def infer()

in src/text_clustering.py [0:0]

• 8 lines of code
• 3 McCabe index (conditional complexity)

    def infer(self, texts, top_k=1):
        embeddings = self.embed(texts)

        dist, neighbours = self.faiss_index.search(embeddings, top_k)
        inferred_labels = []
        for i in tqdm(range(embeddings.shape[0])):
            labels = [self.cluster_labels[doc] for doc in neighbours[i]]
            inferred_labels.append(Counter(labels).most_common(1)[0][0])

        return inferred_labels, embeddings