问题
I am new to the google AutoML, once I trained a model, I want to see the model details, i.e. feature factors and the related coefficients. Any suggestions?
回答1:
Assuming you are talking about an AutoML Vision model (both classification or object detection work similar): You can choose to train an edge model when starting the training. This enables you to download the model afterwards as TensorFlow saved_model.pb
.
With this, you could then e.g. use Netron to visualize the network. Or you can load the model with Python and print details about it with some code like:
import tensorflow as tf
path_mdl = "input/model" # folder to file with saved_model.pb
with tf.Session(graph=tf.Graph()) as sess:
tf.saved_model.loader.load(sess, ["serve"], path_mdl)
graph = tf.get_default_graph()
# print input and output operations
graph.get_operations()
# print infos about all nodes
weight_nodes = [n for n in graph_def.node if n.op == 'Const']
for n in weight_nodes:
print("Name of the node - %s" % n.name)
print("Value - " )
print(tensor_util.MakeNdarray(n.attr['value'].tensor))
来源:https://stackoverflow.com/questions/57340630/how-to-get-google-automl-model-coefficients