Merge branch 'main' of https://github.com/se4ai2324-uniba/DetectionOf…

…OnlineSexism

DockerDocumentation.md

@@ -1,6 +1,8 @@
+# Docker
+
 Docker is a lightweight virtualization tool that allows us to isolate the development environment and ensure the reproducibility of our experiments.
 
-## Docker Components Contents
+## Components
 
 - `Dockerfile`: The Dockerfile contains instructions for creating the Docker image. This file defines the environment and dependencies required to run our machine learning project.
 
@@ -10,7 +12,7 @@ Docker is a lightweight virtualization tool that allows us to isolate the develo
 
 - `data/`: This folder should contain all the training and test data required for the machine learning project. Make sure the data is accessible within the Docker container.
 
-- `notebooks/`: In this folder, you may want to place Jupyter notebooks or other development scripts that we will use during the project. These can be mounted inside the Docker container for execution and editing.
+- `src/`: In this folder, you may want to place our development scripts that we will use during the project. These can be mounted inside the Docker container for execution and editing.
 
 ## Usage
 
@@ -43,7 +45,9 @@ docker-compose up --build
 
 You can now start working inside the container.
 
-5. Access the Docker development environment and begin working on our machine learning projects.
+5. Access the Docker development environment and begin working on our machine learning projects. In our case, we can se the following dashboard:
+
+![Docker Dashboard](./references/images_doc/DockerDashboard.png)
 
 6. When you're done, you can stop the Docker container with the following command:
 

tests/Documentation.md

@@ -1,4 +1,3 @@
-
 # Test documentation
 
 In order to check the correctness of our project, we conducted several tests:
@@ -19,17 +18,11 @@ For more information see [GitHubRepo](https://github.com/se4ai2324-uniba/Detecti
 ## Behavioral tests
 ### Directional Test
 
-This type of test creates messages with distinct tokens and use the predictive models of task A and B to obtain the predicted labels for these messages. 
-
-The primary objective is to assert that the predicted labels for messages featuring different tokens are not identical.
-
-So this test is designed to verify that the model can distinguish between specific tokens and generate varying predictions based on the token used in the input message.
+This type of test creates messages with distinct tokens and use the predictive models of task A and B to obtain the predicted labels for these messages. The primary objective is to assert that the predicted labels for messages featuring different tokens are not identical. So this test is designed to verify that the model can distinguish between specific tokens and generate varying predictions based on the token used in the input message.
 
 ### Invariance Test
 
-The objective of this test is to ensure that the model's classification of a message is not influenced by the choice of synonyms like "send" or "give." The test aims to confirm that the model maintains its ability to recognize and categorize messages consistently even when variations in language are introduced. 
-
-This evaluation is essential for assessing the robustness of the model's discriminatory language detection across different expressions with similar meanings.
+The objective of this test is to ensure that the model's classification of a message is not influenced by the choice of synonyms like "send" or "give." The test aims to confirm that the model maintains its ability to recognize and categorize messages consistently even when variations in language are introduced. This evaluation is essential for assessing the robustness of the model's discriminatory language detection across different expressions with similar meanings.
 
 ### Minimum Functionality Test
 
@@ -45,7 +38,9 @@ It validates the training accuracy ensuring the model approximates a specified t
 
 ## Preprocessing tests
 
-The tests cover scenarios such as punctuation removal, uppercase text transformation, empty input handling, and processing mixed text and numbers. 
+The tests cover scenarios such as punctuation removal, uppercase text transformation, empty input handling, and processing mixed text and numbers. Successful execution confirms the effective preprocessing capability of the `clean_text` function.
+
+## API tests
+Tests concerning APIs implemented with FastAPI are crucial to ensure that the application is robust, reliable, and easy to maintain and develop over time to improve its efficiency. Tests ensure that all the application's functionalities work as expected, verifying that endpoints return the correct data and respond with the appropriate status codes. When changes are made to the code, tests help ensure that existing functionalities are not accidentally disrupted. Finally, by testing various scenarios, including behavior in abnormal conditions or the submission of invalid input, the application can be ensured to be secure and stable.
 
-Successful execution confirms the effective preprocessing capability of the clean_text function.