Merge pull request #1 from pythonlessons/feature/initial-files

Feature/initial files

Author: pythonlessons

.gitignore

@@ -0,0 +1,9 @@
+__pycache__
+*.egg-info
+*.pyc
+venv
+
+Datasets/*
+Models/*
+
+!*.md

.vscode/launch.json

@@ -0,0 +1,16 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "Python: Current File",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "console": "integratedTerminal",
+            "justMyCode": false
+        }
+    ]
+}

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "python.analysis.typeCheckingMode": "off"
+}

Datasets/README.md

@@ -0,0 +1 @@
+# Empty repository to hold the datasets when running Tutorials

Models/README.md

@@ -0,0 +1 @@
+# Empty repository to hold the Models when running Tutorials

README.md

@@ -1,2 +1,14 @@
-# mltu
-Machine Learning Training Utilities (for TensorFlow)
+# MLTU - Machine Learning Training Utilities (TensorFlow)
+Machine Learning Training Utilities with TensorFlow 2.* and Python 3
+
+## Installation:
+Clone the repository and install the requirements:
+```bash
+git clone https://github.com/pythonlessons/mltu.git
+```
+```bash
+pip install .
+```
+
+# Tutorials and Examples:
+...

Tests/README.md

@@ -0,0 +1 @@
+# Repository for unit tests

Tutorials/01_image_to_word/README.md

@@ -0,0 +1,5 @@
+# Text Recognition With TensorFlow and CTC network
+In this tutorial, we will explore how to recognize text from images using TensorFlow and CTC loss with the Neural Networks model
+
+## Introduction
+...

Tutorials/01_image_to_word/configs.py

@@ -0,0 +1,17 @@
+import stow
+from datetime import datetime
+
+from mltu.configs import BaseModelConfigs
+
+class ModelConfigs(BaseModelConfigs):
+    def __init__(self):
+        super().__init__()
+        self.model_path = stow.join('Models/1_image_to_word', datetime.strftime(datetime.now(), "%Y%m%d%H%M"))
+        self.vocab = '0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz'
+        self.height = 32
+        self.width = 128
+        self.max_text_length = 23
+        self.batch_size = 1024
+        self.learning_rate = 1e-4
+        self.train_epochs = 100
+        self.train_workers = 20

Tutorials/01_image_to_word/inferenceModel.py

@@ -0,0 +1,56 @@
+import cv2
+import typing
+import numpy as np
+
+from mltu.inferenceModel import OnnxInferenceModel
+from mltu.utils.text_utils import ctc_decoder, get_cer
+
+class ImageToWordModel(OnnxInferenceModel):
+    def __init__(self, char_list: typing.Union[str, list], *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.char_list = char_list
+
+    def predict(self, image: np.ndarray):
+        image = cv2.resize(image, self.input_shape[:2][::-1])
+
+        image_pred = np.expand_dims(image, axis=0).astype(np.float32)
+
+        preds = self.model.run(None, {self.input_name: image_pred})[0]
+
+        text = ctc_decoder(preds, self.char_list)[0]
+
+        return text
+
+
+if __name__ == "__main__":
+    import pandas as pd
+    from tqdm import tqdm
+    from mltu.configs import BaseModelConfigs
+    
+    configs = BaseModelConfigs.load("Models/1_image_to_word/202211270035/configs.yaml")
+
+    model = ImageToWordModel(model_path=configs.model_path, char_list=configs.vocab)
+
+    df = pd.read_csv("Models/1_image_to_word/202211270035/val.csv").dropna().values.tolist()
+
+    accum_cer = []
+    for image_path, label in tqdm(df[:20]):
+        image = cv2.imread(image_path)
+
+        try:
+            prediction_text = model.predict(image)
+
+            cer = get_cer(prediction_text, label)
+            print(f"Image: {image_path}, Label: {label}, Prediction: {prediction_text}, CER: {cer}")
+
+            # resize image by 3 times for visualization
+            # image = cv2.resize(image, (image.shape[1] * 3, image.shape[0] * 3))
+            # cv2.imshow(prediction_text, image)
+            # cv2.waitKey(0)
+            # cv2.destroyAllWindows()
+        except:
+            continue
+        
+        accum_cer.append(cer)
+
+    print(f"Average CER: {np.average(accum_cer)}")