Releases 0.2.0

Author: FLming

Changelog

@@ -8,6 +8,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Add
 - Use RaggedTensor for train and eval pipeline
 
+## [0.2.0] - 2021-07-29
+### Changed
+- New export script
+- New model build way
+
 ## [0.1.1] - 2021-01-22
 ### Changed
 - Reduce minimum TensorFlow version to 2.2

README.md

@@ -1,6 +1,6 @@
 # Convolutional Recurrent Neural Network for End-to-End Text Recognition - TensorFlow 2
 
-![TensorFlow version](https://img.shields.io/badge/TensorFlow->=2.2-FF6F00?logo=tensorflow)
+![TensorFlow version](https://img.shields.io/badge/TensorFlow->=2.3-FF6F00?logo=tensorflow)
 ![Python version](https://img.shields.io/badge/Python->=3.6-3776AB?logo=python)
 [![Paper](https://img.shields.io/badge/paper-arXiv:1507.05717-B3181B?logo=arXiv)](https://arxiv.org/abs/1507.05717)
 [![Zhihu](https://img.shields.io/badge/知乎-文本识别网络CRNN—实现简述-blue?logo=zhihu)](https://zhuanlan.zhihu.com/p/122512498)
@@ -11,15 +11,20 @@ This is a re-implementation of the CRNN network, build by TensorFlow 2. This rep
 
 This repo aims to build a simple, efficient text recognize network by using the various components of TensorFlow 2. The model build by the Keras API, the data pipeline build by `tf.data`, and training with `model.fit`, so we can use most of the functions provided by TensorFlow 2, such as `Tensorboard`, `Distribution strategy`, `TensorFlow Profiler` etc.
 
+## Installation
+
+```bash
+$ pip install -r requirements.txt
+```
+
 ## Demo
 
 Here I provide an example model that trained on the Mjsynth dataset, this model can only predict 0-9 and a-z(ignore case).
 
-- [百度, 密码: a4ki](https://pan.baidu.com/s/19__FzoQxbCArf1gDm_ptPQ)
-- [Google drive](https://drive.google.com/file/d/1iQZxBovoGT-KDksR5f3PD1Avyiq3QJa-/view?usp=sharing)
-
 ```bash
-$ python tools/demo.py --images example/images/ --config configs/mjsynth.yml --model PATH/TO/MODEL
+$ wget https://github.com/FLming/CRNN.tf2/releases/download/v0.2.0/SavedModel.tgz
+$ tar xzvf SavedModel.tgz
+$ python tools/demo.py --images example/images/ --config configs/mjsynth.yml --model SavedModel
 ```
 
 Then, You will see output like this:
@@ -38,12 +43,6 @@ About decode methods, sometimes the beam search method will be better than the g
 
 Before you start training, maybe you should [prepare](#Data-prepare) data first. All predictable characters are defined by the [table.txt](example/table.txt) file. The configuration of the training process is defined by the [yml](configs/mjsynth.yml) file.
 
-### Installation
-
-```bash
-$ pip install -r requirements.txt
-```
-
 This training script uses all GPUs by default, if you want to use a specific GPU, please set the `CUDA_VISIBLE_DEVICES` parameter.
 
 ```bash
@@ -89,15 +88,15 @@ We should write the image path and its corresponding label to a text file in a c
 ## Eval
 
 ```bash
-$ python crnn/eval.py --config PATH/TO/CONFIG_FILE --model PATH/TO/MODEL
+$ python crnn/eval.py --config PATH/TO/CONFIG_FILE --weight PATH/TO/MODEL_WEIGHT
 ```
 
 ## Converte & Ecosystem
 
 There are many components here to help us do other things. For example, deploy by `Tensorflow serving`. Before you deploy, you can pick up a good weight, and convertes model to `SavedModel` format by this command, it will add the post processing layer in the last and cull the optimizer:
 
 ```bash
-$ python tools/export.py --model PATH/TO/MODEL --config PATH/TO/CONFIG_FILE --post greedy --output PATH/TO/OUTPUT
+$ python tools/export.py --config PATH/TO/CONFIG_FILE --weight PATH/TO/MODEL_WEIGHT --pre rescale --post greedy --output PATH/TO/OUTPUT
 ```
 
 And now `Tensorflow lite` also can convert this model, that means you can deploy it to Android, iOS etc.

requirements.txt

@@ -1,2 +1,2 @@
 pyyaml
-tensorflow >= 2.2
+tensorflow >= 2.3

tools/demo.py

@@ -5,8 +5,6 @@
 import tensorflow as tf
 from tensorflow import keras
 
-import decoders
-
 parser = argparse.ArgumentParser()
 parser.add_argument('--images', type=str, required=True, 
                     help='Image file or folder path.')
@@ -35,19 +33,12 @@ def read_img_and_resize(path, shape):
 
 
 model = keras.models.load_model(args.model, compile=False)
-decoder = decoders.CTCGreedyDecoder(config['table_path'])
 
 p = Path(args.images)
-if p.is_dir():
-    img_paths = p.iterdir()
-else:
-    img_paths = [p]
-
+img_paths = p.iterdir() if p.is_dir() else [p]
 for img_path in img_paths:
     img = read_img_and_resize(str(img_path), config['img_shape'])
     img = tf.expand_dims(img, 0)
     outputs = model(img)
-    if not isinstance(outputs, tuple):
-        outputs = decoder(outputs)
     print(f'Path: {img_path}, y_pred: {outputs[0].numpy()}, '
           f'probability: {outputs[1].numpy()}')