ankonzoid / artificio

Deep Learning Computer Vision Algorithms for Real-World Use

artificio: A suite of computer vision deep learning algorithms

We provide here a suite of deep learning computer vision algorithms that are ready for real-world use:

Image Retrieval (via Transfer Learning)

Image Retrieval (via Autoencoders)

Image Processing Tools

Google Images Scraper

Authors

Anson Wong

artificio's Issues

run the model but get the error, help please!

Hello! Thanks for your sharing, I get a TyoeError when I run the model, and I do not know how to solve it.
On ubuntu14.04, python3.5.2
Detail as following:
Traceback (most recent call last):
File "image_rec.py", line 357, in
main()
File "image_rec.py", line 204, in main
MODEL.load_model(dictfn)
File "/home/naver/line-brain/lixun/cnn-based/image_rec/algo/autoencoders/AE.py", line 137, in load_model
self.encoder = load_model(encoder_filename)
File "/usr/local/lib/python3.5/dist-packages/keras/models.py", line 240, in load_model
with h5py.File(filepath, mode='r') as f:
File "/usr/local/lib/python3.5/dist-packages/h5py/_hl/files.py", line 271, in init
fid = make_fid(name, mode, userblock_size, fapl, swmr=swmr)
File "/usr/local/lib/python3.5/dist-packages/h5py/_hl/files.py", line 101, in make_fid
fid = h5f.open(name, flags, fapl=fapl)
File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper (/tmp/pip-huypgcah-build/h5py/_objects.c:2840)
File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper (/tmp/pip-huypgcah-build/h5py/_objects.c:2798)
File "h5py/h5f.pyx", line 78, in h5py.h5f.open (/tmp/pip-huypgcah-build/h5py/h5f.c:2117)
OSError: Unable to open file (Unable to open file: name = 'db/models/convae_encoder.h5', errno = 2, error message = 'no such file or directory', flags = 0, o_flags = 0)
Exception ignored in: <bound method BaseSession.del of <tensorflow.python.client.session.Session object at 0x7f8d737435f8>>
Traceback (most recent call last):
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 595, in del
TypeError: 'NoneType' object is not callable

List Index out of range

At :
Shape_img = imgs_train[0].shape

Seems to require keras 2.2.0

I was getting the usual error on 2.4.0

https://stackoverflow.com/questions/49113140/importerror-cannot-import-name-obtain-input-shape-from-keras

$ python3 ./similar_images_TL.py 
Using TensorFlow backend.
Traceback (most recent call last):
  File "./similar_images_TL.py", line 17, in <module>
    from vgg19 import VGG19
  File "src/vgg19.py", line 26, in <module>
    from keras.applications.imagenet_utils import _obtain_input_shape
ImportError: cannot import name '_obtain_input_shape'

uninstalling and switching to 2.2.0 fixed it for me

pip install keras==2.2.0

so could be something missing from the documentation? thanks for the interesting project!

I met this error when run

Hi, @zdyang @ouzoh @ankonzoid

I met this error when run similar_images_TL.py .

elxAir:similar_images_TL elx$ python3 similar_images_TL.py
Using TensorFlow backend.
Traceback (most recent call last):
File "similar_images_TL.py", line 17, in
from vgg19 import VGG19
File "src/vgg19.py", line 26, in
from keras.applications.imagenet_utils import _obtain_input_shape
ImportError: cannot import name '_obtain_input_shape' from 'keras.applications.imagenet_utils' (/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/keras/applications/imagenet_utils.py)

What's wrong to me?

Error: 'config': model.optimizer.get_config()

Hi
Great example ! I am having some issues early on:

File "image_retrieval.py", line 116, in <module>
    model.save_models()
  File "/home/jacques/work/fontf/artificio/image_retrieval/src/autoencoder.py", line 131, in save_models
    self.encoder.save(self.info["encoderFile"])
  File "/home/jacques/.virtualenvs/ml/lib/python3.6/site-packages/tensorflow/python/keras/_impl/keras/engine/network.py", line 1129, in save
    save_model(self, filepath, overwrite, include_optimizer)
  File "/home/jacques/.virtualenvs/ml/lib/python3.6/site-packages/tensorflow/python/keras/_impl/keras/engine/saving.py", line 149, in save_model
    'config': model.optimizer.get_config()
AttributeError: 'NoneType' object has no attribute 'get_config'

Query the kNN with new pictures

Hi, I'm trying to use the code to train an kNN model with a set of images. Then I'd like to get a new image (not included in the model) and get the most similar images.

Is it possible to do something like this? I tried using knn.predict(new_img_features) but I get an error and that could be because of the kNN used here is not a classifier.

Can this method be searched in svg in a svg dataset?

Error when running code on custom dataset

Hi @ankonzoid . Thanks for the code.

I am trying to run your code on my dataset which contains 500 random animal images of size 512x 512.
Code abrupty stops with this error.

Reading train images from '.....\dataset\data2\train'...
Reading test images from '......\dataset\data2\test'...
Image shape = (512, 512, 3)
Loading VGG19 pre-trained model...
2020-08-07 23:02:22.275737: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2
input_shape_model = (512, 512, 3)
output_shape_model = (16, 16, 512)
Applying image transformer to training images...
Applying image transformer to test images...
-> X_train.shape = (200, 512, 512, 3)
-> X_test.shape = (5, 512, 512, 3)
Inferencing embeddings using pre-trained model...
2020-08-07 23:02:59.016060: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 2147483648 exceeds 10% of system memory.
2020-08-07 23:03:45.605916: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 2147483648 exceeds 10% of system memory.

Please help!!

Question I am not able to find module autoencoders.AE and utilities?

custom image size for train image_retrieval

how to train with arbitrary image sizes in image retrieval?

Close figure after saving to improve memory and speed of similar_images

After approximately 20 images, the code uses a lot of memory (>10GB) and becomes incredibly slow. This is because the figure objects are not closed.

[Help!!]error of spawning new processes while running image_retrieval.py

I am getting the following error of spawning the process while running "python image_retrieval.py"

Brief error message:
RuntimeError:
An attempt has been made to start a new process before the
current process has finished its bootstrapping phase.
This probably means that you are not using fork to start your
child processes and you have forgotten to use the proper idiom
in the main module:
if name == 'main':
freeze_support()

error.txt

Can anyone plz help me saying why I am getting this error and how to get rid of this error?

Please add this to readme as resource

For those that wish to recreate algorithms I have a Computer Vision Development Container:
https://github.com/joehoeller/NVIDIA-GPU-Tensor-Core-Accelerator-PyTorch-OpenCV

Decoders shape is changing

Im getting the below error when I try to fit the model.

ValueError: A target array with shape (73, 458, 690, 3) was passed for an output of shape (None, 460, 692, 3) while using as loss binary_crossentropy. This loss expects targets to have the same shape as the output.

The summary of the autoencoder is,

autoencoder.summary():

---

Layer (type) Output Shape Param #

input_1 (InputLayer) (None, 458, 690, 3) 0

---

conv2d (Conv2D) (None, 458, 690, 16) 448

---

max_pooling2d (MaxPooling2D) (None, 229, 345, 16) 0

---

conv2d_1 (Conv2D) (None, 229, 345, 8) 1160

---

max_pooling2d_1 (MaxPooling2 (None, 115, 173, 8) 0

---

conv2d_2 (Conv2D) (None, 115, 173, 8) 584

---

max_pooling2d_2 (MaxPooling2 (None, 58, 87, 8) 0

---

conv2d_3 (Conv2D) (None, 58, 87, 8) 584

---

up_sampling2d (UpSampling2D) (None, 116, 174, 8) 0

---

conv2d_4 (Conv2D) (None, 116, 174, 8) 584

---

up_sampling2d_1 (UpSampling2 (None, 232, 348, 8) 0

---

conv2d_5 (Conv2D) (None, 230, 346, 16) 1168

---

up_sampling2d_2 (UpSampling2 (None, 460, 692, 16) 0

---

conv2d_6 (Conv2D) (None, 460, 692, 3) 435

Total params: 4,963
Trainable params: 4,963
Non-trainable params: 0

---

None

The decoder shape is changing at some point. Can you help me in fixing this issue.