[TVM Discuss] [Questions] Incorrect generated function after PartitionGraph pass

Wow, perfect timing! Thanks :) --- [Visit Topic](https://discuss.tvm.ai/t/incorrect-generated-function-after-partitiongraph-pass/6380/3) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.ai/email/

[TVM Discuss] [Questions] Incorrect generated function after PartitionGraph pass

Hi @zhiics @comaniac, I am using BYOC to offload transformers to external codegen tools. These transformers are composite functions. I had been using this feature well with my manually-generated annotation passes, but when I merge the latest changes to go through the `AnnotateGraph -> Partiti

[TVM Discuss] [Questions] ONNX type mismatch when building with opt level 1

Hey all, I am working on a model that is written in PyTorch and exported to ONNX. During `relay.build` with opt level = 1, I ran into a type mismatch. The error does not occur when opt level = 0: ``` TypeError: Check failed: a.dtype() == b.dtype(): mismatched types Error during compile functi

[TVM Discuss] [Questions] CUDA FP16 example

Thanks a lot. I've been playing around with this on a BERT model, but I'm hitting some issues when calling `relay.build` with opt level 3. The target is `cuda`. The error message looks like this: ``` unresolved intrinsic sqrt with return type float16x4 ``` It comes from `codegen_c.cc`. Does t

[TVM Discuss] [Questions] CUDA FP16 example

I am trying to find the actual implementation of `Cast` for each device, but am having trouble finding it. @jwfromm do you know where it is? --- [Visit Topic](https://discuss.tvm.ai/t/cuda-fp16-example/6190/8) to respond. You are receiving this because you enabled mailing list mode. To un

[TVM Discuss] [Questions] CUDA FP16 example

Awesome, thanks a lot @jwfromm! Do you have any experience in how this impacts accuracy? For example, I know that CUDA's `__float2half` function has a decent amount of logic. --- [Visit Topic](https://discuss.tvm.ai/t/cuda-fp16-example/6190/6) to respond. You are receiving this because yo

[TVM Discuss] [Questions] CUDA FP16 example

Got it. Is there any plan to do this in the future? --- [Visit Topic](https://discuss.tvm.ai/t/cuda-fp16-example/6190/4) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.ai/email/unsubscribe/e185d7

[TVM Discuss] [Questions] CUDA FP16 example

Is converting a model to FP16 with target = "cuda" supported? If so, is there an example pass I could look at to convert my model? cc @vinx13 @Hzfengsy Thanks! --- [Visit Topic](https://discuss.tvm.ai/t/cuda-fp16-example/6190/1) to respond. You are receiving this because you enabled ma

[TVM Discuss] [Questions] External codegen with CUDA target

Awesome, thanks a lot @trevor-m. One more quick question before I try it out - what data type is DLTensor->data? The `codegen_c` base casts it to the type that the argument to the function is (in my case, input is a `float*` and input_mask is an `int*`). --- [Visit Topic](https://discuss

[TVM Discuss] [Questions] External codegen with CUDA target

Sorry about that, I think I misspoke. I already have the annotation pass set up properly and my codegen is being called. However, when I try to print out one of my inputs from my codegen, the program crashes. I have a feeling that since the target is “cuda”, the data isn’t being moved from G

[TVM Discuss] [Questions] External codegen with CUDA target

Hey @zhiics and @comaniac, I am working on an external codegen that will run on GPU. My external codegen module is a CSourceModule. The code generated in this module will call some CUDA APIs. If I go through the external codegen workflow and set the target to `cuda -libs=cublas,cudnn`, will

[TVM Discuss] [Questions] How to add floormod in relay and topi

@tqchen can you please update on the status of this? --- [Visit Topic](http://tracking.discuss.tvm.ai/tracking/click?d=1gm_cre9B1NJEXcmlvQA2cc9v5uL6uInipbRnWYa6Y8fWGVIsVdDlo2-fQuk0w0QE3rn8OfNFrF858EfaEbQULg44E_C8qCYXCSwqrPALuh8kt3txwaQJXGqgRxBinb5oxMCcEwz5aH23zNq_Y6Sxj-Sm5z5hkfagoCOd2gxFNoF

[TVM Discuss] [Questions] Tutorial for enabling auto-tuning on new op

Thank you both very much for the responses! They are really helpful. I'll work on preparing the PR for adding this tunable template and ping back on here if I have any more questions. --- [Visit Topic](http://tracking.discuss.tvm.ai/tracking/click?d=qI4-0aU7EDSF9uQ4IHPCfh6lQluC2t3vh3FXTdI

[TVM Discuss] [Questions] How to create a generic compute and override per target without autotvm?

I am trying to call cblas libraries when "cblas" is specified in the target libs and the target is x86. This needs to happen in two places: dense and batch_matmul. Dense is straightforward because x86 already has an overridden compute and schedule. However, batch_matmul only has an overridden

[TVM Discuss] [Questions] Regarding ONNX CUDA support

ONNX models get converted to Relay IR, which gets codegenned using Topi operator implementations, which support CUDA. Yes, a model coming in from ONNX should support CUDA. What error are you seeing? --- [Visit Topic](https://discuss.tvm.ai/t/regarding-onnx-cuda-support/3723/2) to respond