Hot Dog ML Classification
hi! I want to make a snapchat lens that displays whether or not a hotdog is present. I made a model with tensorflow keras, but when I try to import the converted .onnx file I get "Resource import failed: Padding layer input Pad_value must be a constant scalar "
This is my tensorflow code:
```import numpy as np
import onnxmltools
from tensorflow.keras.models import load_model
from tensorflow.keras.applications.resnet50 import preprocess_input
from tensorflow.keras.applications import ResNet50
from tensorflow.keras.preprocessing.image import load_img, img_to_array
from learntools.deep_learning.decode_predictions import decode_predictions
image_size = 224
def read_and_prep_images(img_paths, img_height=image_size, img_width=image_size):
imgs = [load_img(img_path, target_size=(img_height, img_width)) for img_path in img_paths]
img_array = np.array([img_to_array(img) for img in imgs])
output = preprocess_input(img_array)
return(output)
class_list_pathh = "/content/drive/My Drive/resnet50/imagenet_class_index.json"
my_model = ResNet50(weights="imagenet")
test_data = read_and_prep_images(img_paths) #test
output_onnx_model = '/content/drive/My Drive/model2.onnx'
# Convert the Keras model h5 into ONNX
onnx_model = onnxmltools.convert_keras(my_model)
onnxmltools.utils.save_model(onnx_model, output_onnx_model)
preds = my_model.predict(test_data) #test
most_likely_labels = decode_predictions(preds, top=3, class_list_path=class_list_pathh) #test```
How would I fix this? Thanks!
Hi Uniepicgamer,
Thank you for your feedback. We are currently looking into your issue. We will be get back to you asap once we resolve this, thanks!
Eric
hey Eric, is it able to email the snap studio team/you? thanks for responding!