【Hackathon No.153】PaddleRS 英文文档 #113
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Bobholamovic
requested changes
Mar 20, 2023
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| ## Contribute code | ||
|
|
||
| This guide starts with the steps necessary to contribute code to PaddleRS, and then goes into detail on new file self-review, code style specification, and testing steps. |
There was a problem hiding this comment.
steps necessary -> necessary steps
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| ## Contribute code | ||
|
|
||
| This guide starts with the steps necessary to contribute code to PaddleRS, and then goes into detail on new file self-review, code style specification, and testing steps. |
There was a problem hiding this comment.
new file self-review -> self-inspection on newly added files
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| This guide starts with the steps necessary to contribute code to PaddleRS, and then goes into detail on new file self-review, code style specification, and testing steps. | ||
|
|
||
| ### 1 Code contribution steps |
There was a problem hiding this comment.
标题每个实词的首字母需要大写。
请同步修改PR中其他部分。
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| ### 1 Code contribution steps | ||
|
|
||
| PaddleRS uses [git](https://git-scm.com/doc) as a version control tool and is hosted on GitHub. This means that you need to be familiar with git before contributing code, and understand [pull request (PR)](https://docs.github.com/cn/pull-requests/collaborating-with-pull-requests/proposing-changes-to-your-work-with-pull-requests/about-pull-requests) based on the GitHub workflow. |
There was a problem hiding this comment.
, and understand这半句建议拆成一个独立的句子。
docs/apis/data_en.md
Outdated
| |Parameter Name|Type|Parameter Description|Default Value| | ||
| |-------|----|--------|-----| | ||
| |`data_dir`|`str`|Dataset storage directory.|| | ||
| |`file_list`|`str`|file list path. file list is a text file, in which each line contains the path infomation of sample.The specific requirements of `CDDataset` on the file list are listed below.|| |
There was a problem hiding this comment.
请注意标点符号之后空一格。
请同步检查PR中的其它部分。后续不再重复。
docs/apis/data_en.md
Outdated
| |`file_list`|`str`|file list path. file list is a text file, in which each line contains the path infomation of sample.The specific requirements of `CDDataset` on the file list are listed below.|| | ||
| |`transforms`|`paddlers.transforms.Compose`|Data transformation operator applied to input data.|| | ||
| |`label_list`|`str` \| `None`|label list file. label list is a text file, in which each line contains the name of class.|`None`| | ||
| |`num_workers`|`int` \| `str`|The number of worker processes used when loading data. If it is set to `'auto'`, the following rules determine the number of processes to use:When the number of CPU cores is greater than 16, 8 data read helper processes are used; Otherwise, use CPU cores to count half as many worker processes.|`'auto'`| |
There was a problem hiding this comment.
这里翻译好像不太对，请检查。另外请同步检查PR中其它涉及num_workers解释的部分。
docs/apis/data_en.md
Outdated
| |`label_list`|`str` \| `None`|label list file. label list is a text file, in which each line contains the name of class.|`None`| | ||
| |`num_workers`|`int` \| `str`|The number of worker processes used when loading data. If it is set to `'auto'`, the following rules determine the number of processes to use:When the number of CPU cores is greater than 16, 8 data read helper processes are used; Otherwise, use CPU cores to count half as many worker processes.|`'auto'`| | ||
| |`shuffle`|`bool`|Whether to randomly shuffle the samples in the dataset.|`False`| | ||
| |`with_seg_labels`|`bool`|Specify this option as `True` when the dataset contains split labels for each phase.|`False`| |
docs/apis/data_en.md
Outdated
| |`num_workers`|`int` \| `str`|The number of worker processes used when loading data. If it is set to `'auto'`, the following rules determine the number of processes to use:When the number of CPU cores is greater than 16, 8 data read helper processes are used; Otherwise, use CPU cores to count half as many worker processes.|`'auto'`| | ||
| |`shuffle`|`bool`|Whether to randomly shuffle the samples in the dataset.|`False`| | ||
| |`with_seg_labels`|`bool`|Specify this option as `True` when the dataset contains split labels for each phase.|`False`| | ||
| |`binarize_labels`|`bool`|If it is `True`, the change label (and split label) is binarized after all data transformation operators except `Arrange` are processed. For example, binarize a tag with the value {0, 255} to {0, 1}.|`False`| |
There was a problem hiding this comment.
一个像素点为一个label，所以label应该是复数（或者用label image）。
docs/apis/data_en.md
Outdated
| |`with_seg_labels`|`bool`|Specify this option as `True` when the dataset contains split labels for each phase.|`False`| | ||
| |`binarize_labels`|`bool`|If it is `True`, the change label (and split label) is binarized after all data transformation operators except `Arrange` are processed. For example, binarize a tag with the value {0, 255} to {0, 1}.|`False`| | ||
|
|
||
| The requirments of `CDDataset` for the file list are as follows: |
There was a problem hiding this comment.
typo: requirments，请同步检查PR中其他部分。
|
您好，已进行修改 |
|
我注意到好像还没有修改完全~ 建议可以逐条回复review意见，以免遗漏。 |
|
这两个不知道目前的翻译是否可以？其余均已按照意见完成修改~ |
Bobholamovic
requested changes
Mar 24, 2023
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| ## Contribute Code | ||
|
|
||
| This guide starts with the necessary steps to contribute code to PaddleRS, and then goes into detail on self-inspection on newly added files, code style specification, and testing steps. |
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| The steps to contribute code to PaddleRS are as follows: | ||
|
|
||
| 1. Fork the official PaddleRS repository on GitHub, clone the code locally, and pull up the latest version of the develop branch. |
docs/CONTRIBUTING_en.md
Outdated
| 2. Write code according to [dev Guide](dev/dev_guide.md) (it is recommended to develop on a new feature branch). | ||
| 3. Install pre-commit hooks to perform code style checks before each commit. Refer to [Code style specification](#3-Code style specification). | ||
| 4. Write unit tests for the new code and make sure all the tests are successful. Refer to [Test related steps](#4-Test related steps)。 | ||
| 5. Create a new PR for your branch to ensure that the CLA protocol is signed and the CI/CE passes. After that, a PaddleRS team member will review the code you contributed. |
docs/CONTRIBUTING_en.md
Outdated
| 5. Create a new PR for your branch to ensure that the CLA protocol is signed and the CI/CE passes. After that, a PaddleRS team member will review the code you contributed. | ||
| 6. Modify the code according to the review and resubmit it until PR is merged or closed. | ||
|
|
||
| If you contribute code that uses a third-party library that PaddleRS does not currently rely on, please explain when you submit your PR and explain why you need to use this third-party library. |
There was a problem hiding this comment.
`your PR. Also, you should explain why this third-party library should be used.
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| #### 2.2 The Order of Module Import | ||
|
|
||
| All global import statements must be at the beginning of the module, after the copyright information. Import packages or modules in the following order: |
docs/apis/data_en.md
Outdated
| |`allow_empty`|`bool`|Whether to add negative samples to the dataset.|`False`| | ||
| |`empty_ratio`|`float`|Negative sample ratio, take effect only if `allow_empty` is `True`. If `empty_ratio` is negative or greater than or equal to 1, all negative samples generated are retained.|`1.0`| | ||
|
|
||
| The requirments of `VOCDetDataset` for the file list are as follows: |
There was a problem hiding this comment.
requirments拼写错误。
请同步检查并修改PR中其它位置。
docs/apis/data_en.md
Outdated
| |`transforms`|`paddlers.transforms.Compose`|Data transformation operator applied to input data.|| | ||
| |`num_workers`|`int` \| `str`|The number of auxiliary processes used when loading data. If it is set to `'auto'`, the following rules determine the number of processes to use:When the number of CPU cores is greater than 16, 8 data read auxiliary processes are used; Otherwise, use CPU cores to count half as many auxiliary processes.|`'auto'`| | ||
| |`shuffle`|`bool`|Whether to randomly shuffle the samples in the dataset.|`False`| | ||
| |`sr_factor`|`int` \| `None`|For super resolution reconstruction task, specify as super resolution multiple; For other tasks, specify as `None`.|`None`| |
There was a problem hiding this comment.
For super resolution reconstruction tasks, this is the scaling factor. For other tasks, please specify sr_factor as None.
docs/apis/data_en.md
Outdated
|
|
||
| Remote sensing images come from various sources and their data formats are very complicated. PaddleRS provides a unified interface for reading remote sensing images of different types and formats. At present, PaddleRS can read common file formats such as .png, .jpg, .bmp and .npy, as well as handle GeoTiff, img and other image formats commonly used in remote sensing. | ||
|
|
||
| According to the actual demand, the user can choose `paddlers.transforms.decode_image()` or `paddlers.transforms.DecodeImg` to read data. `DecodeImg` is one of[Data transformation operators](#Data transformation operators), can be combined with other operators. `decode_image` is the encapsulation of `DecodeImg` operator, which is convenient use in the way of function calls. |
There was a problem hiding this comment.
According to the actual demand -> Depending on the practical demands
docs/apis/data_en.md
Outdated
|
|
||
| Remote sensing images come from various sources and their data formats are very complicated. PaddleRS provides a unified interface for reading remote sensing images of different types and formats. At present, PaddleRS can read common file formats such as .png, .jpg, .bmp and .npy, as well as handle GeoTiff, img and other image formats commonly used in remote sensing. | ||
|
|
||
| According to the actual demand, the user can choose `paddlers.transforms.decode_image()` or `paddlers.transforms.DecodeImg` to read data. `DecodeImg` is one of[Data transformation operators](#Data transformation operators), can be combined with other operators. `decode_image` is the encapsulation of `DecodeImg` operator, which is convenient use in the way of function calls. |
There was a problem hiding this comment.
of 和 Data transformation operators之间缺少空格
docs/apis/data_en.md
Outdated
|
|
||
| According to the actual demand, the user can choose `paddlers.transforms.decode_image()` or `paddlers.transforms.DecodeImg` to read data. `DecodeImg` is one of[Data transformation operators](#Data transformation operators), can be combined with other operators. `decode_image` is the encapsulation of `DecodeImg` operator, which is convenient use in the way of function calls. | ||
|
|
||
| The argument lish of `decode_image()` function is as follows: |
Bobholamovic
requested changes
Mar 24, 2023
docs/CONTRIBUTING_en.md
Outdated
| #### 4.1 model Single Test | ||
|
|
||
| 1. Find the test case definition file corresponding to the task of the model in `tests/rs_models/`, for example, the change detection task corresponding to `tests/rs_models/test_cd_models.py`. | ||
| 2. Define a Test class for the new Model that inherits from `Test{task name}Model` and sets its `MODEL_CLASS` property to the new model, following the example already in the file. |
docs/CONTRIBUTING_en.md
Outdated
|
|
||
| If your contribution includes TIPC, please submit your PR with a log indicating successful execution of the TIPC basic training chain. | ||
|
|
||
| ## Contribution an Example |
There was a problem hiding this comment.
Contribution -> Contribute
docs/apis/data_en.md
Outdated
|
|
||
| ### Scenario Classification Dataset `ClasDataset` | ||
|
|
||
| `ClasDataset` is define in: https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/datasets/clas_dataset.py |
There was a problem hiding this comment.
typo: defined
请同步检查并修改PR中其它位置。
docs/apis/data_en.md
Outdated
| The return format is as follows: | ||
|
|
||
| - If `read_geo_info` is `False`, the image data ([h, w, c] arrangement) is returned in the format of np.ndarray. | ||
| - If `read_geo_info` is `True`, a binary group is returned, in which the first element is the image data read, the second element is a dictionary, in which the key-value pair is the geographic information of the image, such as geographic transformation information, geographic projection information, etc. |
docs/apis/data_en.md
Outdated
| |`decode_bgr`|`bool`|If `True`, automatically parses non-geo format images (such as jpeg images) into BGR format.|`True`| | ||
| |`decode_sar`|`bool`|If `True`, single-channel geo-format images (such as GeoTiff images) are automatically parsed as SAR images.|`True`| | ||
| |`read_geo_info`|`bool`|If `True`, the geographic information is read from the image.|`False`| | ||
| |`use_stretch`|`bool`|Whether to stretch the image brightness by 2% linear. Take effect only if `to_uint8` is `True`.|`False`| |
There was a problem hiding this comment.
Whether to apply a linear stretch to image brightness (with 2% max and min values removed).
docs/apis/data_en.md
Outdated
| T.Resize(target_size=512), | ||
| # Perform a random horizontal flip with a 50% probability | ||
| T.RandomHorizontalFlip(prob=0.5), | ||
| # Normalized data to [-1,1] |
docs/apis/data_en.md
Outdated
| ]) | ||
| ``` | ||
|
|
||
| Generally, in the list of `Compose` object accepted data transform operators, the first element is `paddlers.Transforms.DecodeImg` object, used to read image data; The last element is [`Arrange` Operator](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/transforms/operators.py, used to extract and arrange information from the `sample` dictionary. |
There was a problem hiding this comment.
in the list of Compose object accepted data transform operators -> in the list of data transform operators accepted by a Compose object
docs/apis/data_en.md
Outdated
|
|
||
| Generally, in the list of `Compose` object accepted data transform operators, the first element is `paddlers.Transforms.DecodeImg` object, used to read image data; The last element is [`Arrange` Operator](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/transforms/operators.py, used to extract and arrange information from the `sample` dictionary. | ||
|
|
||
| For the verification set of image segmentation task and change detection task, the `ReloadMask` operator can be inserted before the `Arrange` operator to reload the GT label. The following is an example: |
There was a problem hiding this comment.
『验证集』的翻译为『validation dataset』
请同步检查并修改PR中其它位置
docs/apis/data_en.md
Outdated
|
|
||
| Generally, in the list of `Compose` object accepted data transform operators, the first element is `paddlers.Transforms.DecodeImg` object, used to read image data; The last element is [`Arrange` Operator](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/transforms/operators.py, used to extract and arrange information from the `sample` dictionary. | ||
|
|
||
| For the verification set of image segmentation task and change detection task, the `ReloadMask` operator can be inserted before the `Arrange` operator to reload the GT label. The following is an example: |
docs/data/dataset_summary_en.md
Outdated
There was a problem hiding this comment.
表格标题的大小写风格不统一，如：link是实词，也需要首字母大写
感谢您的审阅~ |
Bobholamovic
requested changes
Mar 25, 2023
docs/CONTRIBUTING_en.md
Outdated
| #### 4.1 model Single Test | ||
|
|
||
| 1. Find the test case definition file corresponding to the task of the model in `tests/rs_models/`, for example, the change detection task corresponding to `tests/rs_models/test_cd_models.py`. | ||
| 2. Define a Test class for the new Model that inherits from `Test{task name}Model` and sets its `MODEL_CLASS` property to the new model, following the example already in the file. |
tutorials/train/README_en.md
Outdated
| python setup.py install | ||
| ``` | ||
|
|
||
| If the download dependence is slow or times out when using `python setup.py install`, you can create `setup.cfg` in the same directory as `setup.py` and input the following content, then the download can be accelerated through Tsinghua source: |
There was a problem hiding this comment.
the download dependence -> the downloading of dependencies
and input the following content -> with the following content
tutorials/train/README_en.md
Outdated
|
|
||
| + (Optional) GDAL installation | ||
|
|
||
| PaddleRS supports reading of various types of satellite data. To fully use PaddleRS remote sensing data reading function, you need to install GDAL as follows: |
There was a problem hiding this comment.
To fully use PaddleRS remote sensing data reading function -> To use the full data reading functionality of PaddleRS
tutorials/train/README_en.md
Outdated
|
|
||
| - Windows | ||
|
|
||
| Windows users can download the Python and system version corresponding to the .whl format installation package from [this](https://www.lfd.uci.edu/~gohlke/pythonlibs/#gdal) to local, take *GDAL‑3.3.3‑cp39‑cp39‑win_amd64.whl* as an example, use the pip tool installation: |
There was a problem hiding this comment.
Windows users can download GDAL wheels from this site. Please choose the wheel according to the Python version and the platform. Take GDAL‑3.3.3‑cp39‑cp39‑win_amd64.whl as an example, run the following command to install:
tutorials/train/README_en.md
Outdated
| docker pull paddlepaddle/paddlers:1.0.0 | ||
| ``` | ||
|
|
||
| - (Optional) Build from scratch. Multiple base images for PaddlePaddle can be selected by setting `PPTAG` to build cpus or different GPU environments: |
There was a problem hiding this comment.
Select the base image for PaddlePaddle by setting PPTAG. You can build the image in a CPU-only environment or in GPU environments
docs/apis/infer_en.md
Outdated
| |`model_dir`|`str`|Model path (must be an exported deployed or quantified model).|| | ||
| |`use_gpu`|`bool`|Whether to use GPU.|`False`| | ||
| |`gpu_id`|`int`|ID of the GPU used.|`0`| | ||
| |`cpu_thread_num`|`int`|The number of threads when inference is performed using the CPU.|`1`| |
docs/apis/infer_en.md
Outdated
|
|
||
| |Parameter Name|Type|Parameter Description|Default Value| | ||
| |-------|----|--------|-----| | ||
| |`img_file`|`list[str\|tuple\|np.ndarray]` \| `str` \| `tuple` \| `np.ndarray`|For scene classification, object detection, image restoration and image segmentation tasks, this parameter can be a single image path, or a decoded image data in [h, w, c] with a float32 type (expressed as NumPy array), or a list of image paths or np.ndarray objects. For the change detection task, the parameter can be a binary group of image path (representing the two time phase image paths respectively), or a binary group composed of two decoded images, or a list composed of one of the above two binary groups.|| |
docs/apis/infer_en.md
Outdated
| |-------|----|--------|-----| | ||
| |`img_file`|`list[str\|tuple\|np.ndarray]` \| `str` \| `tuple` \| `np.ndarray`|For scene classification, object detection, image restoration and image segmentation tasks, this parameter can be a single image path, or a decoded image data in [h, w, c] with a float32 type (expressed as NumPy array), or a list of image paths or np.ndarray objects. For the change detection task, the parameter can be a binary group of image path (representing the two time phase image paths respectively), or a binary group composed of two decoded images, or a list composed of one of the above two binary groups.|| | ||
| |`topk`|`int`|It is used in scenario classification model prediction, indicating that the category with the top `topk` in the output probability of the model is selected as the final result.|`1`| | ||
| |`transforms`|`paddlers.transforms.Compose`\|`None`|Apply data transformation operator to input data. If `None`, the operators read from 'model.yml' is used.|`None`| |
docs/apis/infer_en.md
Outdated
|
|
||
| ### `Predictor.slider_predict()` | ||
|
|
||
| Realize the sliding window inference function. It is used in the same way as `BaseSegmenter` and `slider_predict()` of `BaseChangeDetector`. |
docs/apis/infer_en.md
Outdated
| |-------|----|--------|-----| | ||
| |`img_file`|`str`|Input image path.|| | ||
| |`save_dir`|`str`|Predict results output path.|| | ||
| |`block_size`|`list[int]` \| `tuple[int]` \| `int`|The size of the slide window (specifying the width, height in a list or tuple, or the same width and height in an integer).|| |
There was a problem hiding this comment.
各个文档中表格里的description，有的时候省略句首的定冠词，有的时候没有。请统一一下风格吧
|
|
Bobholamovic
requested changes
Mar 27, 2023
tutorials/train/README_en.md
Outdated
|
|
||
| + PaddleRS installation | ||
|
|
||
| The PaddleRS code will be updated as the development progresses. You can install code for the develop branch to use the latest features as follows: |
There was a problem hiding this comment.
install code for the develop branch -> install the develop branch
docs/apis/train_en.md
Outdated
|
|
||
| All trainers support default parameter construction (that is, no parameters are passed in when the object is constructed), in which case the constructed trainer object applies to three-channel RGB data. | ||
|
|
||
| ### Initialize `BaseChangeDetector` Sub-Class Object |
There was a problem hiding this comment.
Sub-Class中间建议不添加连字符（添加连字符的写法很少见），写为Subclass。请同步检查并修改PR中的其它位置。
docs/apis/train_en.md
Outdated
| @@ -0,0 +1,420 @@ | |||
| # PaddleRS Training API Description | |||
|
|
|||
| **Trainer** encapsulates model training, validation, quantization, and dynamic graph inference, defined in files of `paddlers/tasks/` directory. For user convenience, PaddleRS provides trainer that inherits from the parent class [`BaseModel`](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/base.py) for all supported models, and provides several apis externally. The types of trainers corresponding to change detection, scene classification, target detection, image restoration and image segmentation tasks are respectively `BaseChangeDetector`、`BaseClassifier`、`BaseDetector`、`BaseRestorer` and `BaseSegmenter`。This document describes the initialization function of the trainer and `train()`、`evaluate()` API。 | |||
There was a problem hiding this comment.
provides trainer -> provides trainers
docs/apis/train_en.md
Outdated
| - The `num_classes`、`use_mixed_loss` and `in_channels` parameters are generally supported, indicating the number of model output categories, whether to use preset mixing losses, and the number of input channels, respectively. Some sub-classes, such as `DSIFN`, do not yet support `in_channels`. | ||
| - `use_mixed_loss` will be deprecated in the future, so it is not recommended. | ||
| - Specify the loss function used during model training through the `losses` parameter. `losses` needs to be a dictionary, where the values for the keys `types` and `coef` are two equal-length lists representing the loss function object (a callable object) and the weight of the loss function, respectively. For example: `losses={'types': [LossType1(), LossType2()], 'coef': [1.0, 0.5]}`. It is equivalent to calculating the following loss function in the training process: `1.0*LossType1()(logits, labels)+0.5*LossType2()(logits, labels)`, where `logits` and `labels` are model output and GT labels, respectively. | ||
| - Different sub-classes support model-related input parameters, for details, refer to [Model definition](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/rs_models/cd) and [Trainer definition](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/change_detector.py). |
There was a problem hiding this comment.
refer to [model definitions](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/rs_models/cd) and [trainer definitions](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/change_detector.py)
docs/apis/train_en.md
Outdated
|
|
||
| ### Initialize `BaseDetector` Sub-Class Object | ||
|
|
||
| - Generally, the `num_classes` and `backbone` parameters can be set to indicate the number of output categories of the model and the type of backbone network used respectively. Compared with other tasks, the trainer of object detection task supports more initialization parameters, including network structure, loss function, post-processing strategy and so on. |
docs/apis/train_en.md
Outdated
|
|
||
| - Generally, the `num_classes` and `backbone` parameters can be set to indicate the number of output categories of the model and the type of backbone network used respectively. Compared with other tasks, the trainer of object detection task supports more initialization parameters, including network structure, loss function, post-processing strategy and so on. | ||
| - Different from tasks such as segmentation, classification and change detection, detection tasks do not support the loss function specified through the `losses` parameter. However, for some trainers such as `PPYOLO`, the loss function can be customized by `use_iou_loss` and other parameters. | ||
| - Different sub-classes support model-related input parameters, for details, refer to [Model definition](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/rs_models/det) and [Trainer definition](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/object_detector.py). |
There was a problem hiding this comment.
这里需要拆成两句话写，或者使用从句，要不然有语法错误（一个句子无连词两个谓语）。
请同步检查并修改PR中的其它位置。
docs/apis/train_en.md
Outdated
|
|
||
| ### Initialize `BaseRestorer` Sub-Class Object | ||
|
|
||
| - Generally support setting `sr_factor` parameter, representing super resolution multiple; for models that do not support super resolution rebuild tasks, `sr_factor` is set to `None`. |
There was a problem hiding this comment.
super resolution multiple -> the scaling factor in image super resolution
|
Bobholamovic
requested changes
Mar 28, 2023
docs/dev/dev_guide_en.md
Outdated
| """ | ||
| ``` | ||
|
|
||
| ### 1.3 Edit Trainer |
docs/dev/dev_guide_en.md
Outdated
|
|
||
| 4. Add the class name of the new trainer to the global variable `__all__`. | ||
|
|
||
| It should be noted that for the image restoration task, the forward and reverse logic of the model are implemented in the trainer definition. For GAN and other models that need to use multiple networks, please refer to the following specifications for the preparation of the trainer: |
docs/dev/dev_guide_en.md
Outdated
| - Override the `default_loss()` method to build the loss function. If more than one loss function is required in the training process, it is recommended to organize in the form of a dictionary. | ||
| - Override the `default_optimizer()` method to build one or more optimizers. When `build_net()` returns a value of type `GANAdapter`, `parameters` is a dictionary. Where, `parameters['params_g']` is a list containing the state dict of the various generators in order; `parameters['params_d']` is a list that contains the state dict of the individual discriminators in order. If you build more than one optimizer, you should use the `OptimizerAdapter` wrapper on return. | ||
| - Override the `run_gan()` method that accepts four parameters: `net`, `inputs`, `mode`, and `gan_mode` for one of the subtasks in the training process, e.g. forward calculation of generator, forward calculation of discriminator, etc. | ||
| - Rewrite `train_step()` method to write the specific logic of one iteration during model training. The usual approach is to call `run_gan()` over and over again, constructing different `inputs` to work in different `gan_mode` as needed each time, extracting useful fields (e.g. losses) from the `outputs` dictionary returned each time and summarizing them into the final result. |
There was a problem hiding this comment.
over and over again -> multiple times
docs/dev/dev_guide_en.md
Outdated
|
|
||
| ### 2.2 Add Data Preprocessing/Data Augmentation Operators | ||
|
|
||
| Define new operators in `paddlers/transforms/operators.py`, all operators are inherited from `paddlers.transforms.Transform`. The operator's `apply()` method receives a dictionary `sample` as input, takes out the related objects stored in it, and makes in-place modifications to the dictionary after processing, and finally returns the modified dictionary. Only in rare cases do we need to override the `apply()` method when defining an operator. In most cases, you just need to override the `apply_im()`, `apply_mask()`, `apply_bbox()`, and `apply_segm()` methods to handle the image, split label, target box, and target polygon, respectively. |
There was a problem hiding this comment.
are inherited from -> inherit from
docs/dev/dev_guide_en.md
Outdated
|
|
||
| If processing logic is more complicated, it is recommended that the encapsulated in the first function, added to the `paddlers/transforms/functions.py`, then call the function in the `apply*()` of operators. | ||
|
|
||
| After writing the implementation of the operator, **must write docstring and add the class name in `__all__`.** |
docs/data/rs_data_en.md
Outdated
|
|
||
| Band combination refers to the result obtained by selecting three band data in MSI/HSI to combine and replace the three RGB channels, which is called the color graph (*The result synthesized using the real RGB three bands is called the true color graph, otherwise it is called the false color graph*). The combination of different bands can highlight different features of ground objects. The following figure shows the visual effects of several different combinations: | ||
|
|
||
|  |
docs/data/coco_tools_en.md
Outdated
|
|
||
| ## 1 Tool Description | ||
|
|
||
| coco_tools is a set of tools provided by PaddleRS for handling COCO annotated files. It is located in the `tools/coco_tools/` directory. Because [pycocotools library] (https://pypi.org/project/pycocotools/) can't install in some environment, PaddleRS provides coco_tools as an alternative for some simple file processing. |
There was a problem hiding this comment.
COCO annotated files -> COCO annotation files
docs/data/coco_tools_en.md
Outdated
| ``` | ||
| Among them, | ||
|
|
||
| - The `info` key is a dictionary, there are 6 key-value pairs, the output shows the first 5 pairs; |
There was a problem hiding this comment.
这里的表述好像有些歧义，key本身应该不是字典，应该是key对应的值为字典。可以考虑改成：
info: A dictionary. There are...
这样的格式。
docs/data/coco_tools_en.md
Outdated
| | `--json_path` | (Optional) Path of the json file whose statistics you want to collect. | | | ||
| | `--csv_path` | (Optional) Save path for the statistics table. | `None` | | ||
| | `--png_shape_path` | (Optional) png image saving path. The image content is the two-dimensional distribution of the shape of all target detection frames.| `None` | | ||
| | `--png_shapeRate_path` | (Optional) png Image saving path. The image content is a one-dimensional distribution of shape ratio (width/height) of all target detection boxes.| `None` | |
docs/data/coco_tools_en.md
Outdated
|
|
||
| #### 3.2.3 Result Presentation | ||
|
|
||
| After the preceding command is executed, the following information is displayed: |
|
Bobholamovic
requested changes
Mar 30, 2023
docs/data/rs_data_en.md
Outdated
| @@ -8,7 +8,7 @@ Compared with aerial images, satellite images cover a wider area, so they are us | |||
|
|
|||
| ## 2 Characteristics of Remote Sensing Images | |||
|
|
|||
| Remote sensing technology has the characteristics of macro, multi-band, periodicity and economy. Macro refers to that the higher the remote sensing platform is, the wider the perspective will be, and the wider the ground can be synchronously detected. The multi-band property means that the sensor can detect and record information in different bands such as ultraviolet, visible light, near infrared and microwave. Periodicity means that the remote sensing satellite has the characteristic of acquiring images repeatedly in a certain period, which can carry out repeated observation of the same area in a short time. Economic means that remote sensing technology can be used as a way to obtain large area of surface information without spending too much manpower and material resources. | |||
| Remote sensing technology has the characteristics of macro, multi-band, periodicity and economic. Macro refers to that the higher the remote sensing platform is, the wider the perspective will be, and the wider the ground can be synchronously detected. The multi-band property means that the sensor can detect and record information in different bands such as ultraviolet, visible light, near infrared and microwave. Periodicity means that the remote sensing satellite has the characteristic of acquiring images repeatedly in a certain period, which can carry out repeated observation of the same area in a short time. Economic means that remote sensing technology can be used as a way to obtain large area of surface information without spending too much manpower and material resources. | |||
There was a problem hiding this comment.
Remote sensing technology has the characteristics of macroscopic, multi-band, periodic and economic
Macro -> Macroscoptic
docs/data/coco_tools_en.md
Outdated
| - The `images` key is a list with 5000 elements, and the output shows the first 5; | ||
| - The `annotations` key is a list with 36,781 elements, and the output shows the first five; | ||
| - The `categories` key is a list of 80 elements, with the output showing the first five. | ||
| - `info`: A dictionary. There are 6 key-value pairs, the output shows the first 5 pairs; |
docs/data/coco_tools_en.md
Outdated
| - The `categories` key is a list of 80 elements, with the output showing the first five. | ||
| - `info`: A dictionary. There are 6 key-value pairs, the output shows the first 5 pairs; | ||
| - `licenses`: A list with eight elements, and the output shows the first five; | ||
| - `images`: A list with 5000 elements, and the output shows the first 5; |
There was a problem hiding this comment.
first 5 -> first five
与其他句子统一~
docs/data/rs_data_en.md
Outdated
|
|
||
| In order to introduce the imaging principle of remote sensing image, the concept of raster should be introduced first. Raster is a pixel-based data format that effectively represents continuous surfaces. The information in the raster is stored in a grid structure, and each information cell or pixel has the same size and shape, but different values. Digital photographs, orthophoto images and satellite images can all be stored in this format. | ||
|
|
||
| Raster formats are ideal for analysis that looks at spatial and temporal changes because each data value has a grid-based accessible location. This allows us to access the same geographic location in two or more different grids and compare their values. |
There was a problem hiding this comment.
looks at -> concentrates on
docs/data/rs_data_en.md
Outdated
|
|
||
| Raster formats are ideal for analysis that looks at spatial and temporal changes because each data value has a grid-based accessible location. This allows us to access the same geographic location in two or more different grids and compare their values. | ||
|
|
||
| When the Earth observation satellite takes a picture, the sensor will record the DN value (Digital Number) of different wavelength electromagnetic wave in the grid pixel. Through DN value, the irradiance rate and reflectance of ground objects can be inversely calculated. The relationship between them is shown in the following formula, where $gain$ and $bias$ refer to the gain and offset of the sensor respectively; $L$ is irradiance, also known as radiant brightness value; $\rho$ is the reflectance of ground objects; $d_{s}$、$E_{0}$ and $\theta$ respectively represent the distance between solar and earth astronomical units, solar irradiance and solar zenith Angle. |
There was a problem hiding this comment.
Digital Number 是DN的缩写，所以这个括号应该在DN之后。
另外英文中时候number已有value之意，所以直接说DN就可以，不必写DN value。
docs/data/tools_en.md
Outdated
|
|
||
| - `image_path`: Path of the image to be split. | ||
| - `mask_path`: Path of the label image to be split together, default is `None`. | ||
| - `block_size`: Size of the split image blocks, default is 512. |
There was a problem hiding this comment.
这里的512好像忘记用反引号圈起来了，也辛苦一并修改一下中文文档吧~
docs/data/tools_en.md
Outdated
| - `image_path`: Path of the image to be split. | ||
| - `mask_path`: Path of the label image to be split together, default is `None`. | ||
| - `block_size`: Size of the split image blocks, default is 512. | ||
| - `save_dir`: folder path to save the split results, default is `output`. |
There was a problem hiding this comment.
首字母需要大写。另外，Folder path -> Directory
docs/data/tools_en.md
Outdated
| - `max_level`: The maximum level of scale to retrieve. If it is `None`, it means that there is no limit to the level. Default is `None`. | ||
| - `include_bg`: If specified, also save the image blocks that only contain the background class and do not contain the target class. | ||
| - `--nonzero_ratio`: Specify a threshold. For any source image, if the ratio of nonzero pixels in the image block is less than this threshold, the image block will be discarded. If it is `None`, no filtering will be performed. Default is `None`. | ||
| - `--visualize`: If specified, after the program is executed, the image `./vis_quadtree.png` will be generated, which contains the visualization result of the node situation in the quadtree. An example is shown in the following figure: |
There was a problem hiding this comment.
contains the visualization result of the node situation -> visualizes the nodes
docs/data/tools_en.md
Outdated
|
|
||
| Among them: | ||
|
|
||
| - `im_paths`: Path of the source image(s), can specify multiple paths. |
docs/data/tools_en.md
Outdated
| Among them: | ||
|
|
||
| - `im_paths`: Path of the source image(s), can specify multiple paths. | ||
| - `mask_path`: Path to the Ground-truth label. |
|
Bobholamovic
requested changes
Mar 30, 2023
docs/apis/data_en.md
Outdated
| @@ -152,7 +152,7 @@ In PaddleRS a series of classes are defined that, when instantiated, perform cer | |||
|
|
|||
| ### `Transform` | |||
|
|
|||
| The `__call__` method of the `Transform` object takes a unique argument `sample`. `sample` must be a dictionary or a sequence of dictionaries. When `sample` is a sequence, perform data transformations for each dictionary in `sample` and return the results sequentially stored in a Python build-in list; when `sample` is a dictionary, the `Transform` object extracts input from some of its key-value pairs (these keys are called "input keys"), performs the transformation, and writes the results as key-value pairs into `sample`(these keys are called "output keys"). It should be noted that many of the `Transform` objects in PaddleRS overwrite key-value pairs, that is, there is an intersection between the input key and the output key. The common keys in `sample` and their meanings are as follows: | |||
| The `__call__` method of the `Transform` object takes a unique argument `sample`. `sample` must be a dictionary or a sequence of dictionaries. When `sample` is a sequence, perform data transformations for each dictionary in `sample` and return the results sequentially stored in a python build-in list; when `sample` is a dictionary, the `Transform` object extracts input from some of its key-value pairs (these keys are called "input keys"), performs the transformation, and writes the results as key-value pairs into `sample`(these keys are called "output keys"). It should be noted that many of the `Transform` objects in PaddleRS overwrite key-value pairs, that is, there is an intersection between the input key and the output key. The common keys in `sample` and their meanings are as follows: | |||
docs/apis/infer_en.md
Outdated
| @@ -145,7 +145,7 @@ Considering the large-scale nature of remote sensing image, PaddleRS provides sl | |||
| 2. Users can customize the size and stride of the sliding window. Meanwhile, PaddleRS supports sliding window overlapping. For the overlapping parts between windows, PaddleRS will automatically fuse the model's predicted results. | |||
| 3. The inference results can be saved in GeoTiff format, and the reading and writing of geographic transformation information and geographic projection information is supported. | |||
|
|
|||
| Currently, the image segmentation trainer ([`BaseSegmenter`](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/segmenter.py) and subclasses) and change detection trainer ([`BaseChangeDetector`](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/change_detector.py) and subclasses)have dynamic graph sliding window inference API. Take the API of image segmentation task as an example, the explanation is as follows: | |||
| Currently, the image segmentation trainer ([`BaseSegmenter`](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/segmenter.py) and subclasses) and change detection trainer ([`BaseChangeDetector`](https://github.com/PaddlePaddle/PaddleRS/blob/develop/paddlers/tasks/change_detector.py) and subclasses)have dynamic graph sliding window inference api. Take the api of image segmentation task as an example, the explanation is as follows: | |||
docs/data/dataset_en.md
Outdated
| @@ -1,6 +1,6 @@ | |||
| # Remote Sensing Open Source Dataset | |||
|
|
|||
| PaddleRS has collected and summarized the most commonly used **open source** deep learning data sets in the field of remote sensing, providing the following information for each data set: dataset description, image information, annotation information, source address, and AI Studio backup link. According to the task type, these data sets can be divided into **image classification, image segmentation, change detection, object detection, object tracking, multi-label classification, image generation** and other types. Currently, the collected remote sensing datasets include: | |||
| PaddleRS has collected and summarized the most commonly used **open source** deep learning data sets in the field of remote sensing, providing the following information for each data set: dataset description, image information, annotation information, source address, and aistudio backup link. According to the task type, these data sets can be divided into **image classification, image segmentation, change detection, object detection, object tracking, multi-label classification, image generation** and other types. Currently, the collected remote sensing datasets include: | |||
docs/data/dataset_en.md
Outdated
| @@ -16,4 +16,4 @@ Visit [Remote Sensing Data Set Summary](./dataset_summary.md) for more informati | |||
|
|
|||
| ## Dataset of Special Contribution | |||
|
|
|||
| * Sample data of typical urban roads in China of 2020(CHN6-CUG), provided by Professor [Qiqi Zhu](http://grzy.cug.edu.cn/zhuqiqi), China University of Geosciences. Please refer to [here](http://grzy.cug.edu.cn/zhuqiqi/zh_CN/yjgk/32368/content/1733.htm) for more information and download information. | |||
| * Sample data of typical urban roads in China of 2020(CHN6-CUG), provided by professor [Qiqi Zhu](http://grzy.cug.edu.cn/zhuqiqi), China University of Geosciences. Please refer to [here](http://grzy.cug.edu.cn/zhuqiqi/zh_CN/yjgk/32368/content/1733.htm) for more information and download information. | |||
docs/data/tools_en.md
Outdated
|
|
||
| The main function of `geojson2mask.py` is to convert the geoJson-formatted labels to a .tif raster format. The usage is as follows: | ||
| The main function of `geojson2mask.py` is to convert the GeoJson-formatted labels to a .tif raster format. The usage is as follows: |
docs/dev/dev_guide_en.md
Outdated
| @@ -97,4 +97,4 @@ After writing the implementation of the operator, **you must write docstring and | |||
|
|
|||
| Remote sensing image processing tools are stored in the `tools/` directory. Each tool should be a relatively independent script, independent of the contents of the `paddlers/` directory, which can be executed by the user without installing PaddleRS. | |||
|
|
|||
| When writing the script, use the Python standard library `argparse` to process the command-line arguments entered by the user and execute the specific logic in the `if __name__ == '__main__':` code block. If you have multiple tools that use the same function or class, define these common components in `tools/utils`. | |||
| When writing the script, use the python standard library `argparse` to process the command-line arguments entered by the user and execute the specific logic in the `if __name__ == '__main__':` code block. If you have multiple tools that use the same function or class, define these common components in `tools/utils`. | |||
docs/dev/docstring_en.md
Outdated
| @@ -166,7 +166,7 @@ Multi examples: | |||
|
|
|||
| Examples 2: | |||
|
|
|||
| # Use it as a Python decorator. | |||
| # Use it as a python decorator. | |||
已修改~ |
|
我们将在CI通过后合入此PR。 再次感谢你为PaddleRS贡献文档，以及对PR review工作的配合。 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
Others
PR changes
Others
Description