Contextual_top2vec error

[kirrat975]

@ddangelov
An error is occuring since new contextual_top2vec is added.When i am using embedding_model='doc2vec' then at time of finding topics this error occurs:
2024-11-14 14:56:44,195 - top2vec - INFO - Pre-processing documents for training
INFO:top2vec:Pre-processing documents for training
2024-11-14 14:56:45,567 - top2vec - INFO - Creating joint document/word embedding
INFO:top2vec:Creating joint document/word embedding
2024-11-14 15:00:05,393 - top2vec - INFO - Creating lower dimension embedding of documents
INFO:top2vec:Creating lower dimension embedding of documents
/usr/local/lib/python3.10/dist-packages/umap/umap_.py:1952: UserWarning: n_jobs value 1 overridden to 1 by setting random_state. Use no seed for parallelism.
warn(
2024-11-14 15:00:20,966 - top2vec - INFO - Finding dense areas of documents
INFO:top2vec:Finding dense areas of documents
2024-11-14 15:00:21,130 - top2vec - INFO - Finding topics
INFO:top2vec:Finding topics
AttributeError Traceback (most recent call last)

in

/usr/local/lib/python3.10/dist-packages/top2vec/top2vec.py in init(self, documents, contextual_top2vec, c_top2vec_smoothing_window, min_count, topic_merge_delta, ngram_vocab, ngram_vocab_args, embedding_model, embedding_model_path, embedding_batch_size, split_documents, document_chunker, chunk_length, max_num_chunks, chunk_overlap_ratio, chunk_len_coverage_ratio, sentencizer, speed, use_corpus_file, document_ids, keep_documents, workers, tokenizer, use_embedding_model_tokenizer, umap_args, gpu_umap, hdbscan_args, gpu_hdbscan, index_topics, verbose)
780 self.topics_indexed = False
781
--> 782 self.compute_topics(umap_args=umap_args,
783 hdbscan_args=hdbscan_args,
784 topic_merge_delta=topic_merge_delta,

/usr/local/lib/python3.10/dist-packages/top2vec/top2vec.py in compute_topics(self, umap_args, hdbscan_args, topic_merge_delta, gpu_umap, gpu_hdbscan, index_topics, contextual_top2vec, c_top2vec_smoothing_window)
1597 self.hierarchy = None
1598
-> 1599 if self.contextual_top2vec & contextual_top2vec:
1600
1601 # smooth document token embeddings

AttributeError: 'Top2Vec' object has no attribute 'contextual_top2vec'.
KINDLY GIVE ME A WAY TO RESOLVE THIS I AM USING THIS MODEL IN PROJECT AND DEADLINE IS NEAR.

[ddangelov]

I just pushed a fix, good catch. Any specific reason you are using doc2vec as embedding model?

[kirrat975]

@ddangelov at first i was using universalsentence encoder but it was causing import error so i used this,can you suggest an embedding model that is good for domain specific topics(technical)?