Spaces:
Runtime error
Runtime error
| # -*- coding: utf-8 -*- | |
| """multilingual_Semantic_Search.ipynb | |
| Automatically generated by Colaboratory. | |
| Original file is located at | |
| https://colab.research.google.com/drive/1Wg8tD1NJqY0lnvSnsZQhB66pAvxSu65h | |
| # Multilingual Semantic Search | |
| Language models give computers the ability to search by meaning and go beyond searching by matching keywords. This capability is called semantic search. | |
|  | |
| In this notebook, we'll build a simple semantic search engine. The applications of semantic search go beyond building a web search engine. They can empower a private search engine for internal documents or records. It can also be used to power features like StackOverflow's "similar questions" feature. | |
| 1. Get the archive of questions | |
| 2. [Embed](https://docs.cohere.ai/embed-reference/) the archive | |
| 3. Search using an index and nearest neighbor search | |
| 4. Visualize the archive based on the embeddings | |
| """ | |
| # Install Cohere for embeddings, Umap to reduce embeddings to 2 dimensions, | |
| # Altair for visualization, Annoy for approximate nearest neighbor search | |
| #!pip install cohere umap-learn altair annoy datasets tqdm | |
| """Get your Cohere API key by [signing up here](https://os.cohere.ai/register). Paste it in the cell below.""" | |
| #pip install umap | |
| #@title Import libraries (Run this cell to execute required code) {display-mode: "form"} | |
| import cohere | |
| import numpy as np | |
| import re | |
| import pandas as pd | |
| from tqdm import tqdm | |
| from datasets import load_dataset | |
| import umap | |
| import altair as alt | |
| from sklearn.metrics.pairwise import cosine_similarity | |
| from annoy import AnnoyIndex | |
| import warnings | |
| warnings.filterwarnings('ignore') | |
| pd.set_option('display.max_colwidth', None) | |
| """You'll need your API key for this next cell. [Sign up to Cohere](https://os.cohere.ai/) and get one if you haven't yet.""" | |
| # Paste your API key here. Remember to not share publicly | |
| api_key = 'twdqnY8kzEsMnu3N0bTX2JsqFUWybVczDDNZTjpd' | |
| # Create and retrieve a Cohere API key from os.cohere.ai | |
| co = cohere.Client(api_key) | |
| """## 1. Get The Archive of Questions | |
| We'll use the [trec](https://www.tensorflow.org/datasets/catalog/trec) dataset which is made up of questions and their categories. | |
| """ | |
| # # Get dataset | |
| # dataset = load_dataset("trec", split="train") | |
| # # Import into a pandas dataframe, take only the first 1000 rows | |
| # df = pd.DataFrame(dataset)[:1000] | |
| # # Preview the data to ensure it has loaded correctly | |
| # df.head(10) | |
| import pandas as pd | |
| # Get dataset | |
| # dataset = load_dataset("trec", split="train") | |
| # https://www.shanelynn.ie/pandas-csv-error-error-tokenizing-data-c-error-eof-inside-string-starting-at-line/ | |
| df = pd.read_excel("/content/news_articles_dataset.xlsx") | |
| df.head() | |
| df.columns | |
| # combine columns , 'summary' | |
| cols = ['Title ', 'News'] | |
| df['text'] = df[cols].apply(lambda row: ' \n '.join(row.values.astype(str)), axis=1) | |
| df['text'].head() | |
| """## 2. Embed the archive | |
| The next step is to embed the text of the questions. | |
|  | |
| To get a thousand embeddings of this length should take about fifteen seconds. | |
| """ | |
| # Get the embeddings | |
| embeds = co.embed(texts=list(df['text']),model="multilingual-22-12",truncate="LEFT").embeddings | |
| # Check the dimensions of the embeddings | |
| embeds = np.array(embeds) | |
| print(embeds.shape) | |
| print(embeds) | |
| print(df['text'][0]) | |
| print(embeds[0]) | |
| print(embeds.shape) | |
| """## 3. Search using an index and nearest neighbor search | |
|  | |
| Let's now use [Annoy](https://github.com/spotify/annoy) to build an index that stores the embeddings in a way that is optimized for fast search. This approach scales well to a large number of texts (other options include [Faiss](https://github.com/facebookresearch/faiss), [ScaNN](https://github.com/google-research/google-research/tree/master/scann), and [PyNNDescent](https://github.com/lmcinnes/pynndescent)). | |
| After building the index, we can use it to retrieve the nearest neighbors either of existing questions (section 3.1), or of new questions that we embed (section 3.2). | |
| """ | |
| # Create the search index, pass the size of embedding | |
| search_index = AnnoyIndex(embeds.shape[1], 'angular') | |
| print(search_index) | |
| # Add all the vectors to the search index | |
| for i in range(len(embeds)): | |
| search_index.add_item(i, embeds[i]) | |
| print(search_index) | |
| search_index.build(10) # 10 trees | |
| search_index.save('test.ann') | |
| """### 3.1. Find the neighbors of an example from the dataset | |
| If we're only interested in measuring the distance between the questions in the dataset (no outside queries), a simple way is to calculate the distance between every pair of embeddings we have. | |
| """ | |
| # Choose an example (we'll retrieve others similar to it) | |
| example_id = 5 | |
| # Retrieve nearest neighbors | |
| similar_item_ids = search_index.get_nns_by_item(example_id,10, | |
| include_distances=True) | |
| # Format and print the text and distances | |
| results = pd.DataFrame(data={'texts': df.iloc[similar_item_ids[0]]['text'], | |
| 'distance': similar_item_ids[1]}).drop(example_id) | |
| print(f"Question:'{df.iloc[example_id]['text']}'\nNearest neighbors:") | |
| results | |
| """### 3.2. Find the neighbors of a user query | |
| We're not limited to searching using existing items. If we get a query, we can embed it and find its nearest neighbors from the dataset. | |
| """ | |
| # query = "skin care ayurveda" | |
| # query = "how much money did skin care ayurveda raise" | |
| # query = "semelso wife arrest" | |
| # query = "avatar 2 movie collection" | |
| # query = "బాలయ్య మాస్ ట్రీట్" | |
| def multilingual_semantic_search(query): | |
| # query = "is messi the best footballer of all time?" | |
| # Get the query's embedding | |
| query_embed = co.embed(texts=[query], | |
| model="multilingual-22-12", | |
| truncate="LEFT").embeddings | |
| # Retrieve the nearest neighbors | |
| similar_item_ids = search_index.get_nns_by_vector(query_embed[0],10, | |
| include_distances=True) | |
| # Format the results | |
| # results = pd.DataFrame(data={'texts': df.iloc[similar_item_ids[0]]['text'], | |
| # 'distance': similar_item_ids[1]}) | |
| results = pd.DataFrame(data={'title': df.iloc[similar_item_ids[0]]['Title '], | |
| 'news': df.iloc[similar_item_ids[0]]['News'], | |
| 'distance': similar_item_ids[1]}) | |
| response = {} | |
| # JSON response | |
| # for i in similar_item_ids[0]: | |
| # # print(i) | |
| # response[i] = \ | |
| # { \ | |
| # "title": df.iloc[i]['Title '], \ | |
| # "news": df.iloc[i]['News'] | |
| # } | |
| response = """ """ | |
| for i in similar_item_ids[0]: | |
| # print(i) | |
| response += "Title: " + df.iloc[i]['Title '] + " \n " +"Short News: "+ df.iloc[i]['News'] + "\n\n" | |
| # print(similar_item_ids) | |
| # print(similar_item_ids[0]) | |
| # print(similar_item_ids[1]) | |
| # print(f"Query:'{query}'\nNearest neighbors:") | |
| # print(results) | |
| # print("----------------------") | |
| # print(type(response)) | |
| print(response) | |
| return response | |
| multilingual_semantic_search("is messi the best footballer of all time?") | |
| #!pip install gradio | |
| import gradio as gr | |
| # demo = gr.Interface(fn=multilingual_semantic_search, inputs="text", outputs="text") | |
| with gr.Blocks() as demo: | |
| gr.Markdown("🌍 This app uses a multilingual semantic model from COhere to 🚀 revolutionize the media and news industry in multilingual markets like India, allowing anyone to track 📰 regional news in real-time without the need for translation or understanding of other regional languages. 🙌") | |
| name = gr.Textbox(label="*Semantic search enable! Search for a news...") | |
| output = gr.Textbox(label="Semantic search results") | |
| greet_btn = gr.Button("Search") | |
| theme="darkpeach" | |
| greet_btn.click(fn=multilingual_semantic_search, inputs=name, outputs=output) | |
| demo.launch() | |
| #!pip install gradio | |
| """## 4. Visualizing the archive | |
| Finally, let's plot out all the questions onto a 2D chart so you're able to visualize the semantic similarities of this dataset! | |
| """ | |
| #@title Plot the archive {display-mode: "form"} | |
| # UMAP reduces the dimensions from 1024 to 2 dimensions that we can plot | |
| reducer = umap.UMAP(n_neighbors=20) | |
| umap_embeds = reducer.fit_transform(embeds) | |
| # Prepare the data to plot and interactive visualization | |
| # using Altair | |
| df_explore = pd.DataFrame(data={'text': df['text']}) | |
| df_explore['x'] = umap_embeds[:,0] | |
| df_explore['y'] = umap_embeds[:,1] | |
| # Plot | |
| chart = alt.Chart(df_explore).mark_circle(size=60).encode( | |
| x=#'x', | |
| alt.X('x', | |
| scale=alt.Scale(zero=False) | |
| ), | |
| y= | |
| alt.Y('y', | |
| scale=alt.Scale(zero=False) | |
| ), | |
| tooltip=['text'] | |
| ).properties( | |
| width=700, | |
| height=400 | |
| ) | |
| chart.interactive() | |
| """Hover over the points to read the text. Do you see some of the patterns in clustered points? Similar questions, or questions asking about similar topics? | |
| This concludes this introductory guide to semantic search using sentence embeddings. As you continue the path of building a search product additional considerations arise (like dealing with long texts, or finetuning to better improve the embeddings for a specific use case). | |
| We can’t wait to see what you start building! Share your projects or find support at [community.cohere.ai](https://community.cohere.ai). | |
| """ |