常連

環境
window10 home
python3
アナコンダ3のジュピターノートブックを使用

pythonでpandasを使ってエクセル(CSV)ファイルを読み込んだ後そのファイルの
最初の行の日時・始値列内の最初の行の値・高値列内の最大値・安値列内の最小値・終値列内の最後の行の値
を取り出してcsvファイルを作り出し保存するプログラムを作ったのですが
何故か最初に書いた4行だけ保存されていません

```python
import pandas as pd

#csvを読み込ませる
a=pd.read_csv(r"Desktop\FX\原本\ユロドル\08\EURUSD_20080103.csv",encoding="shift-jis")

#作ったリストをhour_barに入れデータフレームにしてファイルを保存
cols=["日時","Open","High","Low","Close"]
hour_bar=pd.DataFrame(index=[], columns=cols)
hour_bar["日時"]=pd.to_datetime(str(a["日時"][0])[:8])
hour_bar["Open"]=a["始値"][0]
hour_bar["High"]=a["高値"].max()
hour_bar["Low"]=a["安値"].min()
hour_bar["Close"]=a["終値"].tail(1)
print(hour_bar)
hour_bar.to_csv(r"Desktop\FX\テスト足\08年日足\20080103.csv", encoding="shift-jis", index=False)
```

これを実行するとprint(hour_bar)の部分で確認できますが
Closeの終値のみが代入されていて他の日時・open・High・Lowは何故か代入されていません
更に不可解なのがこの代入作業を下記の様に2回連続でやらせると普通に書き込まれています


```python
import pandas as pd

#csvを読み込ませる
a=pd.read_csv(r"Desktop\FX\原本\ユロドル\08\EURUSD_20080103.csv",encoding="shift-jis")

#作ったリストをhour_barに入れデータフレームにしてファイルを保存
cols=["日時","Open","High","Low","Close"]
hour_bar=pd.DataFrame(index=[], columns=cols)
hour_bar["日時"]=pd.to_datetime(str(a["日時"][0])[:8])
hour_bar["Open"]=a["始値"][0]
hour_bar["High"]=a["高値"].max()
hour_bar["Low"]=a["安値"].min()
hour_bar["Close"]=a["終値"].tail(1)
hour_bar["日時"]=pd.to_datetime(str(a["日時"][0])[:8])
hour_bar["Open"]=a["始値"][0]
hour_bar["High"]=a["高値"].max()
hour_bar["Low"]=a["安値"].min()
hour_bar["Close"]=a["終値"].tail(1)
print(hour_bar)
hour_bar.to_csv(r"Desktop\FX\テスト足\08年日足\20080103.csv", encoding="shift-jis", index=False)
```

これは一体どういう事なのでしょうか？
※読み込んだCSVファイルは一部省略してますが下記です


|日時	始値	高値	安値	終値
20080103070000 	1.4711	1.4714	1.4711	1.4713
20080103070100 	1.4713	1.4714	1.4712	1.4712
20080103070200 	1.4712	1.4714	1.471	1.4714
20080103070300 	1.4714	1.4715	1.4714	1.4714
20080103070400 	1.4714	1.4715	1.4713	1.4714
20080103070500 	1.4714	1.4715	1.4713	1.4713
20080103070600 	1.4713	1.4713	1.4711	1.4712
20080103070700 	1.4712	1.4713	1.4712	1.4712
20080103070800 	1.4712	1.4713	1.4712	1.4713
20080103070900 	1.4713	1.4715	1.4712	1.4715
20080103071000 	1.4716	1.472	1.4716	1.4719
20080103071100 	1.4719	1.4719	1.4718	1.4718
20080103071200 	1.4718	1.4719	1.4715	1.4718
20080103071300 	1.4717	1.4718	1.4715	1.4715
20080103071400 	1.4715	1.4716	1.4711	1.4713
20080103071500 	1.4713	1.4713	1.471	1.4711
20080103071600 	1.4711	1.4712	1.4711	1.4711
20080103071700 	1.4711	1.4713	1.4711	1.4712
20080103071800 	1.4713	1.4713	1.4713	1.4713
20080103071900 	1.4713	1.4716	1.4713	1.4714
20080103072000 	1.4714	1.4715	1.4712	1.4712
20080103072100 	1.4712	1.4712	1.4711	1.4712
20080103072200 	1.4712	1.4712	1.471	1.4711
20080103072300 	1.4711	1.4713	1.4711	1.4713
20080103072400 	1.4713	1.4713	1.471	1.4712
20080103072500 	1.4712	1.4714	1.4712	1.4714
20080103072600 	1.4714	1.4714	1.4714	1.4714
20080103072700 	1.4714	1.4714	1.4714	1.4714
20080103072800 	1.4714	1.4714	1.4714	1.4714
20080103072900 	1.4714	1.4714	1.4714	1.4714
20080103073000 	1.4714	1.4714	1.4713	1.4713
20080103073100 	1.4713	1.4714	1.4711	1.4714
20080103073200 	1.4714	1.4716	1.4711	1.4716
20080103073300 	1.4716	1.472	1.4716	1.4717
20080103073400 	1.4717	1.4722	1.4717	1.4717
20080103073500 	1.4717	1.4722	1.4717	1.4722
20080103073600 	1.4722	1.4722	1.4721	1.4721
20080103073700 	1.4721	1.4721	1.4717	1.4719
20080103073800 	1.4719	1.472	1.4717	1.4719
20080103073900 	1.4719	1.4719	1.4714	1.4715
20080103074000 	1.4715	1.4715	1.4715	1.4715
20080103074100 	1.4716	1.4716	1.4716	1.4716
20080103074200 	1.4716	1.4716	1.4714	1.4715
20080103074300 	1.4715	1.4716	1.4715	1.4716
20080103074400 	1.4716	1.4717	1.4716	1.4716
20080103074500 	1.4716	1.4717	1.4716	1.4717
20080103074600 	1.4717	1.4717	1.4717	1.4717
20080103074700 	1.4717	1.4717	1.4717	1.4717
20080103074800 	1.4717	1.4717	1.4717	1.4717
20080103074900 	1.4717	1.4717	1.4717	1.4717
20080103075000 	1.4717	1.4718	1.4717	1.4717
20080103075100 	1.4717	1.472	1.4717	1.4719
20080103075200 	1.4719	1.4719	1.4718	1.4718
20080103075300 	1.4719	1.4719	1.4719	1.4719
20080103075400 	1.4719	1.4719	1.4719	1.4719
20080103075500 	1.4719	1.4719	1.4719	1.4719
20080103075600 	1.4719	1.4721	1.4719	1.4719
20080103075700 	1.472	1.4721	1.4719	1.472
20080103075800 	1.472	1.472	1.472	1.472
20080103075900 	1.472	1.472	1.472	1.472
20080103080000 	1.472	1.4721	1.472	1.4721
20080103080100 	1.4721	1.4723	1.4721	1.4723
20080103080200 	1.4722	1.4722	1.4721	1.4722
20080103080300 	1.4722	1.4722	1.4721	1.4721
20080103080400 	1.4721	1.4721	1.4721	1.4721
20080103080500 	1.4721	1.4721	1.4721	1.4721
20080103080600 	1.4721	1.4721	1.472	1.472
20080103080700 	1.472	1.4721	1.472	1.4721
20080103080800 	1.4721	1.4722	1.4721	1.4721
20080103080900 	1.4721	1.4721	1.4719	1.472
20080103081000 	1.472	1.472	1.4719	1.472
20080103081100 	1.472	1.472	1.4719	1.4719
20080103081200 	1.4719	1.4721	1.4719	1.472
20080103081300 	1.472	1.4721	1.472	1.4721
20080103081400 	1.4721	1.4721	1.472	1.472
20080103081500 	1.472	1.472	1.472	1.472
20080103081600 	1.472	1.4722	1.472	1.472
20080103081700 	1.472	1.4723	1.472	1.4723
20080103081800 	1.4723	1.4725	1.4723	1.4723
20080103081900 	1.4723	1.4723	1.4719	1.472
20080103082000 	1.472	1.4721	1.472	1.4721
20080103082100 	1.4721	1.4721	1.472	1.4721
20080103082200 	1.4721	1.4723	1.4721	1.4721
20080103082300 	1.4721	1.4722	1.4721	1.4721
20080103082400 	1.4722	1.4722	1.4721	1.4722
20080103082500 	1.4722	1.4722	1.472	1.4721
20080103082600 	1.4721	1.4721	1.4721	1.4721
20080103082700 	1.4721	1.4721	1.472	1.4721
20080103082800 	1.4721	1.4721	1.4721	1.4721
20080103082900 	1.4721	1.4724	1.4721	1.4722
20080103083000 	1.4722	1.4724	1.4722	1.4723
20080103083100 	1.4723	1.4723	1.4722	1.4722
20080103083200 	1.4722	1.4723	1.4722	1.4723
20080103083300 	1.4723	1.4724	1.4721	1.4722
20080103083400 	1.4722	1.4723	1.4722	1.4722
20080103083500 	1.4722	1.4723	1.472	1.4722
20080103083600 	1.4722	1.4722	1.4721	1.4721
20080103083700 	1.4721	1.4722	1.472	1.4721
20080103083800 	1.4721	1.4721	1.4716	1.4717
20080103083900 	1.4717	1.4717	1.4714	1.4716
20080103084000 	1.4716	1.4719	1.4716	1.4718
20080103084100 	1.4718	1.4718	1.4716	1.4716
20080103084200 	1.4716	1.4716	1.4711	1.4713
20080103084300 	1.4713	1.4715	1.4712	1.4714
20080103084400 	1.4714	1.4714	1.4711	1.4711
20080103084500 	1.4711	1.4713	1.4711	1.4713
20080103084600 	1.4713	1.4714	1.4711	1.4714
20080103084700 	1.4714	1.4716	1.4712	1.4716
20080103084800 	1.4716	1.4718	1.4716	1.4717
20080103084900 	1.4717	1.472	1.4717	1.4719
20080103085000 	1.4719	1.472	1.4716	1.4717
20080103085100 	1.4717	1.4719	1.4717	1.4719
20080103085200 	1.4719	1.4719	1.4717	1.4719
20080103085300 	1.4719	1.4721	1.4718	1.4719
20080103085400 	1.4719	1.4719	1.4718	1.4719
20080103085500 	1.4719	1.4722	1.4719	1.4722
20080103085600 	1.4722	1.4722	1.4718	1.4718
20080103085700 	1.4717	1.4717	1.4714	1.4714
20080103085800 	1.4714	1.4715	1.4712	1.4714
20080103085900 	1.4714	1.4715	1.4714	1.4715
20080103090000 	1.4715	1.4716	1.4714	1.4714
----------------------------------------------
20080104060100 	1.4749	1.4749	1.4749	1.4749
20080104060200 	1.4749	1.4749	1.4748	1.4749
20080104060300 	1.4749	1.475	1.4749	1.475
20080104060400 	1.475	1.475	1.4749	1.4749
20080104060500 	1.4749	1.475	1.4749	1.475
20080104060600 	1.475	1.475	1.4749	1.475
20080104060700 	1.475	1.475	1.4748	1.4749
20080104060800 	1.4749	1.4749	1.4748	1.4748
20080104060900 	1.4748	1.4748	1.4747	1.4747
20080104061000 	1.4747	1.4748	1.4746	1.4746
20080104061100 	1.4746	1.4746	1.4743	1.4743
20080104061200 	1.4743	1.4745	1.4743	1.4745
20080104061300 	1.4745	1.4746	1.4744	1.4746
20080104061400 	1.4746	1.4746	1.4745	1.4745
20080104061500 	1.4746	1.4746	1.4745	1.4746
20080104061600 	1.4746	1.4747	1.4745	1.4747
20080104061700 	1.4747	1.4747	1.4746	1.4746
20080104061800 	1.4746	1.4746	1.4745	1.4745
20080104061900 	1.4745	1.4747	1.4745	1.4745
20080104062000 	1.4745	1.4746	1.4745	1.4745
20080104062100 	1.4745	1.4747	1.4745	1.4747
20080104062200 	1.4747	1.4747	1.4746	1.4747
20080104062300 	1.4747	1.4747	1.4747	1.4747
20080104062400 	1.4747	1.4747	1.4747	1.4747
20080104062500 	1.4747	1.4747	1.4745	1.4745
20080104062600 	1.4745	1.4745	1.4744	1.4744
20080104062700 	1.4744	1.4744	1.4742	1.4743
20080104062800 	1.4743	1.4743	1.4743	1.4743
20080104062900 	1.4743	1.4744	1.4743	1.4743
20080104063000 	1.4743	1.4744	1.4743	1.4744
20080104063100 	1.4743	1.4743	1.4743	1.4743
20080104063200 	1.4743	1.4743	1.4743	1.4743
20080104063300 	1.4743	1.4744	1.4743	1.4743
20080104063400 	1.4744	1.4744	1.4743	1.4743
20080104063500 	1.4743	1.4744	1.4743	1.4744
20080104063600 	1.4744	1.4745	1.4744	1.4745
20080104063700 	1.4745	1.4745	1.4745	1.4745
20080104063800 	1.4745	1.4745	1.4745	1.4745
20080104063900 	1.4745	1.4745	1.4744	1.4745
20080104064000 	1.4745	1.4746	1.4745	1.4746
20080104064100 	1.4746	1.4746	1.4745	1.4745
20080104064200 	1.4745	1.4745	1.4745	1.4745
20080104064300 	1.4745	1.4745	1.4745	1.4745
20080104064400 	1.4745	1.4746	1.4743	1.4744
20080104064500 	1.4744	1.4744	1.4743	1.4743
20080104064600 	1.4743	1.4744	1.4743	1.4743
20080104064700 	1.4743	1.4745	1.4743	1.4744
20080104064800 	1.4744	1.4745	1.4744	1.4745
20080104064900 	1.4745	1.4746	1.4745	1.4746
20080104065000 	1.4746	1.4746	1.4744	1.4744
20080104065100 	1.4744	1.4745	1.4743	1.4744
20080104065200 	1.4744	1.4745	1.4744	1.4744
20080104065300 	1.4744	1.4746	1.4743	1.4745
20080104065400 	1.4745	1.4745	1.4743	1.4744
20080104065500 	1.4745	1.4746	1.4744	1.4746
20080104065600 	1.4746	1.4746	1.4745	1.4746
20080104065700 	1.4746	1.4746	1.4744	1.4745
20080104065800 	1.4745	1.4746	1.4744	1.4744
20080104065900 	1.4744	1.4748	1.4744	1.4748
||||

python pandasでデータフレームの書き込みを何故か2回行わないと書き込まれない

k近傍法において、最近傍点を増やしていくとモデルがシンプルになるとはなぜですか？
感覚的にはより多くの点を考慮しているので複雑になっていると思ってしまいます

k近傍法におけるモデルがシンプルとは

提示コードですがmakeを実行すると以下のエラーが出ますこれはなぜでしょうか？参考サイトを参考にライブラリの順序を確認しまたが正しいです。
`adding symbols: DSO missing from command line collect2: error: ld returned 1 exit status`とはどういう意味なのでしょうか？

# 試したこと
参考サイトを参考にライブラリの順序を確認



参考サイト：[https://askubuntu.com/questions/1098073/g-fails-with-an-undefined-reference-to-symbol-dlcloseglibc-2-2-5-even-with](https://askubuntu.com/questions/1098073/g-fails-with-an-undefined-reference-to-symbol-dlcloseglibc-2-2-5-even-with)
参考サイト：[https://discourse.panda3d.org/t/solved-undefined-ref-pthread-getspecific-glibc-2/15283/2](https://discourse.panda3d.org/t/solved-undefined-ref-pthread-getspecific-glibc-2/15283/2)
参考サイト：[https://stackoverflow.com/questions/45135/why-does-the-order-in-which-libraries-are-linked-sometimes-cause-errors-in-gcc](https://stackoverflow.com/questions/45135/why-does-the-order-in-which-libraries-are-linked-sometimes-cause-errors-in-gcc)


```
$ make
g++ obj/Ellipse.o obj/Line.o obj/Transform_2D.o obj/Circle.o obj/Rectangle.o obj/Triangle.o obj/Window.o obj/Camera.o obj/Main.o obj/Init.o obj/Resource.o obj/Text.o obj/Shader.o obj/Help.o obj/Point.o obj/Render.o obj/FrameWork.o obj/Simple_Draw.o -o GL -lGL -lGLEW -lglfw3 -L/usr/local/lib -lfreetype 
/usr/bin/ld: /usr/local/lib/libglfw3.a(posix_thread.c.o): undefined reference to symbol 'pthread_getspecific@@GLIBC_2.2.5'
/usr/bin/ld: /lib/x86_64-linux-gnu/libpthread.so.0: error adding symbols: DSO missing from command line
collect2: error: ld returned 1 exit status
make: *** [Makefile:14: GL] エラー 1
```

```
PRG         :=GL
SRC_DIR     :=source
OBJ_DIR     :=obj
DEP_DIR     :=obj
DEP        	:=$(wildcard $(DEP_DIR)/*.d)
SRC         :=$(wildcard $(SRC_DIR)/*.cpp)
OBJ         :=$(addprefix $(OBJ_DIR)/,$(patsubst %.cpp,%.o,$(notdir $(SRC))))
LDFLAGS 	:=`pkg-config --libs=opengl glew glfw3 freetype2`



$(PRG): $(OBJ)
    #$(CXX)  $^ -l ncurses -o $@
	$(CXX) $^ -o $@ -lGL -lGLEW -lglfw3 -L/usr/local/lib -lfreetype 


$(OBJ_DIR)/%.o: source/%.cpp
	#$(CXX) -std=c++2a -I/usr/local/include/freetype2 -I/usr/include/libpng16 -c -MMD -MP $< -o $@

	$(CXX) -std=c++2a -c -MMD -MP $< -o $@
	-include $(DEP)
	

clean:
	rm -f ./$(OBJ_DIR)/*.o *.out ./$(OBJ_DIR)/*.d *.a $(PRG)
```

「adding symbols: DSO missing from command line」 の原因が知りたい。

### 前提・実現したいこと

Aizu Online Judgeのアルゴリズムの問題で、与えられた数字をマージソートで昇順に並べ替えようとしています。
問題文
「n個の整数を含む数列Sを上の疑似コードに従ったマージソートで昇順に整列するプログラムを作成してください。また、mergeにおける比較回数の総数を報告してください。

入力
１行目にn、２行目にSを表すn個の整数が与えられます。

出力
１行目に整列済みの数列Sを出力してください。数列の隣り合う要素は１つの空白で区切ってください。２行目に比較回数を出力してください。」
url(https://onlinejudge.u-aizu.ac.jp/courses/lesson/1/ALDS1/5/ALDS1_5_B)

### 発生している問題・エラーメッセージ
要素数が少ない場合は問題なく機能するのですが、最後の500000個の数字が与えられた場合のみ以下のエラーが発生してしまいます。

```
timeout: the monitored command dumped core 0.14user 0.00system 0:00.34elapsed 42%CPU (0avgtext+0avgdata 5060maxresident)k 0inputs+8outputs (0major+784minor)pagefaults 0swaps
```

### 該当のソースコード

```C++
#include <bits/stdc++.h>
#define INF 10000000
using namespace std;
void marge(vector<int>&S,int left, int right, int mid, int &cnt){
  int n1 = mid - left;
  int n2 = right - mid;
  vector<int> L(n1+1),R(n2+1);
  for(int i=0;i<n1;i++)L[i]=S[left+i];
  for(int i=0;i<n2;i++)R[i]=S[mid+i];
  int i=0,j=0;
  L[n1]=INF,R[n2]=INF;
  for(int k=left; k<right; k++){
    if(L[i] <= R[j]){
      cnt ++ ;
      S[k]=L[i];
      i++;
    }
    else{
      cnt ++ ;
      S[k]=R[j];
      j++;
    }
  }
}

void margeSort(vector<int>&S,int left, int right, int &cnt){
  if(left+1<right){
    int mid=(left+right)/2;
    margeSort(S,left,mid,cnt);
    margeSort(S,mid,right,cnt);
    marge(S,left,right,mid,cnt);
  }
  return;
}
      
int main(){
  int count=0,n;
  cin >> n;
  vector<int> S(n);
  for(int i=0;i<n;i++)cin >> S[i];
  margeSort(S, 0, n, count);
  for(int i=0;i<n;i++){
    if(i!=n-1)cout << S[i] << " ";
    else cout << S[i] << endl;
  }
  cout << count << endl;
}
```

### 試したこと

LRの両者のn1,n2部分に代入する数値を定義した。


### 補足情報（FW/ツールのバージョンなど）

当方、文系学生で、基礎的な数学の知識も持ち合わせていないので、可能であれば、答えより着眼点・原因発見のヒントやどう言うふうに考えるといいと言うようなアドバイスをいただけると今後の学習に活かせそうなのでありがたいです！！

C++でマージソートのコーディングがうまく行きません！！

下記のようなコードで画像分類器を作成したいと考えています。
しかし、エラーが発生してしまい実行できません。
コードは同じ本から採用しています。
解決方法がありましたら、教えていただけますでしょうか。
よろしくお願い致します。

model_and_weight.h5と1.jpgというファイルは、同階層に保存済みです。
```python
from PIL import Image
import os, glob
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True

from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D
from keras.layers import Activation, Dropout, Flatten, Dense
from keras.optimizers import RMSprop
from keras.utils import np_utils
import keras
import numpy as np

import sys, os
from keras.models import load_model

imsize = (150, 150)

keras_param = "model_and_weight.h5"
testpic     = "1.jpg"

def load_image(path):
    img = Image.open(path)
    img = img.convert('RGB')
    img = img.resize(imsize)
    img = np.asarray(img)
    img = img / 255.0
    return img

model = load_model(keras_param)
img = load_image(testpic)
prd = model.predict(np.array([img]))
print(prd) 
prelabel = np.argmax(prd, axis=1)
if prelabel == 0:
    print(">>> 犬")
elif prelabel == 1:
    print(">>> 猫")
```

下記に記載したのがエラー内容

```python
WARNING:tensorflow:Model was constructed with shape (None, 4, 4, 512) for input KerasTensor(type_spec=TensorSpec(shape=(None, 4, 4, 512), dtype=tf.float32, name='flatten_input'), name='flatten_input', description="created by layer 'flatten_input'"), but it was called on an input with incompatible shape (None, 150, 150, 3).
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-12-baa28c867e4e> in <module>
     35 model = load_model(keras_param)
     36 img = load_image(testpic)
---> 37 prd = model.predict(np.array([img]))
     38 # print(prd) # 精度の表示
     39 prelabel = np.argmax(prd, axis=1)
```
学習済みファイルの作成コード
```python
from keras.applications.vgg16 import VGG16
from keras.preprocessing.image import ImageDataGenerator
from keras.layers import Input
import numpy as np
import os

num_train =2000
num_validation=800
img_h,img_w=150,150
channels=3
batch_size=32
train_data_dir='data/train'
validation_data_dir='data/validation'
result_dir='results'

if not os.path.exists(result_dir):
    os.mkdir(result_dir)

def save_VGG16_outputs():
    model=VGG16(
    include_top=False,
    weights='imagenet',
    input_shape=(img_h,img_w,channels))
    
    model.summary()

    datagen=ImageDataGenerator(rescale=1.0/255)

    train_generator=datagen.flow_from_directory(
        train_data_dir,
        target_size=(img_w,img_h),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False)

    print('train-label:',train_generator.class_indices)

    vgg16_train=model.predict_generator(
        train_generator,
        verbose=1)

    np.save(os.path.join(result_dir,'vgg16_train.npy'),vgg16_train)

    validation_generator=datagen.flow_from_directory(
        validation_data_dir,
        target_size=(img_w,img_h),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False)

    print('test-label:',validation_generator.class_indices)

    vgg16_test=model.predict(
        validation_generator,
        verbose=1)

    np.save(os.path.join(result_dir,'vgg16_test.npy'),vgg16_test)

save_VGG16_outputs()

import os
import numpy as np
from keras.models import Sequential
from keras import optimizers
from keras.layers import Activation,Dropout,Flatten,Dense

def train_FClayer():
    
    train_data=np.load(
        os.path.join(result_dir,'vgg16_train.npy'))
    
    print(train_data.shape)
    
    train_labels=np.array(
    [0]*int(num_train/2)+[1]*int(num_train/2))
    
    validation_data=np.load(
    os.path.join(result_dir,'vgg16_test.npy'))
    
    print(validation_data.shape)
    
    validation_labels=np.array(
    [0]*int(num_validation/2)+[1]*int(num_validation/2))
    
    model=Sequential()
    
    model.add(Flatten(input_shape=train_data.shape[1:]))
    
    model.add(Dense(256,activation='relu'))
    
    model.add(Dense(1,activation='sigmoid'))
    
    model.compile(
        loss='binary_crossentropy',
        metrics=['accuracy'],
        optimizer=optimizers.SGD(lr=1e-4,momentum=0.9),)
    
    epoch=60
    batch_size=32
    history=model.fit(train_data,
                      train_labels,
                      epochs=epoch,
                      batch_size=batch_size,
                      verbose=1,
                      validation_data=(validation_data,validation_labels))
    with open('model.json','w') as json_file:
        json_file.write(model.to_json())
    model.save('model_and_weight.h5')
    
    return history

history=train_FClayer()
```

画像の処理コード
```python
from keras.applications.vgg16 import VGG16
from keras.preprocessing.image import ImageDataGenerator
from keras.layers import Input
import numpy as np
import os

num_eva =1
img_h,img_w=150,150
channels=3
batch_size=32
eva_data_dir='data/evaluation'
result_dir='results'

if not os.path.exists(result_dir):
    os.mkdir(result_dir)

def save_VGG16_outputs():
    model=VGG16(
    include_top=False,
    weights='imagenet',
    input_shape=(img_h,img_w,channels))
    
    model.summary()

    datagen=ImageDataGenerator(rescale=1.0/255)

    eva_generator=datagen.flow_from_directory(
        eva_data_dir,
        target_size=(img_w,img_h),
        batch_size=batch_size,
        class_mode=None,
        shuffle=False)

    print('eva-label:',eva_generator.class_indices)

    vgg16_eva=model.predict_generator(
        eva_generator,
        verbose=1)

    np.save(os.path.join(result_dir,'vgg16_eva.npy'),vgg16_eva)

    print(vgg16_eva.shape)

save_VGG16_outputs()

Model: "vgg16"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
input_2 (InputLayer)         [(None, 150, 150, 3)]     0         
_________________________________________________________________
block1_conv1 (Conv2D)        (None, 150, 150, 64)      1792      
_________________________________________________________________
block1_conv2 (Conv2D)        (None, 150, 150, 64)      36928     
_________________________________________________________________
block1_pool (MaxPooling2D)   (None, 75, 75, 64)        0         
_________________________________________________________________
block2_conv1 (Conv2D)        (None, 75, 75, 128)       73856     
_________________________________________________________________
block2_conv2 (Conv2D)        (None, 75, 75, 128)       147584    
_________________________________________________________________
block2_pool (MaxPooling2D)   (None, 37, 37, 128)       0         
_________________________________________________________________
block3_conv1 (Conv2D)        (None, 37, 37, 256)       295168    
_________________________________________________________________
block3_conv2 (Conv2D)        (None, 37, 37, 256)       590080    
_________________________________________________________________
block3_conv3 (Conv2D)        (None, 37, 37, 256)       590080    
_________________________________________________________________
block3_pool (MaxPooling2D)   (None, 18, 18, 256)       0         
_________________________________________________________________
block4_conv1 (Conv2D)        (None, 18, 18, 512)       1180160   
_________________________________________________________________
block4_conv2 (Conv2D)        (None, 18, 18, 512)       2359808   
_________________________________________________________________
block4_conv3 (Conv2D)        (None, 18, 18, 512)       2359808   
_________________________________________________________________
block4_pool (MaxPooling2D)   (None, 9, 9, 512)         0         
_________________________________________________________________
block5_conv1 (Conv2D)        (None, 9, 9, 512)         2359808   
_________________________________________________________________
block5_conv2 (Conv2D)        (None, 9, 9, 512)         2359808   
_________________________________________________________________
block5_conv3 (Conv2D)        (None, 9, 9, 512)         2359808   
_________________________________________________________________
block5_pool (MaxPooling2D)   (None, 4, 4, 512)         0         
=================================================================
Total params: 14,714,688
Trainable params: 14,714,688
Non-trainable params: 0
_________________________________________________________________
Found 1 images belonging to 1 classes.
eva-label: {'1': 0}
1/1 [==============================] - 2s 2s/step
(1, 4, 4, 512)
```