ivandrian11
commited on
Commit
•
a0294fc
1
Parent(s):
2231cef
Update README.md
Browse files
README.md
CHANGED
@@ -35,17 +35,23 @@ It achieves the following results on the evaluation set:
|
|
35 |
- Loss: 1.0194
|
36 |
- Accuracy: 0.9699
|
37 |
|
38 |
-
##
|
39 |
-
|
40 |
-
More information needed
|
41 |
|
42 |
-
|
43 |
|
44 |
-
|
|
|
|
|
|
|
45 |
|
46 |
-
|
|
|
|
|
|
|
47 |
|
48 |
-
|
|
|
|
|
49 |
|
50 |
## Training procedure
|
51 |
|
|
|
35 |
- Loss: 1.0194
|
36 |
- Accuracy: 0.9699
|
37 |
|
38 |
+
## Training and evaluation data
|
|
|
|
|
39 |
|
40 |
+
This model was fine-tuned on [the Fruits-30 dataset](https://huggingface.co/datasets/VinayHajare/Fruits-30), a collection of images featuring 30 different types of fruits. Each image has been preprocessed and standardized to a size of 224x224 pixels for uniformity.
|
41 |
|
42 |
+
### Dataset Composition
|
43 |
+
- Number of Classes: 30
|
44 |
+
- Image Resolution: 224x224 pixels
|
45 |
+
- Total Images: 826
|
46 |
|
47 |
+
### Training and Evaluation Split
|
48 |
+
The dataset was split into training and evaluation sets using dataset.train_test_split function with a 80/20 train-test split, resulting in:
|
49 |
+
- Training Set: 660 images
|
50 |
+
- Evaluation Set: 166 images
|
51 |
|
52 |
+
### Splitting Strategy
|
53 |
+
- The data was shuffled (shuffle=True) before splitting to ensure a random distribution of classes across the training and evaluation sets.
|
54 |
+
- Additionally, stratification was applied based on the "label" column (stratify_by_column='label') to maintain a balanced class distribution across both sets. This helps prevent the model from biasing towards classes with more samples in the training data.
|
55 |
|
56 |
## Training procedure
|
57 |
|