README.md

---
license: mit
tags:
- image-classification
- cifar100
- geometric-learning
- fractal-encoding
- trained
- no-attention
- no-cross-entropy
datasets:
- cifar100
metrics:
- accuracy
library_name: pytorch
pipeline_tag: image-classification
model-index:
- name: geo-beatrix-resnet34-step20-feats1000
  results:
  - task:
      type: image-classification
      name: Image Classification
    dataset:
      name: CIFAR-100
      type: cifar100
    metrics:
    - type: accuracy
      value: 56.12
      name: Test Accuracy
      verified: false
---

# geo-beatrix-resnet34-step20-feats1000

**Geometric Basin Classification for CIFAR-100**

🎉 **Training Complete** 🎉

Final Status: Epoch 200/200

---

## Current Performance

| Metric | Value |
|--------|-------|
| **Best Test Accuracy** | **56.12%** |
| **Best Epoch** | 160 |
| **Current Train Accuracy** | 59.29% |
| **Current Test Accuracy** | 51.51% |
| **Current α (Cantor param)** | 0.4031 |
| **Total Parameters** | 28,561,101 |
| **Training Time** | 0:27:18 |

### All Training Runs

Autogen bug, they all have different test accs.

| Timestamp | Status | Best Epoch | Test Acc | Train Acc | α |
|-----------|--------|------------|----------|-----------|---|
| `20251010_203717` | ✅ | 160 | **56.12%** | 67.82% | 0.4481 |
| `20251010_211210` | 🔄 | 160 | **56.12%** | 16.21% | 0.3879 |
| `20251010_213807` | ✅ | 160 | **56.12%** | 64.44% | 0.4419 |
| `20251010_230300` | ✅ | 160 | **56.12%** | 52.13% | 0.4997 |
| `20251010_234239` | ✅ | 160 | **56.12%** | 73.34% | 0.4882 |
| `20251011_002858` | ✅ | 160 | **56.12%** | 46.05% | 0.4712 |
| `20251011_012453` | ✅ | 160 | **56.12%** | 40.18% | 0.4963 |
| `20251011_023128` | ✅ | 160 | **56.12%** | 54.65% | 0.5005 |
| `20251011_025919` | ✅ | 160 | **56.12%** | 57.80% | 0.4994 |
| `20251011_032343` | ✅ | 160 | **56.12%** | 53.80% | 0.4377 |
| `20251011_034748` | ✅ | 160 | **56.12%** | 65.10% | 0.4326 |
| `20251011_041716` | ✅ | 160 | **56.12%** | 59.29% | 0.4031 |
| `20251010_200842` | ✅ | 180 | **53.61%** | 67.53% | 0.4442 |
| `20251010_185133` | ✅ | 200 | **52.97%** | 69.87% | 0.4452 |

### Comparison to State-of-the-Art

| Model | Accuracy | Status |
|-------|----------|--------|
| **geo-beatrix (this model)** | **56.12%** | ✅ Complete |
| geo-beatrix (50M params) | 69.0% | Geometric Basin CONV architecture |

🎯 **Current target**: Beat geo-beatrix (69.0%) - Currently -12.88%

---

## Architecture

- **Base**: ResNet34 (torchvision)
- **Pretrained**: From scratch
- **Features**: 512-dim from ResNet34
- **Positional Encoding**: Devil's Staircase (Cantor function, 1883)
- **PE Levels**: 20
- **PE Features/Level**: 1000
- **Classification**: Geometric Basin Compatibility (NO cross-entropy)
- **Attention Mechanisms**: NONE
- **Mixing**: Standard (single patch)

---

## Training Configuration

```json
{
  "model_name": "geo-beatrix-resnet34-step20-feats1000",
  "model_type": "geometric_basin_classifier",
  "num_classes": 100,
  "batch_size": 512,
  "num_epochs": 200,
  "base_learning_rate": 0.001,
  "weight_decay": 0.05,
  "warmup_epochs": 10,
  "pe_levels": 20,
  "pe_features_per_level": 1000,
  "dropout": 0.1,
  "pretrained_resnet": false,
  "frozen_resnet": false,
  "a100_optimizations": {
    "mixed_precision": true,
    "torch_compile": false,
    "channels_last": true,
    "gradient_checkpointing": false
  },
  "alphamix": {
    "enabled": true,
    "fractal_mode": false,
    "range": [
      0.3,
      0.7
    ],
    "spatial_ratio": 0.1,
    "curriculum_start": 0.0,
    "curriculum_end": 0.75,
    "fractal_steps": [
      1,
      3
    ],
    "fractal_scales": [
      0.3333333333333333,
      0.1111111111111111,
      0.037037037037037035
    ]
  },
  "architecture": "ResNet34 + Devil's Staircase PE",
  "loss_function": "Geometric Basin Compatibility",
  "cross_entropy": false,
  "attention_mechanisms": false,
  "timestamp": "20251011_041716"
}
```

---

## Files Structure

```
├── model.pt                 (BEST overall model - easy access!)
├── model.safetensors        (BEST overall model - easy access!)
├── best_model_info.json     (which epoch/run this came from)
├── runs_history.json        (all training runs and their results)
├── README.md
├── weights/geo-beatrix-resnet34-step20-feats1000/20251011_041716/
│   ├── model.pt                 (best from this training run)
│   ├── model.safetensors        (best from this training run)
│   ├── config.json
│   ├── training_log.txt
│   └── checkpoints/
│       ├── checkpoint_epoch_50.safetensors
│       ├── checkpoint_epoch_100.safetensors
│       └── checkpoint_epoch_150.safetensors
│       (snapshots every 10 epochs)
└── runs/geo-beatrix-resnet34-step20-feats1000/20251011_041716/
    ├── events.out.tfevents.*    (TensorBoard logs)
    └── metrics.csv              (training metrics)
```

**Note**: The root `model.pt` and `model.safetensors` always contain the best model across all training runs!

---

## Usage

```python
from huggingface_hub import hf_hub_download
import torch

# EASIEST: Download BEST overall model from root (recommended!)
from safetensors.torch import load_file
model_path = hf_hub_download(
    repo_id="AbstractPhil/geo-beatrix-resnet",
    filename="model.safetensors"
)
state_dict = load_file(model_path)
# model.load_state_dict(state_dict)

# Check which epoch/run the best model came from
info_path = hf_hub_download(
    repo_id="AbstractPhil/geo-beatrix-resnet",
    filename="best_model_info.json"
)
with open(info_path) as f:
    best_info = json.load(f)
    print(f"Best model: epoch {best_info['epoch']}, {best_info['test_accuracy']:.2f}%")

# Or download from specific training run
model_path = hf_hub_download(
    repo_id="AbstractPhil/geo-beatrix-resnet",
    filename="weights/geo-beatrix-resnet34-step20-feats1000/20251011_041716/model.safetensors"
)

# Download specific epoch checkpoint
epoch_checkpoint = hf_hub_download(
    repo_id="AbstractPhil/geo-beatrix-resnet",
    filename="weights/geo-beatrix-resnet34-step20-feats1000/20251011_041716/checkpoints/checkpoint_epoch_100.safetensors"
)
```

---

## Training History

### Best Checkpoint
- Epoch: 160
- Train Acc: 59.43%
- Test Acc: 51.64%
- Alpha: 0.4071
- Loss: 0.7570

### Latest 5 Epochs

- **Epoch 196**: Train 62.03%, Test 0.00%, α=0.4032, Loss=0.7300
- **Epoch 197**: Train 59.02%, Test 0.00%, α=0.4031, Loss=0.6201
- **Epoch 198**: Train 58.49%, Test 0.00%, α=0.4031, Loss=0.6571
- **Epoch 199**: Train 59.32%, Test 0.00%, α=0.4031, Loss=0.6543
- **Epoch 200**: Train 59.29%, Test 51.51%, α=0.4031, Loss=0.6505

### Training Milestones
- 🎯 **50% Accuracy** reached at epoch 120
- 📊 **α ≥ 0.40** reached at epoch 17

---

## Innovation

✅ **NO attention mechanisms**  
✅ **NO cross-entropy loss**  
✅ **Fractal positional encoding** (Cantor function from 1883)  
✅ **Geometric compatibility classification**  
✅ **ResNet34 backbone** (proven CNN architecture)


---

**Repository**: https://huggingface.co/AbstractPhil/geo-beatrix-resnet  
**Author**: AbstractPhil  
**Framework**: PyTorch