You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Verify that all system components function correctly:
38
42
```bash
39
43
docker run --rm image-classification make test
40
-
```
44
+
```-->
41
45
42
-
## Usage
46
+
<!--## Usage
43
47
44
48
### Docker Container Execution
45
49
Execute the project within Docker, automating the full workflow from preprocessing to postprocessing:
@@ -62,24 +66,28 @@ Execute the project within Docker, automating the full workflow from preprocessi
62
66
Alternatively, you can execute the entire pipeline using the default command:
63
67
```bash
64
68
docker run --rm -v $(pwd)/input_raw:/input_raw -v $(pwd)/output:/output image-classification
65
-
```
69
+
```-->
66
70
67
71
## Makefile Details
68
72
The project provides two Makefiles:
69
73
70
74
### Makefile (Build System)
71
75
This Makefile is used to compile all binary executables and/or libraries necessary for the project. The Dockerfile relies on this Makefile to build the project inside a Docker image. It includes the following targets:
72
76
77
+
-**all**: Installs all dependencies and compiles all executables, i.e. steps prereqs and build.
73
78
-**prereqs**: Installs all dependencies, compiles, and installs necessary components.
74
79
-**build**: Compiles all necessary executables, including preprocessing (`preprocessing.out`) and postprocessing (`postprocessing.out`) executables.
75
80
-**test**: Tests all steps of the pipeline, including verifying the correctness of preprocessing, processing, and postprocessing.
81
+
-**clean**: Clean target to remove built files or unnecessary directories.
82
+
76
83
77
84
### Makefile.run (Entry Point for Docker Image)
78
85
This Makefile serves as an entry point for running the pipeline in the Docker container. It orchestrates the following steps:
79
86
80
87
-**preprocess**: Converts all files in `/input_raw` to a standardized format in `/input`.
81
-
-**process**: Uses the neural network model to classify images from `/input` and writes results to `/output_raw`.
88
+
-**inference**: Uses the neural network model to classify images from `/input` and writes results to `/output_raw`.
82
89
-**postprocess**: Annotates images using results from `/output_raw` and saves them in `/output`.
90
+
-**run**: Run the whole pipeline, i.e. it will execute preprocess, then inference, and, finally, postprocess.
83
91
84
92
## Example Input and Output
85
93
-**Input**: The input is a raw image (`image.jpg`), which is first downscaled to `224x224` pixels in the preprocessing step.
0 commit comments