Skip to content

Epoch 51: early stopping #28

Open
Open

Description

@Zain-ul-abdin0

Hi!
Model ended training with Epoch 51: early stopping
. I am training with dataset 1040 images? I am training with my 1040 labled images given below but when I try to predict text it shows empty string.
Code ::
image = cv2.imread('./6wf4ef.jpg')
prediction_text = model.predict(image) // return empty string
print(f"Predicted Text: {prediction_text}")
2a4m2e
2a5bpc
2a8722
2d8dny
2de8g

2a2gn2

Prediction is Empty like this :
image

Activity

  1. pythonlessons commented on Aug 6, 2023

    @pythonlessons
    Owner

    your code doesn't match image, not clear what is model

  2. pythonlessons commented on Aug 6, 2023

    @pythonlessons
    Owner

    This trained model is not predicting my images sir.

    show your code, how you do this prediction, because I can't tell why without code...

  3. Zain-ul-abdin0 commented on Aug 6, 2023

    @Zain-ul-abdin0
    Author

    when I run train.py, it stopped training after sometime and give this message Epoch 51: early stopping and after that when I try to predict image. it did not predict. Below 3 lines I added, it worked with different data set nut not with this.

    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/02_captcha_to_text/202308060003/configs.yaml")
    model = ImageToWordModel(model_path=configs.model_path, char_list=configs.vocab)

    df = pd.read_csv("./Models/02_captcha_to_text/202308060003/val.csv").values.tolist()

    accum_cer = []

    for image_path, label in tqdm(df):

    image = cv2.imread(image_path)

    prediction_text = model.predict(image)

    cer = get_cer(prediction_text, label)

    print(f"Image: {image_path}, Label: {label}, Prediction: {prediction_text}, CER: {cer}")

    accum_cer.append(cer)

    image = cv2.imread('./6wf4ef.jpg')
    prediction_text = model.predict(image)
    print(f"Predicted Text: {prediction_text}")

  4. pythonlessons commented on Aug 7, 2023

    @pythonlessons
    Owner

    what was your CER while training model on validation data, I think your model was not trained

  5. Zain-ul-abdin0 commented on Aug 8, 2023

    @Zain-ul-abdin0
    Author

    is your model is able to be train from this data below , and if I share path to my labeled folder. May you please train your model on my data.

    2a8722

  6. pythonlessons commented on Aug 8, 2023

    @pythonlessons
    Owner

    Yes, I can train you a model for $$. Otherwise, I can't train a model for everyone. If you are interested write me an email to pythonlessons0@gmail.com

  7. zainulabdinsystems commented on Aug 11, 2023

    @zainulabdinsystems

    I can give you $$ but how much? I need to crack all images of this type.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions