tvm

tvm copied to clipboard

[CI Problem] TVMError: Check failed: ret == 0 (-1 vs. 0)

1

I'm now using some script in Esp-net (https://github.com/espnet/espnet) based on tvm. I'm using virtual environment in conda with Python 3.7 and Cuda 10.0 However, while running the script, I received...

chlorane

[Relax] Implement Function.check_for_special_case

3

If a dynamic model is frequently called with specific arguments or shapes of arguments, performance may be improved by generating to specialized versions of the model. Previously, specialized versions of...

Lunderberg

This example demonstrates the process of building and executing a Keyword Spotting (KWS) model using TVM and ESP-IDF Python tools specifically for the Seeed Studio XIAO ESP32-C3 board with an...

Aleksei-grovety

[DEV][Announcement] Main now transitions to Unity

2

Dear community: Following the vote to [transition main to unity branch](https://github.com/apache/tvm/issues/16434). We are working to transition main to unity. I am happy to announce that as of now, the main...

tqchen

[SVE] Implement scalable vectors in TVM

2

This prototype is to accompany the open [SVE RFC](https://github.com/apache/tvm-rfcs/pull/104). It implements the design outlined in the RFC. The main changes to the stack include: 1. `tir.split` can accept an expression...

ekalda

How to use slicing of data for NDArray structure?

1

How to get something similar to output[:,0,:] if output is of type of NDArray? I am using for example int64_t X = logits_on_device->shape[0]; int64_t Y = logits_on_device->shape[1]; int64_t Z =...

musram

[COLLAGE][ADRENO] Enable adreno texture and benchmark

1

1. Enable adreno texture in collage 2. Integrate implicit call of BYOC preprocessing module into collage tunning module 3. Enable benchmark script for adreno targets. Benchmark results: Networks | OpenCL...

krishnaraj36

[OpenCL][Texture] Improved texture memory planning

12

Motivated form the fact that textures can be allocated over a clBuffer object and the size of backing clBuffer can be computed based on hardware image pitch alignment. This optimizes...

srkreddy1238

[Unity][Tracking Issue] In-place operations

3

Per the [discussion on in-place updates](https://discuss.tvm.apache.org/t/discuss-inplace-update-in-dataflow-block/14669), this is a tracking issue to discuss the steps and implementation details. - [x] Implement a basic version of the `call_tir_inplace` operator. This will...

slyubomirsky

[apps][bundle_deploy][RISC-V] Extend bundle build for RISC-V

3

Add cross-compilation options and example to build and run for qemu-riscv64

Aleksei-grovety