Pavan Yalamanchili pavanky

## broadcast_test.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                pavanky
                / broadcast_test.ipynb
            
            
              Created
              September 23, 2021 23:17
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## cpu_gpu_copy.py
import torch
import numpy as np

import time

elapsed = 0


for _ in range(10):
  a1 = np.random.random((8192, 8192)).astype(np.float32)

## gpu_xgboost

xgb_params = {
  "booster": "gbtree",
  "objective": "binary:logistic",
  "eta": 0.1,
  "gamma": 0,
  "min_child_weight": 200,
  "max_depth": 6,
  "eval_train": 1,
  "tree_method": "gpu_hist",

## recusrive_mean.cpp
#include <vector>
#include <utility>
#include <cstdio>


std::pair<float, int> recursive_mean(
    const std::vector<float> &vec,
    const int first,
    const int last) {

## dummy_keras_estimator.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                pavanky
                / dummy_keras_estimator.ipynb
            
            
              Created
              April 28, 2020 23:56
            
              
                dummy_keras_estimator.ipynb
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## dummy_estimator_model.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                pavanky
                / dummy_estimator_model.ipynb
            
            
              Created
              April 28, 2020 22:40
            
              
                dummy_estimator_model.ipynb
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## dummy_keras_model.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                pavanky
                / dummy_keras_model.ipynb
            
            
              Created
              April 28, 2020 22:40
            
              
                dummy_keras_model.ipynb
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## Dockerfile
FROM centos:7

RUN yum makecache
RUN yum install -y centos-release-SCL which yum-utils centos-release-scl-rh epel-release
RUN yum install -y  python27  \
		python27-python-devel \
		python27-python-wheel \
                java-1.8.0-openjdk-devel \
		devtoolset-8-toolchain \
		devtoolset-6-toolchain \

## TfLoader.scala
package com.twitter.sample.lib

import java.io.File
import org.apache.commons.io.{FileUtils, IOUtils}
import org.tensorflow.TensorFlow
import com.twitter.logging.Logger
import java.nio.file.{Files, Paths}

object TfLoader {

## sparse_model.py
import numpy as np
import tensorflow.compat.v2 as tf

class Sparse(tf.keras.layers.Dense):
  def call(self, inputs):
    outputs = tf.sparse.sparse_dense_matmul(inputs, self.kernel)
    if self.use_bias:
      outputs = tf.nn.bias_add(outputs, self.bias)
    return outputs
	import torch
	import numpy as np

	import time

	elapsed = 0


	for _ in range(10):
	a1 = np.random.random((8192, 8192)).astype(np.float32)

	xgb_params = {
	"booster": "gbtree",
	"objective": "binary:logistic",
	"eta": 0.1,
	"gamma": 0,
	"min_child_weight": 200,
	"max_depth": 6,
	"eval_train": 1,
	"tree_method": "gpu_hist",
	#include <vector>
	#include <utility>
	#include <cstdio>


	std::pair<float, int> recursive_mean(
	const std::vector<float> &vec,
	const int first,
	const int last) {
	FROM centos:7

	RUN yum makecache
	RUN yum install -y centos-release-SCL which yum-utils centos-release-scl-rh epel-release
	RUN yum install -y python27 \
	python27-python-devel \
	python27-python-wheel \
	java-1.8.0-openjdk-devel \
	devtoolset-8-toolchain \
	devtoolset-6-toolchain \
	package com.twitter.sample.lib

	import java.io.File
	import org.apache.commons.io.{FileUtils, IOUtils}
	import org.tensorflow.TensorFlow
	import com.twitter.logging.Logger
	import java.nio.file.{Files, Paths}

	object TfLoader {
	import numpy as np
	import tensorflow.compat.v2 as tf

	class Sparse(tf.keras.layers.Dense):
	def call(self, inputs):
	outputs = tf.sparse.sparse_dense_matmul(inputs, self.kernel)
	if self.use_bias:
	outputs = tf.nn.bias_add(outputs, self.bias)
	return outputs