Skip to main content

How to clone/duplicate a TensorFlow Probability neural network model

I have a TensorFlow Probability model that is built similar to models described in this YouTube Video.

I'm using

  • python==3.8.11
  • tensorflow==2.10.0
  • tensorflow-probability==0.18.0

Here's the code to build the model:

def posterior_mean_field(kernel_size: int, bias_size: int, dtype: Any) -> tf.keras.Model:
    n = kernel_size + bias_size
    c = np.log(np.expm1(1.))

    return tf.keras.Sequential([
        tfp.layers.VariableLayer(2 * n, dtype=dtype),
        tfp.layers.DistributionLambda(lambda t: tfd.Independent(tfd.Normal(loc=t[..., :n],
                                                                           scale=1e-5 + tf.nn.softplus(c + t[..., n:])),
                                                                reinterpreted_batch_ndims=1)),
    ])


def prior_trainable(kernel_size: int, bias_size: int, dtype: Any) -> tf.keras.Model:
    n = kernel_size + bias_size
    return tf.keras.Sequential([
        tfp.layers.VariableLayer(n, dtype=dtype),
        tfp.layers.DistributionLambda(lambda t: tfd.Independent(
            tfd.Normal(loc=t, scale=1),
            reinterpreted_batch_ndims=1)),
    ])


def build_model():
    model = keras.Sequential([
        tfp.layers.DenseVariational(64, activation='relu', input_shape=[len(train_dataset.keys())],
                                    make_posterior_fn=posterior_mean_field,
                                    make_prior_fn=prior_trainable),
        layers.Dense(64, activation='relu'),
        layers.Dense(1),
    ])
    optimizer = tf.keras.optimizers.RMSprop(0.001)
    model.compile(loss='mse', optimizer=optimizer, metrics=['mae', 'mse'])
    return model


model = build_model()
model.build((3, 10))

When I remove the TensorFlow Probability layer (1st layer) in the model, I can clone the model and copy its weights like this:

import copy
from tensorflow.keras.models import clone_model
model_weights = copy.deepcopy(model.get_weights())
model_copy = clone_model(model)
model_copy.set_weights(model_weights)

However, when the TensorFlow Probability layer is present I get this error:

Traceback (most recent call last):
  File "/Users/jisom/opt/miniconda3/envs/ic-hours/lib/python3.8/site-packages/IPython/core/interactiveshell.py", line 3398, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-6-349eb0e7c1e5>", line 1, in <cell line: 1>
    model_new = clone_model(model)
  File "/Users/jisom/opt/miniconda3/envs/ic-hours/lib/python3.8/site-packages/keras/models.py", line 448, in clone_model
    return _clone_sequential_model(
  File "/Users/jisom/opt/miniconda3/envs/ic-hours/lib/python3.8/site-packages/keras/models.py", line 326, in _clone_sequential_model
    if isinstance(layer, InputLayer) else layer_fn(layer))
  File "/Users/jisom/opt/miniconda3/envs/ic-hours/lib/python3.8/site-packages/keras/models.py", line 56, in _clone_layer
    return layer.__class__.from_config(layer.get_config())
  File "/Users/jisom/opt/miniconda3/envs/ic-hours/lib/python3.8/site-packages/keras/engine/base_layer.py", line 727, in get_config
    raise NotImplementedError('Layer %s has arguments in `__init__` and '
NotImplementedError: Layer DenseVariational has arguments in `__init__` and therefore must override `get_config`.

I can see some information about how to deal with this error in this StackOverflow question, but in that question there's a custom-built transformer class that can be modified. I'm trying to use the clone_model function in keras, which I don't directly control. And, the error seems to be coming from the TFP DenseVariational layer that doesn't override get_config. How can I clone/duplicate a model, including its weights, if the model includes TensorFlow Probability layers?

Or should I be creating an issue in the TensorFlow Probability Github repo to get this functionality added to the DenseVariational layer?



source https://stackoverflow.com/questions/73832532/how-to-clone-duplicate-a-tensorflow-probability-neural-network-model
Labels:

Comments

Post a Comment

Popular posts from this blog

Where and how is this Laravel kernel constructor called? [closed]

Where and how is this Laravel kernel constructor called? public fucntion __construct(Application $app, $Router $roouter) { } I have read the documentation and some online tutorial but I can find any clear explanation. I am learning Laravel and I am wondering where does this kernel constructor receives its arguments from. "POSTMOTERM" CLARIFICATION: Here is more clarity.I have checked the boostrap/app.php and it is only used for boostrapping the interfaces into the container class. What is not clear to me is where and how the Kernel class is instatiated and the arguments passed to the object calling the constructor.Something similar to; obj = new kernel(arg1,arg2) or, is the framework using some magic functions somewhere? Special gratitude to those who burn their eyeballs and brain cells on this trivia before it goes into a full blown menopause alias "MARKED AS DUPLICATE". To some of the itchy-finger keyboard warriors, a.k.a The mods,because I believe in th...
Post a Comment
Read more

Why is my reports service not connecting?

I am trying to pull some data from a Postgres database using Node.js and node-postures but I can't figure out why my service isn't connecting. my routes/index.js file: const express = require('express'); const router = express.Router(); const ordersCountController = require('../controllers/ordersCountController'); const ordersController = require('../controllers/ordersController'); const weeklyReportsController = require('../controllers/weeklyReportsController'); router.get('/orders_count', ordersCountController); router.get('/orders', ordersController); router.get('/weekly_reports', weeklyReportsController); module.exports = router; My controllers/weeklyReportsController.js file: const weeklyReportsService = require('../services/weeklyReportsService'); const weeklyReportsController = async (req, res) => { try { const data = await weeklyReportsService; res.json({data}) console...
Post a Comment
Read more

How to show number of registered users in Laravel based on usertype?

i'm trying to display data from the database in the admin dashboard i used this: <?php use Illuminate\Support\Facades\DB; $users = DB::table('users')->count(); echo $users; ?> and i have successfully get the correct data from the database but what if i want to display a specific data for example in this user table there is "usertype" that specify if the user is normal user or admin i want to user the same code above but to display a specific usertype i tried this: <?php use Illuminate\Support\Facades\DB; $users = DB::table('users')->count()->WHERE usertype =admin; echo $users; ?> but it didn't work, what am i doing wrong? source https://stackoverflow.com/questions/68199726/how-to-show-number-of-registered-users-in-laravel-based-on-usertype
Post a Comment
Read more