This repository has been archived by the owner on Jul 11, 2024. It is now read-only.
forked from jsmok-psnc/DEEP-OC-integrated_plant_protection
-
Notifications
You must be signed in to change notification settings - Fork 0
/
metadata.json
50 lines (49 loc) · 2.37 KB
/
metadata.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
{
"title": "Integrated Plant Protection",
"summary": "A Torch model to classify plant.",
"description": [
"Integrated Plant Protection is an application using the DEEPaaS API.\n",
"The module contains 4 trained models based on CNN architecture with reduced network capacity due to the size of datasets.",
"The pipeline is implemented in [PyTorch](https://pytorch.org/).\n",
"For rye, the first classifier in the normal approach is rye-3 and for beets it is beet-52.\n",
"In the second approach, the impact of using preprocessing before the classifier was checked.\n",
"For this purpose, masks were collected and models were trained with U-Net architecture for segmentation. rye-1 for rye and beet-1 for beet.\n",
"Two more classifiers, namely rye-7 and beet-54, were used to preprocess the corresponding models.\n",
"It was possible to train new models without and with preprocessing.\n",
"The same scheme is then used for prediction, with the difference that preprocessing is done immediately before GPU inference.\n"
],
"keywords": [
"docker",
"api-v2",
"pytorch",
"deep learning",
"pre-trained",
"trainable",
"imagine classification",
"vo.ai4eosc.eu"
],
"license": "MIT",
"date_creation": "2023-12-20",
"sources": {
"dockerfile_repo": "https://github.com/ai4eosc-psnc/DEEP-OC-integrated_plant_protection",
"docker_registry_repo": "deephdc/uc-ai4eosc-psnc-deep-oc-integrated_plant_protection",
"code": "https://github.com/ai4eosc-psnc/integrated_plant_protection"
},
"continuous_integration": {
"build_status_badge": "https://jenkins.indigo-datacloud.eu/buildStatus/icon?job=Pipeline-as-code/DEEP-OC-org/UC-ai4eosc-psnc-DEEP-OC-integrated_plant_protection/master",
"build_status_url": "https://jenkins.indigo-datacloud.eu/job/Pipeline-as-code/job/DEEP-OC-org/job/UC-ai4eosc-psnc-DEEP-OC-integrated_plant_protection/job/master"
},
"tosca": [
{
"title": "Mesos (CPU)",
"url": "https://raw.githubusercontent.com/indigo-dc/tosca-templates/master/deep-oc/deep-oc-marathon-webdav.yml",
"inputs": [
"rclone_conf",
"rclone_url",
"rclone_vendor",
"rclone_user",
"rclone_pass"
]
}
]
}