pythonlessons

# Create a data provider for the dataset
data_provider = DataProvider(
    dataset=dataset,
    skip_validation=True,
    batch_size=configs.batch_size,
    data_preprocessors=[ImageReader()],
    transformers=[
        # ImageShowCV2(), # uncomment to show images during training
        ImageResizer(configs.width, configs.height, keep_aspect_ratio=False),
        LabelIndexer(configs.vocab),

pythonlessons

dataset, vocab, max_len = [], set(), 0

# Preprocess the dataset by the specific IAM_Words dataset file structure
words = open(os.path.join(dataset_path, "words.txt"), "r").readlines()
for line in tqdm(words):
    if line.startswith("#"):
        continue

    line_split = line.split(" ")
    if line_split[1] == "err":

pythonlessons

def download_and_unzip(url, extract_to='Datasets', chunk_size=1024*1024):
    http_response = urlopen(url)

    data = b''
    iterations = http_response.length // chunk_size + 1
    for _ in tqdm(range(iterations)):
        data += http_response.read(chunk_size)

    zipfile = ZipFile(BytesIO(data))
    zipfile.extractall(path=extract_to)

pythonlessons

import os
import tarfile
from tqdm import tqdm
from io import BytesIO
from zipfile import ZipFile
from urllib.request import urlopen

import torch
import torch.optim as optim
from torchsummaryX import summary

pythonlessons

...
Image: Datasets/IAM_Words/words/b05/b05-017/b05-017-03-03.png, Label: won't, Prediction: won't, CER: 0.0
Image: Datasets/IAM_Words/words/a01/a01-049u/a01-049u-07-06.png, Label: session, Prediction: sessicn, CER: 0.14285714285714285
Image: Datasets/IAM_Words/words/a02/a02-000/a02-000-07-00.png, Label: but, Prediction: but, CER: 0.0
Image: Datasets/IAM_Words/words/m02/m02-087/m02-087-06-02.png, Label: as, Prediction: as, CER: 0.0
Image: Datasets/IAM_Words/words/g06/g06-037j/g06-037j-07-01.png, Label: his, Prediction: his, CER: 0.0
Image: Datasets/IAM_Words/words/g06/g06-047i/g06-047i-02-09.png, Label: human, Prediction: human, CER: 0.0
Image: Datasets/IAM_Words/words/c03/c03-094c/c03-094c-08-03.png, Label: gaudy, Prediction: gaudy, CER: 0.0
Image: Datasets/IAM_Words/words/e04/e04-132/e04-132-01-04.png, Label: on, Prediction: on, CER: 0.0
Image: Datasets/IAM_Words/words/k02/k02-018/k02-018-04-01.png, Label: surprised, Prediction: supised, CER: 0.2222222222222222

pythonlessons

===================================================================================
                                Kernel Shape      Output Shape    Params  \
Layer
0_rb1.convb1.Conv2d_conv       [3, 16, 3, 3]  [1, 16, 32, 128]     448.0
1_rb1.convb1.BatchNorm2d_bn             [16]  [1, 16, 32, 128]      32.0
2_rb1.LeakyReLU_act1                       -  [1, 16, 32, 128]         -
3_rb1.convb2.Conv2d_conv      [16, 16, 3, 3]  [1, 16, 32, 128]     2.32k
4_rb1.convb2.BatchNorm2d_bn             [16]  [1, 16, 32, 128]      32.0
5_rb1.Conv2d_shortcut          [3, 16, 1, 1]  [1, 16, 32, 128]      64.0
6_rb1.LeakyReLU_act2                       -  [1, 16, 32, 128]         -

pythonlessons

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, *args, **kwargs):
        super().__init__(*args, **kwargs)

pythonlessons

Serving TensorBoard on localhost; to expose to the network, use a proxy or pass --bind_all
TensorBoard 2.10.1 at http://localhost:6006/ (Press CTRL+C to quit)

pythonlessons

Tensorboard --logdir Models\08_handwriting_recognition_torch\202303142139\logs

pythonlessons

# create callbacks
earlyStopping = EarlyStopping(monitor='val_CER', patience=20, mode="min", verbose=1)
modelCheckpoint = ModelCheckpoint(configs.model_path + '/model.pt', monitor='val_CER', mode="min", save_best_only=True, verbose=1)
tb_callback = TensorBoard(configs.model_path + '/logs')
reduce_lr = ReduceLROnPlateau(monitor='val_CER', factor=0.9, patience=10, verbose=1, mode='min', min_lr=1e-6)
model2onnx = Model2onnx(
    saved_model_path=configs.model_path + '/model.pt', 
    input_shape=(1, configs.height, configs.width, 3), 
    verbose=1,
    metadata={"vocab": configs.vocab}