ['B', 'G', 'R', 'U', 'W']
+b = [1, 0, 0, 0, 0]
+bw = [1, 0, 0, 0, 1]
+gru = [0, 1, 1, 1, 0]
+# and so on
+```
+
 # SetFit with sentence-transformers/paraphrase-mpnet-base-v2
 
 This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2) as the Sentence Transformer embedding model. A OneVsRestClassifier instance is used for classification.
@@ -89,6 +112,48 @@ The model has been trained using an efficient few-shot learning technique that i
 |:--------|:---------|
 | **all** | 0.7146   |
 
+### Classification Report
+```text
+              precision    recall  f1-score   support
+
+                   0.80      0.77      0.78       594
+           B       0.81      0.76      0.78       821
+          BG       0.42      0.56      0.48        63
+         BGR       0.46      0.55      0.50        22
+        BGRU       0.00      0.00      0.00         0
+       BGRUW       0.73      0.33      0.46        24
+        BGRW       0.00      0.00      0.00         0
+         BGU       0.27      0.38      0.32         8
+        BGUW       0.12      1.00      0.22         1
+         BGW       0.14      0.33      0.19         9
+          BR       0.41      0.59      0.48        80
+         BRU       0.55      0.50      0.52        24
+        BRUW       0.00      0.00      0.00         0
+         BRW       0.29      0.36      0.32        14
+          BU       0.53      0.56      0.54        91
+         BUW       0.29      0.43      0.34        14
+          BW       0.36      0.37      0.37        73
+           G       0.77      0.76      0.77       791
+          GR       0.42      0.46      0.44        85
+         GRU       0.14      0.22      0.17         9
+        GRUW       0.00      0.00      0.00         0
+         GRW       0.27      0.50      0.35        18
+          GU       0.48      0.49      0.49        69
+         GUW       0.15      0.27      0.20        15
+          GW       0.40      0.43      0.41        89
+           R       0.81      0.77      0.79       803
+          RU       0.43      0.51      0.47        68
+         RUW       0.20      0.43      0.27         7
+          RW       0.47      0.49      0.48        80
+           U       0.83      0.81      0.82       818
+          UW       0.37      0.43      0.40        86
+           W       0.77      0.72      0.74       777
+
+    accuracy                           0.71      5553
+   macro avg       0.40      0.46      0.41      5553
+weighted avg       0.73      0.71      0.72      5553
+```
+
 ## Uses
 
 ### Direct Use for Inference
@@ -105,7 +170,7 @@ 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("joshuasundance/mtg-coloridentity-multilabel-classification")
 # Run inference
 preds = model("Scattershot Archer