Hugging Face's logo

AbstractPhil
/
geo-beatrix-resnet

Image Classification
PyTorch
TensorBoard
Safetensors
cifar100
geometric-learning
fractal-encoding
trained
no-attention
no-cross-entropy
Eval Results (legacy)
Model card Files
xet
Metrics Community
AbstractPhil commited on
Commit
a1920ce
·
verified ·
1 Parent(s): 7cb6bd9

Epoch 100: 50.72%

Browse files
Files changed (1) hide show
  1. README.md +15 -13
README.md CHANGED
@@ -36,7 +36,7 @@ model-index:
36
 
37
  🚧 **Training in Progress** 🚧
38
 
39
- Current Status: Epoch 50/200
40
 
41
  ---
42
 
@@ -46,11 +46,11 @@ Current Status: Epoch 50/200
46
  |--------|-------|
47
  | **Best Test Accuracy** | **56.12%** |
48
  | **Best Epoch** | 160 |
49
- | **Current Train Accuracy** | 42.89% |
50
- | **Current Test Accuracy** | 43.93% |
51
- | **Current α (Cantor param)** | 0.4310 |
52
  | **Total Parameters** | 27,846,901 |
53
- | **Training Time** | 0:06:50 |
54
 
55
  ### All Training Runs
56
 
@@ -58,7 +58,7 @@ Current Status: Epoch 50/200
58
  |-----------|--------|------------|----------|-----------|---|
59
  | `20251010_203717` | ✅ | 160 | **56.12%** | 67.82% | 0.4481 |
60
  | `20251010_211210` | 🔄 | 160 | **56.12%** | 16.21% | 0.3879 |
61
- | `20251010_213807` | 🔄 | 160 | **56.12%** | 42.89% | 0.4310 |
62
  | `20251010_200842` | ✅ | 180 | **53.61%** | 67.53% | 0.4442 |
63
  | `20251010_185133` | ✅ | 200 | **52.97%** | 69.87% | 0.4452 |
64
 
@@ -210,21 +210,23 @@ epoch_checkpoint = hf_hub_download(
210
 
211
  ### Best Checkpoint
212
  - Epoch: 160
213
- - Train Acc: 42.89%
214
  - Test Acc: 56.12%
215
- - Alpha: 0.4310
216
  - Loss: 0.0000
217
 
218
  ### Latest 5 Epochs
219
 
220
- - **Epoch 46**: Train 41.81%, Test 0.00%, α=0.4305, Loss=1.6292
221
- - **Epoch 47**: Train 42.20%, Test 0.00%, α=0.4324, Loss=1.5935
222
- - **Epoch 48**: Train 43.09%, Test 0.00%, α=0.4302, Loss=1.6075
223
- - **Epoch 49**: Train 42.97%, Test 0.00%, α=0.4346, Loss=1.5375
224
- - **Epoch 50**: Train 42.89%, Test 43.93%, α=0.4310, Loss=1.4858
225
 
226
  ### Training Milestones
 
227
  - 📊 **α ≥ 0.40** reached at epoch 12
 
228
 
229
  ---
230
 
 
36
 
37
  🚧 **Training in Progress** 🚧
38
 
39
+ Current Status: Epoch 100/200
40
 
41
  ---
42
 
 
46
  |--------|-------|
47
  | **Best Test Accuracy** | **56.12%** |
48
  | **Best Epoch** | 160 |
49
+ | **Current Train Accuracy** | 54.90% |
50
+ | **Current Test Accuracy** | 50.72% |
51
+ | **Current α (Cantor param)** | 0.4377 |
52
  | **Total Parameters** | 27,846,901 |
53
+ | **Training Time** | 0:13:47 |
54
 
55
  ### All Training Runs
56
 
 
58
  |-----------|--------|------------|----------|-----------|---|
59
  | `20251010_203717` | ✅ | 160 | **56.12%** | 67.82% | 0.4481 |
60
  | `20251010_211210` | 🔄 | 160 | **56.12%** | 16.21% | 0.3879 |
61
+ | `20251010_213807` | 🔄 | 160 | **56.12%** | 54.90% | 0.4377 |
62
  | `20251010_200842` | ✅ | 180 | **53.61%** | 67.53% | 0.4442 |
63
  | `20251010_185133` | ✅ | 200 | **52.97%** | 69.87% | 0.4452 |
64
 
 
210
 
211
  ### Best Checkpoint
212
  - Epoch: 160
213
+ - Train Acc: 54.90%
214
  - Test Acc: 56.12%
215
+ - Alpha: 0.4377
216
  - Loss: 0.0000
217
 
218
  ### Latest 5 Epochs
219
 
220
+ - **Epoch 96**: Train 54.56%, Test 0.00%, α=0.4394, Loss=1.1654
221
+ - **Epoch 97**: Train 53.43%, Test 0.00%, α=0.4392, Loss=1.1241
222
+ - **Epoch 98**: Train 53.11%, Test 0.00%, α=0.4387, Loss=1.1020
223
+ - **Epoch 99**: Train 54.46%, Test 0.00%, α=0.4444, Loss=1.1417
224
+ - **Epoch 100**: Train 54.90%, Test 50.72%, α=0.4377, Loss=1.0984
225
 
226
  ### Training Milestones
227
+ - 🎯 **50% Accuracy** reached at epoch 85
228
  - 📊 **α ≥ 0.40** reached at epoch 12
229
+ - 📊 **α ≥ 0.44** (near triadic equilibrium) at epoch 64
230
 
231
  ---
232