I would like to develop a TensorFlow probability regression model locally and deploy as Sagemaker endpoint. I have deployed standard XGB models like this previously and understand that one can deploy TensorFlow model like so:
from sagemaker.tensorflow.model import TensorFlowModel
tensorflow_model = TensorFlowModel(
name=tensorflow_model_name,
source_dir='code',
entry_point='inference.py',
model_data=<TENSORFLOW_MODEL_S3_URI>,
role=role,
framework_version='<TENSORFLOW_VERSION>')
tensorflow_model.deploy(endpoint_name=<ENDPOINT_NAME>,
initial_instance_count=1,
instance_type='ml.m5.4xlarge',
wait=False)
However, I do not think this will cover for example the dependency:
import tensorflow_probability as tfp
Do I need to use script mode or Docker instead? Any pointer would be very much appreciated. Thanks.
Another way would be to add "tensorflow_probability" to "requirements.txt" and both your code and requirements into dependencies if you are not using source_dir:
Model(entry_point='inference.py',
dependencies=['code', 'requirements.txt'])
You can create a requirements.txt in your source_dir(code) and place tensorflow-probability in it. Sagemaker will install the dependencies listed in requirements.txt before running your script.
Related
I am running a very basic sentiment analysis pipeline utilising the XLM-Roberta model on Huggingface. I am trying to ensure I am utilising the M1 chip as I will be looping over ~10e7 entries.
So as to be consistent I am running a fresh install of PyTorch following the yml file and step outlined in this (very useful) video, I subsequently pip install sentence-piece and protobuf (version 3.2.0) to deal with a few subsequent errors. When running a simple pipeline model I am however faced with the below:
# Imports
import pandas as pd
import datetime as dt
import itertools
from transformers import pipeline, AutoTokenizer
sentiment_model = pipeline(model="cardiffnlp/twitter-xlm-roberta-base-sentiment", return_all_scores = True)
ValueError: google.__spec__ is None
Interesting following the install methods for Tensorflow from the same channel runs fine but does not access the M1 chip and simply runs on CPU.
Has anyone faced this prior or have a method such that I can run PyTorch?
Many thanks in advance.
I am trying to save a model with tensorflow-agents. First I define the following:
collect_policy = tf_agent.collect_policy
saver = PolicySaver(collect_policy, batch_size=None)
and then save the model like this:
saver.save('my_directory/')
This works ok in google colab but I am getting the following error in my local PC.
AttributeError: module 'tensorflow.python.saved_model.nested_structure_coder' has no attribute 'StructureCoder'
These are the library versions I am using:
tensorflow 2.9.1
tf-agents 0.11.0
Tl;dr
Make sure you have the right tensorflow-probability version that is compatible for 2.9.x and tf-agents 0.11.0
pip uninstall tensorflow-probability
pip install tensorflow-probability==0.17.0
(0.19.0 for TF 2.11, 0.18.0 for TF 2.10 or look at the release notes)
Also make sure to restart your kernel from the notebook.
What the problem was
StructureCoder has been moved to tensorflow API. Therefore, other dependent libraries have made changes like this in tf-agent and like this in tensorflow-probability. Your machine is somehow picking up an older version that depends on the previous version of nested_structure_coder.
For me, I was using
tensorflow 2.9.0
tf-agents 0.13.0
tensorflow-probabilities 0.17.0
Try making an explicit import in your notebook:
import tensorflow_probability
print(tensorflow_probability.__version__) # was 0.17.0 for me
I want to use FaceNet as a embedding layer (which won't be trainable).
I tried loading FaceNet like so :
tf.keras.models.load_model('./path/tf_facenet')
where directory ./path/tf_facenet contains 4 files that can be downloaded at https://drive.google.com/file/d/0B5MzpY9kBtDVZ2RpVDYwWmxoSUk/edit
but a message error shows up :
OSError: SavedModel file does not exist at: ./path/tf_facenet/{saved_model.pbtxt|saved_model.pb}
And the h5 files downloaded from https://github.com/nyoki-mtl/keras-facenet doesn't seem to work either (they use tensorflow 1.3)
I had issued like you when load model facenet-keras. Maybe you python env missing h5py modules.
So you should install that conda install h5py
Hope you success!!!
I am trying to convert .onxx model to .pb model. I have written the code but i am getting error:
#tf_func(tf.ceil)AttributeError: module 'tensorflow' has no attribute 'ceil'
Code:
import onnx
from tensorflow.python.tools.import_pb_to_tensorboard import import_to_tensorboard
from onnx_tf.backend import prepare
onnx_model = onnx.load("original_3dlm.onnx")
tf_rep = prepare(onnx_model)
tf_rep.export_graph("model_var.pb")
import_to_tensorboard("model_var.pb", "tb_log")
How to resolve this issue? Is there any other way to convert Onxx to Tensorflow?
I solve this issue with this.
Tensorflow Backend for ONNX.
Let me know if you have any issue.
Change from tensorflow 2.0 to 1.14.Maybe solve the problem.
your code as far as I can tell should be fine. The problem probably lies in the onnx-tf version you currently use. pip currently installs a version that only supports TensorFlow <= 1.15.
run this in the terminal to install a more up-to-date version of onnx-tf.
pip uninstall onnx_tf
pip install git+https://github.com/onnx/onnx-tensorflow.git
refer to this issue for further details
Can the code which is written on keras using tensorflow as backend can run on another keras environment were theano is the backend? Is there any computational benefits when running on any of them?
To avoid dimensional errors just add
"image_data_format": "channels_last"
in the Keras.Json configuration file at
~\.keras
If there is not such a file you can create it yourself.