v. 0.9.8 - AbstractIntegratedModule New release

IntegratedPipeline v0.9.8 - PIP Version release

🎯 Overview

IntegratedPipeline is a standalone Custom AI Agent Library for memory-augmented agentic frameworks, designed for continuous learning with MANN-Type Architecture designed for long term operation.

📦 The provided Files below includes:

• aarch64 with musllinux (v. 1.2+) and manylinux (v. 2.17+) arch. Python Wheels (3.10, 3.11, 3.12 only)
• x86_64 with musllinux and manylinux arch. Python Wheels (3.10 -> 3.12 only)
• macOS Python Wheels (3.10 ->3.12 only)

✨ use pip for downloading the correct wheels for your setup:

• pip install abstractintegratedmodule --extra-index-url https://Micro-Novelty.github.io/abstract-modules/whl/ --break-system-packages
  # ensures proper installation by bypassing pip strict external download setup. 

✨ What's New and Included

• All Asynchronous handling Pipeline blocks have been Thoroughly Tested and robustness is ensured from dangerous Memory leak and hanging threads.
• Fixed multiple rate limiter bugs, ex: clock skew can corrupt buckets.
• fixed edge cases where poison text injection can happen in Input sanitizer block, ensuring robustness during P2P.

Core Features

• Memory-Augmented Neural Networks (MANN) - External dynamic memory module for persistent learning
• Abstract Weight Encoder (AWE) - Eigenvalue-based weight shaping for noise robustness
• Specialized MLP - Noise-robust classification with geometric weight adaptation
• Optimized Transformer - Alpha-based computing for stable contextual understanding
• LSTM Architectures - Efficient LSTM Using AWE Weight in LSTM init.
• Hybrid Ensemble - Dynamic weighting of MLP + Transformer predictions
• SQLite Integration - Local database for memory, attention weights, and predictions
• Peer-to-Peer Coordination - Distributed agent communication with SSL security, Cohesive Interaction, and Compact multi-modal ensemble weighting from peer.

Platform Support

• ✅ Windows OS.
• ✅ Linux (x86_64)
• ✅ macOS architecture
• ✅ aarch64/Raspberry Pi (ARM64 v7/v8)

📦 Library requirements when using AbstractIntegratedModule tarball.

• numpy
• scikit-learn
• pandas
• aiohttp
• psutil
• cargotraphy
• Note: this will usually automatically downloaded by the system.

🚀 Quick Start

from AbstractIntegratedModule import IntegratedPipeline
from AbstractIntegratedModule import PipelinePredictionManager
import numpy as np

memory_name = 'agent_memory'
cert_file = <your_cert_file_dir> # your .crt file
key_file = <your_key_file_dir> # your .key file

main_model = IntegratedPipeline(memory_name=memory_name, use_async=True, agent_port=5000, ssl_cert_file=cert_file, ssl_key_file=key_file)
 # provide cert_file path or key_file path (optional)
main_prediction = PipelinePredictionManager(main_model, 
               label_csv='C:/users/yourdevice/example_manual_training.txt', 
                  # or /home/yourdevice/example_manual_training.txt.
                  #your path dir that contains the .txt file that contains CSV format.
               target_title='window_title', label='label')
# example_manual_training is a .txt file that contain csv format like above example.

example_rules = [
                     # === WORK / PRODUCTIVITY ===
                     (r'code|programming|develop|debug|compile|script', 'focused_work'),
                     (r'vscode|visual_studio|ide|terminal|shell', 'focused_work'),
                     (r'notion|evernote|onenote|notes|todo|task', 'productive'),
                     (r'slack|teams|discord|zoom|meeting|call', 'communication'),
                     (r'email|gmail|outlook|inbox|mail', 'communication'),
                     
                     # === ENTERTAINMENT ===
                     (r'youtube|netflix|twitch|stream|video', 'entertainment'),
                     (r'music|spotify|soundcloud|audio|player', 'entertainment'),
                     (r'game|gaming|steam|epic|play', 'gaming'),
                     (r'facebook|instagram|tiktok|social|post', 'social_media'),
                     
                     # === BROWSING ===
                     (r'chrome|firefox|edge|safari|browser', 'browsing'),
                     (r'google|search|wiki|wiki|article', 'information'),
                     (r'stackoverflow|github|docs|documentation', 'research'),

                     # more rules
                 ]
# activate explainability capability to explain uncertainty:
main_model.show_explainability_details = True
main_model.distribution.predict_manager = main_prediction # set PipelinePredictionManager to AgentDistributedInference for asynchronous prediction later

# test samples with more sophisticated rules and more complex titles for prediction
# (title, intent)
test_titles = [
 ("Opening Thesis.docx", "slight_work"),
 ("Watching YouTube and Google Chrome", "distracted"),
 ("Watching Slack", "communication"),
 ("Programming in Visual Studio Code", "focused_work"),
 ("Watching netflix.com - Chrome", "break"),
# more titles 
 ]  
            
titles, y, label_map = main_prediction.load_labels_from_csv(
    <C:/your/path/to/example_manual_training.txt>, # your actual .txt file that contains CSV format for trainings.  same like the above initialized label_csv directory, can also use the provided ManualsTraining.txt below for quick use.
    <target_title>, <target_label>)
# small training with simple titles
main_model.train(titles, y)
   
results, chosen_label, confidence = main_prediction.advanced_prediction_method(test_titles, label_map, example_rules,
                             X=None, y=None,
                             show_proba=False, top_k=3, 
                             use_transformer=True,
                             return_attention=False,
                             save_results=True)

# ... more features/wrapper you can add

)