Commit 3c04b0c6 authored by  Omar  Todd's avatar Omar Todd


parent f9c11e81
......@@ -3,6 +3,8 @@
The notebooks with our experiments on MLP, SVR and Gaussian Processes as well as the Attention LSTM can be found
in the ```Experiments``` directory.
```Data``` directory was used to store different embeddings to save time across experiments
`````` is used for transforming the sentences into encodings. It can be either
inside other scripts or as a standalone script.
......@@ -58,9 +60,9 @@ python3 -m laserembeddings download-models
Laser is a port of Facebook Research's LASER (Language-Agnostic SEntence Representations).
It is able to compute multilingual sentence embedding.
### 4 General Information
### 4 General Word Information
In order to perform the encoding inside a CNN, the methods ```get_english_word_vecs``` and ```get_german_word_vecs``` can be called to output a list
In order to perform the encoding inside the LSTM, the methods ```get_english_word_vecs``` and ```get_german_word_vecs``` can be called to output a list
of dictionaries, containing the word string, the Spacy word vector, the Part-of-Speech tag and a 'is stopword' flag.
......@@ -100,7 +102,7 @@ bert_de_encodings_dev = embed_german_bert(prefix="dev")
laser_en_encodings_test = embed_english_laser(prefix="test")
For information:
For information about the words in the sentence:
english_info_train = get_english_word_vecs()
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment