profile
viewpoint
Zhaoyu Zhang dreamibor ARM Cambridge http://www.cloverio.com/ Software Engineer @ ARM.

dreamibor/GoogleHashCode2017 2

Google Hash Code 2017 First Round

dreamibor/algorithms_and_data_structures 1

Implementing basic algorithms and data structures using Python and C++.

dreamibor/Antbot_New 1

The old code (03-2016) of the Antbot by Leonard Eberding

dreamibor/finance_math_resources 1

A collection of online finance math resources.

dreamibor/hik 1

Python Solutions to HackerRank Interview Kit

dreamibor/ml_resources 1

A collection of awsome machine learming resources: blogs, books and courses.

dreamibor/Python-Face-Recognition-Simple-Example 1

A simple Python example of face recognition based on Face++ API.

dreamibor/requirements-to-table 1

Convert Python environmrnt file reuqirements.txt to MarkDown / HTML table.

dreamibor/Algorithm_Interview_Notes-Chinese 0

2018/2019/校招/春招/秋招/算法/机器学习(Machine Learning)/深度学习(Deep Learning)/自然语言处理(NLP)/C/C++/Python/面试笔记

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha ed10ddf361a307ed44f697c7db45645d011e1636

Update README.md

view details

push time in 22 days

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 02a73d679fab1b3a469299fa87abd04012d27694

Update README.md

view details

push time in 23 days

startedSplode/pomotroid

started time in 23 days

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 1508107036e6d9651b9ebf7f881d375422086659

Update README.md

view details

push time in 24 days

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 784820ab9ea7f6e44c59272f67706a804c300f12

Update README.md

view details

push time in 24 days

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 34c304248f317e62619a5f9283acb6d9c2123e74

Update README.md

view details

push time in a month

startedlilujunai/UCAS-Course

started time in a month

fork dreamibor/tensorflow

An Open Source Machine Learning Framework for Everyone

https://tensorflow.org

fork in a month

startedDeepTecher/awesome-autonomous-vehicle

started time in a month

startedalibaba/ai-matrix

started time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 69319560c6499808500a92e8c44655f3b44a0409

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 5181bf270d7b5821eb4d7328a31797230f6a3859

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 2fd8b00edee45ff580e544fc41bbe1d246c911e7

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 024617cc24b85b85f217c71db3a0980caea0e1ed

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 92a2bb2c90e787848594980557294805fc09a79f

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha f1da5dedaa11bf46fa663520fe57401e47f8870e

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha d58152ceafbb66d00ee1ee1f8fae9180fc5a114c

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha c94b627926dcc54814f8ccf30bcb03222a5cfea5

Update README.md

view details

push time in a month

push eventdreamibor/ml_resources

Jason Zhang

commit sha 028cd826578afb167c1cdac4b21e68fd3e45fa91

add ml interview resources

view details

push time in 2 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 5f86a43e30b84da66302423b04e1a08a9bf3b9b3

Update README.md

view details

push time in 2 months

startedtoandaominh1997/EfficientDet.Pytorch

started time in 2 months

create barnchdreamibor/ml_project_template

branch : master

created branch time in 3 months

created repositorydreamibor/ml_project_template

A template for ML projects in Python

created time in 3 months

startedichitenfont/I.Ming

started time in 3 months

startedichitenfont/inheritedglyphs

started time in 3 months

startedLingDong-/wenyan-lang

started time in 3 months

startedgoogle/mediapipe

started time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 9d7ace574e906e8053bd10c605ce0bd81a25c1d1

Update README.md

view details

push time in 4 months

issue commentAkojimaSLP/Neural-mask-estimation

Beamformed Samples Missing

Update: a straight-forward example is the beamformed audio under the /result folder, enhacement_all_channels has file size of 205,356 bytes while the original wav file such as F02_011C021A_BUS.CH1.wav 206,114 bytes)

dreamibor

comment created time in 4 months

issue closedAkojimaSLP/Neural-mask-estimation

Dataset

Hi, is there a way to create the training dataset? I mean the approach that you take to get seperate speech and noise data?

closed time in 4 months

dreamibor

issue commentAkojimaSLP/Neural-mask-estimation

Dataset

Thank you for your response! I think your answer solved my problem and I will close the issue.

dreamibor

comment created time in 4 months

issue openedAkojimaSLP/Neural-mask-estimation

Beamformed Data Sample Missing

Hi @AkojimaSLP , When I am exploring the repo, I found that the number of input samples and output samples are different, for example, if I got a 50-second wav audio (800000 samples) and the output will be only 799744 samples. I see it's related to the MVDR beamforming part, do you know why it's happening? Thank you very much!

created time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha f0b1920cbf2f484be49f59da50cc965d503317c7

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 392dc26ba86498c592a9da1ee3e84c1b065a1d45

Update README.md

view details

push time in 4 months

issue commentAkojimaSLP/Neural-mask-estimation

bug in generate_validate_data

Just close the two image windows should finish the program.

hadaso

comment created time in 4 months

startedvoice-engine/make-a-smart-speaker

started time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha b64dfe1c65f7e559909a11d89581357820574e3e

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha ce8758caf41a0559ead5fcc2a3042fdf78635ff6

Update README.md

view details

push time in 4 months

issue commentAkojimaSLP/Neural-mask-estimation

bug in generate_validate_data

Hi, I got the same error, did you manage to solve it? Thanks!

hadaso

comment created time in 4 months

startedfuncwj/chime4-nn-mask

started time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha edd76fd1fda7770525b034e4a121da267e2fdcbe

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 51aaab599150878f74a383a647bcfe1c017d2851

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha e94e558668f479366d805a7321cc745e8e1eab2d

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha cb6d1dd3dac74c6d0728f0e3ff6f43f9fdc7a1d9

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha d9acc95b0a68b1eff9031058c443ec7ca597ad41

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha e7621b2464d4aef0e95d729d1f851cc8bd9cf444

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha bdb29ece3e8c15277d6dbc7ade389905e1ea6ebd

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha a7f9dd23f0d4649332ad47ccc8fd74dc68f1fff3

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Jason Zhang

commit sha dfb57ff465e02c573cc01e963113e0c05626f39f

add other people's collection as submoudle

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 50ff203667a1b1d6e15444fc9a1ec58fc73af0b7

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha d5211e4edbe580cef1bd47b40201f2aff1a73c76

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 4593bc9e7f7ca6b658486ad1d320b1069445092b

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 18e5731d2ef2078f5a489e8638a52da26fd2ad1e

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 4a7d380f9bf9a61da623dee6556c03bf93fe04aa

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha ee09fd5ccf16a2bdd8545dfeeaa5ee109fd3dd25

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 26b440583d610b5d4ddcf3a368c36f9d90a3d121

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 79f87484a5421f2159e5115fba9a96118eb511f5

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 6e3b3bdaab61b4a63b0c89a81c926d654d59c0a3

Update README.md

view details

push time in 4 months

startedchenjiandongx/cutecharts

started time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha e94038707b5837062cd91aac5074f0916a41c4c3

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha eccc926bcfd2f4d00dc96123da394e0a6d7ff36b

Update README.md

view details

push time in 4 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 5cc11b68f6a38bebaacbce16fc89c5c2a9e46567

Update README.md

view details

push time in 4 months

issue openedfgnt/nn-gev

Clean speech only training

Hi, you mentioned in the paper, It's also possible to train on clean speech only. I checked the source code but couldn't find any relevant code about this. Could you point me where are you doing that? Or what's the way of doing that if the code is not included?

created time in 5 months

issue closedfgnt/nn-gev

CHiME-5 update

Hi, do you have any plans to use the CHiME-5 dataset to re-train the model and evaluate on CHiME-5?

closed time in 5 months

dreamibor

issue openedAkojimaSLP/Neural-mask-estimation

Dataset

Hi, is there a way to create the training dataset? I mean the approach that you take to get seperate speech and noise data?

created time in 5 months

startedAkojimaSLP/Neural-mask-estimation

started time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 3e976a37b740ac8f4a4c3de3296ee60da3c25459

Update README.md

view details

push time in 5 months

startediawia002/annie

started time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha dd693b1a9493dec3925a3850ea5334275e98028a

Update README.md

view details

push time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha f7aab4645f194c83cdccc7375ca3387416c18c85

Update README.md

view details

push time in 5 months

startedfgnt/nn-gev

started time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 0712adab5b1a28db116f37f20b1ed9719866fda6

Update README.md

view details

push time in 5 months

issue openedfgnt/nn-gev

CHiME-5 update

Hi, do you have any plans to use the CHiME-5 dataset to re-train the model and evaluate on CHiME-5?

created time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 78b2eeb64dff9a5c8f006879939e7869bc94fa7d

Update README.md

view details

push time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha 0e90a63b7a99f71ebf2c56328a4b6e4187187c20

Update README.md

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Jason Zhang

commit sha ad964a9b28ceb7bdfa25f9e87207238ad02257ef

add google code style guide

view details

push time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha ada47b580a87022ad69d662435a6264e7c8569b2

Update README.md

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha f7779816f8ecac8e60567f69c138e14f3825d4e8

add smallest subarray

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Jason Zhang

commit sha 0e8455c174d17ffc21c7715710a76e9f3bd6b128

add maximum of subarrays using sliding window

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha 4b9e66a373b3ad266329ce8911b7b6ca40efe4ee

Update problem_list.md

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha a68f7d40a3a8048a9547e0902825311dd4f8c094

Update problem_list.md

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha df81fd62525eb2d48bbb3c60c8fac12368ad7f3c

Update problem_list.md

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha 51d5d721383d5a8019cedb23631aed781f8857d9

Update problem_list.md

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha bd47293c214f4dbea22384bf0361757d9ee1f232

problem ticks

view details

push time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha 236a897d23d9fa3f3d76824b1c8b9a1db232f62e

add sliding window introduction example

view details

push time in 5 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha a66d86c1291368aa5113262e4723f686bc656ec0

Update README.md

view details

push time in 5 months

issue openedtensorflow/tensorflow

[tflite] Support INT8 quantization for PACK with TFLITE_BUILTINS_INT8 OpsSet

<em>Please make sure that this is a bug. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:bug_template</em>

System information

  • Have I written custom code (as opposed to using a stock example script provided in TensorFlow): Yes
  • OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Linux Ubuntu 18.04
  • TensorFlow installed from (source or binary): binary
  • TensorFlow version (use command below):1.14
  • Python version: 3.6

Describe the current behavior Similar to the UNPACK node issue in https://github.com/tensorflow/tensorflow/issues/31902, the new TFLiteConverter post-training quantisation flow, as described in https://www.tensorflow.org/lite/performance/post_training_quantization#full_integer_quantization_of_weights_and_activations, does not support quantization of PACK/STACK operation when only integer operations are requested in the output model. When such conversion is attempted the following error is reported:

RuntimeError: Quantization not yet supported for op: UNPACK

Code to reproduce the issue For example, the script below: ` import tensorflow as tf import numpy as np

def representative_dataset_gen(): input_1 = np.ones([1, 10],dtype=np.float32) input_2 = np.ones([1, 10],dtype=np.float32) for _ in range(10): yield [input_1, input_2]

tf Graph Input

foo = tf.compat.v1.placeholder("float32", [1, 10]) bar = tf.compat.v1.placeholder("float32", [1, 10]) out_stacked = tf.stack([foo, bar], axis=0)

with tf.compat.v1.Session() as sess: tf.io.write_graph(tf.compat.v1.get_default_graph(), '.','pack.pb', as_text=False)

input_name = ["Placeholder", "Placeholder_1"] output_name = ["stack"]

tflite_model_name = "int8_pack.tflite" converter = tf.lite.TFLiteConverter.from_frozen_graph("pack.pb", input_name, output_name) converter.optimizations = [tf.lite.Optimize.DEFAULT] converter.target_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8] converter.representative_dataset = representative_dataset_gen tflite_model = converter.convert() open(tflite_model_name, "wb").write(tflite_model)

Load TFLite model and allocate tensors.

interpreter = tf.lite.Interpreter(tflite_model_name) interpreter.allocate_tensors()

Get input and output tensors.

input_details = interpreter.get_input_details() output_details = interpreter.get_output_details()

Test model on random input data.

input_shape = input_details[0]['shape'] input_data = np.array(np.random.random_sample(input_shape), dtype=np.float32) interpreter.set_tensor(input_details[0]['index'], input_data) interpreter.invoke() produces errors as follows: 2019-10-21 14:02:33.682706: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA 2019-10-21 14:02:33.708278: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3408000000 Hz 2019-10-21 14:02:33.708892: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x33f5a60 executing computations on platform Host. Devices: 2019-10-21 14:02:33.708924: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (0): <undefined>, <undefined> 2019-10-21 14:02:33.717107: I tensorflow/core/grappler/devices.cc:60] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0 (Note: TensorFlow was not compiled with CUDA support) 2019-10-21 14:02:33.717228: I tensorflow/core/grappler/clusters/single_machine.cc:359] Starting new session INFO: Initialized TensorFlow Lite runtime. Traceback (most recent call last): File "pack_example.py", line 26, in <module> tflite_model = converter.convert() File "/home/jaszha02/Work/venvs/audio/lib/python3.6/site-packages/tensorflow/lite/python/lite.py", line 908, in convert inference_output_type) File "/home/jaszha02/Work/venvs/audio/lib/python3.6/site-packages/tensorflow/lite/python/lite.py", line 200, in _calibrate_quantize_model inference_output_type, allow_float) File "/home/jaszha02/Work/venvs/audio/lib/python3.6/site-packages/tensorflow/lite/python/optimize/calibrator.py", line 78, in calibrate_and_quantize np.dtype(output_type.as_numpy_dtype()).num, allow_float) File "/home/jaszha02/Work/venvs/audio/lib/python3.6/site-packages/tensorflow/lite/python/optimize/tensorflow_lite_wrap_calibration_wrapper.py", line 115, in QuantizeModel return _tensorflow_lite_wrap_calibration_wrapper.CalibrationWrapper_QuantizeModel(self, input_py_type, output_py_type, allow_float) RuntimeError: Quantization not yet supported for op: PACK ` Both kTfLiteUInt8 and kTfLiteInt8 version of the PACK operator is already implemented in TFLite (see pack.cc), so it should be straightforward to support PACK as well in the TFLite Converter.

created time in 5 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha 0baf7c958326a6fb678152ddb18c145b02b7c2b9

add dynamic array

view details

push time in 6 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha 88745ce8166d0a281839b7584d64c13f1db62835

add dynamic array

view details

push time in 6 months

startedCSTR-Edinburgh/mlpractical

started time in 6 months

push eventdreamibor/ml_resources

Zhaoyu Zhang

commit sha b821e3db4ef3f762ece72c8af927096626150b2b

Update README.md

view details

push time in 6 months

push eventdreamibor/algorithms_and_data_structures

Zhaoyu Zhang

commit sha 02774d59da340827abbb8674818cd09e26111b20

add problem list

view details

push time in 6 months

push eventdreamibor/algorithms_and_data_structures

Jason Zhang

commit sha 1052e7ef91a29d822ca31fa786e8d51bf5b836a5

add distributed system material

view details

push time in 6 months

push eventdreamibor/Python-Face-Recognition-Simple-Example

Zhaoyu Zhang

commit sha 6cc4478454ec594c17be03e051719f5babb3e325

Update README.md

view details

push time in 6 months

push eventdreamibor/algorithms_and_data_structures

Jason Zhang

commit sha cf1462f449d5d784c601c2940a9f15a1a97e066e

ignore pycharm setting folder

view details

push time in 6 months

push eventdreamibor/algorithms_and_data_structures

Jason Zhang

commit sha 272c11166cc9dc60000927b1c77b3e362193c2d1

add system design problem

view details

push time in 6 months

more