Autoeval config (#4234)

* autoeval config added

* autoeval config added

* Added autonlp config changes

https://github.com/huggingface/autonlp-backend/issues/414

* multi-input text classification as task id instead of category

* improve metadata validation:
- support YAML keys with dashes
- add train-eval-index validation

* revert debugging stuff

* fix tests

* style

* Update metadata.py

* Update metadata.py

Co-authored-by: Quentin Lhoest <lhoest.q@gmail.com>
Co-authored-by: Quentin Lhoest <42851186+lhoestq@users.noreply.github.com>

 Commit from 6af556b64d

README.md

@@ -19,6 +19,55 @@ task_categories:
 task_ids:
 - sentiment-classification
 paperswithcode_id: imdb-movie-reviews
+train-eval-index:
+- config: plain_text
+  task: text-classification
+  task_id: binary_classification
+  splits:
+    train_split: train
+    eval_split: test
+  col_mapping:
+    text: text
+    label: target
+  metrics:
+  - type: accuracy
+  - name: Accuracy
+  - type: f1
+    name: F1 macro
+    args:
+      average: macro
+  - type: f1
+    name: F1 micro
+    args:
+      average: micro  
+  - type: f1
+    name: F1 weighted
+    args:
+      average: weighted
+  - type: precision
+    name: Precision macro
+    args:
+      average: macro  
+  - type: precision
+    name: Precision micro
+    args:
+      average: micro  
+  - type: precision
+    name: Precision weighted
+    args:
+      average: weighted  
+  - type: recall
+    name: Recall macro
+    args:
+      average: macro  
+  - type: recall
+    name: Recall micro
+    args:
+      average: micro  
+  - type: recall
+    name: Recall weighted
+    args:
+      average: weighted 
 ---
 
 # Dataset Card for "imdb"