Downloads · 30 days
0
bwsk-mobilenetv2
tzervas/bwsk-mobilenetv2
bwsk-mobilenetv2 is a image classification model from tzervas. Use it when you need a label for an image. The card lists the license as mit.
MobileNetV2 (3M params) trained in 6 variants (3 BWSK modes x 2 experiments) on CIFAR-10 with full convergence training and early stopping.
Access
Public
Updated Mar 5, 2026
Repo size
55 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt55 MB · 97%
At a glance
- Task
- Image Classification
- License
- mit
- Access
- Public
- Created
- Mar 5, 2026
- Updated
- Mar 5, 2026
- SHA
- 353ff69d
Base models
Datasets
Model card
From the Hugging Face model README
BWSK MobileNetV2
MobileNetV2 (3M params) trained in 6 variants (3 BWSK modes x 2 experiments) on CIFAR-10 with full convergence training and early stopping.
This repo contains all model weights, configs, and training results in a single consolidated repository.
What is BWSK?
BWSK is a framework that classifies every neural network operation as S-type (information-preserving, reversible, coordination-free) or K-type (information-erasing, synchronization point) using combinator logic. This classification enables reversible backpropagation through S-phases to save memory, and CALM-based parallelism analysis.
Model Overview
| Property | Value |
|---|---|
| Base Model | google/mobilenet_v2_1.0_224 |
| Architecture | Cnn (image_cls) |
| Parameters | 3M |
| Dataset | CIFAR-10 |
| Eval Metric | Accuracy |
S/K Classification
| Type | Ratio |
|---|---|
| S-type (information-preserving) | 34.0% |
| K-type (information-erasing) | 66.0% |
Fine-tune Results
| Mode | Final Loss | Val Accuracy | Test Accuracy | Peak Memory | Time | Epochs |
|---|---|---|---|---|---|---|
| Conventional | 0.3677 | 84.9% | 84.4% | 2.4 GB | 1.0m | 2 |
| BWSK Analyzed | 0.0471 | 92.9% | 92.6% | 2.4 GB | 4.2m | 6 |
| BWSK Reversible | 0.4448 | 85.4% | 84.4% | 2.4 GB | 48s | 2 |
Memory savings (reversible vs conventional): 0.0%
From Scratch Results
| Mode | Final Loss | Val Accuracy | Test Accuracy | Peak Memory | Time | Epochs |
|---|---|---|---|---|---|---|
| Conventional | 0.2578 | 85.6% | 84.9% | 2.4 GB | 7.4m | 10 |
| BWSK Analyzed | 0.7148 | 70.7% | 69.9% | 2.4 GB | 2.6m | 4 |
| BWSK Reversible | 1.0967 | 77.3% | 77.4% | 2.4 GB | 3.7m | 5 |
Memory savings (reversible vs conventional): 0.0%
Repository Structure
├── README.md
├── results.json
├── finetune-conventional/
│ ├── model.safetensors
│ ├── config.json
│ └── training_results.json
├── finetune-bwsk-analyzed/
│ ├── model.safetensors
│ ├── config.json
│ └── training_results.json
├── finetune-bwsk-reversible/
│ ├── model.safetensors
│ ├── config.json
│ └── training_results.json
├── scratch-conventional/
│ ├── model.safetensors
│ ├── config.json
│ └── training_results.json
├── scratch-bwsk-analyzed/
│ ├── model.safetensors
│ ├── config.json
│ └── training_results.json
├── scratch-bwsk-reversible/
│ ├── model.safetensors
│ ├── config.json
│ └── training_results.json
Usage
Load a specific variant:
import torch
# Load fine-tuned conventional variant
# Weights are in the finetune-conventional/ subdirectory
Training Configuration
| Setting | Value |
|---|---|
| Optimizer | AdamW |
| LR (fine-tune) | 1e-03 |
| LR (from-scratch) | 5e-03 |
| LR Schedule | Cosine with warmup |
| Max Grad Norm | 1.0 |
| Mixed Precision | AMP (float16) |
| Early Stopping | Patience 3 |
| Batch Size | 32 |
Links
Citation
@software{zervas2026bwsk,
author = {Zervas, Tyler},
title = {BWSK: Combinator-Typed Neural Network Analysis},
year = {2026},
url = {https://github.com/tzervas/ai-s-combinator},
}
License
MIT
- Task
- Image Classification
- License
- mit
- Created
- Mar 5, 2026
- Updated
- Mar 5, 2026