Add files via upload

Author: MainakRepositor

__pycache__/func_file.cpython-310.pyc

@@ -0,0 +1,45 @@
+import streamlit as st
+import numpy as np
+import pandas as pd
+import os
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+from func_file import *
+from string import punctuation
+from nltk.corpus import stopwords
+
+#####
+df = pd.read_csv('prog_book.csv')
+import nltk
+nltk.download('stopwords')
+stop = stopwords.words('english')
+stop = set(stop)
+
+df['clean_Book_title']=df['Book_title'].apply(clean_text)
+df['clean_Description']=df['Description'].apply(clean_text)
+
+vectorizer = TfidfVectorizer(analyzer='word', lowercase=False)
+X = vectorizer.fit_transform(df['clean_Book_title'])
+title_vectors = X.toarray()
+
+tk = 0
+st.title('Book Recommendation System')
+
+col1, col2 = st.columns(2)
+#taking book name as input
+with col1:
+    book = st.text_input('Enter book name that you liked : ')
+    
+#taking multiple fiels to get similarity
+with col2:
+    feat = st.selectbox("Select Mode : ",['Book_title', 'Rating', 'Price'])
+    if st.button('Search'):
+        tk = 1
+
+#st.dataframe(df.head(10))
+if tk == 1:
+    st.success('Recommending books similar to '+book)
+    rec = st.empty()
+    rec = st.dataframe(get_recommendations(book, 'Book_title', df, title_vectors, feat), width=700, height=76)
+
+#print(get_recommendations())

app.py

@@ -0,0 +1,223 @@
+# To add a new cell, type '# %%'
+# To add a new markdown cell, type '# %% [markdown]'
+# %%
+# This Python 3 environment comes with many helpful analytics libraries installed
+# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python
+# For example, here's several helpful packages to load
+
+import numpy as np # linear algebra
+import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
+
+# Input data files are available in the read-only "../input/" directory
+# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory
+
+import os
+#for dirname, _, filenames in os.walk('/kaggle/input'):
+#    for filename in filenames:
+#        print(os.path.join(dirname, filename))
+
+# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using "Save & Run All" 
+# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session
+
+
+# %%
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+
+
+# %%
+df = pd.read_csv('prog_book.csv')
+#/kaggle/input/top-270-rated-computer-science-programing-books/
+
+# %% [markdown]
+# Let's take a look at our programming books dataset:
+
+# %%
+df.head()
+
+# %% [markdown]
+# We can use "Book_title" and "Description" columns to find books similar to each other.
+# %% [markdown]
+# # Text preprocessing
+
+# %%
+import nltk
+nltk.download('stopwords')
+from nltk.corpus import stopwords
+stop = stopwords.words('english')
+
+# Set of stopwords to remove
+stop = set(stop)
+
+# Set of punctuation signs to remove
+from string import punctuation
+
+# %% [markdown]
+# We'll be using this small set of functions for text preprocessing:
+
+# %%
+import re
+
+def lower(text):
+    return text.lower()
+
+def remove_punctuation(text):
+    return text.translate(str.maketrans('','', punctuation))
+
+def remove_stopwords(text):
+    return " ".join([word for word in str(text).split() if word not in stop])
+
+# Removing all words with digits and standalone digits
+def remove_digits(text):
+    return re.sub(r'\d+', '', text)
+
+# One function to clean it all
+def clean_text(text):
+    text = lower(text)
+    text = remove_punctuation(text)
+    text = remove_stopwords(text)
+    text = remove_digits(text)
+    return text
+
+# %% [markdown]
+# And then, we'll create new columns with cleaned "Book_title" and "Description" texts:
+
+# %%
+df['clean_Book_title']=df['Book_title'].apply(clean_text)
+df.head()
+
+
+# %%
+df['clean_Description']=df['Description'].apply(clean_text)
+df.head()
+
+# %% [markdown]
+# # Creating features
+# Now, we need to transform text from "Book_title" to vectors array:
+
+# %%
+# Initializing vectorizer
+vectorizer = TfidfVectorizer(analyzer='word', lowercase=False)
+
+# Applying vectorized to clean text
+X = vectorizer.fit_transform(df['clean_Book_title'])
+
+# Getting array with vectorized titles
+title_vectors = X.toarray()
+title_vectors
+
+# %% [markdown]
+# Let's do the same with "Description" column:
+
+# %%
+desc_vectorizer = TfidfVectorizer(analyzer='word', lowercase=False)
+Y = desc_vectorizer.fit_transform(df['clean_Description'])
+desc_vectors = Y.toarray()
+desc_vectors
+
+# %% [markdown]
+# And now we have two arrays of vectors ready for work.
+
+# %%
+# List of titles for use
+# df['Book_title'].tolist()
+
+# %% [markdown]
+# # Recommendation system
+# 
+
+# %%
+def get_recommendations(value_of_element, feature_locate, df, vectors_array, feature_show):
+    """Returns DataFrame with particular feature of target and the same feature of five objects similar to it.
+
+    value_of_element     - unique value of target object
+    feature_locate       - name of the feature which this unique value belongs to
+    df                   - DataFrame with feautures
+    vectors_array        - array of vectorized text used to find similarity
+    feature_show         - feature that will be shown in final DataFrame
+    """
+    
+    # Locating target element by its specific value
+    index_of_element = df[df[feature_locate]==value_of_element].index.values[0]
+    
+    # Finding its value to show
+    show_value_of_element = df.iloc[index_of_element][feature_show]
+
+    # Dropping target element from df
+    df_without = df.drop(index_of_element).reset_index().drop(['index'], axis=1)
+
+    # Dropping target element from vectors array
+    vectors_array = list(vectors_array)
+    target = vectors_array.pop(index_of_element).reshape(1,-1)
+    vectors_array = np.array(vectors_array)
+
+    # Finding cosine similarity between vectors
+    most_similar_sklearn = cosine_similarity(target, vectors_array)[0]
+
+    # Sorting coefs in desc order 
+    idx = (-most_similar_sklearn).argsort()
+
+    # Finding features of similar objects by index
+    all_values = df_without[[feature_show]]
+    for index in idx:
+        simular = all_values.values[idx]
+     
+    recommendations_df = pd.DataFrame({feature_show: show_value_of_element,
+                                    "rec_1": simular[0][0],
+                                    "rec_2": simular[1][0],
+                                    "rec_3": simular[2][0],
+                                    "rec_4": simular[3][0],
+                                    "rec_5": simular[4][0]}, index=[0])
+    
+
+    return recommendations_df
+
+# %% [markdown]
+# Ok, let's find books similar to "Algorithms" book based on the title:
+
+# %%
+get_recommendations("Algorithms", 'Book_title', df, title_vectors, 'Book_title')
+
+# %% [markdown]
+# We can also look at their prices:
+
+# %%
+get_recommendations("Algorithms", 'Book_title', df, title_vectors, 'Price')
+
+# %% [markdown]
+# Or ratings:
+
+# %%
+get_recommendations("Algorithms", 'Book_title', df, title_vectors, 'Rating')
+
+# %% [markdown]
+# Now, let's find books similar to "Algorithms" book based on the description:
+
+# %%
+get_recommendations("Algorithms", 'Book_title', df, desc_vectors, 'Book_title')
+
+# %% [markdown]
+# As you can see, recommendations based on description are different from title-based recommendations in some ways.
+
+# %%
+get_recommendations("Unity in Action", 'Book_title', df, desc_vectors, 'Book_title')
+
+
+# %%
+get_recommendations("Unity in Action", 'Book_title', df, title_vectors, 'Book_title')
+
+# %% [markdown]
+# We can also access some book by any unique value, for example, by number of reviwes (or, more logically, ID of the book, if there's some):
+
+# %%
+get_recommendations("1,406", 'Reviews',  df, title_vectors, 'Book_title')
+
+
+# %%
+get_recommendations("The Information: A History, a Theory, a Flood", 'Book_title', df, title_vectors, 'Book_title')
+
+
+# %%
+
+
+