I've trained a simple CNN model on Cifar-10 in tensorflow with fake quantization (https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/quantize). I then generated a .tflite file using toco. Now I want to use a python interpreter to test the tflite model.
Since I used tf.image.per_image_standardization to subtract mean and divide by variance during training. I need to do the same thing to the testing data right? But, the problem is, my model is already fully quantized by tflite, and it only takes uint8 data as inputs. To do image standardization, I need to convert my image to float32. So how do I convert it back to uint8, or is image standardization even necessary for the testing data in this case? Thanks.
So, it turns out I need to do standardization on the testing data for a good accuracy.
To do it, I directly feed uint8 input images to the tf.image.per_image_standardization function. The function would convert the uint8 data to float32, and then do standardization (subtract mean, divide by std). You can find source code of the function here: https://github.com/tensorflow/tensorflow/blob/r1.11/tensorflow/python/ops/image_ops_impl.py
Now, I have the standardized float32 input images. What I did is writing a quantization function to quantize the float32 images back to uint8. The math comes from this paper: https://arxiv.org/abs/1803.08607
Now, I have the standardized uint8 input images, I then use tflite interpreter python API to test the model. It works as expected.
Related
Is there a general statement if or how much the accuracy and runtime at the inference decrease when using a TFLite model (.tflite) instead the original TensorFlow model (.h5)?
The simple answer is no. This really relies on how the model looks like along with what sort of optimization was applied on converting the H5 to tflite. But generally if you do not apply any sort of additional optimization but just performed the float32 conversion to tflite, I'd say most of the time tflite would generate the equivalent level of accuracy compared to the original model.
I am training a tensorflow model with with multiple images as input and a segmentation mask at the output. I wanted to perform random rotation augmentation in my dataset pipeline.
I have a list of parallel image file names (input output files aligned) for which I convert into tf dataset object using tf.data.Dataset.from_generator and then and use Dataset.map function to load images with tf.image.decode_png and tf.io.read_file commands.
How can I perform random rotations on the input-output images. I tried using random_transform function of ImageDataGenerator class but it expects numpy data as input and does not work on Tensors (Since tensorflow does not support eager execution in data pipeline I cannot convert it into numpy as well). I suppose I can use tf.numpy_function but I expect there should be some simple solution for this simple problem.
I am trying to convert DeepLab trained on the Cityscapes dataset from here to TFLite. From viewing the frozen graph in Netron, the input and output tensors both are of type uint8. I was able to use the default DeepLab model provided for the TFLite GPU delegate, which had float32 input and output tensors. I didn't think the model was supposed to be quantized, so when trying the following code without the commented lines, I got this error:
F tensorflow/lite/toco/tooling_util.cc:2241] Check failed: array.data_type == array.final_data_type Array "ImageTensor" has mis-matching actual and final data types (data_type=uint8, final_data_type=float).
After this, I found that I should try to quantize the model. I inserted the commented lines to use uint8 instead of float32, but I got this error, which seems like an unsupported op.
F ./tensorflow/lite/toco/toco_tooling.h:38] Check failed: s.ok() Unimplemented: this graph contains anoperator of type Cast for which the quantized form is not yet implemented. Sorry, and patches welcome (that's a relatively fun patch to write, mostly providing the actual quantized arithmetic code for this op).
Is it right to use the quantized script? The off-the-shelf TFLite DeepLab model provided uses float32. Thanks!
I have a model much like the tensorflow speech command demo except it takes a variable sized 1D array as input. Now I find it difficult to convert this model to TF lite using tflite_convert which requires input_shape for input.
It's said that tf lite requires fixed size input for efficiency and you can resize input during inference as part of your model. However, I think it would involve truncating the input which I don't want. Is there any way to make this work with TF lite?
You can convert your model using a fixed shape as in --input_shape=64, then at inference-time you would do:
interpreter->ResizeInputTensor(interpreter->inputs()[0], {128});
interpreter->AllocateTensors();
// ... populate your input tensors with 128 entries ...
interpreter->Invoke();
// ... read your output tensor ...
I'm using Tensorflow's 1.3 Estimator API to perform some image classification. Since I have a considerable amount of data, I gave the TFRecords a go. Saved the file and can read the examples to a Dataset using a parser function inside the input_fn of the estimator model. So far so good.
The issue is when I want to do some image augmentation (rotating and shearing in this case).
1) I tried using the tf.contrib.keras.preprocessing.image.random_shearand the likes. Turns out Keras doesn't like the format of TF's shape ('Dimension') and I can't cast it to a list because its arguments are the axis indexes not the actual value.
2) Then I tried using the tf.contrib.image.rotate and tf.contrib.image.transform with random values in my chosen range. This time I get an error of NotFoundError: Op type not registered 'ImageProjectiveTransform' in binary running on MYPC. Make sure the Op and Kernel are registered in the binary running in this process. which is an open issue (https://github.com/tensorflow/tensorflow/issues/9672). At the moment I can't move from Windows, so I would very interested in possible alternatives.
3) Searched for a way to read TFRecords and transform it to numpy array and do the augmentation with other tools, but can't find a way from within the input_fn from where I can't access the session.
Thanks!
Have you tried using function from the answer to the question below?tensorflow: how to rotate an image for data augmentation?