# Team members: Yifei Shen, Katherine Tang # My own contribution: I was responsible for handling Task3 in the coding tasks and deployed our web app to Hugging Face: https://huggingface.co/spaces/yifeis02/Search_Based_Retrieval_Demo ## Mini Project 1 - Part 1: Getting Familiar with Word Embeddings. # This assignment introduces students to text similarity measures using cosine similarity and sentence embeddings. # Students will implement and compare different methods for computing and analyzing text similarity using GloVe and Sentence Transformers. #Learning Objectives #By the end of this assignment, students will: #Understand how cosine similarity is used to measure text similarity. #Learn to encode sentences using GloVe embeddings and Sentence Transformers. #Compare the performance of different embedding techniques. #Create a Web interface for your model # Context: In this part, you are going to play around with some commonly used pretrained text embeddings for text search. For example, GloVe is an unsupervised learning algorithm for obtaining vector representations for words. Pretrained on # 2 billion tweets with vocabulary size of 1.2 million. Download from [Stanford NLP](http://nlp.stanford.edu/data/glove.twitter.27B.zip). # Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. *GloVe: Global Vectors for Word Representation*. ### Import necessary libraries: here you will use streamlit library to run a text search demo, please make sure to install it. import streamlit as st import numpy as np import numpy.linalg as la import pickle import os os.system("pip install gdown") import gdown from sentence_transformers import SentenceTransformer import matplotlib.pyplot as plt import math ### Some predefined utility functions for you to load the text embeddings # Function to Load Glove Embeddings def load_glove_embeddings(glove_path="Data/embeddings.pkl"): with open(glove_path, "rb") as f: embeddings_dict = pickle.load(f, encoding="latin1") return embeddings_dict def get_model_id_gdrive(model_type): if model_type == "25d": word_index_id = "13qMXs3-oB9C6kfSRMwbAtzda9xuAUtt8" embeddings_id = "1-RXcfBvWyE-Av3ZHLcyJVsps0RYRRr_2" elif model_type == "50d": embeddings_id = "1DBaVpJsitQ1qxtUvV1Kz7ThDc3az16kZ" word_index_id = "1rB4ksHyHZ9skes-fJHMa2Z8J1Qa7awQ9" elif model_type == "100d": word_index_id = "1-oWV0LqG3fmrozRZ7WB1jzeTJHRUI3mq" embeddings_id = "1SRHfX130_6Znz7zbdfqboKosz-PfNvNp" return word_index_id, embeddings_id def download_glove_embeddings_gdrive(model_type): # Get glove embeddings from google drive word_index_id, embeddings_id = get_model_id_gdrive(model_type) # Use gdown to get files from google drive embeddings_temp = "embeddings_" + str(model_type) + "_temp.npy" word_index_temp = "word_index_dict_" + str(model_type) + "_temp.pkl" # Download word_index pickle file print("Downloading word index dictionary....\n") gdown.download(id=word_index_id, output=word_index_temp, quiet=False) # Download embeddings numpy file print("Donwloading embedings...\n\n") gdown.download(id=embeddings_id, output=embeddings_temp, quiet=False) # @st.cache_data() def load_glove_embeddings_gdrive(model_type): word_index_temp = "word_index_dict_" + str(model_type) + "_temp.pkl" embeddings_temp = "embeddings_" + str(model_type) + "_temp.npy" # Load word index dictionary word_index_dict = pickle.load(open(word_index_temp, "rb"), encoding="latin") # Load embeddings numpy embeddings = np.load(embeddings_temp) return word_index_dict, embeddings @st.cache_resource() def load_sentence_transformer_model(model_name): sentenceTransformer = SentenceTransformer(model_name) return sentenceTransformer def get_sentence_transformer_embeddings(sentence, model_name="all-MiniLM-L6-v2"): """ Get sentence transformer embeddings for a sentence """ # 384 dimensional embedding # Default model: https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2 sentenceTransformer = load_sentence_transformer_model(model_name) try: return sentenceTransformer.encode(sentence) except: if model_name == "all-MiniLM-L6-v2": return np.zeros(384) else: return np.zeros(512) def get_glove_embeddings(word, word_index_dict, embeddings, model_type): """ Get glove embedding for a single word """ if word.lower() in word_index_dict: return embeddings[word_index_dict[word.lower()]] else: return np.zeros(int(model_type.split("d")[0])) def get_category_embeddings(embeddings_metadata): """ Get embeddings for each category 1. Split categories into words 2. Get embeddings for each word """ model_name = embeddings_metadata["model_name"] st.session_state["cat_embed_" + model_name] = {} for category in st.session_state.categories.split(" "): if model_name: if not category in st.session_state["cat_embed_" + model_name]: st.session_state["cat_embed_" + model_name][category] = get_sentence_transformer_embeddings(category, model_name=model_name) else: if not category in st.session_state["cat_embed_" + model_name]: st.session_state["cat_embed_" + model_name][category] = get_sentence_transformer_embeddings(category) def update_category_embeddings(embeddings_metadata): """ Update embeddings for each category """ get_category_embeddings(embeddings_metadata) ### Plotting utility functions def plot_piechart(sorted_cosine_scores_items): sorted_cosine_scores = np.array([ sorted_cosine_scores_items[index][1] for index in range(len(sorted_cosine_scores_items)) ] ) categories = st.session_state.categories.split(" ") categories_sorted = [ categories[sorted_cosine_scores_items[index][0]] for index in range(len(sorted_cosine_scores_items)) ] fig, ax = plt.subplots() ax.pie(sorted_cosine_scores, labels=categories_sorted, autopct="%1.1f%%") st.pyplot(fig) # Figure def plot_piechart_helper(sorted_cosine_scores_items): sorted_cosine_scores = np.array( [ sorted_cosine_scores_items[index][1] for index in range(len(sorted_cosine_scores_items)) ] ) categories = st.session_state.categories.split(" ") categories_sorted = [ categories[sorted_cosine_scores_items[index][0]] for index in range(len(sorted_cosine_scores_items)) ] fig, ax = plt.subplots(figsize=(3, 3)) my_explode = np.zeros(len(categories_sorted)) my_explode[0] = 0.2 if len(categories_sorted) == 3: my_explode[1] = 0.1 # explode this by 0.2 elif len(categories_sorted) > 3: my_explode[2] = 0.05 ax.pie( sorted_cosine_scores, labels=categories_sorted, autopct="%1.1f%%", explode=my_explode, ) return fig def plot_piecharts(sorted_cosine_scores_models): scores_list = [] categories = st.session_state.categories.split(" ") index = 0 for model in sorted_cosine_scores_models: scores_list.append(sorted_cosine_scores_models[model]) # scores_list[index] = np.array([scores_list[index][ind2][1] for ind2 in range(len(scores_list[index]))]) index += 1 if len(sorted_cosine_scores_models) == 2: fig, (ax1, ax2) = plt.subplots(2) categories_sorted = [ categories[scores_list[0][index][0]] for index in range(len(scores_list[0])) ] sorted_scores = np.array( [scores_list[0][index][1] for index in range(len(scores_list[0]))] ) ax1.pie(sorted_scores, labels=categories_sorted, autopct="%1.1f%%") categories_sorted = [ categories[scores_list[1][index][0]] for index in range(len(scores_list[1])) ] sorted_scores = np.array( [scores_list[1][index][1] for index in range(len(scores_list[1]))] ) ax2.pie(sorted_scores, labels=categories_sorted, autopct="%1.1f%%") st.pyplot(fig) def plot_alatirchart(sorted_cosine_scores_models): models = list(sorted_cosine_scores_models.keys()) tabs = st.tabs(models) figs = {} for model in models: figs[model] = plot_piechart_helper(sorted_cosine_scores_models[model]) for index in range(len(tabs)): with tabs[index]: st.pyplot(figs[models[index]]) ### Your Part To Complete: Follow the instructions in each function below to complete the similarity calculation between text embeddings # Task I: Compute Cosine Similarity def cosine_similarity(x, y): """ Exponentiated cosine similarity 1. Compute cosine similarity 2. Exponentiate cosine similarity 3. Return exponentiated cosine similarity (20 pts) """ ################################## ### TODO: Add code here ########## ################################## # Compute cosine similarity cos_sim = np.dot(x, y) / (la.norm(x) * la.norm(y) + 1e-6) # Exponentiate cosine similarity return math.exp(cos_sim) # Task II: Average Glove Embedding Calculation def averaged_glove_embeddings_gdrive(sentence, word_index_dict, embeddings, model_type=50): """ Get averaged glove embeddings for a sentence 1. Split sentence into words 2. Get embeddings for each word 3. Add embeddings for each word 4. Divide by number of words 5. Return averaged embeddings (30 pts) """ embedding = np.zeros(int(model_type.split("d")[0])) ################################## ##### TODO: Add code here ######## ################################## words = sentence.split() count = 0 for word in words: word_embedding = get_glove_embeddings(word, word_index_dict, embeddings, model_type) if word_embedding is not None: embedding += word_embedding count += 1 if count > 0: embedding /= count return embedding # Task III: Sort the cosine similarity def get_sorted_cosine_similarity(_, embeddings_metadata): """ Get sorted cosine similarity between input sentence and categories Steps: 1. Get embeddings for input sentence 2. Get embeddings for categories (if not found, update category embeddings) 3. Compute cosine similarity between input sentence and categories 4. Sort cosine similarity 5. Return sorted cosine similarity (50 pts) """ categories = st.session_state.categories.split(" ") cosine_sim = {} if embeddings_metadata["embedding_model"] == "glove": word_index_dict = embeddings_metadata["word_index_dict"] embeddings = embeddings_metadata["embeddings"] model_type = embeddings_metadata["model_type"] input_embedding = averaged_glove_embeddings_gdrive(st.session_state.text_search, word_index_dict, embeddings, model_type) ########################################## ## TODO: Get embeddings for categories ### ########################################## for index, category in enumerate(categories): category_embedding = averaged_glove_embeddings_gdrive( category, word_index_dict, embeddings, model_type ) sim_val = cosine_similarity(input_embedding, category_embedding) cosine_sim[index] = sim_val else: model_name = embeddings_metadata["model_name"] if not "cat_embed_" + model_name in st.session_state: get_category_embeddings(embeddings_metadata) category_embeddings = st.session_state["cat_embed_" + model_name] print("text_search = ", st.session_state.text_search) if model_name: input_embedding = get_sentence_transformer_embeddings(st.session_state.text_search, model_name=model_name) else: input_embedding = get_sentence_transformer_embeddings(st.session_state.text_search) for index in range(len(categories)): ########################################## # TODO: Compute cosine similarity between input sentence and categories # TODO: Update category embeddings if category not found ########################################## category = categories[index] if category not in category_embeddings: if model_name: category_embeddings[category] = get_sentence_transformer_embeddings(category, model_name=model_name) else: category_embeddings[category] = get_sentence_transformer_embeddings(category) sim_val = cosine_similarity(input_embedding, category_embeddings[category]) cosine_sim[index] = sim_val sorted_cosine_sim = sorted(cosine_sim.items(), key=lambda x: x[1], reverse=True) return sorted_cosine_sim ### Below is the main function, creating the app demo for text search engine using the text embeddings. if __name__ == "__main__": ### Text Search ### ### There will be Bonus marks of 10% for the teams that submit a URL for your deployed web app. ### Bonus: You can also submit a publicly accessible link to the deployed web app. ### This is our deployed web app: https://huggingface.co/spaces/yifeis02/Search_Based_Retrieval_Demo st.sidebar.title("GloVe Twitter") st.sidebar.markdown( """ GloVe is an unsupervised learning algorithm for obtaining vector representations for words. Pretrained on 2 billion tweets with vocabulary size of 1.2 million. Download from [Stanford NLP](http://nlp.stanford.edu/data/glove.twitter.27B.zip). Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. *GloVe: Global Vectors for Word Representation*. """ ) model_type = st.sidebar.selectbox("Choose the model", ("25d", "50d", "100d"), index=1) st.title("Search Based Retrieval Demo") st.subheader( "Pass in space separated categories you want this search demo to be about." ) # st.selectbox(label="Pick the categories you want this search demo to be about...", # options=("Flowers Colors Cars Weather Food", "Chocolate Milk", "Anger Joy Sad Frustration Worry Happiness", "Positive Negative"), # key="categories" # ) st.text_input( label="Categories", key="categories", value="Flowers Colors Cars Weather Food" ) print(st.session_state["categories"]) print(type(st.session_state["categories"])) # print("Categories = ", categories) # st.session_state.categories = categories st.subheader("Pass in an input word or even a sentence") text_search = st.text_input( label="Input your sentence", key="text_search", value="Roses are red, trucks are blue, and Seattle is grey right now", ) # st.session_state.text_search = text_search # Download glove embeddings if it doesn't exist embeddings_path = "embeddings_" + str(model_type) + "_temp.npy" word_index_dict_path = "word_index_dict_" + str(model_type) + "_temp.pkl" if not os.path.isfile(embeddings_path) or not os.path.isfile(word_index_dict_path): print("Model type = ", model_type) glove_path = "Data/glove_" + str(model_type) + ".pkl" print("glove_path = ", glove_path) # Download embeddings from google drive with st.spinner("Downloading glove embeddings..."): download_glove_embeddings_gdrive(model_type) # Load glove embeddings word_index_dict, embeddings = load_glove_embeddings_gdrive(model_type) # Find closest word to an input word if st.session_state.text_search: # Glove embeddings print("Glove Embedding") embeddings_metadata = { "embedding_model": "glove", "word_index_dict": word_index_dict, "embeddings": embeddings, "model_type": model_type, } with st.spinner("Obtaining Cosine similarity for Glove..."): sorted_cosine_sim_glove = get_sorted_cosine_similarity( st.session_state.text_search, embeddings_metadata ) # Sentence transformer embeddings print("Sentence Transformer Embedding") embeddings_metadata = {"embedding_model": "transformers", "model_name": ""} with st.spinner("Obtaining Cosine similarity for 384d sentence transformer..."): sorted_cosine_sim_transformer = get_sorted_cosine_similarity( st.session_state.text_search, embeddings_metadata ) # Results and Plot Pie Chart for Glove print("Categories are: ", st.session_state.categories) st.subheader( "Closest word I have between: " + st.session_state.categories + " as per different Embeddings" ) print(sorted_cosine_sim_glove) print(sorted_cosine_sim_transformer) # print(sorted_distilbert) # Altair Chart for all models plot_alatirchart( { "glove_" + str(model_type): sorted_cosine_sim_glove, "sentence_transformer_384": sorted_cosine_sim_transformer, } ) # "distilbert_512": sorted_distilbert}) st.write("") st.write( "Demo developed by [Yifei Shen](https://www.linkedin.com/in/yifei-shen-7b7973270/)" )