Practical-Deep-Learning-Book

Practical-Deep-Learning-Book copied to clipboard

sidgan

Create data dir

5

I get an error when I run the notebook. I cannot write file because this directory './data' does not exist. This pr fixes that. To reproduce the error you can...

tbarton16

Predicted Class Name not plottted in Ch2-what-does-my-neural-network-think.ipynb

1

Hi, Thanks for the amazing book. I recently finished the book and started trying out exercises. In this notebook [1], the predictions like "10% cardigan" are not plotted in the...

CloudaYolla

ML Engine example: bad data preprocessing results in bad predictions

1

The [image-to-json.py](https://github.com/PracticalDL/Practical-Deep-Learning-Book/blob/d14a375379ab31874e23363d5fa009e1b3ea3ad4/code/chapter-9/image-to-json.py#L36) conversion script used in chapter 13 does not preprocess the data, instead leaving all inputs as values between 0-255. If used with the model from chapter 3, the...

fdb

The summary of changes included are as follows: chapter 2: - tensorflow 2.0.0 does not run on python 3.10 (version used by colab currently) - tensorflow gpu removed: https://github.com/tensorflow/tensorflow/releases/tag/v2.12.0 -...

manavrmoorthy

chapter-4/1_feature_extraction.ipynb using default batch size 32 instead of defined batch_size 128

1

``` batch_size = 128 datagen = tensorflow.keras.preprocessing.image.ImageDataGenerator(preprocessing_function=preprocess_input) generator = datagen.flow_from_directory(root_dir, target_size=(224, 224), class_mode=None, shuffle=False) ``` uses default batch_size = 32. `num_epochs = int(math.ceil(num_images / batch_size))` calculates 68 epochs based on...

gitgithan

chapter 9. infer.py does not work, but i fixed it.

1

here`s working code: ``` import flask import tensorflow as tf from tensorflow import keras from keras.utils import CustomObjectScope from keras.models import load_model from tensorflow.keras.utils import img_to_array from PIL import Image...

zoldaten

``` signature = tf.saved_model.signature_def_utils.predict_signature_def( inputs={'image_bytes': model.input}, outputs={'scores': model.output}) builder = tf.saved_model.builder.SavedModelBuilder('model.pb') builder.add_meta_graph_and_variables( sess=K.get_session(), tags=[tf.saved_model.tag_constants.SERVING], signature_def_map={ tf.saved_model.signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY: signature }) builder.save() ``` gives error: ``` AttributeError Traceback (most recent call last) Cell...

zoldaten

when train model (transfer learning) ``` # select the percentage of layers to be trained while using the transfer learning # technique. The selected layers will be close to the...

zoldaten

chapter ends with MobileNetV2 example. i extend a bit using MobileNetV3, as it is a bit fresher. but perfomance results shows strange things. ``` def predict2(img_path): img = image.load_img(img_path, target_size=(224,...

zoldaten