update

app.py

@@ -1,8 +1,7 @@
-
 import gradio as gr
 from huggingface_hub import InferenceClient
 
-#STEP 1 FROM SEMANTIC SEARCH
+# STEP 1 FROM SEMANTIC SEARCH
 from sentence_transformers import SentenceTransformer
 import torch 
 
@@ -10,144 +9,99 @@ theme = gr.themes.Soft(
     primary_hue="rose",
     secondary_hue="zinc",
     neutral_hue="pink"
-    )
+)
 
 custom_css = """
-:root { /* This applies to the light mode */
-    --background-fill-primary: #FFB6C2 !important; /* Light pink */
+:root {
+    --background-fill-primary: #FFB6C2 !important;
 }
-.dark { /* This applies to the dark mode */
-    --background-fill-primary: #FFB6C1 !important; /* Light pink */
+.dark {
+    --background-fill-primary: #FFB6C1 !important;
 }
 """
 
-#STEP 2 FROM SEMANTIC SEARCH 
-# Open the water_cycle.txt file in read mode with UTF-8 encoding
+# STEP 2: Load research file
 with open("research.txt", "r", encoding="utf-8") as file:
-  # Read the entire contents of the file and store it in a variable
-  research_text = file.read()
+    research_text = file.read()
 
-# Print the text below
-print(research_text)
-
-# STEP 3 FROM SEMANTIC SEARCH 
+# STEP 3: Preprocess text
 def preprocess_text(text):
-  # Strip extra whitespace from the beginning and the end of the text
-  cleaned_text = text.strip()
-  # Split the cleaned_text by every newline character (\n)
-  chunks = cleaned_text.split("\n")
-  # Create an empty list to store cleaned chunks
-  cleaned_chunks = []
-  # Write your for-in loop below to clean each chunk and add it to the cleaned_chunks list
-  for chunk in chunks:
-    stripped_chunk = chunk.strip()
-    if stripped_chunk != "":
-      cleaned_chunks.append(stripped_chunk)
-  # Print cleaned_chunks
-  print(cleaned_chunks)
-  # Print the length of cleaned_chunks
-  print(len(cleaned_chunks))
-  # Return the cleaned_chunks
-  return cleaned_chunks
-# Call the preprocess_text function and store the result in a cleaned_chunks variable
-cleaned_chunks =  preprocess_text(research_text) # Complete this line
-
-# STEP 4 FROM SEMANTIC SEARCH 
-# Load the pre-trained embedding model that converts text to vectors
-model = SentenceTransformer('all-MiniLM-L6-v2')
+    cleaned_text = text.strip()
+    chunks = cleaned_text.split("\n")
+    cleaned_chunks = [chunk.strip() for chunk in chunks if chunk.strip() != ""]
+    return cleaned_chunks
 
-def create_embeddings(text_chunks):
-  # Convert each text chunk into a vector embedding and store as a tensor
-  chunk_embeddings = model.encode(text_chunks, convert_to_tensor=True) # Replace ... with the text_chunks list
-
-  # Print the chunk embeddings
-  print(chunk_embeddings)
+cleaned_chunks = preprocess_text(research_text)
 
-  # Print the shape of chunk_embeddings
-  print(chunk_embeddings.shape)
+# STEP 4: Create embeddings
+model = SentenceTransformer('all-MiniLM-L6-v2')
 
-  # Return the chunk_embeddings
-  return chunk_embeddings
+def create_embeddings(text_chunks):
+    chunk_embeddings = model.encode(text_chunks, convert_to_tensor=True)
+    return chunk_embeddings
 
-# Call the create_embeddings function and store the result in a new chunk_embeddings variable
-chunk_embeddings = create_embeddings(cleaned_chunks) # Complete this line
+chunk_embeddings = create_embeddings(cleaned_chunks)
 
-#STEP 5 IN SEMANTIC SEARCH
+# STEP 5: Get top chunks
 def get_top_chunks(query, chunk_embeddings, text_chunks):
-  # Convert the query text into a vector embedding
-  query_embedding = model.encode(query,convert_to_tensor=True) # Complete this line
-
-  # Normalize the query embedding to unit length for accurate similarity comparison
-  query_embedding_normalized = query_embedding / query_embedding.norm()
-
-  # Normalize all chunk embeddings to unit length for consistent comparison
-  chunk_embeddings_normalized = chunk_embeddings / chunk_embeddings.norm(dim=1, keepdim=True)
-
-  # Calculate cosine similarity between query and all chunks using matrix multiplication
-  similarities = torch.matmul(chunk_embeddings_normalized, query_embedding_normalized) # Complete this line
-
-  # Print the similarities
-  print(similarities)
-
-  # Find the indices of the 3 chunks with highest similarity scores
-  top_indices = torch.topk(similarities, k=3).indices
-
-  # Print the top indices
-  print(top_indices)
+    query_embedding = model.encode(query, convert_to_tensor=True)
+    query_embedding_normalized = query_embedding / query_embedding.norm()
+    chunk_embeddings_normalized = chunk_embeddings / chunk_embeddings.norm(dim=1, keepdim=True)
+    similarities = torch.matmul(chunk_embeddings_normalized, query_embedding_normalized)
+    top_indices = torch.topk(similarities, k=3).indices
+    top_chunks = [text_chunks[i] for i in top_indices]
+    return top_chunks
 
-  # Create an empty list to store the most relevant chunks
-  top_chunks = []
+# STEP 6: Inference client
+client = InferenceClient("Qwen/Qwen2.5-72B-Instruct")
 
-  # Loop through the top indices and retrieve the corresponding text chunks
-  for i in top_indices:
-    top_chunks.append(text_chunks[i])
-
-  # Return the list of most relevant chunks
-  return top_chunks
-
-#STEP 6 IN SEMANTIC SEARCH 
-# Call the get_top_chunks function with the original query
-top_results = ('How does water get into sky?',chunk_embeddings, cleaned_chunks) # Complete this line
-
-# Print the top results
-print(top_results)
-
-client=InferenceClient("Qwen/Qwen2.5-72B-Instruct")
 def respond(message, history):
-    top_results= get_top_chunks(message, chunk_embeddings, cleaned_chunks)
-    str_top_results= '\n'.join(top_results)
-    messages = [{'role':'system', 'content': f'You are a chatbot. Complete all your sentences, do not be blunt, and do not cut yourself off. The word limit is 100 words. Start off by giving a career in a complete, kind sentence, and then if prompted by the user provide more information like salary, college course,etc. Base your response on the provided context:\n{str_top_results}'}]
+    top_results = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
+    str_top_results = '\n'.join(top_results)
+    messages = [
+        {'role': 'system', 'content': f'You are a chatbot. Complete all your sentences, do not be blunt, and do not cut yourself off. The word limit is 100 words. Start off by giving a career in a complete, kind sentence, and then if prompted by the user provide more information like salary, college course,etc. Base your response on the provided context:\n{str_top_results}'}
+    ]
     if history:
         messages.extend(history)
-
-    messages.append({'role':'user','content':message})
+    messages.append({'role': 'user', 'content': message})
 
     response = client.chat_completion(
         messages,
-        max_tokens = 1000,
-        temperature = 0.2
+        max_tokens=1000,
+        temperature=0.2
     )
     return response['choices'][0]['message']['content'].strip()
 
 def display_image():
     return "KWKbanner.png"
 
+# EXPLORE PAGE DATA
 def show_info(topic):
     responses = {
-        "Highest Paying STEM Jobs": " 1. AI/Machine Learning Engineer – $171,774 \n 2. Cloud Solutions Architect – $150,241 \n 3. Quantitative Analyst (Quant) – $139,949 \n 4. Data Scientist – $128,115 \n 5. Actuary – $128,147",
-        "Most Flexible STEM Jobs": " 1. Software Developer \n 2. Cloud Solutions Architect \n 3. Data Scientist \n 4. Cybersecurity Analyst \n 5. Statistician",
-        "Most Creative STEM Jobs": " 1. Software Developer \n 2. AI/Machine Learning Engineer \n 3. Biomedical Engineer \n 4, Mechanical Engineer \n 5. Biochemist", 
-        "Fastest Growing STEM Jobs": " 1. AI/Machine Learning Engineer \n 2. Cybersecurity Analyst \n 3. Data Scientist \n 4. Software Developer \n 5. Cloud Solutions Architect", 
-        "Low-Stress STEM Jobs": " 1. Statistician \n 2. Mathematician \n 3. Operations Research Analyst \n 4. Environmental Scientist \n 5. Biochemist" 
+        "Highest Paying STEM Jobs": "1. AI/Machine Learning Engineer – $171,774\n2. Cloud Solutions Architect – $150,241\n3. Quantitative Analyst (Quant) – $139,949\n4. Data Scientist – $128,115\n5. Actuary – $128,147",
+        "Most Flexible STEM Jobs": "1. Software Developer\n2. Cloud Solutions Architect\n3. Data Scientist\n4. Cybersecurity Analyst\n5. Statistician",
+        "Most Creative STEM Jobs": "1. Software Developer\n2. AI/Machine Learning Engineer\n3. Biomedical Engineer\n4. Mechanical Engineer\n5. Biochemist", 
+        "Fastest Growing STEM Jobs": "1. AI/Machine Learning Engineer\n2. Cybersecurity Analyst\n3. Data Scientist\n4. Software Developer\n5. Cloud Solutions Architect", 
+        "Low-Stress STEM Jobs": "1. Statistician\n2. Mathematician\n3. Operations Research Analyst\n4. Environmental Scientist\n5. Biochemist"
     }
     return responses.get(topic, "Select a category to see the corresponding careers.")
 
+# RESOURCES PAGE DATA
+def resource_links(career):
+    links = {
+        "AI/Machine Learning Engineer": "• [DeepLearning.AI](https://www.deeplearning.ai)\n• [Fast.ai](https://www.fast.ai)\n• [Stanford CS229](https://cs229.stanford.edu/)",
+        "Data Scientist": "• [Kaggle Learn](https://www.kaggle.com/learn)\n• [Harvard’s Data Science](https://online-learning.harvard.edu/series/data-science)\n• [DataCamp](https://www.datacamp.com)",
+        "Cloud Solutions Architect": "• [AWS Training](https://aws.amazon.com/training/)\n• [Azure Certification Paths](https://learn.microsoft.com/en-us/certifications/)\n• [Google Cloud Skills Boost](https://cloudskillsboost.google/)",
+        "Cybersecurity Analyst": "• [Cybrary](https://www.cybrary.it)\n• [MITRE ATT&CK](https://attack.mitre.org/)\n• [TryHackMe](https://tryhackme.com)",
+        "Statisticians": "• [Coursera: Statistics with R](https://www.coursera.org/specializations/statistics)\n• [Khan Academy: Statistics](https://www.khanacademy.org/math/statistics-probability)",
+        "Biomedical Engineer": "• [Johns Hopkins BME](https://www.bme.jhu.edu/)\n• [edX Biomedical Courses](https://www.edx.org/learn/biomedical-engineering)\n• [NIH Biomedical Research](https://www.nih.gov/)"
+    }
+    return links.get(career, "Select a career to see recommended learning platforms and resources.")
 
+# UI
 with gr.Blocks(theme=theme, css=custom_css) as chatbot:
     gr.Image(display_image)
 
-    explore_page= "Select a cateogry to see the corresponding careers."
-
     with gr.Tabs():
         with gr.Tab("ChatBot"):
             gr.ChatInterface(
@@ -157,23 +111,36 @@ with gr.Blocks(theme=theme, css=custom_css) as chatbot:
                 textbox=gr.Textbox(placeholder="Share your interests and explore more on your career of choice!"),
                 description="This tool provides information on STEM Careers."
             )
-        with gr.Tab("Explore Page"): 
-            gr.Markdown("")
-                
-
-
-    with gr.TabItem("Explore Page"): 
-        gr.Markdown(explore_page)
-        #gr.Markdown(explore_page)
-        dropdown = gr.Dropdown(choices=["Highest Paying STEM Jobs", "Most Flexible STEM Jobs", "Most Creative STEM Jobs", "Fastest Growing STEM Jobs", "Low-Stress STEM Jobs"],
-        label="Choose Your Category"
-                                  )
-        output = gr.Textbox(label="Top-Ranked Jobs", interactive=False)
-    
-    dropdown.change(fn=show_info, inputs=dropdown, outputs=output)
-     #with gr.Tab("Educational PDFs"):
-       # gr.Markdown("### 📘 Helpful Resources")
-
 
+        with gr.Tab("Explore Page"): 
+            gr.Markdown("### 🌟 Explore STEM Career Categories")
+            dropdown_explore = gr.Dropdown(
+                choices=[
+                    "Highest Paying STEM Jobs",
+                    "Most Flexible STEM Jobs",
+                    "Most Creative STEM Jobs",
+                    "Fastest Growing STEM Jobs",
+                    "Low-Stress STEM Jobs"
+                ],
+                label="Choose a Category"
+            )
+            output_explore = gr.Textbox(label="Top-Ranked Jobs", interactive=False)
+            dropdown_explore.change(fn=show_info, inputs=dropdown_explore, outputs=output_explore)
+
+        with gr.Tab("Resources Page"): 
+            gr.Markdown("### 📚 Career-Specific Educational Resources")
+            dropdown_resources = gr.Dropdown(
+                choices=[
+                    "AI/Machine Learning Engineer",
+                    "Data Scientist",
+                    "Cloud Solutions Architect",
+                    "Cybersecurity Analyst",
+                    "Statisticians",
+                    "Biomedical Engineer"
+                ],
+                label="Choose a Career"
+            )
+            output_resources = gr.Textbox(label="Recommended Resources", interactive=False)
+            dropdown_resources.change(fn=resource_links, inputs=dropdown_resources, outputs=output_resources)
 
-chatbot.launch() 
+chatbot.launch()