Upload truthful_qa_eval.ipynb

truthful_qa_eval.ipynb

@@ -753,7 +753,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 6,
+      "execution_count": 7,
       "metadata": {
         "id": "V0qUtme9m6gp",
         "outputId": "e8596f0c-6c9f-4ef8-c49a-58fefb458a6a"
@@ -763,14 +763,14 @@
           "name": "stdout",
           "output_type": "stream",
           "text": [
-            "{'bleu': 0.07843873724059973, 'precisions': [0.16914540670403844, 0.08830763693082969, 0.0596619576185671, 0.042478229076646945], 'brevity_penalty': 1.0, 'length_ratio': 2.219936507936508, 'translation_length': 17482, 'reference_length': 7875}\n"
+            "{'rouge1': 0.24958893911586305, 'rouge2': 0.13816924470734443, 'rougeL': 0.22814075909017828, 'rougeLsum': 0.22836556528082447}\n"
           ]
         }
       ],
       "source": [
         "import evaluate\n",
         "\n",
-        "rouge = evaluate.load('bleu') # bleu\n",
+        "rouge = evaluate.load('rouge') # bleu\n",
         "\n",
         "references = dataset.best_answer.to_list()\n",
         "predictions = dataset.models_responses.to_list()\n",