Créé avec Colaboratory

colab/Sentiment_Analysis_+_Back_translation.ipynb

@@ -36,7 +36,7 @@
  "source": [
  "# Introduction\n",
  "\n",
- "In this colab, we walk you through how to use our code in `vietai/back_translate` to augment an example of a Vietnamese dataset. The dataset is relatively small: it consists of only 27,000 supervised examples in its training set - perfect for some experiments on data augmentation!\n",
+ "In this colab, we walk you through how to use our code in `vietai/dab` to augment an example of a Vietnamese dataset. The dataset is relatively small: it consists of only 27,000 supervised examples in its training set - perfect for some experiments on data augmentation!\n",
  "\n",
  "For this task, we have already trained vi --> en and en --> vi translation models and placed them on Google Cloud Storage. There is no need to train new ones unless you have more data or data that fits the downstream task better -- more on this later.\n",
  "\n"
@@ -525,8 +525,8 @@
  "metadata": {
  "id": "4ngordrxKCWk",
  "colab_type": "code",
- "colab": {},
- "cellView": "form"
+ "cellView": "form",
+ "colab": {}
  },
  "source": [
  "output_file = os.path.join(foody_data_dir, 'foody_paraphrased_train.txt')\n",