Commit
•
6981863
1
Parent(s):
1bcd0fe
Update README.md
Browse files
README.md
CHANGED
|
@@ -13,7 +13,7 @@ model-index:
|
|
| 13 |
name: Image Classification
|
| 14 |
type: image-classification
|
| 15 |
dataset:
|
| 16 |
-
name:
|
| 17 |
type: image_folder
|
| 18 |
config: default
|
| 19 |
split: train
|
|
@@ -21,7 +21,7 @@ model-index:
|
|
| 21 |
metrics:
|
| 22 |
- name: Accuracy
|
| 23 |
type: accuracy
|
| 24 |
-
value: 0.
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
This model is a fine-tuned version of [google/mobilenet_v2_1.0_224](https://huggingface.co/google/mobilenet_v2_1.0_224) on the [Kaggle version](https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset) of the [Plant Village dataset](https://github.com/spMohanty/PlantVillage-Dataset).
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
-
- Cross Entropy Loss:
|
| 35 |
-
- Accuracy: 0.
|
| 36 |
|
| 37 |
Will be further training it (such as finding optimal hyperparameters) better to achieve much better accuracy.
|
| 38 |
|
|
@@ -47,28 +47,13 @@ The plant village dataset consists of 38 classes of diseases in common crops (in
|
|
| 47 |
### Training hyperparameters
|
| 48 |
|
| 49 |
The following hyperparameters were used during training:
|
| 50 |
-
- learning_rate:
|
| 51 |
- train_batch_size: 256
|
| 52 |
- eval_batch_size: 256
|
| 53 |
-
- seed: 42
|
| 54 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 55 |
- lr_scheduler_type: linear
|
| 56 |
-
- lr_scheduler_warmup_ratio: 0.
|
| 57 |
-
- num_epochs:
|
| 58 |
-
|
| 59 |
-
### Training results
|
| 60 |
-
|
| 61 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 62 |
-
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 63 |
-
| 2.9265 | 1.0 | 248 | 2.7159 | 0.4703 |
|
| 64 |
-
| 1.9734 | 2.0 | 496 | 1.7668 | 0.6649 |
|
| 65 |
-
| 1.7206 | 3.0 | 744 | 1.4012 | 0.7206 |
|
| 66 |
-
| 1.6406 | 4.0 | 992 | 1.2514 | 0.7644 |
|
| 67 |
-
| 1.6075 | 5.0 | 1240 | 1.2934 | 0.7094 |
|
| 68 |
-
| 1.5932 | 6.0 | 1488 | 1.2093 | 0.7257 |
|
| 69 |
-
| 1.5203 | 7.0 | 1736 | 1.0461 | 0.7858 |
|
| 70 |
-
| 1.5076 | 8.0 | 1984 | 1.0580 | 0.7848 |
|
| 71 |
-
|
| 72 |
|
| 73 |
### Framework versions
|
| 74 |
|
|
|
|
| 13 |
name: Image Classification
|
| 14 |
type: image-classification
|
| 15 |
dataset:
|
| 16 |
+
name: New Plant Diseases Dataset
|
| 17 |
type: image_folder
|
| 18 |
config: default
|
| 19 |
split: train
|
|
|
|
| 21 |
metrics:
|
| 22 |
- name: Accuracy
|
| 23 |
type: accuracy
|
| 24 |
+
value: 0.9541
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
This model is a fine-tuned version of [google/mobilenet_v2_1.0_224](https://huggingface.co/google/mobilenet_v2_1.0_224) on the [Kaggle version](https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset) of the [Plant Village dataset](https://github.com/spMohanty/PlantVillage-Dataset).
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
+
- Cross Entropy Loss: 0.15
|
| 35 |
+
- Accuracy: 0.9541
|
| 36 |
|
| 37 |
Will be further training it (such as finding optimal hyperparameters) better to achieve much better accuracy.
|
| 38 |
|
|
|
|
| 47 |
### Training hyperparameters
|
| 48 |
|
| 49 |
The following hyperparameters were used during training:
|
| 50 |
+
- learning_rate: 5e-5
|
| 51 |
- train_batch_size: 256
|
| 52 |
- eval_batch_size: 256
|
|
|
|
| 53 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
- lr_scheduler_type: linear
|
| 55 |
+
- lr_scheduler_warmup_ratio: 0.2
|
| 56 |
+
- num_epochs: 6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
### Framework versions
|
| 59 |
|