import torch
from torch import nn
import IStudio as iv
studio = iv.AIStudio("pytorch")
model = nn.Sequential()
model.add_module('W0', nn.Linear(8, 16))
model.add_module('tanh', nn.Tanh())
model.add_module('W1', nn.Linear(16, 1))
x = torch.randn(1, 8)
studio.RecordGraphDef(model(x), "speech",
dict(list(model.named_parameters()) + [('x', x)]))
def main(_):
"""get ai studio with tensorflow as backend"""
studio = iv.AIStudio("tensorflow")
# Import data
mnist = input_data.read_data_sets(studio.DataFolder, one_hot=True)
# Create the model
x = tf.placeholder(tf.float32, [None, 784], name="x")
W = tf.Variable(tf.zeros([784, 10]), name="W")
b = tf.Variable(tf.zeros([10]), name="b")
y = tf.matmul(x, W) + b
# Define loss and optimizer
y_ = tf.placeholder(tf.float32, [None, 10])
# The raw formulation of cross-entropy,
#
# tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(tf.nn.softmax(y)),
# reduction_indices=[1]))
#
# can be numerically unstable.
#
# So here we use tf.nn.softmax_cross_entropy_with_logits on the raw
# outputs of 'y', and then average across the batch.
cross_entropy = tf.reduce_mean(
tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y))
train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)
sess = tf.InteractiveSession()
tf.global_variables_initializer().run()
#record computation graph
studio.RecordGraphDef(sess.graph_def, "softmax")
studio.AddRecordVariable(W, "w")
studio.AddRecordImage(x, "input image")
studio.AddRecordScalar(cross_entropy, "loss")
summary = studio.GetRecordList()
# Train
for _ in range(20):
batch_xs, batch_ys = mnist.train.next_batch(100)
c, ww, _ = sess.run([cross_entropy, W, train_step], feed_dict={x: batch_xs, y_: batch_ys})
"""record data for real-time view"""
studio.RecordVariable(ww, "w")
studio.RecordImage(batch_xs, "input image")
studio.RecordScalar(c, "loss")
#studio.SaveRecordList(summary_result)
# Test trained model
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
ac = sess.run(accuracy, feed_dict={x: mnist.test.images,
y_: mnist.test.labels})
#record results for plot
studio.RecordScalar(ac, "accuracy(true)")
studio.RecordScalar(1.0 - ac, "accuracy(false)")
#inference
draw = iv.IPDraw("Draw2D-1")
image3 = iv.IPGraph("Image-3")
plot3 = image3.Plots(0)
print("Start inference, draw number using draw pad")
for k in range(600):
# get draw image
image = draw.GetDrawImage(28*28, 28, 28, 1, False)
image.astype(np.float32)
image /= (-image.max())
image += 1
#draw the normalized image
plot3.ImageColor(image, 28, 28)
xx = np.expand_dims(image, axis=0)
inference = tf.argmax(y, 1)
yy, results = sess.run([y, inference], feed_dict={x: xx})
print("You drawing is:", results)
sleep(1)
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
import sys
import tensorflow as tf
import numpy as np
import IStudio as iv
studio = iv.AIStudio("tensorflow")
filename = inspect.getframeinfo(inspect.currentframe()).filename
dir_path = os.path.dirname(os.path.abspath(filename))
mobile_file = dir_path + "\\rcnn_ssdlite_mobile.pb"
resnet_file = dir_path + "\\rcnn_resnet50.pb"
maskrnn_file = dir_path + "\\mask_rcnn_inceptionv2.pb"
label_file = dir_path + "\\mscoco_label_map.txt"
height, width = 800, 800
NUM_CLASSES = 90
max_boxes_to_draw = 100
#load label
with open(label_file) as f:
label_list = f.read().splitlines()
camera = iv.IPCamera("Camera-1")
camera.ClearBoundingBox()
input_img = np.zeros((1, height, width, 3), dtype=np.uint8)
#load forzen graph
studio.LoadGraphDef(maskrnn_file)
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import sys
import IStudio as vi
import numpy as np
s = vi.AIStudio("tensorflow")
print(sys.path)