How to use _output_summary method in stestr

Best Python code snippet using stestr_python

utils_nn.py

Source:utils_nn.py

...87	tf.summary.histogram(params.op.name, params)88	tf.summary.histogram(params.op.name + '/spartisty', tf.nn.zero_fraction(params))89def _image_summary(params):90	tf.summary.image(params.op.name, params, max_outputs=1)91def _output_summary(outputs):92	tf.summary.histogram(outputs.op.name + '/outputs', outputs)93	tf.summary.scalar(outputs.op.name + '/outputs_sparsity',94		tf.nn.zero_fraction(outputs))95def get_weight_placeholders():96    return tf.get_collection('weight_placeholder')97def get_mask_placeholders():98    return tf.get_collection('mask_placeholder')99#################################################################################100#LAYERS101#################################################################################102def fully_connected_layer(x, units, name, gpu, config):103	'''104	Args:105		x: inputs from the previous layer, shape = [None, x.shape[0]].106		units: # of nodes of this layer, scalar.107		name: the name of this layer in graph.108		config: yaml file for the configuration.109	Returns:110		h: outputs, 1D tensor, shape = [None, units]111	'''112	with tf.variable_scope(name) as scope:113		# initialize layer parameters114		weights = _weights_initializer(name='weights',115			shape=[x.get_shape().as_list()[1], units],116			stddev=np.sqrt(2/units),117			gpu=gpu,118			config=config)119		biases = _biases_initializer(name='biases',120			shape=[units],121			val=0.1,122			gpu=gpu,123			config=config)124		# initilize and apply mask if specified to125		if config['use_mask']:126			mask = _mask_initializer(name='mask',127				shape=weights.get_shape().as_list(),128				gpu=gpu)129			# Apply mask to the weights130			masked_weights = tf.multiply(mask, weights, name='masked_weights')131			_parameter_summary(masked_weights)132			new_shape = masked_weights.get_shape().as_list()133			new_shape.insert(0,1)134			new_shape.append(1)135			_image_summary(tf.reshape(masked_weights, new_shape))136			# Calculate the output of the fully connected layer137			h = tf.add(tf.matmul(x, masked_weights), biases, name=name)138		else:139			h = tf.add(tf.matmul(x, weights), biases, name=name)140		# Adds summary to the weights and outputs141		_parameter_summary(weights)142		_parameter_summary(mask)143		_output_summary(h)144		if config['use_owl'] | config['use_growl'] | config['use_group_lasso']:145			# create interface for OWL146			w_placeholder = tf.placeholder(tf.float32, [x.get_shape().as_list()[1], units])147			assign_op_w = tf.assign(weights, w_placeholder, validate_shape=True)148			tf.add_to_collection('weight_placeholder', (weights, w_placeholder, assign_op_w))149			mask_placeholder = tf.placeholder(tf.float32, [x.get_shape().as_list()[1], units])150			assign_op_m = tf.assign(mask, mask_placeholder, validate_shape=True)151			tf.add_to_collection('mask_placeholder', (mask, mask_placeholder, assign_op_m))152	return h153def relu_layer(x, name):154	'''155	Args:156		x: inputs from the previous layer.157	Returns:158		h: activation.159	'''160	with tf.variable_scope(name) as scope:161		h = tf.nn.relu(x, name=name)162		_output_summary(h)163	return h164def batch_normalization_layer(x, axis, phase, name):165	'''166	Args:167		x: inputs.168		phase: boolean, true for training, false for testing.169	Returns:170		h: results.171	'''172	h = tf.layers.batch_normalization(x, axis=axis, training=phase, name=name)...

utils_nn-checkpoint.py

Source:utils_nn-checkpoint.py

...87	tf.summary.histogram(params.op.name, params)88	tf.summary.histogram(params.op.name + '/spartisty', tf.nn.zero_fraction(params))89def _image_summary(params):90	tf.summary.image(params.op.name, params, max_outputs=1)91def _output_summary(outputs):92	tf.summary.histogram(outputs.op.name + '/outputs', outputs)93	tf.summary.scalar(outputs.op.name + '/outputs_sparsity',94		tf.nn.zero_fraction(outputs))95def get_weight_placeholders():96    return tf.get_collection('weight_placeholder')97def get_mask_placeholders():98    return tf.get_collection('mask_placeholder')99#################################################################################100#LAYERS101#################################################################################102def fully_connected_layer(x, units, name, gpu, config):103	'''104	Args:105		x: inputs from the previous layer, shape = [None, x.shape[0]].106		units: # of nodes of this layer, scalar.107		name: the name of this layer in graph.108		config: yaml file for the configuration.109	Returns:110		h: outputs, 1D tensor, shape = [None, units]111	'''112	with tf.variable_scope(name) as scope:113		# initialize layer parameters114		weights = _weights_initializer(name='weights',115			shape=[x.get_shape().as_list()[1], units],116			stddev=np.sqrt(2/units),117			gpu=gpu,118			config=config)119		biases = _biases_initializer(name='biases',120			shape=[units],121			val=0.1,122			gpu=gpu,123			config=config)124		# initilize and apply mask if specified to125		if config['use_mask']:126			mask = _mask_initializer(name='mask',127				shape=weights.get_shape().as_list(),128				gpu=gpu)129			# Apply mask to the weights130			masked_weights = tf.multiply(mask, weights, name='masked_weights')131			_parameter_summary(masked_weights)132			new_shape = masked_weights.get_shape().as_list()133			new_shape.insert(0,1)134			new_shape.append(1)135			_image_summary(tf.reshape(masked_weights, new_shape))136			# Calculate the output of the fully connected layer137			h = tf.add(tf.matmul(x, masked_weights), biases, name=name)138		else:139			h = tf.add(tf.matmul(x, weights), biases, name=name)140		# Adds summary to the weights and outputs141		_parameter_summary(weights)142		_parameter_summary(mask)143		_output_summary(h)144		if config['use_owl'] | config['use_growl'] | config['use_group_lasso']:145			# create interface for OWL146			w_placeholder = tf.placeholder(tf.float32, [x.get_shape().as_list()[1], units])147			assign_op_w = tf.assign(weights, w_placeholder, validate_shape=True)148			tf.add_to_collection('weight_placeholder', (weights, w_placeholder, assign_op_w))149			mask_placeholder = tf.placeholder(tf.float32, [x.get_shape().as_list()[1], units])150			assign_op_m = tf.assign(mask, mask_placeholder, validate_shape=True)151			tf.add_to_collection('mask_placeholder', (mask, mask_placeholder, assign_op_m))152	return h153def relu_layer(x, name):154	'''155	Args:156		x: inputs from the previous layer.157	Returns:158		h: activation.159	'''160	with tf.variable_scope(name) as scope:161		h = tf.nn.relu(x, name=name)162		_output_summary(h)163	return h164def batch_normalization_layer(x, axis, phase, name):165	'''166	Args:167		x: inputs.168		phase: boolean, true for training, false for testing.169	Returns:170		h: results.171	'''172	h = tf.layers.batch_normalization(x, axis=axis, training=phase, name=name)...

tiny_svr.py

Source:tiny_svr.py

...92            return False93    def _reload(self):94        self.warn("reloading ...")95        self._clear_timer()96        self._output_summary()97        self._handler.close()98        reload(svr_util)99        reload(svr_conf)100        self.on_reload()101        self._start()102        self._is_reload = False103    def _close(self):104        self.warn("close ...")105        self._clear_timer()106        self._output_summary()107        self._is_run = False108        self._handler.close()109    def _start(self):110        self._conf = svr_conf.CONF111        self._get_logger()112        self._handler = self.get_handler()113        self._handler.start()114        self._init_timer()115        self.on_start()116    def _signal(self, sig, _):117        self.warn('signal %s' % sig)118        if sig in (signal.SIGINT, signal.SIGTERM):119            self._is_close = True120        elif sig == signal.SIGUSR1:121            self._is_reload = True122    def _get_logger(self):123        self._logger = svr_util.get_logger(self.get_name(), self._conf)124        self.info = self._logger.info125        self.warn = self._logger.warn126        self.error = self._logger.error127    def _states_check(self):128        if not self._last_report_time:129            return130        report_time = self._conf['svr.timer.run_status_check_time_span']131        delta_time = time.time() - self._last_report_time132        if delta_time < report_time:133            return134        self.error('run_states_check: not receive report in %s s, start auto-exit!' % report_time)135        # close safely136        os.kill(os.getpid(), signal.SIGINT)137        # if can't close safely force_exit138        delay = self._conf['svr.close.force_close_delay']139        threading.Timer(interval=delay, function=svr_util.force_exit, args=[-3]).start()140    def _output_summary(self):141        self.warn('<SUMMARY> %s' % self._handler.get_summary())142    def _clear_timer(self):143        self._timer.clear()144        self._thread_timer.clear()145    def _init_timer(self):146        self._timer = TinyTimerObserver(self._logger)147        self._thread_timer = TinyThreadTimerObserver(self._logger, self._conf['svr.timer.min_span'])148        check_timer = PeriodTimer(self._states_check, self._conf['svr.timer.run_status_check_time_span'])149        summary_timer = FixedPeriodTimer(self._output_summary, self._conf['svr.timer.summary_output_time_point'])150        self._thread_timer.add('check_timer', check_timer)151        self._thread_timer.add('summary_timer', summary_timer)...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.