how to load a pretrained model?

[lijiaman]

I want to load the pretrained AlexNet model bvlc_alexnet.npy, and I know using np.load('bvlc_alexnet.npy') can get some numpy array of weights and bias. (Like net = np.load('bvlc_alexnet.npy'), net["conv1"][0] represent the weights of conv1) But there are some problems loading them into Conv2d layer.

Then while running, it gives error:
Variable cannot be callable.

Could someone give an example to load weights of pretrained model?By the way, I tried to install TensorLayer from git, while inputting command "pip install . -e", it said -e need an argument.. so how to install it properly if I want to modify some codes of it? Hope someone can help me. Thanks very much.

[zsdonghao]

Hi, you should use pip install -e ..
In term of loading model, I think you can find the answer from here : http://tensorlayer.readthedocs.io/en/latest/user/more.html#fqa

[lijiaman]

Thanks~

[wagamamaz]

@lijiaman your W_int ann b_int are wrong, this is not the way to use TensorLayer.

The network.all_params is the list of all parameters in network.

you can find an example to load a list of numpy array into tensorlayer as follow:

1. http://tensorlayer.readthedocs.io/en/latest/_modules/tensorlayer/files.html#load_npz
2. main_restore_embedding_layer()
   https://github.com/zsdonghao/tensorlayer/blob/master/tutorial_generate_text.py

[zsdonghao]

This is a good example provided by @wagamamaz

#! /usr/bin/python
# -*- coding: utf8 -*-

import numpy as np
import tensorflow as tf
import tensorlayer as tl
from tensorlayer.layers import set_keep
import time

is_test_only = False # if True, restore and test without training

X_train, y_train, X_val, y_val, X_test, y_test = \
                tl.files.load_mnist_dataset(shape=(-1, 28, 28, 1))

X_train = np.asarray(X_train, dtype=np.float32)[0:10000]#<-- small training set for fast debugging
y_train = np.asarray(y_train, dtype=np.int64)[0:10000]
X_val = np.asarray(X_val, dtype=np.float32)
y_val = np.asarray(y_val, dtype=np.int64)
X_test = np.asarray(X_test, dtype=np.float32)
y_test = np.asarray(y_test, dtype=np.int64)

print('X_train.shape', X_train.shape)
print('y_train.shape', y_train.shape)
print('X_val.shape', X_val.shape)
print('y_val.shape', y_val.shape)
print('X_test.shape', X_test.shape)
print('y_test.shape', y_test.shape)
print('X %s   y %s' % (X_test.dtype, y_test.dtype))

sess = tf.InteractiveSession()

batch_size = 128
x = tf.placeholder(tf.float32, shape=[None, 28, 28, 1])
y_ = tf.placeholder(tf.int64, shape=[None,])

def inference(x, is_train, reuse=None):
    # gamma_init = tf.random_normal_initializer(1., 0.02)
    with tf.variable_scope("CNN", reuse=reuse):
        tl.layers.set_name_reuse(reuse)
        network = tl.layers.InputLayer(x, name='input_layer')

        network = tl.layers.Conv2d(network, n_filter=32, filter_size=(5, 5), strides=(1, 1),
                act=None, b_init=None, padding='SAME', name='cnn_layer1')
        network = tl.layers.BatchNormLayer(network, act=tf.identity,#tf.nn.relu,
                # gamma_init=gamma_init,
                is_train=is_train, name='batch1')
        check = network.outputs
        network.outputs = tf.nn.relu(network.outputs)
        network = tl.layers.MaxPool2d(network, filter_size=(2, 2), strides=(2, 2),
                padding='SAME', name='pool_layer1')

        network = tl.layers.Conv2d(network, n_filter=64, filter_size=(5, 5), strides=(1, 1),
                act=None, b_init=None, padding='SAME', name='cnn_layer2')
        network = tl.layers.BatchNormLayer(network, act=tf.nn.relu,
                # gamma_init=gamma_init,
                is_train=is_train, name='batch2')
        network = tl.layers.MaxPool2d(network, filter_size=(2, 2), strides=(2, 2),
                padding='SAME', name='pool_layer2')

        ## end of conv
        network = tl.layers.FlattenLayer(network, name='flatten_layer')
        # if is_train:
        #     network = tl.layers.DropoutLayer(network, keep=0.5, is_fix=True, name='drop1')
        network = tl.layers.DenseLayer(network, n_units=256,
                                        act = tf.nn.relu, name='relu1')
        # if is_train:
        #     network = tl.layers.DropoutLayer(network, keep=0.5, is_fix=True, name='drop2')
        network = tl.layers.DenseLayer(network, n_units=10,
                                        act = tf.identity, name='output_layer')
    return network, check


# train phase
network, check = inference(x, is_train=True, reuse=False)
y = network.outputs
cost = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(y, y_))
correct_prediction = tf.equal(tf.argmax(y, 1), y_)
acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

# test phase
network_test, check_t = inference(x, is_train=False, reuse=True)
y_t = network_test.outputs
cost_t = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(y_t, y_))
correct_prediction_t = tf.equal(tf.argmax(y_t, 1), y_)
acc_t = tf.reduce_mean(tf.cast(correct_prediction_t, tf.float32))

# train
n_epoch = 5
learning_rate = 0.0001
print_freq = 1

train_params = network.all_params
train_op = tf.train.AdamOptimizer(learning_rate, beta1=0.9, beta2=0.999,
    epsilon=1e-08, use_locking=False).minimize(cost, var_list=train_params)

tl.layers.initialize_global_variables(sess)

if is_test_only:
    load_params = tl.files.load_npz(name='_model_test.npz')
    tl.files.assign_params(sess, load_params, network)

# network.print_params(True)
network.print_layers()

# tl.layers.print_all_variables(train_only=True)

for i, p in enumerate(tf.all_variables()):
    print("  Before {:3}: {:15} (mean: {:<18}, median: {:<18}, std: {:<18})   {}".format(i, str(p.eval().shape), p.eval().mean(), np.median(p.eval()), p.eval().std(), p.name))

print('   learning_rate: %f' % learning_rate)
print('   batch_size: %d' % batch_size)

if not is_test_only:
    for epoch in range(n_epoch):
        start_time = time.time()
        for X_train_a, y_train_a in tl.iterate.minibatches(
                                    X_train, y_train, batch_size, shuffle=True):
            _, c = sess.run([train_op, check], feed_dict={x: X_train_a, y_: y_train_a})
            # print('bn out train:', np.mean(c), np.std(c))

        if epoch + 1 == 1 or (epoch + 1) % print_freq == 0:
            print("Epoch %d of %d took %fs" % (epoch + 1, n_epoch, time.time() - start_time))
            train_loss, train_acc, n_batch = 0, 0, 0
            for X_train_a, y_train_a in tl.iterate.minibatches(
                                    X_train, y_train, batch_size, shuffle=True):
                err, ac = sess.run([cost_t, acc_t], feed_dict={x: X_train_a, y_: y_train_a})
                train_loss += err; train_acc += ac; n_batch += 1
            print("   train loss: %f" % (train_loss/ n_batch))
            print("   train acc: %f" % (train_acc/ n_batch))
            val_loss, val_acc, n_batch = 0, 0, 0
            for X_val_a, y_val_a in tl.iterate.minibatches(
                                        X_val, y_val, batch_size, shuffle=True):
                err, ac = sess.run([cost_t, acc_t], feed_dict={x: X_val_a, y_: y_val_a})
                val_loss += err; val_acc += ac; n_batch += 1
            print("   val loss: %f" % (val_loss/ n_batch))
            print("   val acc: %f" % (val_acc/ n_batch))

print('Evaluation')
test_loss, test_acc, n_batch = 0, 0, 0
for X_test_a, y_test_a in tl.iterate.minibatches(
                            X_test, y_test, batch_size=1, shuffle=True):
    err, ac, c = sess.run([cost_t, acc_t, check_t], feed_dict={x: X_test_a, y_: y_test_a})
    # print('bn out test:', np.mean(c), np.std(c))
    test_loss += err; test_acc += ac; n_batch += 1
print("   test loss: %f" % (test_loss/n_batch))
print("   test acc: %f" % (test_acc/n_batch))

# network.print_param#! /usr/bin/python
# -*- coding: utf8 -*-

import numpy as np
import tensorflow as tf
import tensorlayer as tl
from tensorlayer.layers import set_keep
import time

is_test_only = False # if True, restore and test without training

X_train, y_train, X_val, y_val, X_test, y_test = \
                tl.files.load_mnist_dataset(shape=(-1, 28, 28, 1))

X_train = np.asarray(X_train, dtype=np.float32)[0:10000]#<-- small training set for fast debugging
y_train = np.asarray(y_train, dtype=np.int64)[0:10000]
X_val = np.asarray(X_val, dtype=np.float32)
y_val = np.asarray(y_val, dtype=np.int64)
X_test = np.asarray(X_test, dtype=np.float32)
y_test = np.asarray(y_test, dtype=np.int64)

print('X_train.shape', X_train.shape)
print('y_train.shape', y_train.shape)
print('X_val.shape', X_val.shape)
print('y_val.shape', y_val.shape)
print('X_test.shape', X_test.shape)
print('y_test.shape', y_test.shape)
print('X %s   y %s' % (X_test.dtype, y_test.dtype))

sess = tf.InteractiveSession()

batch_size = 128
x = tf.placeholder(tf.float32, shape=[None, 28, 28, 1])
y_ = tf.placeholder(tf.int64, shape=[None,])

def inference(x, is_train, reuse=None):
    # gamma_init = tf.random_normal_initializer(1., 0.02)
    with tf.variable_scope("CNN", reuse=reuse):
        tl.layers.set_name_reuse(reuse)
        network = tl.layers.InputLayer(x, name='input_layer')

        network = tl.layers.Conv2d(network, n_filter=32, filter_size=(5, 5), strides=(1, 1),
                act=None, b_init=None, padding='SAME', name='cnn_layer1')
        network = tl.layers.BatchNormLayer(network, act=tf.identity,#tf.nn.relu,
                # gamma_init=gamma_init,
                is_train=is_train, name='batch1')
        check = network.outputs
        network.outputs = tf.nn.relu(network.outputs)
        network = tl.layers.MaxPool2d(network, filter_size=(2, 2), strides=(2, 2),
                padding='SAME', name='pool_layer1')

        network = tl.layers.Conv2d(network, n_filter=64, filter_size=(5, 5), strides=(1, 1),
                act=None, b_init=None, padding='SAME', name='cnn_layer2')
        network = tl.layers.BatchNormLayer(network, act=tf.nn.relu,
                # gamma_init=gamma_init,
                is_train=is_train, name='batch2')
        network = tl.layers.MaxPool2d(network, filter_size=(2, 2), strides=(2, 2),
                padding='SAME', name='pool_layer2')

        ## end of conv
        network = tl.layers.FlattenLayer(network, name='flatten_layer')
        # if is_train:
        #     network = tl.layers.DropoutLayer(network, keep=0.5, is_fix=True, name='drop1')
        network = tl.layers.DenseLayer(network, n_units=256,
                                        act = tf.nn.relu, name='relu1')
        # if is_train:
        #     network = tl.layers.DropoutLayer(network, keep=0.5, is_fix=True, name='drop2')
        network = tl.layers.DenseLayer(network, n_units=10,
                                        act = tf.identity, name='output_layer')
    return network, check


# train phase
network, check = inference(x, is_train=True, reuse=False)
y = network.outputs
cost = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(y, y_))
correct_prediction = tf.equal(tf.argmax(y, 1), y_)
acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

# test phase
network_test, check_t = inference(x, is_train=False, reuse=True)
y_t = network_test.outputs
cost_t = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(y_t, y_))
correct_prediction_t = tf.equal(tf.argmax(y_t, 1), y_)
acc_t = tf.reduce_mean(tf.cast(correct_prediction_t, tf.float32))

# train
n_epoch = 5
learning_rate = 0.0001
print_freq = 1

train_params = network.all_params
train_op = tf.train.AdamOptimizer(learning_rate, beta1=0.9, beta2=0.999,
    epsilon=1e-08, use_locking=False).minimize(cost, var_list=train_params)

tl.layers.initialize_global_variables(sess)

if is_test_only:
    load_params = tl.files.load_npz(name='_model_test.npz')
    tl.files.assign_params(sess, load_params, network)

# network.print_params(True)
network.print_layers()

# tl.layers.print_all_variables(train_only=True)

for i, p in enumerate(tf.all_variables()):
    print("  Before {:3}: {:15} (mean: {:<18}, median: {:<18}, std: {:<18})   {}".format(i, str(p.eval().shape), p.eval().mean(), np.median(p.eval()), p.eval().std(), p.name))

print('   learning_rate: %f' % learning_rate)
print('   batch_size: %d' % batch_size)

if not is_test_only:
    for epoch in range(n_epoch):
        start_time = time.time()
        for X_train_a, y_train_a in tl.iterate.minibatches(
                                    X_train, y_train, batch_size, shuffle=True):
            _, c = sess.run([train_op, check], feed_dict={x: X_train_a, y_: y_train_a})
            # print('bn out train:', np.mean(c), np.std(c))

        if epoch + 1 == 1 or (epoch + 1) % print_freq == 0:
            print("Epoch %d of %d took %fs" % (epoch + 1, n_epoch, time.time() - start_time))
            train_loss, train_acc, n_batch = 0, 0, 0
            for X_train_a, y_train_a in tl.iterate.minibatches(
                                    X_train, y_train, batch_size, shuffle=True):
                err, ac = sess.run([cost_t, acc_t], feed_dict={x: X_train_a, y_: y_train_a})
                train_loss += err; train_acc += ac; n_batch += 1
            print("   train loss: %f" % (train_loss/ n_batch))
            print("   train acc: %f" % (train_acc/ n_batch))
            val_loss, val_acc, n_batch = 0, 0, 0
            for X_val_a, y_val_a in tl.iterate.minibatches(
                                        X_val, y_val, batch_size, shuffle=True):
                err, ac = sess.run([cost_t, acc_t], feed_dict={x: X_val_a, y_: y_val_a})
                val_loss += err; val_acc += ac; n_batch += 1
            print("   val loss: %f" % (val_loss/ n_batch))
            print("   val acc: %f" % (val_acc/ n_batch))

print('Evaluation')
test_loss, test_acc, n_batch = 0, 0, 0
for X_test_a, y_test_a in tl.iterate.minibatches(
                            X_test, y_test, batch_size=1, shuffle=True):
    err, ac, c = sess.run([cost_t, acc_t, check_t], feed_dict={x: X_test_a, y_: y_test_a})
    # print('bn out test:', np.mean(c), np.std(c))
    test_loss += err; test_acc += ac; n_batch += 1
print("   test loss: %f" % (test_loss/n_batch))
print("   test acc: %f" % (test_acc/n_batch))

# network.print_params(True)

if not is_test_only:
    tl.files.save_npz(network.all_params, name='_model_test.npz', sess=sess)

for i, p in enumerate(tf.all_variables()):
    print("  After {:3}: {:15} (mean: {:<18}, median: {:<18}, std: {:<18})   {}".format(i, str(p.eval().shape), p.eval().mean(), np.median(p.eval()), p.eval().std(), p.name))
s(True)

if not is_test_only:
    tl.files.save_npz(network.all_params, name='_model_test.npz', sess=sess)

for i, p in enumerate(tf.all_variables()):
    print("  After {:3}: {:15} (mean: {:<18}, median: {:<18}, std: {:<18})   {}".format(i, str(p.eval().shape), p.eval().mean(), np.median(p.eval()), p.eval().std(), p.name))