Update README.md
Browse files
README.md
CHANGED
|
@@ -100,11 +100,9 @@ requirements:
|
|
| 100 |
---
|
| 101 |
|
| 102 |
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
precision recall f1-score support
|
| 108 |
|
| 109 |
High 0.31 0.43 0.36 1609
|
| 110 |
Low 0.40 0.32 0.36 2599
|
|
@@ -114,11 +112,11 @@ Very_High 0.36 0.54 0.43 1438
|
|
| 114 |
Very_Low 0.80 0.69 0.74 8448
|
| 115 |
Water 0.67 0.93 0.78 584
|
| 116 |
|
| 117 |
-
accuracy
|
| 118 |
macro avg 0.51 0.57 0.53 21541
|
| 119 |
weighted avg 0.64 0.63 0.63 21541
|
| 120 |
|
| 121 |
-
|
| 122 |
[[ 686 74 221 54 451 107 16]
|
| 123 |
[ 317 842 302 76 296 733 33]
|
| 124 |
[ 578 202 276 58 402 237 19]
|
|
@@ -127,25 +125,25 @@ weighted avg 0.64 0.63 0.63 21541
|
|
| 127 |
[ 282 877 329 786 213 5829 132]
|
| 128 |
[ 7 7 1 8 2 15 544]]
|
| 129 |
|
| 130 |
-
|
| 131 |
|
| 132 |
-
|
| 133 |
|
| 134 |
-
|
| 135 |
|
| 136 |
-
|
| 137 |
|
| 138 |
-
|
| 139 |
|
| 140 |
-
|
| 141 |
|
| 142 |
-
|
| 143 |
|
| 144 |
-
|
| 145 |
|
| 146 |
-
|
| 147 |
|
| 148 |
-
|
| 149 |
|
| 150 |
**DISCLOSURES**
|
| 151 |
-Model was developed with assistance from Github Copilot Only for the augmentation techniques. (MIXUP, CUTMIX)
|
|
|
|
| 100 |
---
|
| 101 |
|
| 102 |
|
| 103 |
+
Evaluation Results
|
| 104 |
+
Classification Report
|
| 105 |
+
precision recall f1-score support
|
|
|
|
|
|
|
| 106 |
|
| 107 |
High 0.31 0.43 0.36 1609
|
| 108 |
Low 0.40 0.32 0.36 2599
|
|
|
|
| 112 |
Very_Low 0.80 0.69 0.74 8448
|
| 113 |
Water 0.67 0.93 0.78 584
|
| 114 |
|
| 115 |
+
accuracy 0.63 21541
|
| 116 |
macro avg 0.51 0.57 0.53 21541
|
| 117 |
weighted avg 0.64 0.63 0.63 21541
|
| 118 |
|
| 119 |
+
Confusion Matrix
|
| 120 |
[[ 686 74 221 54 451 107 16]
|
| 121 |
[ 317 842 302 76 296 733 33]
|
| 122 |
[ 578 202 276 58 402 237 19]
|
|
|
|
| 125 |
[ 282 877 329 786 213 5829 132]
|
| 126 |
[ 7 7 1 8 2 15 544]]
|
| 127 |
|
| 128 |
+
Notes
|
| 129 |
|
| 130 |
+
Overall accuracy: 0.63
|
| 131 |
|
| 132 |
+
Best performing classes:
|
| 133 |
|
| 134 |
+
Non-burnable (F1: 0.85)
|
| 135 |
|
| 136 |
+
Very_Low (F1: 0.74)
|
| 137 |
|
| 138 |
+
Water (F1: 0.78)
|
| 139 |
|
| 140 |
+
Weaker performance on:
|
| 141 |
|
| 142 |
+
Moderate
|
| 143 |
|
| 144 |
+
High
|
| 145 |
|
| 146 |
+
Low
|
| 147 |
|
| 148 |
**DISCLOSURES**
|
| 149 |
-Model was developed with assistance from Github Copilot Only for the augmentation techniques. (MIXUP, CUTMIX)
|