dragen1860

import torch
import torch.nn as nn
from torch.autograd import Variable
import numpy as np
import torch.utils.data as data_utils
import pandas as pd
import os
from sklearn.preprocessing import MinMaxScaler
from sklearn.model_selection import train_test_split
import pickle as pkl

dragen1860

name: "ResNet-101"
input: "data"
input_dim: 1
input_dim: 3
input_dim: 224
input_dim: 224

layer {
	bottom: "data"
	top: "conv1"

dragen1860

1. improve CSML-TF version and find it hard to debug
2. re-develop CSML-Pytorch
3. follow latest reasoning paper: Compositional Attention Networks for Machine Reasoning

Next Week:
1. work on CSML-Pytorch
2. rethink the machanism of machine reasoning
3. prepare for meta-learning presentation

dragen1860

1. Finish debug of csml-pytorch
2. meta-learning review
3. papers on memory relative topic

This week
1. NIPS

dragen1860

MOV SI, Table_A     ;POINTER TO TABLE_A.
MOV DI, Table_B     ;POINTER TO TABLE_B.
MOV CX, 3           ;ARRAY LENGTH.
REPEAT:       
    MOV AL, [SI]
    MOV [DI], AL
    INC SI
    INC DI
    LOOP REPEAT     ;CX-1. IF CX>0 JUMP TO REPEAT.

dragen1860

1. Presentation
2. re-run the code on memory related papers and understand its limitations
3. propose an idea based on memory: learning to rethink

Next:
1. research on the new idea

dragen1860

bool copy_(void* src, void* dest)
{
  void* psrc = src;
  void* pdest = dest;
  while(*psrc!=0)
  {
    *pdest = *psrc;
    psrc++;
    pdest++;
  }