To use the embedding, I prefer to use the same segment method with training since this could help reduce OOV words.
By the way, it would be great if you can provide a small simple for some large file. This gives me a chance to check the format and tokens before a long downloading.
To use the embedding, I prefer to use the same segment method with training since this could help reduce OOV words.
By the way, it would be great if you can provide a small simple for some large file. This gives me a chance to check the format and tokens before a long downloading.