Get prediction evaluation working

syntaxgym.py

@@ -185,7 +185,18 @@ class SyntaxGymMetric(datasets.Metric):
                         # up the aggregation output
                         print("Warning: exceeded ", token)
 
-        region_totals = {c: dict(totals) for c, totals in region_totals.items()}
+        region_totals = {(condition_name, region_number): float(total)
+                         for condition_name, totals in region_totals.items()
+                         for region_number, total in totals.items()}
+
+        results = {
+            "prediction_results": [
+                p.formula(region_totals) for p in self.predictions
+            ],
+
+            "region_totals": region_totals
+        }
+        return results
 
     def get_region_edges(self, item_number, condition_name):
         """

test.py

@@ -1,3 +1,6 @@
+import itertools
+from typing import List
+
 import datasets
 import numpy as np
 import transformers
@@ -18,6 +21,11 @@ model = transformers.AutoModelForCausalLM.from_pretrained(model_ref)
 model.eval()
 
 
+# all_sentences: List[List[str]] = [item["conditions"]["content"] for item in dataset["test"]]
+# all_sentences_flat = list(itertools.chain.from_iterable(all_sentences))
+
+tokenized = tokenizer(all_sentences_flat,
+                      return_tensors="pt", padding=True, return_offsets_mapping=True)
 for item in dataset["test"]:
     # TODO full preprocessing setup
     condition_names = item["conditions"]["condition_name"]