Skip to content

Commit c660a50

Browse files
aalborovAlexanderDokuchaev
aalborov
and
AlexanderDokuchaev
committed
Documentation Review (open-edge-platform#211)
* 1/2 of README's * 4 more README's * fix * fix * fixes * Documentation Review * Update tensorflow_toolkit/ssd_detector/README.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Update pytorch_toolkit/action_recognition/README.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Update pytorch_toolkit/action_recognition/README.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Update pytorch_toolkit/face_recognition/README.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Update tensorflow_toolkit/ssd_detector/README.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Update pytorch_toolkit/nncf/README.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * some additional fixes * some additional fixes * Update README.md * Update pytorch_toolkit/human_pose_estimation/README_single.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Update pytorch_toolkit/human_pose_estimation/README_single.md Co-Authored-By: Alexander Dokuchaev <[email protected]> * Remove extra spaces * Remove extra spaces * Fix path in doc * Fix decode ascii * Remove extra spaces * Remove extra spaces Co-authored-by: Alexander Dokuchaev <[email protected]>
1 parent 440bbba commit c660a50

File tree

59 files changed

+1893
-1779
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

59 files changed

+1893
-1779
lines changed

‎.github/ISSUE_TEMPLATE/bug_report.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -29,13 +29,13 @@ assignees: AlexanderDokuchaev
2929

3030

3131
**Environment:**
32-
- OS: <!--[e.g. Linux Ubuntu 16.04]-->
32+
- OS: <!--[for example, Linux Ubuntu 16.04]-->
3333
- Framework version: <!--[TensorFlow or PyTorch]-->
3434
- Python version:
3535
- OpenVINO version:
3636
- CUDA/cuDNN version:
3737
- GPU model and memory:
3838
<!--
39-
Include as many relevant details about the environment with which you experienced the bug.
39+
Include as many relevant details about the environment in which you experienced the bug as you can.
4040
-->
4141

‎.github/ISSUE_TEMPLATE/feature_request.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -13,10 +13,10 @@ assignees: AlexanderDokuchaev
1313

1414
**Is your feature request related to a problem? Please describe.**
1515
<!--
16-
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
16+
A clear and concise description of what the problem is. For example: I'm always frustrated when [...]
1717
-->
1818

19-
**Describe the solution you'd like**
19+
**Describe the solution you'd like to propose.**
2020
<!--
2121
A clear and concise description of what you want to happen.
2222
-->

‎.github/ISSUE_TEMPLATE/question.md‎

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -10,6 +10,5 @@ assignees: AlexanderDokuchaev
1010
<!--
1111
Thank you very much for contributing to this project by creating an issue!
1212
13-
Try to clear and concise describe your quastion.
14-
Do not forget to specify the context.
13+
Try to describe your issue clearly and concisely and include the context.
1514
-->

‎README.md‎

Lines changed: 29 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -1,58 +1,59 @@
1-
# OpenVINO Training Extensions
1+
# OpenVINO™ Training Extensions
22

3-
OpenVINO Training Extensions provide a convenient environment to train
4-
Deep Learning models and convert them using [OpenVINO™
5-
Toolkit](https://software.intel.com/en-us/openvino-toolkit) for optimized
3+
OpenVINO™ Training Extensions provide a convenient environment to train
4+
Deep Learning models and convert them using the [OpenVINO™
5+
toolkit](https://software.intel.com/en-us/openvino-toolkit) for optimized
66
inference.
77

8-
98
# Quick Start Guide
109

11-
## Setup OpenVINO Training Extensions
10+
## Setup OpenVINO™ Training Extensions
1211

13-
1. Clone repository in the working directory
12+
1. Clone repository in the working directory by running the following:
1413

15-
```
16-
cd /<path_to_working_dir>
17-
git clone https://github.com/opencv/openvino_training_extensions.git
18-
```
14+
```
15+
git clone https://github.com/opencv/openvino_training_extensions.git
16+
```
1917
20-
2. Install prerequisites
18+
2. Install prerequisites by running the following:
2119
22-
```
23-
sudo apt-get install libturbojpeg python3-tk python3-pip virtualenv
24-
```
20+
```
21+
sudo apt-get install libturbojpeg python3-tk python3-pip virtualenv
22+
```
2523
2624
2725
# Models
2826
29-
* [PyTorch](pytorch_toolkit)
27+
* [PyTorch\*](pytorch_toolkit)
3028
31-
* [Action recognition](pytorch_toolkit/action_recognition)
32-
* [Face recognition](pytorch_toolkit/face_recognition)
33-
* [Person reidentification](pytorch_toolkit/person_reidentification)
34-
* [Human pose estimation](pytorch_toolkit/human_pose_estimation)
35-
* [Instance segmentation](pytorch_toolkit/instance_segmentation)
29+
* [Action Recognition](pytorch_toolkit/action_recognition)
30+
* [Face Recognition](pytorch_toolkit/face_recognition)
31+
* [Person Reidentification](pytorch_toolkit/person_reidentification)
32+
* [Human Pose Estimation](pytorch_toolkit/human_pose_estimation)
33+
* [Instance Segmentation](pytorch_toolkit/instance_segmentation)
3634
* [Object Detection](pytorch_toolkit/object_detection)
3735
- [Face Detection](pytorch_toolkit/object_detection/face_detection.md)
3836
- [Person Vehicle Bike Detector](pytorch_toolkit/object_detection/person_vehicle_bike_detection.md)
39-
* [Segmentation of thoracic organs](pytorch_toolkit/segthor)
40-
* [Super resolution](pytorch_toolkit/super_resolution)
37+
* [Segmentation of Thoracic Organs](pytorch_toolkit/segthor)
38+
* [Super Resolution](pytorch_toolkit/super_resolution)
4139
42-
* [TensorFlow](tensorflow_toolkit)
40+
* [TensorFlow\*](tensorflow_toolkit)
4341
4442
* [Action Detection](tensorflow_toolkit/action_detection)
45-
* [Image retrieval](tensorflow_toolkit/image_retrieval)
43+
* [Image Retrieval](tensorflow_toolkit/image_retrieval)
4644
* [License Plate Recognition](tensorflow_toolkit/lpr)
4745
* [Person Vehicle Bike Detector](tensorflow_toolkit/person_vehicle_bike_detector)
4846
* [SSD MobileNet FPN 602](tensorflow_toolkit/ssd_mobilenet_fpn_602)
4947
* [SSD Object Detection](tensorflow_toolkit/ssd_detector)
50-
* [Text detection](tensorflow_toolkit/text_detection)
51-
* [Text recognition](tensorflow_toolkit/text_recognition)
48+
* [Text Detection](tensorflow_toolkit/text_detection)
49+
* [Text Recognition](tensorflow_toolkit/text_recognition)
5250
* [Vehicle Attributes](tensorflow_toolkit/vehicle_attributes)
5351
5452
# Tools
5553
56-
* [PyTorch](pytorch_toolkit)
54+
* [PyTorch\*](pytorch_toolkit)
5755
5856
* [Neural Networks Compression Framework](pytorch_toolkit/nncf)
57+
58+
---
59+
\* Other names and brands may be claimed as the property of others.

‎data/bitvehicle/README.md‎

Lines changed: 77 additions & 77 deletions
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,9 @@
1-
BIT-Vehicle Dataset: http://iitlab.bit.edu.cn/mcislab/vehicledb/
1+
# BIT-Vehicle Dataset
22

3-
# Prepare the dataset
3+
## Prepare the Dataset
44

5-
1. Download the dataset from link above.
6-
2. Unpack dataset in the directory `images`.
5+
1. Download the [dataset](http://iitlab.bit.edu.cn/mcislab/vehicledb/).
6+
2. Unpack the dataset in the `images` directory:
77
```
88
$ tree
99
.
@@ -15,84 +15,84 @@ BIT-Vehicle Dataset: http://iitlab.bit.edu.cn/mcislab/vehicledb/
1515
| ...
1616
└── README.md
1717
```
18-
3. Downscale images to increase training speed.
18+
3. Downscale images to increase training speed:
1919
```
2020
python <training_toolbox_tensorflow>/tools/downscale_images.py -target_size 512 <training_toolbox_tensorflow>/data/bitvehicle/images
2121
2222
```
2323
24-
# Annotation structure
24+
## Annotation Structure
2525
26-
Json files contain annotation in a fairly straightforward structure. There’s 3
27-
top-level arrays: "images", "annotations" and "categories"
26+
The structure of JSON files is fairly straightforward. There are three
27+
top-level arrays: `images`, `annotations` and `categories`.
2828
29-
1. “images” has records, like,
30-
```
31-
{
32-
"dataset": "BitVehicle",
33-
"height": 1080,
34-
"id": 4,
35-
"width": 1920,
36-
"file_name":
37-
"vehicle_0000005.jpg",
38-
"coco_url": null,
39-
"license": null,
40-
"flickr_url": null,
41-
"image": "./images/vehicle_0000005.jpg",
42-
"date_captured": null
43-
}
44-
```
45-
2. “annotation” has records, like,
46-
```
47-
[
48-
{
49-
"area": 199023.0,
50-
"id": 10,
51-
"iscrowd": 0,
52-
"category_id": 1,
53-
"is_occluded": false,
54-
"image_id": 4,
55-
"segmentation": null,
56-
"bbox": [512.0, 346.0, 407.0, 489.0],
57-
"attributes": {}
58-
},
59-
{
60-
"area": 2668.0,
61-
"id": 11,
62-
"iscrowd": 0,
63-
"category_id": 2,
64-
"is_occluded": false,
65-
"image_id": 4,
66-
"segmentation": null,
67-
"bbox": [638.0, 773.0, 92.0, 29.0],
68-
"attributes": {}
69-
},
70-
{
71-
"area": 5187.0,
72-
"id": 12,
73-
"iscrowd": 0,
74-
"category_id": 1,
75-
"is_occluded": true,
76-
"image_id": 4,
77-
"segmentation": null,
78-
"bbox": [1023.0, 0.0, 273.0, 19.0],
79-
"attributes": {}
80-
}
81-
]
82-
```
83-
3. “categories” has just 3 records:
84-
```
85-
[
86-
{"id": 0, "name": "bg", "supercategory": ""},
87-
{"id": 1, "name": "vehicle", "supercategory": ""},
88-
{"id": 2, "name": "plate", "supercategory": ""}
89-
]
90-
```
29+
**Example of arrays and their fields**
9130
92-
In the example above, the image with the file name `vehicle_0000005.jpg` has
93-
`id` equal to 4. Then the records in `annotation` array say that this image
94-
(`image_id` is equal to 4) has 3 bounding boxes – 2 of them have `category_id`
95-
set to 1 and one has `category_id` set to 2. Now, the last array -
96-
`categories` - tells us that it means that there’s 2 cars and 1 plate in that
97-
image (their coordinates are given in the `bbox` field – `[x, y, width,
98-
height]`).
31+
1. `images` records:
32+
```
33+
{
34+
"dataset": "BitVehicle",
35+
"height": 1080,
36+
"id": 4,
37+
"width": 1920,
38+
"file_name":
39+
"vehicle_0000005.jpg",
40+
"coco_url": null,
41+
"license": null,
42+
"flickr_url": null,
43+
"image": "./images/vehicle_0000005.jpg",
44+
"date_captured": null
45+
}
46+
```
47+
2. `annotation` records:
48+
```
49+
[
50+
{
51+
"area": 199023.0,
52+
"id": 10,
53+
"iscrowd": 0,
54+
"category_id": 1,
55+
"is_occluded": false,
56+
"image_id": 4,
57+
"segmentation": null,
58+
"bbox": [512.0, 346.0, 407.0, 489.0],
59+
"attributes": {}
60+
},
61+
{
62+
"area": 2668.0,
63+
"id": 11,
64+
"iscrowd": 0,
65+
"category_id": 2,
66+
"is_occluded": false,
67+
"image_id": 4,
68+
"segmentation": null,
69+
"bbox": [638.0, 773.0, 92.0, 29.0],
70+
"attributes": {}
71+
},
72+
{
73+
"area": 5187.0,
74+
"id": 12,
75+
"iscrowd": 0,
76+
"category_id": 1,
77+
"is_occluded": true,
78+
"image_id": 4,
79+
"segmentation": null,
80+
"bbox": [1023.0, 0.0, 273.0, 19.0],
81+
"attributes": {}
82+
}
83+
]
84+
```
85+
3. `categories` records:
86+
```
87+
[
88+
{"id": 0, "name": "bg", "supercategory": ""},
89+
{"id": 1, "name": "vehicle", "supercategory": ""},
90+
{"id": 2, "name": "plate", "supercategory": ""}
91+
]
92+
```
93+
94+
In the example above, the ID of the `vehicle_0000005.jpg` image
95+
equals to `4`, which is represented by `"image_id": 4` that has in the `annotation` array, and has three bounding boxes: two of them have the `category_id`
96+
set to `1` and one has the `category_id` set to `2`.
97+
From the `categories` array, we learn that there are two vehicles and one plate in that
98+
image (their coordinates are given in the `annotation` array in the `bbox` field with the `[x, y, width, height]` format).

‎data/coco/README.md‎

Lines changed: 5 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,10 @@
1-
COCO (Common Objects in Context) Dataset: http://cocodataset.org
1+
# Common Objects in Context (COCO) Dataset
22

33
Prepare the dataset:
4-
1. Download the dataset from link above.
5-
2. Update annotation with script `add_full_image_path.py`.
4+
5+
1. Download the [dataset](http://cocodataset.org).
6+
7+
2. Update the annotation with the `add_full_image_path.py` script:
68
```
79
python add_full_image_path.py instances_train2017.json ./train2017 instances_train2017_full_paths.json
810

0 commit comments

Comments
 (0)