Merge probability branch.

Author: cottrell

Dockerfile

@@ -1,4 +1,4 @@
-FROM pytorch/pytorch
+FROM docker.io/pytorch/pytorch
 
 # if you forked EasyOCR, you can pass in your own GitHub username to use your fork
 # i.e. gh_username=myname

MANIFEST.in

@@ -4,3 +4,5 @@ include README.md
 include easyocr/model/*
 include easyocr/character/*
 include easyocr/dict/*
+include easyocr/scripts/compile_dbnet_dcn.py
+recursive-include easyocr/DBNet *

README.md

@@ -14,33 +14,8 @@ Integrated into [Huggingface Spaces 🤗](https://huggingface.co/spaces) using [
 
 
 ## What's new
-- 2 June 2022 - Version 1.5.0
-    - Add trainer for CRAFT detection model (thanks[@gmuffiness](https://github.com/gmuffiness), see [PR](https://github.com/JaidedAI/EasyOCR/pull/739))
-- 9 April 2022 - Version 1.4.2
-    - Update dependencies (opencv and pillow issues)
-- 11 September 2021 - Version 1.4.1
-    - Add trainer folder
-    - Add `readtextlang` method (thanks[@arkya-art](https://github.com/arkya-art), see [PR](https://github.com/JaidedAI/EasyOCR/pull/525))
-    - Extend `rotation_info` argument to support all possible angles (thanks[abde0103](https://github.com/abde0103), see [PR](https://github.com/JaidedAI/EasyOCR/pull/515))
-- 29 June 2021 - Version 1.4
-    - [Instructions](https://github.com/JaidedAI/EasyOCR/blob/master/custom_model.md) on training/using custom recognition models
-    - Example [dataset](https://www.jaided.ai/easyocr/modelhub) for model training
-    - Batched image inference for GPUs (thanks [@SamSamhuns](https://github.com/SamSamhuns), see [PR](https://github.com/JaidedAI/EasyOCR/pull/458))
-    - Vertical text support (thanks [@interactivetech](https://github.com/interactivetech)). This is for rotated text, not to be confused with vertical Chinese or Japanese text. (see [PR](https://github.com/JaidedAI/EasyOCR/pull/450))
-    - Output in dictionary format (thanks [@A2va](https://github.com/A2va), see [PR](https://github.com/JaidedAI/EasyOCR/pull/441))
-- 30 May 2021 - Version 1.3.2
-    - Faster greedy decoder (thanks [@samayala22](https://github.com/samayala22))
-    - Fix bug when a text box's aspect ratio is disproportional (thanks [iQuartic](https://iquartic.com/) for bug report)
-- 20 April 2021 - Version 1.3.1
-    - Add support for PIL image (thanks [@prays](https://github.com/prays))
-    - Add Tajik language (tjk)
-    - Update argument setting for command line
-    - Add `x_ths` and `y_ths` to control merging behavior when `paragraph=True`
-- 21 March 2021 - Version 1.3
-    - Second-generation models: multiple times smaller size, multiple times faster inference, additional characters and comparable accuracy to the first generation models.
-    EasyOCR will choose the latest model by default but you can also specify which model to use by passing `recog_network` argument when creating a `Reader` instance.
-    For example, `reader = easyocr.Reader(['en','fr'], recog_network='latin_g1')` will use the 1st generation Latin model
-    - List of all models: [Model hub](https://www.jaided.ai/easyocr/modelhub)
+- 24 September 2024 - Version 1.7.2
+    - Fix several compatibilities
 
 - [Read all release notes](https://github.com/JaidedAI/EasyOCR/blob/master/releasenotes.md)
 
@@ -145,7 +120,7 @@ The api should be as easy as
 ``` python
 reader = easyocr.Reader(['en'], detection='DB', recognition = 'Transformer')
 ```
-The idea is to be able to plug-in any state-of-the-art model into EasyOCR. There are a lot of geniuses trying to make better detection/recognition models, but we are not trying to be geniuses here. We just want to make their works quickly accessible to the public ... for free. (well, we believe most geniuses want their work to create a positive impact as fast/big as possible) The pipeline should be something like the below diagram. Grey slots are placeholders for changeable light blue modules.
+The idea is to be able to plug in any state-of-the-art model into EasyOCR. There are a lot of geniuses trying to make better detection/recognition models, but we are not trying to be geniuses here. We just want to make their works quickly accessible to the public ... for free. (well, we believe most geniuses want their work to create a positive impact as fast/big as possible) The pipeline should be something like the below diagram. Grey slots are placeholders for changeable light blue modules.
 
 ![plan](examples/easyocr_framework.jpeg)
 

custom_model.md

@@ -1,12 +1,8 @@
-# Custom model
+# Custom recognition models
 
 ## How to train your custom model
 
-There are 2 options to train your own recognition model:
-
-**1. Open-source approach**
-
-For the open-source approach, you can use your own data or generate your own dataset. To generate your own data, we recommend using
+You can use your own data or generate your own dataset. To generate your own data, we recommend using
 [TextRecognitionDataGenerator](https://github.com/Belval/TextRecognitionDataGenerator). We provide an example of a dataset [here](https://jaided.ai/easyocr/modelhub/).
 After you have a dataset, you can train your own model by following this repository
 [deep-text-recognition-benchmark](https://github.com/clovaai/deep-text-recognition-benchmark).
@@ -18,13 +14,9 @@ Please do not create an issue about data generation and model training in this r
 
 Note: We also provide our version of a training script [here](https://github.com/JaidedAI/EasyOCR/tree/master/trainer). It is a modified version from [deep-text-recognition-benchmark](https://github.com/clovaai/deep-text-recognition-benchmark).
 
-**2. Web-based approach**
-
-Jaided AI provides a web-based (paid) service for training your own model [here](https://jaided.ai/). You can train your model on the cloud and export it for local deployment. All 3 files are downloadable once the model has finished training.
-
 ## How to use your custom model
 
-To use your own recognition model, you need the three files from the open-source or web-based approach above. These three files have to share the same name (i.e. `yourmodel.pth`, `yourmodel.yaml`, `yourmodel.py`) that you will then use to call your model with in the EasyOCR API.
+To use your own recognition model, you need the three files as explained above. These three files have to share the same name (i.e. `yourmodel.pth`, `yourmodel.yaml`, `yourmodel.py`) that you will then use to call your model with EasyOCR API.
 
 We provide [custom_example.zip](https://jaided.ai/easyocr/modelhub/)
 as an example. Please download, extract and place `custom_example.py`, `custom_example.yaml` in the `user_network_directory` (default = `~/.EasyOCR/user_network`) and place `custom_example.pth` in model directory (default = `~/.EasyOCR/model`)