4 Commits (5fc67ca2c760546f27e57ee518f25104dff6dd21)

Author	SHA1	Message	Date
matt haigh	a4896da2a0	Experimental: Add other threshold types to SemanticChunker (#16807) ... **Description** Adding different threshold types to the semantic chunker. I’ve had much better and predictable performance when using standard deviations instead of percentiles.  For all the documents I’ve tried, the distribution of distances look similar to the above: positively skewed normal distribution. All skews I’ve seen are less than 1 so that explains why standard deviations perform well, but I’ve included IQR if anyone wants something more robust. Also, using the percentile method backwards, you can declare the number of clusters and use semantic chunking to get an ‘optimal’ splitting. --------- Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>	7 months ago
Leonid Ganeline	3f6bf852ea	experimental: docstrings update (#18048) ... Added missed docstrings. Formatted docsctrings to the consistent format.	7 months ago
Giulio Zani	9f0b63dba0	experimental[patch]: Fixes issue #17060 (#17062) ... As described in issue #17060, in the case in which text has only one sentence the following function fails. Checking for that and adding a return case fixed the issue. ```python def split_text(self, text: str) -> List[str]: """Split text into multiple components.""" # Splitting the essay on '.', '?', and '!' single_sentences_list = re.split(r"(?<=[.?!])\s+", text) sentences = [ {"sentence": x, "index": i} for i, x in enumerate(single_sentences_list) ] sentences = combine_sentences(sentences) embeddings = self.embeddings.embed_documents( [x["combined_sentence"] for x in sentences] ) for i, sentence in enumerate(sentences): sentence["combined_sentence_embedding"] = embeddings[i] distances, sentences = calculate_cosine_distances(sentences) start_index = 0 # Create a list to hold the grouped sentences chunks = [] breakpoint_percentile_threshold = 95 breakpoint_distance_threshold = np.percentile( distances, breakpoint_percentile_threshold ) # If you want more chunks, lower the percentile cutoff indices_above_thresh = [ i for i, x in enumerate(distances) if x > breakpoint_distance_threshold ] # The indices of those breakpoints on your list # Iterate through the breakpoints to slice the sentences for index in indices_above_thresh: # The end index is the current breakpoint end_index = index # Slice the sentence_dicts from the current start index to the end index group = sentences[start_index : end_index + 1] combined_text = " ".join([d["sentence"] for d in group]) chunks.append(combined_text) # Update the start index for the next group start_index = index + 1 # The last group, if any sentences remain if start_index < len(sentences): combined_text = " ".join([d["sentence"] for d in sentences[start_index:]]) chunks.append(combined_text) return chunks ``` Co-authored-by: Giulio Zani <salamanderxing@Giulios-MBP.homenet.telecomitalia.it>	8 months ago
Harrison Chase	20abe24819	experimental[minor]: Add semantic chunker (#15799)	9 months ago