Skip to content

Releases: tinyms-ai/tinyms

v0.3.2

09 Feb 01:45
@hellowaywewe hellowaywewe
v0.3.2
88155e2
Compare
Choose a tag to compare
v0.3.2 Latest
Latest

Major Features and Improvements

  • Upgrade the version of mindspore module dependencies from v1.6.0 to the higher. #147
Assets 2
Loading

v0.3.1

28 Jan 14:17
@hellowaywewe hellowaywewe
v0.3.1
d064595
Compare
Choose a tag to compare
v0.3.1

Released 2021-01-28.

Major Features and Improvements

  • Upgrade the version of mindspore module dependencies to v1.6.0 to support running model in Linux Ubuntu 18.04, Window 10 and Mac environment. #126 #127 #136 #142
  • Updated the docker image, the quickstart doc and the quickstart_in_one_minute tutorial to users. #123 #130 #138 #140

Model Park

  • Add 1 model support: DeepFM. #125

API Change

None

Backwards Incompatible Change

None

Bug fixes

  • Fixed the bug of the bert model. #122

Contributors

Great thanks go to these wonderful people:

@kaierlong,@hellowaywewe, @hannibalhuang, @huxiaoman7, @lwdragon, @xinwenh

Assets 2
Loading
0 Join discussion

v0.3.0

31 Dec 04:37
@hellowaywewe hellowaywewe
v0.3.0
e63ee8c
Compare
Choose a tag to compare
v0.3.0

Released 2021-12-31.

Major Features and Improvements

  • Updated test use cases for test_transforms module. #109
  • Upgrade the version of mindspore module dependencies from v1.2.0 to v1.5.0. #103 #107 #118
  • Refactor the vision/transforms module to make it configure parameters flexibly. #113
  • Added the the front and back ends modules to implement the Web UI reasoning visualization feature. #104 #105 #106 #107 #108 #110 #111
  • Combined with OpenCV to add app module to support object detection visualization to make reasoning intuitive. #114
  • Added test/st/app/object_detection system test module to provide static image detection and video image detection dynamically collected by cameras. #115 #116
  • Added Nginx version container to make Web UI reasoning visualization module deploy and use easily. #120

Model Park

  • None

API Change

  • Add a new interface to get specified vision transform parameters from the transform YAML file.
v0.3.0 
from tinyms.vision.transform_config import get_specified_config

configs = get_specified_config(transforms_op='DatasetTransform', yaml_path=None)
  • Add shanshui dataset transform class to process shanshui dataset.
v0.3.0 
from tinyms.vision import ShanshuiTransform

shanshui_transform = ShanshuiTransform()
img = Image.open('shanshui_example.jpg')
img = shanshui_transform(img)

  • Add predict_web method in the serving/server module to provide back-end reasoning function for the request /predict from the web front end page.

  • Add object detection model reasoning visualization module, there are too many new methods to list one by one, please find them in API Documentation tinyms.app menu.

For the detailed API changes, please find TinyMS Python API in API Documentation.

Backwards Incompatible Change

None

Bug fixes

  • Fixed the Linux server startup error of the serving/server module in previous versions. #102

Contributors

Great thanks go to these wonderful people:

@zjuter0126, @Mickls, @leonwanghui, @hannibalhuang, @hellowaywewe, @huxiaoman7

Assets 2
Loading
0 Join discussion

v0.2.1

15 Jul 07:20
@hellowaywewe hellowaywewe
v0.2.1
760204b
Compare
Choose a tag to compare
v0.2.1

Released 2021-07-15.

Major Features and Improvements

  • Fix load_checkpoint interface bug in TinyMS 0.2.0 hub module. #96
Assets 2
Loading

v0.2.0

07 Jun 01:35
@leonwanghui leonwanghui
v0.2.0
157336a
Compare
Choose a tag to compare
v0.2.0

Released 2021-06-07.

Major Features and Improvements

  • Add text module to provide the basic dataset loading and preprocessing in NLP scenarios. #53 #73
  • Upgrade the version of mindspore module dependencies from v1.1.1 to v1.2.0. #81 #84
  • Refactor the Client and Server communication interface in serving module. #76
  • Added server_path, start FlaskServer and add host and port parameters. #77
  • Implement TinyMS hub module to enable loading lots of pre-trained models, incluidng lenet5_v1, resnet50_v1, alexnet_v1, vgg16_v1, mobilenet_v2 and ssd300_v1. #86 #93
  • Publish the TinyMS Hub contributing guidelines in public to welcome pre-trained model from the comunity. #91
  • Refactor the model network entrypoint method to provide the unified interface. #85

Model Park

  • Add 5 models support: AlexNet, DenseNet100, VGG16, SentimentNet, Bert. #59 #89 #63 #67

API Change

  • Refactor the serving entrypoint function with Client and Server class interface.
v0.1.0 v0.2.0 
from tinyms.serving import start_server, server_started, list_servables, predict, shutdown

start_server()
if server_started():
    list_servables()
    predict('example.jpg', 'servable_name', dataset_name='mnist')
shutdown()
 
from tinyms.serving import Client, Server

server = Server()
server.start_server()
client = Client()
client.list_servables()
client.predict('example.jpg', 'servable_name', dataset_name='mnist')
server.shutdown()
  • Add a new interface load in model module to support load MindIR graph directly to perform model inference operation.
v0.2.0 
>>> import tinyms as ts
>>> import tinyms.layers as layers
>>> from tinyms.model import Model, load
>>>
>>> net = layers.Conv2d(1, 1, kernel_size=3)
>>> model = Model(net)
>>> input = ts.ones([1, 1, 3, 3])
>>> model.export(input, "net", file_format="MINDIR")
...
>>> net = load("net.mindir")
>>> print(net(input))
[[[[ 0.02548009  0.04010789  0.03120251]
    [ 0.00268656  0.02353744  0.03807815]
    [-0.00896441 -0.00303641  0.01502199]]]]
  • Add hub.load method to easily load pretrained model and apply model evaluation and inference operation.
v0.2.0 
from PIL import Image
from tinyms import hub
from tinyms.vision import mnist_transform
from tinyms.model import Model

img = Image.open(img_path)
img = mnist_transform(img)

# load LeNet5 pretrained model
net= hub.load('tinyms/0.2/lenet5_v1_mnist', class_num=10)
model = Model(net)

res = model.predict(ts.expand_dims(ts.array(img), 0)).asnumpy()
print("The label is:", mnist_transform.postprocess(res))

For the detailed API changes, please find TinyMS Python API in API Documentation.

Backwards Incompatible Change

None

Bug fixes

  • Fix some bugs when serving in Windows operating system. #74
  • Set batch_norm as True by default in VGG16 to fix the converge problem of accuracy. #90

Contributors

Great thanks go to these wonderful people:

@zjuter0126, @Mickls, @leonwanghui, @hannibalhuang, @hellowaywewe, @huxiaoman7

Assets 2
Loading
0 Join discussion

v0.1.0

28 Mar 09:00
@leonwanghui leonwanghui
v0.1.0
85ec6fb
Compare
Choose a tag to compare
v0.1.0

Released 2021-03-28.

Major Features and Improvements

  • Design the overall framework of TinyMS development toolkit. #3 #5 #12 #13
  • Support install TinyMS binary in Linux Ubuntu 18.04 and Window 10 environment, also provide TinyMS docker image to users. #2 #45
  • Enable document auto-generation using Sphinx. #35
  • Provide several end to end model development and deployment tutorials for machine learning beginners. #11 #24 #26 #34
  • Set up the initial CI pipeline (including cla-assistant, GitHub Actions, readthedocs) for TinyMS project. #1 #49 #50

Model Park

  • Add 5 models support: LeNet5, ResNet50, MobileNetV2, SSD300, CycleGAN. #5 #14 #17 #32

API Change

There is no API change for the first version of TinyMS. Please find TinyMS Python API in API Documentation.

Backwards Incompatible Change

None

Bug fixes

None

Contributors

Great thanks go to these wonderful people:

@leonwanghui, @lyd911, @hannibalhuang, @hellowaywewe, @Yikun, @huxiaoman7

Assets 2
Loading