fix quote
Browse files
README.md
CHANGED
|
@@ -12,6 +12,6 @@ A crude CV to classify retinal images. I started with a pretrained model (resnet
|
|
| 12 |
Limitations stem primarily from a naïve use of the training data. Imbalances exist in the number of training images associated with each condition. In the future, a weighting function could be applied.
|
| 13 |
|
| 14 |
## Training and evaluation data
|
| 15 |
-
|
| 16 |
|
| 17 |
The model was tuned with 50 epochs. accuracy-multi: 91.4%; valid-loss: 26.0%; train-loss: 26.0%
|
|
|
|
| 12 |
Limitations stem primarily from a naïve use of the training data. Imbalances exist in the number of training images associated with each condition. In the future, a weighting function could be applied.
|
| 13 |
|
| 14 |
## Training and evaluation data
|
| 15 |
+
About 400 retinal images were downloaded from STATE (http://cecas.clemson.edu/~ahoover/stare/).
|
| 16 |
|
| 17 |
The model was tuned with 50 epochs. accuracy-multi: 91.4%; valid-loss: 26.0%; train-loss: 26.0%
|