third

language-detector-models/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:279792554b0bc74e77f1bdab5b46c9962ce026ef56a814e7da0bb9edb7911545
+size 2176914

language-detector-models/vectorizer.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc63d514d2c6cf98e7a5c44f6edcab0189be7195878bb87ba7a23015746a1209
+size 517916

pages/2_📞_CONTACT.py

@@ -0,0 +1,38 @@
+import streamlit as st
+st.set_page_config(
+    page_title="NLP WEB APP"
+)
+
+def local_css(file_name):
+    with open(file_name) as f:
+        st.markdown(f"<style>{f.read()}</style>", unsafe_allow_html=True)
+
+local_css("style/style.css")
+
+
+st.title("CONTACT US")
+st.sidebar.success("Select a page above")
+
+with st.container():
+    st.write("-----")
+    st.header("Get in Touch with Me !")
+    st.write("##")
+
+    contact_form="""
+
+     <form action="https://formsubmit.co/cus146126@gmail.com" method="POST">
+     <input type="hidden" name="_captcha" value="false">
+     <input type="text" name="name" placeholder="Your name" required>
+     <input type="email" name="email" placeholder="Your email" required>
+     <textarea name="message" placeholder="Your message here ... " required></textarea>
+
+     <button type="submit">Send</button>
+     </form>
+
+ """
+
+left_col , right_col = st.columns(2)
+with left_col:
+    st.markdown(contact_form, unsafe_allow_html=True)
+with right_col:
+    st.empty()

pages/3_🙎_RESUME.py

@@ -0,0 +1,120 @@
+from pathlib import Path
+
+import streamlit as st
+from PIL import Image
+st.set_page_config(
+    page_title="NLP WEB APP"
+)
+
+
+# --- PATH SETTINGS ---
+current_dir = Path(__file__).parent if "__file__" in locals() else Path.cwd()
+css_file = current_dir / "styles" / "main.css"
+resume_file = current_dir / "assets" / "my_resume.pdf"
+profile_pic = current_dir / "assets" / "profile-pic.png"
+
+
+# --- GENERAL SETTINGS ---
+PAGE_TITLE = "Digital CV | John Doe"
+PAGE_ICON = ":wave:"
+NAME = "SUDHANSHU"
+DESCRIPTION = """
+Aspiring Data Scientist | 18-Year-Old Data Enthusiast | 1 Year of Hands-On Experience | Passionate about Solving Real-World Problems"
+"""
+EMAIL = "gusainsudhanshu43@gmail.com"
+SOCIAL_MEDIA = {
+    "YouTube": "https://youtube.com/",
+    "LinkedIn": "https://www.linkedin.com/in/sudhanshu-gusain-34271028a/",
+    "GitHub": "https://github.com/sudhanshu976",
+    "Twitter": "https://twitter.com",
+}
+PROJECTS = {
+    "🏆 POWER-BI Dashboards - Making interactive and dynamic dashboards": "https://github.com/sudhanshu976/POWER-BI-PROJECTS",
+    "🏆 Potato Disease Classifier using CNN - Checks whether a given potato leaf is healthy , early-blight or late-blight": "https://github.com/sudhanshu976/POTATO-DISEASE-CLASSIFIER-WITH-DEPLOYMENT",
+    "🏆 Combined NLP WEB APP - This web app contains all NLP Projects I have made till date ": "https://github.com/sudhanshu976/NLP_FULL",
+}
+
+
+
+
+# --- LOAD CSS, PDF & PROFIL PIC ---
+with open(css_file) as f:
+    st.markdown("<style>{}</style>".format(f.read()), unsafe_allow_html=True)
+with open(resume_file, "rb") as pdf_file:
+    PDFbyte = pdf_file.read()
+profile_pic = Image.open(profile_pic)
+
+
+# --- HERO SECTION ---
+col1, col2 = st.columns(2, gap="small")
+with col1:
+    st.image(profile_pic, width=230)
+
+with col2:
+    st.title(NAME)
+    st.write(DESCRIPTION)
+    st.download_button(
+        label=" 📄 Download Resume",
+        data=PDFbyte,
+        file_name=resume_file.name,
+        mime="application/octet-stream",
+    )
+    st.write("📫", EMAIL)
+
+
+# --- SOCIAL LINKS ---
+st.write('\n')
+cols = st.columns(len(SOCIAL_MEDIA))
+for index, (platform, link) in enumerate(SOCIAL_MEDIA.items()):
+    cols[index].write(f"[{platform}]({link})")
+
+
+# --- EXPERIENCE & QUALIFICATIONS ---
+st.write('\n')
+st.subheader("Experience & Qulifications")
+st.write(
+    """
+- ✔️ 1 Year expereince of performing various Data Science and NLP tasks
+- ✔️ Strong hands on experience and knowledge in Python , ML , DL and NLP
+- ✔️ Good understanding of statistical principles and their respective applications
+- ✔️ Excellent team-player and displaying strong sense of initiative on tasks
+"""
+)
+
+
+# --- SKILLS ---
+st.write('\n')
+st.subheader("Hard Skills")
+st.write(
+    """
+- 👩‍💻 Programming: Python (Scikit-learn, Pandas , Numpy , Pytorch , Tensorflow)
+- 📊 Data Visulization: PowerBi, Matplotlib , Seaborn
+- 📚 Modeling: Supervised and Unsupervised ML algorithms , ANN , RNN , CNN
+- 🗄️ Databases: MySQL
+- 🗄️ WEB DEPLOYMENT: FLASK , Streamlit , Heroku
+"""
+)
+
+
+# --- WORK HISTORY ---
+st.write('\n')
+st.subheader("Work History")
+st.write("---")
+
+# --- JOB 1
+st.write("🚧", "**Freelancer Data Scientist and NLP Engineer**")
+st.write("05/2023 - Present")
+st.write(
+    """
+- ► Used PowerBI for creating interactive dashboards 
+- ► Solved many ML , DL and NLP problems in various fields like medical , agriculture , etc
+- ► Well versed in solving real life problems especially using NLP
+"""
+)
+
+# --- Projects & Accomplishments ---
+st.write('\n')
+st.subheader("Projects & Accomplishments")
+st.write("---")
+for project, link in PROJECTS.items():
+    st.write(f"[{project}]({link})")

pages/4_LANGUAGE-DETECTOR-MODEL.py

@@ -0,0 +1,88 @@
+import streamlit as st
+import pickle
+import re
+import string
+import nltk
+from nltk.corpus import stopwords
+from nltk.tokenize import word_tokenize
+from nltk.stem.porter import PorterStemmer
+stemmer = PorterStemmer()
+
+st.set_page_config(
+    page_title="NLP WEB APP"
+)
+
+st.title("LANGUAGE DETECTOR MODEL")
+st.sidebar.success("Select a page above")
+nltk.download('stopwords')
+nltk.download('punkt')
+
+def preprocess(text):
+    text = text.lower()
+    text = re.sub(r'\d+', '', text)
+    translator = str.maketrans('', '', string.punctuation)
+    text = text.translate(translator)
+    
+    
+    stop_words = set(stopwords.words("english"))
+    word_tokens = word_tokenize(text)
+    filtered_text = [word for word in word_tokens if word not in stop_words]
+    
+    stems = [stemmer.stem(word) for word in filtered_text]
+    preprocessed_text = ' '.join(stems)
+    return  preprocessed_text
+
+
+
+cv = pickle.load(open('language-detector-models/vectorizer.pkl','rb'))
+model = pickle.load(open('language-detector-models/model.pkl','rb'))
+
+message= st.text_input("ENTER THE MESSAGE")
+ 
+
+if st.button("PREDICT"):
+    # PREPROCESS 
+    transformed_text = preprocess(message)
+
+    # VECTORIZE
+    vector_input = cv.transform([message])
+
+    # PREDICTION
+    result = model.predict(vector_input)[0]
+
+
+    # DISPLAY
+    if result==0:
+       st.header("ARABIC")
+    elif result==1:
+       st.header("DANISH")
+    elif result==2:
+       st.header("DUTCH")
+    elif result==3:
+       st.header("ENGLISH")
+    elif result==4:
+       st.header("FRENCH")
+    elif result==5:
+       st.header("GERMAN")
+    elif result==6:
+       st.header("GREEK")
+    elif result==7:
+       st.header("HINDI")
+    elif result==8:
+       st.header("ITALIAN")
+    elif result==9:
+       st.header("KANNADA")
+    elif result==10:
+       st.header("MALYALAM")
+    elif result==11:
+       st.header("PORTUGESE")
+    elif result==12:
+       st.header("RUSSIAN")
+    elif result==13:
+       st.header("SPANISH")
+    elif result==14:
+       st.header("SWEDISH")
+    elif result==15:
+       st.header("TAMIL")
+    else:
+       st.header("TURKISH")

pages/5_SENTIMENT-ANALYZER.py

@@ -0,0 +1,40 @@
+import streamlit as st
+from transformers import AutoTokenizer , AutoModelForSequenceClassification
+import torch
+
+
+tokenizer = AutoTokenizer.from_pretrained('nlptown/bert-base-multilingual-uncased-sentiment')
+model = AutoModelForSequenceClassification.from_pretrained('nlptown/bert-base-multilingual-uncased-sentiment')
+
+st.set_page_config(
+    page_title="NLP WEB APP"
+)
+
+
+st.title("SENTIMENT ANALYZER")
+st.sidebar.success("Select a page above")
+
+
+message= st.text_input("ENTER THE MESSAGE")
+
+if st.button("PREDICT"):
+    tokens  = tokenizer.encode(message , return_tensors='pt')
+    output = model(tokens)
+    result = int(torch.argmax(output.logits))+1
+
+
+    if result==1:
+       st.header("TOO MUCH NEGATIVE STATEMENT")
+       st.header("RATING : ⭐ ")
+    elif result==2:
+       st.header("NEGATIVE STATEMENT")
+       st.header("RATING : ⭐⭐")
+    elif result==3:
+       st.header("NEUTRAL STATEMENT")
+       st.header("RATING : ⭐⭐⭐")
+    elif result==4:
+       st.header("POSITIVE STATEMENT")
+       st.header("RATING : ⭐⭐⭐⭐ ")
+    elif result==5:
+       st.header("TOO MUCH POSITIVE STATEMENT")
+       st.header("RATING : ⭐⭐⭐⭐⭐ ")

pages/6_SMS-SPAM-CLASSIFIER.py

@@ -0,0 +1,75 @@
+import streamlit as st
+
+st.set_page_config(
+    page_title="NLP WEB APP"
+)
+
+st.title("SMS SPAM CLASSIFIER")
+st.sidebar.success("Select a page above")
+
+
+
+import pickle
+import string
+
+from nltk.corpus import stopwords
+import nltk
+nltk.download('punkt')
+nltk.download('stopwords')
+
+from nltk.stem.porter import PorterStemmer
+ps = PorterStemmer()
+
+def transform_text(text):
+    # Lowering 
+    text = text.lower()
+    
+    #Tokenizing
+    text = nltk.word_tokenize(text)
+    
+    #Removing special characters
+    new_text=[]
+    for word in text:
+        if word.isalnum():
+            new_text.append(word)
+    
+    text = new_text[:]
+    new_text.clear()
+    
+    #Removing stopwords and punctuation
+    for word in text:
+        if word not in stopwords.words('english') and word not in string.punctuation:
+            new_text.append(word)
+            
+            
+    text = new_text[:]
+    new_text.clear()
+    
+    # Stemming
+    for word in text :
+        new_text.append(ps.stem(word))
+        
+    return " ".join(new_text)
+
+tfidf = pickle.load(open('sms-spam-models/vectorizer.pkl','rb'))
+model = pickle.load(open('sms-spam-models/model.pkl','rb'))
+
+sms = st.text_input("ENTER THE MESSAGE")
+
+if st.button("PREDICT"):
+    # PREPROCESS 
+    transformed_sms = transform_text(sms)
+
+    # VECTORIZE
+    vector_input = tfidf.transform([transformed_sms])
+
+    # PREDICTION
+    result = model.predict(vector_input)[0]
+
+
+    # DISPLAY
+
+    if result==1:
+        st.header("SPAM")
+    else:
+        st.header("NOT SPAM")

pages/7_NEXT_WORD_PREDICTOR.py

@@ -0,0 +1,74 @@
+import streamlit as st
+import numpy as np
+
+
+st.set_page_config(
+    page_title="NLP WEB APP"
+)
+
+st.title("NEXT WORD PREDICTOR")
+st.sidebar.success("Select a page above")
+
+
+string1 = st.text_area("Enter the training text   (Note : This may take time depending upon the data size )")
+
+
+
+test = st.text_input("ENTER THE WORD")
+number = st.number_input("Enter the number of next words" )
+number = int(number)
+
+
+import tensorflow as tf
+import numpy as np
+from tensorflow.keras.preprocessing.text import Tokenizer
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+from tensorflow.keras.utils import to_categorical
+from tensorflow.keras.models import Sequential
+from tensorflow.keras.layers import Embedding,Dense,LSTM
+
+if st.button("PREDICT"):
+    tokenizer = Tokenizer()
+    tokenizer.fit_on_texts([string1])
+
+    input_sequences =[]
+    for sentence in string1.split("\n"):
+        tokenized_sentences = tokenizer.texts_to_sequences([sentence])[0]
+        
+        for i in range(1,len(tokenized_sentences)):
+            input_sequences.append(tokenized_sentences[:i+1])
+
+
+    max_len = max([len(x) for x in input_sequences])
+
+    padded_input_sentences = pad_sequences(input_sequences , maxlen = max_len , padding ="pre")
+    X = padded_input_sentences[:,:-1]
+    Y = padded_input_sentences[:,-1]
+    num_class = len(tokenizer.word_index)
+    Y = to_categorical(Y , num_classes=num_class+1)
+
+
+
+    model = Sequential()
+    model.add(Embedding(num_class+1,100,input_shape = (X.shape[1],)))
+
+    model.add(LSTM(250))
+    
+    model.add(Dense(num_class+1,activation ="softmax"))
+
+    model.compile(loss="categorical_crossentropy" , optimizer="adam" , metrics=["accuracy"])
+    
+
+    model.fit(X,Y,epochs=100)
+
+    for i in  range(number):
+    
+
+        output_token = tokenizer.texts_to_sequences([test])[0]
+        padded_token = pad_sequences([output_token] , maxlen=max_len,padding="pre")
+        output = np.argmax(model.predict(padded_token))
+        for word,index in tokenizer.word_index.items():
+            if index == output:
+                test =test + " " + word
+                     
+    st.header(test) 

pages/styles/main.css

@@ -0,0 +1,27 @@
+@import url('https://fonts.googleapis.com/css2?family=Readex+Pro:wght@300;400;500;600;700&display=swap');
+
+
+* {font-family: 'Readex Pro';}
+
+
+a {
+    text-decoration: none;
+    color: white !important;
+    font-weight: 500;
+}
+
+a:hover {
+    color: #d33682 !important;
+    text-decoration: none;
+}
+
+ul {list-style-type: none;}
+
+hr {
+    margin-top: 0px;
+    margin-bottom: 5%;
+}
+
+#MainMenu {visibility: hidden;}
+footer {visibility: hidden;}
+header {visibility: hidden;}

sms-spam-models/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:208d6e1dcb73e5cdaf76e165c69821004bed116c25080eae843df1f4092c138c
+size 96605

sms-spam-models/vectorizer.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6acf5b90ca6f2a5cd6c978ff7101a497849b20bba64844028da4cc0d337ebf01
+size 160670

style/style.css

@@ -0,0 +1,33 @@
+/* CSS Snippet from W3schools: https://www.w3schools.com/howto/howto_css_contact_form.asp */
+/* Style inputs with type="text", select elements and textareas */
+input[type=message], input[type=email], input[type=text], textarea {
+    width: 100%; /* Full width */
+    padding: 12px; /* Some padding */ 
+    border: 1px solid #ccc; /* Gray border */
+    border-radius: 4px; /* Rounded borders */
+    box-sizing: border-box; /* Make sure that padding and width stays in place */
+    margin-top: 6px; /* Add a top margin */
+    margin-bottom: 16px; /* Bottom margin */
+    resize: vertical /* Allow the user to vertically resize the textarea (not horizontally) */
+  }
+  
+  /* Style the submit button with a specific background color etc */
+  button[type=submit] {
+    background-color: #04AA6D;
+    color: white;
+    padding: 12px 20px;
+    border: none;
+    border-radius: 4px;
+    cursor: pointer;
+  }
+  
+  /* When moving the mouse over the submit button, add a darker green color */
+  button[type=submit]:hover {
+    background-color: #45a049;
+  }
+  
+  
+  /* Hide Streamlit Branding */
+  #MainMenu {visibility: hidden;}
+  footer {visibility: hidden;}
+  header {visibility: hidden;}