tensorflow - Freeze tensor_forest graph for Android -

-



i built aimple random forest model in tensorflow, , want freeze & optimize android. used following function building tesnor_forest estimator:

def build_estimator(_model_dir, _num_classes, _num_features, _num_trees, _max_nodes):       params = tensor_forest.foresthparams(       num_classes=_num_classes, num_features=_num_features,       num_trees=_num_trees, max_nodes=_max_nodes, min_split_samples=3)      graph_builder_class = tensor_forest.randomforestgraphs     return random_forest.tensorforestestimator(       params, graph_builder_class=graph_builder_class,       model_dir=_model_dir)

this function stores textual model graph.pbtxt file in specified model directory.

then train using:

est = build_estimator(output_model_dir, 3,np.size(features_eval,1), 5,6) train_x = features_eval.astype(dtype=np.float32) train_y = labels_y.astype(dtype=np.float32) est.fit(x=train_x, y=train_y, batch_size=np.size(features_eval,0))

(in simple example: number of trees = 5, max_nodes=6)

now want freeze model, call function:

def save_model_android(model_path): checkpoint_state_name = "model.ckpt-1" input_graph_name = "graph.pbtxt" output_graph_name = "freezed_model.pb" checkpoint_path = os.path.join(model_path, checkpoint_state_name)  input_graph_path = os.path.join(model_path, input_graph_name) input_saver_def_path = none input_binary = false output_node_names = "output" restore_op_name = "save/restore_all" filename_tensor_name = "save/const:0" output_graph_path = os.path.join(model_path, output_graph_name) clear_devices = true  freeze_graph(input_graph_path, input_saver_def_path,                           input_binary, checkpoint_path,                           output_node_names, restore_op_name,                           filename_tensor_name, output_graph_path,                           clear_devices, "")

and in freezed_model.pb file generated 1 op output node. in console following message when freeze_graph function called:

converted 0 variables const ops. 1 ops in final graph.

does knows why 1 node exported when calling freeze_graph?

i'm using tensorflow version 1.2.1 cuda support , installed sources on linux





wiki

Comments

Post a Comment

Popular posts from this blog

Asterisk AGI Python Script to Dialplan does not work -

-
i'm stuck project have check database of sip extension incoming caller id , dial sip extension. below receive when run script on asterisk cli. -- registered sip '9125' @ 192.168.1.102:50142 > saved useragent "zoiper rv2.8.40" peer 9125 [2017-08-22 13:53:33] notice[5095]: chan_sip.c:24558 handle_response_peerpoke: peer '9125' reachable. (17ms / 2000ms) == using sip rtp tos bits 184 == using sip rtp cos mark 5 -- executing [0112617769@from-trunk:1] answer("sip/obitrunk1-0000000d", "") in new stack > 0x1ca50d0 -- probation passed - setting rtp source address 192.168.1.98:16834 -- executing [0112617769@from-trunk:2] agi("sip/obitrunk1-0000000d", "test.py") in new stack -- launched agi script /var/lib/asterisk/agi-bin/test.py args: ['/var/lib/asterisk/agi-bin/test.py'] env line: agi_request: test.py env line: agi_channel: sip/obitrunk1-0000000d env line: agi_language: en env line: a
Read more

python - Read npy file directly from S3 StreamingBody -

-
i have many .npy (numpy) , .json files saved in s3, need download , load these files. for json files downloading files , loading dictionary using following code: obj = s3.object(bucket, key) response = obj.get() body = response['body'].read().decode('utf-8') data = json.loads(body) that way avoid writing .json file disk , reading it. now want same .npy files, equivalent? currently downloading file, saving temp file on disk , loading np.load avoid if possible. following https://stackoverflow.com/a/28196540/3509999 : obj = s3.object(bucket, key) io.bytesio(obj.get()["body"].read()) f: # rewind file f.seek(0) arr = np.load(f) wiki
Read more

kotlin - Out-projected type in generic interface prohibits the use of metod with generic parameter -

-
i start using kotlin , defined interface this: interface aadapter<vh : recyclerview.viewholder> { fun oncreateaviewholder(parent: viewgroup): vh fun onbindaviewholder(v: vh, position: int) } and when try use on code: class klasa ( private val adapter: aadapter<*> ) { fun dosth(){ //... val vh = this.adapter.oncreateaviewholder(parent) //on below line error adapter.onbindaviewholder(v, position) //... } } i error out-projected type 'aadapter<*>' prohibits use of 'public abstract fun onbindaviewholder(v: t, position: int): unit i tied add "in" or "out" definition i'm confused. how allow it. replace aadapter<*> aadapter<recyclerview.viewholder> . class klasa ( private val adapter: aadapter<recyclerview.viewholder> ) { // ... because have declared aadapter has type parameter vh type subclass of recyclerview.
Read more