wildfire_classifier
Browse files- README.md +62 -0
- all_results.json +12 -0
- config.json +32 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- test_results.json +8 -0
- train_results.json +7 -0
- trainer_state.json +241 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: google/vit-base-patch16-384
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: fine_tuned_wildfire
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# fine_tuned_wildfire
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [google/vit-base-patch16-384](https://huggingface.co/google/vit-base-patch16-384) on the None dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.2329
|
21 |
+
- Accuracy: 0.9202
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.0002
|
41 |
+
- train_batch_size: 16
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: linear
|
46 |
+
- num_epochs: 4
|
47 |
+
|
48 |
+
### Training results
|
49 |
+
|
50 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
51 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
52 |
+
| 0.1208 | 1.28 | 100 | 0.2329 | 0.9202 |
|
53 |
+
| 0.0261 | 2.56 | 200 | 0.2469 | 0.9316 |
|
54 |
+
| 0.0007 | 3.85 | 300 | 0.2358 | 0.9392 |
|
55 |
+
|
56 |
+
|
57 |
+
### Framework versions
|
58 |
+
|
59 |
+
- Transformers 4.33.2
|
60 |
+
- Pytorch 2.0.1+cu117
|
61 |
+
- Datasets 2.14.5
|
62 |
+
- Tokenizers 0.13.3
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9201520912547528,
|
4 |
+
"eval_loss": 0.23285537958145142,
|
5 |
+
"eval_runtime": 53.4625,
|
6 |
+
"eval_samples_per_second": 4.919,
|
7 |
+
"eval_steps_per_second": 0.617,
|
8 |
+
"train_loss": 0.11611736357735936,
|
9 |
+
"train_runtime": 1202.626,
|
10 |
+
"train_samples_per_second": 4.121,
|
11 |
+
"train_steps_per_second": 0.259
|
12 |
+
}
|
config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/vit-base-patch16-384",
|
3 |
+
"architectures": [
|
4 |
+
"ViTForImageClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.0,
|
7 |
+
"encoder_stride": 16,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.0,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"id2label": {
|
12 |
+
"0": "nofire",
|
13 |
+
"1": "fire"
|
14 |
+
},
|
15 |
+
"image_size": 384,
|
16 |
+
"initializer_range": 0.02,
|
17 |
+
"intermediate_size": 3072,
|
18 |
+
"label2id": {
|
19 |
+
"fire": 1,
|
20 |
+
"nofire": 0
|
21 |
+
},
|
22 |
+
"layer_norm_eps": 1e-12,
|
23 |
+
"model_type": "vit",
|
24 |
+
"num_attention_heads": 12,
|
25 |
+
"num_channels": 3,
|
26 |
+
"num_hidden_layers": 12,
|
27 |
+
"patch_size": 16,
|
28 |
+
"problem_type": "single_label_classification",
|
29 |
+
"qkv_bias": true,
|
30 |
+
"torch_dtype": "float32",
|
31 |
+
"transformers_version": "4.33.2"
|
32 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9201520912547528,
|
4 |
+
"eval_loss": 0.23285537958145142,
|
5 |
+
"eval_runtime": 53.4625,
|
6 |
+
"eval_samples_per_second": 4.919,
|
7 |
+
"eval_steps_per_second": 0.617
|
8 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7b830165192a2ba1d0b93c6808e4f7127b382406ef47bbcdddd0b60d978dc2ee
|
3 |
+
size 344391328
|
preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_rescale": true,
|
4 |
+
"do_resize": true,
|
5 |
+
"image_mean": [
|
6 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTFeatureExtractor",
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"resample": 2,
|
17 |
+
"rescale_factor": 0.00392156862745098,
|
18 |
+
"size": {
|
19 |
+
"height": 384,
|
20 |
+
"width": 384
|
21 |
+
}
|
22 |
+
}
|
test_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.966789667896679,
|
4 |
+
"eval_loss": 0.0835316851735115,
|
5 |
+
"eval_runtime": 58.7175,
|
6 |
+
"eval_samples_per_second": 4.615,
|
7 |
+
"eval_steps_per_second": 0.579
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"train_loss": 0.11611736357735936,
|
4 |
+
"train_runtime": 1202.626,
|
5 |
+
"train_samples_per_second": 4.121,
|
6 |
+
"train_steps_per_second": 0.259
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,241 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.23285537958145142,
|
3 |
+
"best_model_checkpoint": "./fine_tuned_wildfire\\checkpoint-100",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 312,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.13,
|
13 |
+
"learning_rate": 0.00019423076923076924,
|
14 |
+
"loss": 0.663,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.26,
|
19 |
+
"learning_rate": 0.00018782051282051283,
|
20 |
+
"loss": 0.2291,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.38,
|
25 |
+
"learning_rate": 0.00018205128205128207,
|
26 |
+
"loss": 0.2045,
|
27 |
+
"step": 30
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.51,
|
31 |
+
"learning_rate": 0.00017564102564102566,
|
32 |
+
"loss": 0.3236,
|
33 |
+
"step": 40
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.64,
|
37 |
+
"learning_rate": 0.00016923076923076923,
|
38 |
+
"loss": 0.2852,
|
39 |
+
"step": 50
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.77,
|
43 |
+
"learning_rate": 0.00016282051282051282,
|
44 |
+
"loss": 0.2129,
|
45 |
+
"step": 60
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.9,
|
49 |
+
"learning_rate": 0.00015641025641025642,
|
50 |
+
"loss": 0.2448,
|
51 |
+
"step": 70
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 1.03,
|
55 |
+
"learning_rate": 0.00015000000000000001,
|
56 |
+
"loss": 0.2508,
|
57 |
+
"step": 80
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 1.15,
|
61 |
+
"learning_rate": 0.0001435897435897436,
|
62 |
+
"loss": 0.1551,
|
63 |
+
"step": 90
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 1.28,
|
67 |
+
"learning_rate": 0.00013717948717948718,
|
68 |
+
"loss": 0.1208,
|
69 |
+
"step": 100
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 1.28,
|
73 |
+
"eval_accuracy": 0.9201520912547528,
|
74 |
+
"eval_loss": 0.23285537958145142,
|
75 |
+
"eval_runtime": 50.1675,
|
76 |
+
"eval_samples_per_second": 5.242,
|
77 |
+
"eval_steps_per_second": 0.658,
|
78 |
+
"step": 100
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 1.41,
|
82 |
+
"learning_rate": 0.00013076923076923077,
|
83 |
+
"loss": 0.087,
|
84 |
+
"step": 110
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 1.54,
|
88 |
+
"learning_rate": 0.00012435897435897437,
|
89 |
+
"loss": 0.1733,
|
90 |
+
"step": 120
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 1.67,
|
94 |
+
"learning_rate": 0.00011794871794871796,
|
95 |
+
"loss": 0.0241,
|
96 |
+
"step": 130
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 1.79,
|
100 |
+
"learning_rate": 0.00011153846153846154,
|
101 |
+
"loss": 0.1283,
|
102 |
+
"step": 140
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 1.92,
|
106 |
+
"learning_rate": 0.00010512820512820514,
|
107 |
+
"loss": 0.281,
|
108 |
+
"step": 150
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 2.05,
|
112 |
+
"learning_rate": 9.871794871794872e-05,
|
113 |
+
"loss": 0.0491,
|
114 |
+
"step": 160
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 2.18,
|
118 |
+
"learning_rate": 9.230769230769232e-05,
|
119 |
+
"loss": 0.0263,
|
120 |
+
"step": 170
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 2.31,
|
124 |
+
"learning_rate": 8.58974358974359e-05,
|
125 |
+
"loss": 0.0276,
|
126 |
+
"step": 180
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 2.44,
|
130 |
+
"learning_rate": 7.948717948717948e-05,
|
131 |
+
"loss": 0.025,
|
132 |
+
"step": 190
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 2.56,
|
136 |
+
"learning_rate": 7.307692307692307e-05,
|
137 |
+
"loss": 0.0261,
|
138 |
+
"step": 200
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 2.56,
|
142 |
+
"eval_accuracy": 0.9315589353612167,
|
143 |
+
"eval_loss": 0.2469397932291031,
|
144 |
+
"eval_runtime": 49.5425,
|
145 |
+
"eval_samples_per_second": 5.309,
|
146 |
+
"eval_steps_per_second": 0.666,
|
147 |
+
"step": 200
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 2.69,
|
151 |
+
"learning_rate": 6.666666666666667e-05,
|
152 |
+
"loss": 0.0239,
|
153 |
+
"step": 210
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 2.82,
|
157 |
+
"learning_rate": 6.025641025641026e-05,
|
158 |
+
"loss": 0.011,
|
159 |
+
"step": 220
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 2.95,
|
163 |
+
"learning_rate": 5.384615384615385e-05,
|
164 |
+
"loss": 0.0276,
|
165 |
+
"step": 230
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 3.08,
|
169 |
+
"learning_rate": 4.7435897435897435e-05,
|
170 |
+
"loss": 0.0018,
|
171 |
+
"step": 240
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 3.21,
|
175 |
+
"learning_rate": 4.1025641025641023e-05,
|
176 |
+
"loss": 0.0013,
|
177 |
+
"step": 250
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 3.33,
|
181 |
+
"learning_rate": 3.461538461538462e-05,
|
182 |
+
"loss": 0.0005,
|
183 |
+
"step": 260
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 3.46,
|
187 |
+
"learning_rate": 2.8205128205128207e-05,
|
188 |
+
"loss": 0.0174,
|
189 |
+
"step": 270
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 3.59,
|
193 |
+
"learning_rate": 2.1794871794871795e-05,
|
194 |
+
"loss": 0.0002,
|
195 |
+
"step": 280
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 3.72,
|
199 |
+
"learning_rate": 1.5384615384615387e-05,
|
200 |
+
"loss": 0.0003,
|
201 |
+
"step": 290
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 3.85,
|
205 |
+
"learning_rate": 8.974358974358976e-06,
|
206 |
+
"loss": 0.0007,
|
207 |
+
"step": 300
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 3.85,
|
211 |
+
"eval_accuracy": 0.9391634980988594,
|
212 |
+
"eval_loss": 0.23577481508255005,
|
213 |
+
"eval_runtime": 49.794,
|
214 |
+
"eval_samples_per_second": 5.282,
|
215 |
+
"eval_steps_per_second": 0.663,
|
216 |
+
"step": 300
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 3.97,
|
220 |
+
"learning_rate": 2.564102564102564e-06,
|
221 |
+
"loss": 0.0006,
|
222 |
+
"step": 310
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 4.0,
|
226 |
+
"step": 312,
|
227 |
+
"total_flos": 1.1324765556186808e+18,
|
228 |
+
"train_loss": 0.11611736357735936,
|
229 |
+
"train_runtime": 1202.626,
|
230 |
+
"train_samples_per_second": 4.121,
|
231 |
+
"train_steps_per_second": 0.259
|
232 |
+
}
|
233 |
+
],
|
234 |
+
"logging_steps": 10,
|
235 |
+
"max_steps": 312,
|
236 |
+
"num_train_epochs": 4,
|
237 |
+
"save_steps": 100,
|
238 |
+
"total_flos": 1.1324765556186808e+18,
|
239 |
+
"trial_name": null,
|
240 |
+
"trial_params": null
|
241 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:11845d4df1cc4bd7ca99e8d1a206179601d32b1523415de2370a26a2a0b29c66
|
3 |
+
size 4027
|