## Dynamic Model in TVM ## AWS AI Presenter: Haichen Shen, Yao Wang Amazon SageMaker Neo, Deep Engine Science ## Models with dynamism • Control flow (if, loop, etc) • Dynamic shapes ☐ Dynamic inputs: loop Limitation of TVM/graph runtime • Cannot compile and run dynamic models ## Support dynamic model in TVM • Support Any-dim in typing • Use shape function to compute the type at runtime • Virtual "data" input_shape = [tvm.relay.Any(), 3, 224, 224] dtype = "float32" block = get_model('resnet50_v1', pretrained=True) mod, params = relay.frontend.from_mxnet(block, shape={input_name:

China 2018 ## Model and Operate Datacenter by Kubernetes at eBay 辛肖刚, Cloud Engineering Manager, ebay 梅岑恺, Senior Operation Manager, ebay ## Agenda About ebay Our fleet ★ Kubernetes makes magic at at ebay Model + Controller How we model our datacenter Operation in large scale Q&A 177M Active buyers worldwide $22.7B Amount of eBay Inc. GMV $2.6B Reported revenue 1.1B Live listings • Network ## Kubernetes • Core components • Addon • Taint ## Operations ## Let's model a datacenter running Kubernetes Onboard ![Image](/uploads/documents/2/4/9/8/249845b422d764ccc2a2fa576344083d/p13_2

. . . . . . . . . . . . . . 2 1.4 The model layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.5 The view layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.6 The template layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.7 Forms . . . . . . . . . . . django-admin commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573 4.3 Writing custom model fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 578 4.4 Custom

## Microsoft Ready ## The Future of Cloud Native Applications with Open Application Model (OAM) and Dapr Mark Russinovich Chief Technology Officer, Microsoft Azure @markrussinovich ## Application models /9/bc29b5fbf1de02f8968a900b84d384e5/p4_1.jpg) Open Application Model (OAM) Open Application Model Platform agnostic application model Distributed Application Runtime (Dapr) dapr Building blocks developing cloud and edge applications Microsoft is introducing two new specs, the Open Application Model and Dapr, with the aim of making building cloud, edge and Kubernetes apps easier. ### f in ☐ ☑ ☒

## +23 ## Distributed Ranges: A Model for Building Distributed Data Structures, Algorithms, and Views ## BENJAMIN BROCK ## Notices and Disclaimers For notices, disclaimers, and details about performance

CakePHP at a Glance • Conventions Over Configuration • The Model Layer • The View Layer • The Controller Layer • CakePHP Request Cycle • Just the Start • Additional Tutorial - Creating the Database • Database Configuration • Creating our First Model • CMS Tutorial - Creating the Articles Controller • Create the Article List Template • CMS Tutorial - Creating the Database • Database Configuration • Creating our First Model • CMS Tutorial - Creating the Articles Controller • Create the Article List Template

Default: 1 ### LSTM.forward() - out, (ht, ct) = lstm(x, [ht_1, ct_1]) x: [seq, b, vec] h/c: [num_layer, b, h] out: [seq, b, h] ## ☐ ☐ ☐ 1 lstm = nn.LSTM(input_size=100, hidden_size=20, num_layers=4) LSTMCell.forward() ■ ht, ct = lstmcell(xt, [ht_1, ct_1]) xt: [b, vec] - ht/ct: [b, h] print('one layer lstm') cell = nn.LSTMCell(input_size=100, hidden_size=20) h = torch.zeros(3, 20) c = torch shape, c.shape) torch.Size([3, 20]) torch.Size([3, 20]) ## Single layer ## ☀️ ☁️ ☁️ ## Two Layers ## ☐ ☐ ☐ print('two layer lstm') cell1 = nn.LSTMCell(input_size=100, hidden_size=30) cell2

## PyTorch ## RNN Layer使用 主讲人：龙良曲 ## Folded model [batch, feature len]@[hidden len, feature len] $ ^{T} $ +[batch, hidden len]@[hidden len, hidden len] $ ^{T} $ [0,0,0 ... ] $$ x_{t}@w_{xh} + h_{t}@w_{hh} h0) x: [seq len, b, word vec] h0/ht: [num layers, b, h dim] out: [seq len, b, h dim] ## Single layer RNN ## ☐ ☐ ☐ rnn = nn.RNN(input_size=100, hidden_size=20, num_layers=1) print(rnn) x = torch.randn(10 h_{t}^{2}@w_{hh}^{2} $$ [0,0,0 ... ] $$ x_{t}@w_{xh}^{1} + h_{t}^{1}@w_{hh}^{1} $$ feature ## 2 layer RNN ## ☐ ☐ ☐ In [17]: rnn=nn.RNN(100, 10, num_layers=2) In [18]: rnn.__parameters.keys() Out[18]:

## +23 ## C++ Memory Model: from C++11 to C++23 ## ALEX DATHSKOVSKY ## 20 23 October 01 - 06 ## About Me: SPEEDATA alex.dathskovsky@speedata.io www.linkedin.com/in/alexdathskovsky https://www.cppnext [Image](/uploads/documents/a/b/8/f/ab8f331d7156e1b708cb39698c0a1f00/p3_1.jpg) ## its a conspiracy man ## Memory Model ## I mportant Question ## Does the processor executes the program you wrote? ## The Answer ## Short

Efficient Mixture-of-Experts Language Model DeepSeek-AI research@deepseek.com ## Abstract We present DeepSeek-V2, a strong Mixture-of-Experts (MoE) language model characterized by economical training DeepSeek-V2 and its chat versions still achieve top-tier performance among open-source models. The model checkpoints are available at https://github.com/deepseek-ai/DeepSeek-V2. ![Image](/uploads/docum A Contributions and Acknowledgments 27 B DeepSeek-V2-Lite: A 16B Model Equipped with MLA and DeepSeekMoE 29 B.1 Model Description ..... 29 B.2 Performance Evaluation ..... 30 C Full Formulas