sample code doesn't work
#1
by andersonbcdefg - opened
The provided code does not run as written after installing mlx-embeddings.
TypeError Traceback (most recent call last)
Cell In[6], line 14
7 # For text embedding
8 sentences = [
9 "task: sentence similarity | query: Nothing really matters.",
10 "task: sentence similarity | query: The dog is barking.",
11 "task: sentence similarity | query: The dog is barking.",
12 ]
---> 14 encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='mlx')
16 # Compute token embeddings
17 input_ids = encoded_input['input_ids']
TypeError: 'TokenizerWrapper' object is not callable
Change the line to this:
encoded_input = tokenizer.batch_encode_plus(
sentences, padding=True, truncation=True, return_tensors="mlx"
)
Source: https://github.com/Blaizzy/mlx-embeddings/tree/main