Epoch 10 - 25.13% acc
Browse files- weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/README.md +1 -1
- weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/best.safetensors +1 -1
- weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/tensorboard/events.out.tfevents.1764480213.c46210e51d85.45850.9 +3 -0
- weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/training_config.json +1 -1
- weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/training_history.json +3 -35
weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/README.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
# Run: run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437
|
| 2 |
|
| 3 |
## Results
|
| 4 |
-
- **Best Accuracy**:
|
| 5 |
- **Dataset**: cifar100
|
| 6 |
- **Epochs**: 200
|
| 7 |
- **Model Version**: V2
|
|
|
|
| 1 |
# Run: run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437
|
| 2 |
|
| 3 |
## Results
|
| 4 |
+
- **Best Accuracy**: 25.13%
|
| 5 |
- **Dataset**: cifar100
|
| 6 |
- **Epochs**: 200
|
| 7 |
- **Model Version**: V2
|
weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/best.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 156046704
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af85cfe4c28d1af5cc4597a0a751550e0c139c1de2d89b566f005fa2fd59f282
|
| 3 |
size 156046704
|
weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/tensorboard/events.out.tfevents.1764480213.c46210e51d85.45850.9
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a953fe66a85fb8eec10dd4fcb67e0311818dc1d62238a4016104d5baa17fb09
|
| 3 |
+
size 6336
|
weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/training_config.json
CHANGED
|
@@ -25,7 +25,7 @@
|
|
| 25 |
"cutmix_alpha": 1.0,
|
| 26 |
"save_interval": 10,
|
| 27 |
"output_dir": "./checkpoints/cifar100_v2",
|
| 28 |
-
"resume_from":
|
| 29 |
"use_tensorboard": true,
|
| 30 |
"log_interval": 50,
|
| 31 |
"log_routing": true,
|
|
|
|
| 25 |
"cutmix_alpha": 1.0,
|
| 26 |
"save_interval": 10,
|
| 27 |
"output_dir": "./checkpoints/cifar100_v2",
|
| 28 |
+
"resume_from": "./checkpoints/cifar100_v2/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/best_model.pt",
|
| 29 |
"use_tensorboard": true,
|
| 30 |
"log_interval": 50,
|
| 31 |
"log_routing": true,
|
weights/run_002_v2_16patch_4tilewormholes_d768_4layer_20251130_045437/training_history.json
CHANGED
|
@@ -1,46 +1,14 @@
|
|
| 1 |
{
|
| 2 |
"loss": [
|
| 3 |
-
|
| 4 |
-
9.826913677117764,
|
| 5 |
-
9.627644357925806,
|
| 6 |
-
9.398329890079987,
|
| 7 |
-
9.207778887870985,
|
| 8 |
-
9.057374726808987,
|
| 9 |
-
9.029630415256207,
|
| 10 |
-
8.938092751380724,
|
| 11 |
-
8.874607647382296
|
| 12 |
],
|
| 13 |
"ce": [
|
| 14 |
-
|
| 15 |
-
4.226625170463171,
|
| 16 |
-
4.139200558417882,
|
| 17 |
-
4.05009743861663,
|
| 18 |
-
3.9672310199493017,
|
| 19 |
-
3.8850555597207483,
|
| 20 |
-
3.8780411903674787,
|
| 21 |
-
3.8434145303872915,
|
| 22 |
-
3.817438381146162
|
| 23 |
],
|
| 24 |
"contrast": [
|
| 25 |
-
|
| 26 |
-
4.307914226483076,
|
| 27 |
-
4.221879853346409,
|
| 28 |
-
4.114024965579693,
|
| 29 |
-
4.031190662506299,
|
| 30 |
-
3.9787070555564683,
|
| 31 |
-
3.9627609515801456,
|
| 32 |
-
3.9189832241107254,
|
| 33 |
-
3.8901301915829
|
| 34 |
],
|
| 35 |
"lr": [
|
| 36 |
-
3.3333333333333335e-05,
|
| 37 |
-
6.666666666666667e-05,
|
| 38 |
-
0.0001,
|
| 39 |
-
0.00013333333333333334,
|
| 40 |
-
0.00016666666666666666,
|
| 41 |
-
0.0002,
|
| 42 |
-
0.00023333333333333333,
|
| 43 |
-
0.0002666666666666667,
|
| 44 |
0.00030000000000000003
|
| 45 |
]
|
| 46 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"loss": [
|
| 3 |
+
8.868392405143151
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
],
|
| 5 |
"ce": [
|
| 6 |
+
3.8159447284845207
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
],
|
| 8 |
"contrast": [
|
| 9 |
+
3.886498171855242
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
],
|
| 11 |
"lr": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
0.00030000000000000003
|
| 13 |
]
|
| 14 |
}
|