2.1. Pre-trained embeddings.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"2.1. Pre-trained embeddings.ipynb","version":"0.3.2","provenance":[],"private_outputs":true},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"metadata":{"id":"kbuGzY-nGv45","colab_type":"code","colab":{}},"cell_type":"code","source":["import tensorflow as tf\n","import tensorflow_hub as hub\n","import numpy as np"],"execution_count":0,"outputs":[]},{"metadata":{"id":"hPsjoGAmm0W5","colab_type":"code","colab":{}},"cell_type":"code","source":["elmo = hub.Module(\"https://tfhub.dev/google/elmo/2\")"],"execution_count":0,"outputs":[]},{"metadata":{"id":"Qs33E6NAOhbR","colab_type":"code","colab":{}},"cell_type":"code","source":["messages = [\n","    \"i went to the bank to withdraw some cash\", \n","    \"i went to the bank of the river\", \n","    \"my neighbour works as a teller in a bank\"]"],"execution_count":0,"outputs":[]},{"metadata":{"id":"VIy3bktfxf0Q","colab_type":"code","colab":{}},"cell_type":"code","source":["with tf.Session() as session:\n","  session.run([tf.global_variables_initializer(), tf.tables_initializer()])\n","\n","  outputs = session.run(elmo(\n","    messages,\n","    signature=\"default\",\n","    as_dict=True))\n","  \n","  word_embeddings = outputs[\"elmo\"] \n","  \n","  for i, message_embedding in enumerate(np.array(word_embeddings).tolist()):\n","    print(\"Message: {}\".format(messages[i]))\n","    print(\"Embedding size: {}\".format(len(message_embedding)))\n","    for e in message_embedding:\n","      print(e)"],"execution_count":0,"outputs":[]},{"metadata":{"id":"-LUEGUNsxpST","colab_type":"text"},"cell_type":"markdown","source":["Tasks:\n","\n","\n","1.   Locate embeddings of the word \"bank\".\n","2.   Check if these embeddings are equal to each other.\n","3.   If not, calculate the distances between embeddings.\n","4.   Come up with your own list of phrases with polysemous words\n","5.   Download embeddings for these words and visualize them in Embeddings Projector.\n","6.   Inspect other keys in **outputs** map\n","\n"]},{"metadata":{"id":"_6KoPWnUmUVc","colab_type":"code","colab":{}},"cell_type":"code","source":[""],"execution_count":0,"outputs":[]}]}