Code / app.py
Artificial-superintelligence's picture
Update app.py
ca9dc4d verified
raw
history blame
13.1 kB
import streamlit as st
import google.generativeai as genai
import requests
import subprocess
import os
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
import torch
import torch.nn as nn
import torch.optim as optim
from transformers import AutoTokenizer, AutoModel, pipeline
import ast
import networkx as nx
import matplotlib.pyplot as plt
import re
import javalang
import clang.cindex
import radon.metrics as radon_metrics
import radon.complexity as radon_complexity
import black
import isort
import autopep8
# Configure the Gemini API
genai.configure(api_key=st.secrets["GOOGLE_API_KEY"])
# Create the model with optimized parameters and enhanced system instructions
generation_config = {
"temperature": 0.7,
"top_p": 0.9,
"top_k": 40,
"max_output_tokens": 32768,
}
model = genai.GenerativeModel(
model_name="gemini-1.5-pro",
generation_config=generation_config,
system_instruction="""
You are Ath, an extremely advanced code assistant with deep expertise in AI, machine learning, software engineering, and multiple programming languages. You provide cutting-edge, optimized, and secure code solutions across various domains. Use your vast knowledge to generate high-quality code, perform advanced analyses, and offer insightful optimizations. Adapt your language and explanations based on the user's expertise level.
"""
)
chat_session = model.start_chat(history=[])
# Load pre-trained models for code understanding and generation
tokenizer = AutoTokenizer.from_pretrained("microsoft/codebert-base")
codebert_model = AutoModel.from_pretrained("microsoft/codebert-base")
code_generation_model = pipeline("text-generation", model="EleutherAI/gpt-neo-2.7B")
class AdvancedCodeImprovement(nn.Module):
def __init__(self, input_dim):
super(AdvancedCodeImprovement, self).__init__()
self.fc1 = nn.Linear(input_dim, 1024)
self.fc2 = nn.Linear(1024, 512)
self.fc3 = nn.Linear(512, 256)
self.fc4 = nn.Linear(256, 128)
self.fc5 = nn.Linear(128, 64)
self.fc6 = nn.Linear(64, 32)
self.fc7 = nn.Linear(32, 16)
self.fc8 = nn.Linear(16, 4) # Multiple classification: style, efficiency, security, maintainability
def forward(self, x):
x = torch.relu(self.fc1(x))
x = torch.relu(self.fc2(x))
x = torch.relu(self.fc3(x))
x = torch.relu(self.fc4(x))
x = torch.relu(self.fc5(x))
x = torch.relu(self.fc6(x))
x = torch.relu(self.fc7(x))
return torch.sigmoid(self.fc8(x))
code_improvement_model = AdvancedCodeImprovement(768) # 768 is BERT's output dimension
optimizer = optim.Adam(code_improvement_model.parameters())
criterion = nn.BCELoss()
def generate_response(user_input):
try:
response = chat_session.send_message(user_input)
return response.text
except Exception as e:
return f"Error in generating response: {str(e)}"
def detect_language(code):
# Simple language detection based on keywords and syntax
if re.search(r'\b(def|class|import)\b', code):
return 'python'
elif re.search(r'\b(function|var|let|const)\b', code):
return 'javascript'
elif re.search(r'\b(public|private|class)\b', code):
return 'java'
elif re.search(r'\b(#include|int main)\b', code):
return 'c++'
else:
return 'unknown'
def validate_and_fix_code(code, language):
if language == 'python':
try:
fixed_code = autopep8.fix_code(code)
fixed_code = isort.SortImports(file_contents=fixed_code).output
fixed_code = black.format_str(fixed_code, mode=black.FileMode())
return fixed_code
except Exception as e:
return code, f"Error in fixing Python code: {str(e)}"
elif language == 'javascript':
# Use a JS beautifier (placeholder)
return code
elif language == 'java':
# Use a Java formatter (placeholder)
return code
elif language == 'c++':
# Use a C++ formatter (placeholder)
return code
else:
return code
def optimize_code(code):
language = detect_language(code)
fixed_code, fix_error = validate_and_fix_code(code, language)
if fix_error:
return fixed_code, fix_error
if language == 'python':
try:
tree = ast.parse(fixed_code)
# Perform advanced Python-specific optimizations
optimizer = PythonCodeOptimizer()
optimized_tree = optimizer.visit(tree)
optimized_code = ast.unparse(optimized_tree)
except SyntaxError as e:
return fixed_code, f"SyntaxError: {str(e)}"
elif language == 'java':
try:
tree = javalang.parse.parse(fixed_code)
# Perform Java-specific optimizations
optimizer = JavaCodeOptimizer()
optimized_code = optimizer.optimize(tree)
except javalang.parser.JavaSyntaxError as e:
return fixed_code, f"JavaSyntaxError: {str(e)}"
elif language == 'c++':
try:
index = clang.cindex.Index.create()
tu = index.parse('temp.cpp', args=['-std=c++14'], unsaved_files=[('temp.cpp', fixed_code)])
# Perform C++-specific optimizations
optimizer = CppCodeOptimizer()
optimized_code = optimizer.optimize(tu)
except Exception as e:
return fixed_code, f"C++ Parsing Error: {str(e)}"
else:
optimized_code = fixed_code # For unsupported languages, return the fixed code
# Run language-specific linter
lint_results = run_linter(optimized_code, language)
return optimized_code, lint_results
def run_linter(code, language):
if language == 'python':
with open("temp_code.py", "w") as file:
file.write(code)
result = subprocess.run(["pylint", "temp_code.py"], capture_output=True, text=True)
os.remove("temp_code.py")
return result.stdout
elif language == 'javascript':
# Run ESLint (placeholder)
return "JavaScript linting not implemented"
elif language == 'java':
# Run CheckStyle (placeholder)
return "Java linting not implemented"
elif language == 'c++':
# Run cppcheck (placeholder)
return "C++ linting not implemented"
else:
return "Linting not available for the detected language"
def fetch_from_github(query):
headers = {"Authorization": f"token {st.secrets['GITHUB_TOKEN']}"}
response = requests.get(f"https://api.github.com/search/code?q={query}", headers=headers)
if response.status_code == 200:
return response.json()['items'][:5] # Return top 5 results
return []
def analyze_code_quality(code):
inputs = tokenizer(code, return_tensors="pt", truncation=True, max_length=512, padding="max_length")
with torch.no_grad():
outputs = codebert_model(**inputs)
cls_embedding = outputs.last_hidden_state[:, 0, :]
predictions = code_improvement_model(cls_embedding)
quality_scores = {
"style": predictions[0][0].item(),
"efficiency": predictions[0][1].item(),
"security": predictions[0][2].item(),
"maintainability": predictions[0][3].item()
}
# Calculate additional metrics
language = detect_language(code)
if language == 'python':
complexity = radon_complexity.cc_visit(code)
maintainability = radon_metrics.mi_visit(code, True)
quality_scores["cyclomatic_complexity"] = complexity[0].complexity
quality_scores["maintainability_index"] = maintainability
return quality_scores
def visualize_code_structure(code):
try:
tree = ast.parse(code)
graph = nx.DiGraph()
def add_nodes_edges(node, parent=None):
node_id = id(node)
graph.add_node(node_id, label=f"{type(node).__name__}\n{ast.unparse(node)[:20]}")
if parent:
graph.add_edge(id(parent), node_id)
for child in ast.iter_child_nodes(node):
add_nodes_edges(child, node)
add_nodes_edges(tree)
plt.figure(figsize=(15, 10))
pos = nx.spring_layout(graph, k=0.9, iterations=50)
nx.draw(graph, pos, with_labels=True, node_color='lightblue', node_size=2000, font_size=8, font_weight='bold', arrows=True)
labels = nx.get_node_attributes(graph, 'label')
nx.draw_networkx_labels(graph, pos, labels, font_size=6)
return plt
except SyntaxError:
return None
def suggest_improvements(code, quality_scores):
suggestions = []
if quality_scores["style"] < 0.7:
suggestions.append("Consider improving code style for better readability.")
if quality_scores["efficiency"] < 0.7:
suggestions.append("There might be room for optimizing the code's efficiency.")
if quality_scores["security"] < 0.8:
suggestions.append("Review the code for potential security vulnerabilities.")
if quality_scores["maintainability"] < 0.7:
suggestions.append("The code could be refactored to improve maintainability.")
if "cyclomatic_complexity" in quality_scores and quality_scores["cyclomatic_complexity"] > 10:
suggestions.append("Consider breaking down complex functions to reduce cyclomatic complexity.")
return suggestions
# Streamlit UI setup
st.set_page_config(page_title="Highly Advanced AI Code Assistant", page_icon="πŸš€", layout="wide")
# ... (keep the existing CSS styles) ...
st.markdown('<div class="main-container">', unsafe_allow_html=True)
st.title("πŸš€ Highly Advanced AI Code Assistant")
st.markdown('<p class="subtitle">Powered by Advanced AI & Multi-Domain Expertise</p>', unsafe_allow_html=True)
prompt = st.text_area("What advanced code task can I assist you with today?", height=120)
if st.button("Generate Advanced Code"):
if prompt.strip() == "":
st.error("Please enter a valid prompt.")
else:
with st.spinner("Generating and analyzing code..."):
completed_text = generate_response(prompt)
if "Error in generating response" in completed_text:
st.error(completed_text)
else:
optimized_code, lint_results = optimize_code(completed_text)
if "Error" in lint_results:
st.warning(f"Issues detected in the generated code. Attempting to fix...")
st.code(optimized_code)
st.info("Please review the code above. It may contain errors or be incomplete.")
else:
quality_scores = analyze_code_quality(optimized_code)
overall_quality = sum(quality_scores.values()) / len(quality_scores)
st.success(f"Code generated and optimized successfully! Overall Quality Score: {overall_quality:.2f}")
st.markdown('<div class="output-container">', unsafe_allow_html=True)
st.markdown('<div class="code-block">', unsafe_allow_html=True)
st.code(optimized_code)
st.markdown('</div>', unsafe_allow_html=True)
col1, col2 = st.columns(2)
with col1:
st.subheader("Code Quality Metrics")
for metric, score in quality_scores.items():
st.metric(metric.capitalize(), f"{score:.2f}")
with col2:
st.subheader("Improvement Suggestions")
suggestions = suggest_improvements(optimized_code, quality_scores)
for suggestion in suggestions:
st.info(suggestion)
visualization = visualize_code_structure(optimized_code)
if visualization:
with st.expander("View Advanced Code Structure Visualization"):
st.pyplot(visualization)
else:
st.warning("Unable to generate code structure visualization.")
with st.expander("View Detailed Lint Results"):
st.text(lint_results)
with st.expander("Explore Similar Code from GitHub"):
github_results = fetch_from_github(prompt)
for item in github_results:
st.markdown(f"[{item['name']}]({item['html_url']})")
st.markdown('</div>', unsafe_allow_html=True)
st.markdown("""
<div style='text-align: center; margin-top: 2rem; color: #4a5568;'>
Crafted with πŸš€ by Your Highly Advanced AI Code Assistant
</div>
""", unsafe_allow_html=True)
st.markdown('</div>', unsafe_allow_html=True)