Update the readme files and configuration file examples.

Signed-off-by: khaoula boutiche <[email protected]>

Author: KBOUSTM

image_classification/src/README.md

@@ -211,7 +211,7 @@ data_augmentation:
 
 The data augmentation functions are applied to the input images in their order of appearance in the configuration file. If an argument of a given function is omitted, the default value is used. There are no constraints on the number of functions, types of functions and order of functions. 
 
-Please refer to [the data augmentation documentation](https://codex.cro.st.com/plugins/git/external-model-zoo/external_model_zoo?a=blob&hb=8e976faa7f22a7d9954295f0e74cdb99fbb6572a&h=27f56a5b1d5a6b96d256afbb889b8e5e1c8d196b&f=image_classification%2Fsrc%2Fdata_augmentation%2FREADME.md) for a list of functions that are available and the transforms they apply to the input images.
+Please refer to [the data augmentation documentation](data_augmentation/README.md) for a list of functions that are available and the transforms they apply to the input images.
 
 #### <a id="2-6">2.6 Set the training parameters</a>
 
@@ -354,7 +354,7 @@ Finally, in the `deployment` section, users must provide information about the h
 
 #### <a id="2-10">2.10 Hydra and MLflow settings</a>
 
-The `mlflow` and `hydra` sections must always be present in the YAML configuration file. The `hydra` section can be used to specify the name of the directory where experiment directories are saved and/or the pattern used to name experiment directories. In the YAML code below, it is set to save the outputs as explained in the section <a id="4">visualize the chained services results</a>:
+The `mlflow` and `hydra` sections must always be present in the YAML configuration file. The `hydra` section can be used to specify the name of the directory where experiment directories are saved and/or the pattern used to name experiment directories. In the YAML code below, it is set to save the outputs as explained in the section <a href="#4">visualize the chained services results</a> :
 
 ```yaml
 hydra:

image_classification/src/config_file_examples/chain_eqe_config.yaml

@@ -6,7 +6,7 @@ operation_mode: chain_eqe
 dataset:
   name: flowers
   class_names: [daisy, dandelion, roses, sunflowers, tulips]
-  test_path: ../datasets/flower_photos  
+  test_path: ../datasets/flower_photos_testset
   quantization_path: ../datasets/flower_photos
 
 preprocessing:

image_classification/src/config_file_examples/chain_qd_config.yaml

@@ -1,10 +1,51 @@
 general:
    model_path: ../pretrained_models/squeezenetv1.1/ST_pretrainedmodel_public_dataset/plant-village/squeezenetv1.1_128_tfs/squeezenetv1.1_128_tfs.h5
-operation_mode: chain_eqeb
+
+operation_mode: chain_qd
 
 dataset:
-   quantization_path: ../datasets/plant-village
-   quantization_split: 0.5
+  class_names:
+  - Apple___Apple_scab
+  - Apple___Black_rot
+  - Apple___Cedar_apple_rust
+  - Apple___healthy
+  - Background_without_leaves
+  - Blueberry___healthy
+  - Cherry___healthy
+  - Cherry___Powdery_mildew
+  - Corn___Cercospora_leaf_spot Gray_leaf_spot
+  - Corn___Common_rust
+  - Corn___healthy
+  - Corn___Northern_Leaf_Blight
+  - Grape___Black_rot
+  - Grape___Esca_(Black_Measles)
+  - Grape___healthy
+  - Grape___Leaf_blight_(Isariopsis_Leaf_Spot)
+  - Orange___Haunglongbing_(Citrus_greening)
+  - Peach___Bacterial_spot
+  - Peach___healthy
+  - Pepper,_bell___Bacterial_spot
+  - Pepper,_bell___healthy
+  - Potato___Early_blight
+  - Potato___healthy
+  - Potato___Late_blight
+  - Raspberry___healthy
+  - Soybean___healthy
+  - Squash___Powdery_mildew
+  - Strawberry___healthy
+  - Strawberry___Leaf_scorch
+  - Tomato___Bacterial_spot
+  - Tomato___Early_blight
+  - Tomato___healthy
+  - Tomato___Late_blight
+  - Tomato___Leaf_Mold
+  - Tomato___Septoria_leaf_spot
+  - Tomato___Spider_mites Two-spotted_spider_mite
+  - Tomato___Target_Spot
+  - Tomato___Tomato_mosaic_virus
+  - Tomato___Tomato_Yellow_Leaf_Curl_Virus
+  quantization_path: ../datasets/Plant_leave_diseases_dataset_without_augmentation
+  quantization_split: 0.5
 
 preprocessing:
    rescaling:

image_classification/src/config_file_examples/evaluation_config.yaml

@@ -4,7 +4,7 @@ general:
 operation_mode: evaluation
 
 dataset:
-   test_path: ../datasets/flower_test_set
+   test_path: ../datasets/flower_photos_testset
 
 preprocessing:
    rescaling:

image_classification/src/config_file_examples/quantization_config.yaml

@@ -6,7 +6,6 @@ operation_mode: quantization
 dataset:
   name: flowers
   class_names: [daisy, dandelion, roses, sunflowers, tulips]
-  test_path:
   quantization_path: ../datasets/flower_photos
   quantization_split: 0.4
 

image_classification/src/evaluation/README.md

@@ -25,7 +25,7 @@ Alternatively, you can follow the tutorial below, which shows how to evaluate yo
 
 ### <a id="1-1">1.1 Setting the model and the operation mode</a>
 
-As mentioned previously, all the sections of the YAML file must be set in accordance with this **[README.md](../config_file_examples/evaluation_config.yaml)**.
+As mentioned previously, all the sections of the YAML file must be set in accordance with this **[evaluation_config.yaml](../config_file_examples/evaluation_config.yaml)**.
 In particular, `operation_mode` should be set to evaluation and the `evaluation` section should be filled as in the following example: 
 
 ```yaml

image_classification/src/training/README.md

@@ -223,7 +223,7 @@ training:
          decay_steps: 170
          alpha: 0.001
 ```
-Refer to <a href="#A">Appendix A: Learning rate schedulers</a> for a list of the available learning rate schedulers.
+Refer to [Appendix A: Learning rate schedulers](#A) for a list of the available learning rate schedulers.
 
 ### <a id="2-8">2.8 Hydra and MLflow settings</a>
 

object_detection/pretrained_models/README.md

@@ -5,7 +5,7 @@ The STM32 model zoo includes several models for object detection use case pre-tr
 Under each model directory, you can find the following model categories:
 
 - `Public_pretrainedmodel_public_dataset` contains public object detection models trained on public datasets.
-- `ST_pretrainedmodel_custom_dataset` contains different object detection models trained on ST custom datasets using our [training scripts](../src/training/README.md). 
+- `ST_pretrainedmodel_custom_dataset` contains different object detection models trained on ST custom datasets using our [training scripts](../src/training). 
 - `ST_pretrainedmodel_public_dataset` contains different object detection models trained on various public datasets following the [training section](../src/training/README.md) in STM32 model zoo.
 
 

object_detection/src/config_file_examples/chain_eqe_config.yaml

@@ -6,8 +6,8 @@ operation_mode: chain_eqe
 dataset:
   name: coco_2017_person
   class_names: [person]
-  test_path: ../datasets/coco_2017_person/
-  quantization_path: ../datasets/coco_2017_person/
+  test_path: ../datasets/coco_2017_person/test
+  quantization_path: ../datasets/coco_2017_person/train
 
 preprocessing:
    rescaling:

object_detection/src/config_file_examples/chain_eqeb_config.yaml

@@ -5,9 +5,9 @@ operation_mode: chain_eqeb
 
 dataset:
   class_names: [person]
-  training_path: ../datasets/coco_person_2017/
+  training_path: ../datasets/coco_person_2017/train
   validation_split: 0.1
-  quantization_path: ../datasets/coco_person_2017/
+  quantization_path: ../datasets/coco_person_2017/train
   quantization_split: 0.05
 
 preprocessing: