Skip to content

Commit

Permalink
large update docs (#33)
Browse files Browse the repository at this point in the history
* Add tasks page

* Update mkdocs yml

* restructuring docs
  • Loading branch information
ishaansingh22 authored Aug 17, 2024
1 parent 1bb595e commit fc633c1
Show file tree
Hide file tree
Showing 9 changed files with 48 additions and 2 deletions.
Empty file added docs/datasets.md
Empty file.
File renamed without changes.
35 changes: 35 additions & 0 deletions docs/logging/logging.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,35 @@
# Logging Documentation

This document provides a detailed overview of what is being logged in the Sonar setup.

<!-- ## Table of Contents
1. [Overview](#overview)
2. [Logging Types](#logging-types)
3. [Log Sources](#log-sources)
4. [Log Details](#log-details) -->

## Overview

This documentation aims to provide transparency on the logging mechanisms implemented in the Sonar project. It includes information on the types of data being logged, their sources, formats, and purposes.

## Logging Types

- **DEBUG:** Detailed information, typically of interest only when diagnosing problems.
- **INFO:** Confirmation that things are working as expected.
- **Tensorboard logging**: Logging specific metrics, images, and other data to TensorBoard for visualization and analysis.
- Console Logging: Logs a message to the console.
- Scalar Logging: Logs scalar values to TensorBoard for tracking metrics(loss, accuracy) over time.
- Image Logging: Logs images to both a file and TensorBoard for visual analysis.

## Log Sources

| Component/Module | Data Logged | Log Level | Format | Storage Location | Frequency/Trigger |
|--------------------|--------------------------------------------------|---------------|-------------|-------------------------------------|----------------------------------------|
| Model Training (FL) | Aggregated model metrics, client updates | INFO, DEBUG | Plain text | `./expt_dump/<experiment_name>/logs/client_<client_index>/summary.txt` | On every FL round

<!-- ## Log Details
### Federated Learning
Logs aggregated model metrics (loss and accuracy) and updates from clients to track the overall progress and performance of the federated learning process. Additionally, logs include training loss and accuracy from individual clients. Also logs communication events between the server and clients to monitor interactions and data exchange.
-->
Empty file added docs/models.md
Empty file.
3 changes: 3 additions & 0 deletions docs/tasks.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# Tasks

Welcome to the **Tasks** section of our repository! Here, you'll find resources and guides for three core tasks in machine learning: Image Classification, Object Detection, and Text Classification.
1 change: 1 addition & 0 deletions docs/tasks/image-classification.md
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
# Image Classification
1 change: 1 addition & 0 deletions docs/tasks/object-detection.md
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
# Object Detection
1 change: 1 addition & 0 deletions docs/tasks/text-classification.md
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
# Text Classification
9 changes: 7 additions & 2 deletions mkdocs.yml
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@ nav:
- Home:
- Home: index.md
- Logging:
- Main: logging.md
- Log Details: log_details.md
- Main: logging/logging.md
- Log Details: logging/log_details.md
# - [Overview](#overview)
# - [Logging Types](#logging-types)
# - [Log Sources](#log-sources)
Expand All @@ -29,6 +29,11 @@ nav:
- Contribution Guidelines: contributing/contribution-guidelines.md
- Documentation: contributing/documentation.md

- Tasks:
- Main: tasks.md
- Image Classification: tasks/image-classification.md
- Object Detection: tasks/object-detection.md
- Text Classification: tasks/text-classification.md
#
# Theme Settings
#
Expand Down

0 comments on commit fc633c1

Please sign in to comment.