edadaltocg commited on
Commit
0c27122
1 Parent(s): 6dc0737

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +87 -0
README.md ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ license: mit
4
+ library_name: timm
5
+ tags:
6
+ - image-classification
7
+ - densenet121
8
+ - cifar100
9
+ datasets: cifar100
10
+ metrics:
11
+ - accuracy
12
+ model-index:
13
+ - name: densenet121_cifar100
14
+ results:
15
+ - task:
16
+ type: image-classification
17
+ dataset:
18
+ name: CIFAR-100
19
+ type: cifar100
20
+ metrics:
21
+ - type: accuracy
22
+ value: 0.7619
23
+ ---
24
+
25
+ # Model Card for Model ID
26
+
27
+ This model is a small densenet121 trained on cifar100.
28
+
29
+ - **Test Accuracy:** 0.7619
30
+ - **License:** MIT
31
+
32
+ ## How to Get Started with the Model
33
+
34
+ Use the code below to get started with the model.
35
+
36
+ ```python
37
+ import detectors
38
+ import timm
39
+
40
+ model = timm.create_model("densenet121_cifar100", pretrained=True)
41
+ ```
42
+
43
+ ## Training Data
44
+
45
+ Training data is cifar100.
46
+
47
+ ## Training Hyperparameters
48
+
49
+
50
+ - **config**: `scripts/train_configs/cifar100.json`
51
+
52
+ - **model**: `densenet121_cifar100`
53
+
54
+ - **dataset**: `cifar100`
55
+
56
+ - **batch_size**: `128`
57
+
58
+ - **epochs**: `300`
59
+
60
+ - **validation_frequency**: `5`
61
+
62
+ - **seed**: `1`
63
+
64
+ - **criterion**: `CrossEntropyLoss`
65
+
66
+ - **criterion_kwargs**: `{}`
67
+
68
+ - **optimizer**: `SGD`
69
+
70
+ - **lr**: `0.1`
71
+
72
+ - **optimizer_kwargs**: `{'momentum': 0.9, 'weight_decay': 0.0005}`
73
+
74
+ - **scheduler**: `CosineAnnealingLR`
75
+
76
+ - **scheduler_kwargs**: `{'T_max': 280}`
77
+
78
+ - **debug**: `False`
79
+
80
+
81
+ ## Testing Data
82
+
83
+ Testing data is cifar100.
84
+
85
+ ---
86
+
87
+ This model card was created by Eduardo Dadalto.