Home.md

@@ -31,6 +31,19 @@ Performance of individual checkpoints on 100 labeled test sentences.
 
 Perplexity of individual checkpoints on each test set (wikipedia, tiger and 10kGNAD/German News corpus). 
 
+
+## DE trained from scratched on diverse data
+
+The model is trained on 70% of the Wikipedia dataset (same as the "normal" DE model) and a 90% fraction of the datasets originally ment for only validation: Tiger, 10kGNAD (news corpus), Europarl.
+The model is trained for 20 epochs (= 122400 steps). Final eval perplexities for the validation datasets (remaining 10% of the datasets mentioned above + 5% of wikipedia) are:
+
+| DATASET | PPL |
+| --------- | --------- | 
+| Wikipedia | 81 |
+| Tiger | 2599 |
+| 10kGNAD | 534 |
+| Europarl | 855 |
+
 # LINKS
 
 https://wandb.ai/susannb/huggingface?workspace=user-susannb