Add table

README.md

@@ -10,4 +10,18 @@ app_file: app.py
 
 # AutoEvaluate
 
-The Space to submit evaluation jobs from the Hub
+> Submit evaluation jobs to AutoTrain from the Hugging Face Hub
+
+## Supported tasks
+
+The table below shows which tasks are currently supported for evaluation in the AutoTrain backend:
+
+| Task                            | Supported | Sample prediction repository                                                        |
+|:--------------------------------|:---------:|:------------------------------------------------------------------------------------|
+| `binary_classification`         |     ✅     | [`eval-staging-835`](https://huggingface.co/datasets/autoevaluate/eval-staging-835) |
+| `multi_class_classification`    |     ✅     | [`eval-staging-822`](https://huggingface.co/datasets/autoevaluate/eval-staging-822) |
+| `multi_label_classification`    |     ❌     |                                                                                     |
+| `entity_extraction`             |     ✅     | [`eval-staging-838`](https://huggingface.co/datasets/autoevaluate/eval-staging-838) |
+| `extractive_question_answering` |     ❌     |                                                                                     |
+| `translation`                   |     ❌     |                                                                                     |
+| `summarization`                 |     ❌     |                                                                                     |

app.py

@@ -26,7 +26,7 @@ TASK_TO_ID = {
 }
 
 # TODO: remove this hardcorded logic and accept any dataset on the Hub
-DATASETS_TO_EVALUATE = ["emotion", "conll2003", "imdb", "squad", "xsum", "ncbi_disease"]
+DATASETS_TO_EVALUATE = ["emotion", "conll2003", "imdb", "squad", "xsum", "ncbi_disease", "go_emotions"]
 
 ###########
 ### APP ###