renxida

import gc
import sys
import torch
import torch_mlir

batch_size = 1
seq_len = 3
input_size = 5
hidden_size = 5
kernel_size = 3

renxida

Add
AveragePool
BatchNormalization
Cast
Clip
Concat
Constant
ConstantOfShape
Conv

renxida

// torch-mlir/.vscode/settings.json
{
    "files.associations": {
        "*.inc": "cpp",
        "ranges": "cpp",
        "regex": "cpp",
        "functional": "cpp",
        "chrono": "cpp",
        "__functional_03": "cpp",
        "target": "cpp",

renxida

OVERVIEW: MLIR modular optimizer driver

Available Dialects: builtin, chlo, complex, func, linalg, memref, ml_program, scf, sparse_tensor, stablehlo, tensor, tm_tensor, torch, torch_c, tosa, vhlo
USAGE: torch-mlir-opt [options] <input file>

OPTIONS:

Color Options:

  --color                                                  - Use colors in output (default=autodetect)

renxida

// RUN: torch-mlir-opt <%s -split-input-file -verify-diagnostics -convert-torch-onnx-to-torch
// FB OPT OPS from https://github.com/llvm/torch-mlir/issues/2689

// -----
func.func @cast_operation(%arg0: !torch.vtensor<[?,?,?,?],si64>) -> !torch.vtensor<[?,?,?,?],f32> attributes {torch.onnx_meta.ir_version = 9 : si64, torch.onnx_meta.opset_version = 18 : si64, torch.onnx_meta.producer_name = "backend-test", torch.onnx_meta.producer_version = ""} {
  // expected-error @+1 {{failed to legalize operation 'torch.operator'}}
  %208 = torch.operator "onnx.Cast"(%arg0) {
    torch.onnx.to = 1 : si64
  } : (!torch.vtensor<[?,?,?,?],si64>) -> !torch.vtensor<[?,?,?,?],f32>
  return %208 : !torch.vtensor<[?,?,?,?],f32>

renxida

/*===- TableGen'erated file -------------------------------------*- C++ -*-===*\
|*                                                                            *|
|* Op Declarations                                                            *|
|*                                                                            *|
|* Automatically generated file, do not edit!                                 *|
|* From: TorchOps.td                                                          *|
|*                                                                            *|
\*===----------------------------------------------------------------------===*/

#if defined(GET_OP_CLASSES) || defined(GET_OP_FWD_DEFINES)

renxida

#!/bin/bash
# Xida: I tested this on a fresh GCP machine. It should work on Debian and Ubuntu though.

# deps
sudo apt install wget git -y

# Specify the Miniconda version you want to install
MINICONDA_VERSION="latest"
# For a specific version, replace 'latest' with the version number, e.g., "Miniconda3-4.7.12-Linux-x86_64.sh"

renxida

From @visakanv and @billyisyoung


https://twitter.com/visakanv/status/1037754098328600576?t=mwXfUbS-Mxjr-4zODZjQNw&s=19

The courage to be disliked, Ichiro Kishimi and Fumitake Koga. Heard good things, let’s dig in

Youth tells philosopher, life is hard and complicated and full of struggle. Philosopher says life is simple. Debate ensues

renxida

#!/usr/bin/env python

"""
tokc: A Token Counter Utility

Description:
-------------
tokc is a tool designed to count tokens in text files or input streams. It's especially useful for calculating the cost of using text with various models.

Installation:

renxida

#!/bin/bash

# Install Miniconda
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
bash miniconda.sh -b -p $HOME/miniconda
rm miniconda.sh

# Add Miniconda to PATH
export PATH="$HOME/miniconda/bin:$PATH"