1# galdric
 2machine learning server, for image classification
 3
 4
 5 - Reads all the datasets in the folder /dataset
 6 - Runs a server that allows to upload images to classify
 7 - Accepts PNG, JPG, JPEG
 8 - Each image is resized to the same size and converted to PNG type, configured in the config.json
 9 - For the input images, calculates the euclidean distances
10 - Applyies KNN (K-Nearest Neighbours algorithm) to classify the images
11 - Server returns the classification result, that is the label of the object in the image
12
13
14
15### Instructions
16
17Put dataset in /dataset directory with subdirectories, where each subdirectory contains images of one element.
18
19For example:
20```
21dataset/
22    leopard/
23        img01.png
24        img02.png
25        img03.png
26        ...
27    laptop/
28        img01.png
29        img02.png
30        ...
31    camera/
32        img01.png
33        img02.png
34        ...
35```
36So, we have each image and to which element category is (the name of subdirectory).
37
38
39Then, run the server:
40```
41    >./galdric
42```
43
44Now, just need to perform petitions with new images, to get the response from the server classifying them:
45```bash
46    curl -F file=@./testimage.png http://127.0.0.1:3055/image
47```
48And the server will return:
49```
50    seems to be a leopard
51```
52
53Can perform some tests with the test.sh file:
54```
55    bash test.sh
56```
57
58
59-------------
60#### Useful commands
61
62send file over ssh:
63```
64scp dataset.tar.gz root@SERVERIP:/root/galdric
65```
66
67on the server, untar file:
68```
69tar -xvzf dataset.tar.gz
70```