small fixes

app.py

@@ -1,5 +1,5 @@
 import evaluate
 from gradio_tst import launch_gradio_widget2
 
-module = evaluate.load("textgen_evaluator.py")
+module = evaluate.load("generation_evaluator.py")
 launch_gradio_widget2(module)

generation_evaluator.py

@@ -159,7 +159,7 @@ class GenerationEvaluator(evaluate.Metric):
             predictions=predictions, references=references
         )
         
-        bert_score = evaluate.load("bert_score")
+        bert_score = evaluate.load("bertscore")
         bert_score_results = bert_score.compute(
             predictions=predictions, references=references,
             lang="en"

requirements.txt

@@ -2,3 +2,5 @@ evaluate
 datasets
 scikit-learn
 gradio
+bert_score
+git+https://github.com/google-research/bleurt.git