[TVM Discuss] [Questions] Relay graph representation to networkx and vise versa

Hi All, I am wondering if it is possible to convert the relay expressions of network graph to networks? and vise versa? The networks is here: https://networkx.github.io/ --- [Visit Topic](https://discuss.tvm.ai/t/relay-graph-representation-to-networkx-and-vise-versa/6738/1) to respond

[TVM Discuss] [Questions] Traversing Relay Graph order (source to sink, sink to source)

Hi All, I was able to traverse the graph using [ `post_order_visit` ](https://docs.tvm.ai/api/python/relay/analysis.html#tvm.relay.analysis.post_order_visit)(expr, fvisit) However, this function starts from the source of DAG. I want to traverse the DAG from the reverse order. Is there a wa

[TVM Discuss] [Questions] Why LSTM is implemented repeatedly for every front-end?

Hi, I notice that LSTM is implemented repeatedly for every front-end(Keras, ONNX, etc) and the implementations look similar to each other. I'm just wondering if we can make LSTM an Op in relay, or at least use unified implementation across different front-ends? I understand that different fro

[TVM Discuss] [Questions] Java bindings for AST

Hiya. I'm thinking about putting together a front-end in Scala. Is there a JVM AST binding planned? I see the JVM binding there, but looks like it's only for runtime. --- [Visit Topic](https://discuss.tvm.ai/t/java-bindings-for-ast/6735/1) to respond. You are receiving this because you en

[TVM Discuss] [Questions] Arch(sm_xy) is not passed, and we cannot detect it from env

See https://discuss.tvm.ai/t/solved-compile-error-related-to-autotvm/804/11?u=gasgallo --- [Visit Topic](https://discuss.tvm.ai/t/arch-sm-xy-is-not-passed-and-we-cannot-detect-it-from-env/4490/2) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe fr

[TVM Discuss] [Questions] CUDA -libs=cudnn performance

conv2d, conv3d and softmax. --- [Visit Topic](https://discuss.tvm.ai/t/cuda-libs-cudnn-performance/6700/4) 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/03c8b3ed49200976017

[TVM Discuss] [Questions] Can not build android_rpc

You can use this way to avoid it: in tvm/apps/android_rpc/app/src/main/jni/make/config.mk modify : APP_ABI = all to APP_ABI = armeabi-v7a #choise your android abi in (armeabi-v7a arm64-v8a x86 x86_64 mips) And at the last line: ADD_LDLIBS = /home/tvm/apps/android_rpc/app/src/main/jni/libtvm_r

[TVM Discuss] [Questions] CUDA -libs=cudnn performance

Yes, thank you! Which ops do we offload to CUDNN then? --- [Visit Topic](https://discuss.tvm.ai/t/cuda-libs-cudnn-performance/6700/3) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.ai/email/unsu

[TVM Discuss] [Questions] CUDA -libs=cudnn performance

conv_transpose won't be run on cudnn even if you specify `-libs=cudnn`. Does this answer your question? --- [Visit Topic](https://discuss.tvm.ai/t/cuda-libs-cudnn-performance/6700/2) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these email

[TVM Discuss] [Questions] MTCNN port on tvm

Have look at: [mtcnn on tvm](https://github.com/thongdk8/realtime-facereccpp) --- [Visit Topic](https://discuss.tvm.ai/t/mtcnn-port-on-tvm/469/3) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.ai

[TVM Discuss] [Questions] Issue with static tensor array

Thank you. Looking forward to it. --- [Visit Topic](https://discuss.tvm.ai/t/issue-with-static-tensor-array/6333/13) 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/3225551da