tvm
tvm copied to clipboard
apache
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Hi, I’m trying to index the tensor in the following fashion: ```py x[0, mask] ``` where x is the input tensor (e.g. (2, 3) shape), while the mask is a...
Fix pylint issues in https://github.com/apache/tvm/issues/11414 List of files to complete: - [x] tests/python/frontend/caffe/test_forward.py - [x] tests/python/frontend/caffe2/model_zoo/init.py - [x] tests/python/frontend/caffe2/test_forward.py - [x] tests/python/frontend/coreml/model_zoo/init.py - [x] tests/python/frontend/coreml/test_forward.py - [x] tests/python/frontend/darknet/test_forward.py - [x]...
(This is on top of https://github.com/apache/tvm/pull/12319 - please review that one first 😸 ) This is clean up to use the new `target.features` instead of `IsaAnalyzer`.
(This is on top of #12319 - please review that one first 😸 ) Previously `CMSISNNFlags` was derived using logic specific to the external code generator, this converts the external...
In this PR problem of MXNet dot for all tensor dimensions is fixed. This PR is related to #11691 issue. Previously, it was only provided the case of 2D tensors,...
This PR makes a few code optimizations, 1. utilize the `np.cumsum` API rather than inefficient for loops 2. use `multi dot` operator to perform matrix multiplication using a single API...
sched_setaffinity should use pid_t instread of pthread_t https://man7.org/linux/man-pages/man2/sched_setaffinity.2.html cc @areusch
I encountered this when trying to run [this script](https://github.com/tlc-pack/relax/blob/relax/apps/relax_examples/e2e_auto_tir.py) over RPC on machines with v100's. Though it was done using Relax, @zxybazh says he thinks this can probably be triggered...
This re-runs the build with newly built images (so the build + test runs twice) to ensure that Docker image changes won't break anything once they're actually deployed. Testing changes...
This is a global tracking issue for landing **UMA**. The original RFC can be found [here](https://github.com/boschresearch/tvm-rfcs/blob/rfc_uma/rfcs/00xx_UMA_Unified_Modular_Accelerator_Interface.md). Please feel free to bring in more questions to the corresponding thread. To make...
