diff --git a/docs_and_specs/keysight_operating_service.pdf b/docs_and_specs/keysight_operating_service.pdf
index dc75f1ae..a72c8998 100644
Binary files a/docs_and_specs/keysight_operating_service.pdf and b/docs_and_specs/keysight_operating_service.pdf differ
diff --git a/test/data/meta.xml b/test/data/meta.xml
index ceabb419..91f10b28 100644
--- a/test/data/meta.xml
+++ b/test/data/meta.xml
@@ -4900,4 +4900,32 @@
 		<voltage>1.200000</voltage>
 		<file>/Users/lwh/CLionProjects/ce_device/test/data/1_7_100_12_1628863407</file>
 	</test>
+	<test>
+		<chip>1</chip>
+		<cell>2</cell>
+		<frequency>100</frequency>
+		<voltage>1.200000</voltage>
+		<file>/Users/lwh/CLionProjects/ce_device/test/data/1_2_100_12_1631603739</file>
+	</test>
+	<test>
+		<chip>1</chip>
+		<cell>2</cell>
+		<frequency>100</frequency>
+		<voltage>1.200000</voltage>
+		<file>/Users/lwh/CLionProjects/ce_device/test/data/1_2_100_12_1631604369</file>
+	</test>
+	<test>
+		<chip>1</chip>
+		<cell>2</cell>
+		<frequency>100</frequency>
+		<voltage>1.200000</voltage>
+		<file>/Users/lwh/CLionProjects/ce_device/test/data/1_2_100_12_1631604406</file>
+	</test>
+	<test>
+		<chip>1</chip>
+		<cell>2</cell>
+		<frequency>100</frequency>
+		<voltage>1.200000</voltage>
+		<file>/Users/lwh/CLionProjects/ce_device/test/data/1_2_100_12_1631604451</file>
+	</test>
 </tests>
diff --git a/test/plot.ipynb b/test/plot.ipynb
index 8813e3f5..4f3c1c5f 100644
--- a/test/plot.ipynb
+++ b/test/plot.ipynb
@@ -2,48 +2,314 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 334,
+   "execution_count": 217,
    "id": "b5e66da9",
    "metadata": {},
    "outputs": [],
    "source": [
+    "# SECENGUP\n",
     "import xml.etree.ElementTree as ET\n",
     "import pandas as pd\n",
     "import matplotlib.pyplot as plt\n",
-    "import re"
+    "import re\n",
+    "\n",
+    "# test_file = 'data/1_2_100_12_1631603739'\n",
+    "# test_file = '~/Downloads/2_1_100_12_1631609594'\n",
+    "# test_file = '~/Downloads/2_4_500_12_1631610487'\n",
+    "# test_file = '~/Downloads/2_7_500_12_1631610820'\n",
+    "test_file = '/Users/lwh/Downloads/test_data/10000/2_13_10000_12_1631611741'\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 335,
+   "execution_count": 218,
    "id": "7647d63c",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2000\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "def plot_single_file(filename):\n",
-    "    data = pd.read_csv(filename)\n",
-    "    plt.plot(data['time'], data['input'], label=filename)\n",
-    "    plt.plot(data['time'], data['output'], label=filename)"
+    "    data = pd.read_csv(filename)[0:2000]\n",
+    "    print(len(data))\n",
+    "    plt.scatter(data['time'], data['input'], s=1, label=filename)\n",
+    "    plt.scatter(data['time'], data['output'], s=1, label=filename)\n",
+    "plot_single_file(test_file)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 336,
+   "execution_count": 219,
    "id": "f2f5ce11",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "def plot_file(filename):\n",
     "    data = pd.read_csv(filename) # input\n",
     "    plt.title(filename)\n",
     "    plt.scatter(data['input'], data['output'], s=1, label=filename)\n",
     "\n",
-    "# plot_file('data/1_4_100_12_1628855074')"
+    "plot_file(test_file)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 220,
+   "id": "6b114604",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import math\n",
+    "\n",
+    "def get_zero_points(filename):\n",
+    "    data = pd.read_csv(filename)\n",
+    "    input_data = data['input']\n",
+    "    is_positive = input_data[0] > 0\n",
+    "    zero_points = []\n",
+    "    for i in range(len(data)):\n",
+    "        new_is_positive = input_data[i] > 0\n",
+    "        if (is_positive != new_is_positive):\n",
+    "            # met a zero points\n",
+    "            zero_points.append(i)\n",
+    "            is_positive = new_is_positive\n",
+    "    return zero_points\n",
+    "\n",
+    "# test_zero_points = get_zero_points(test_file)\n",
+    "# print(test_zero_points, len(test_zero_points))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 243,
+   "id": "a432fe3a",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "200\n",
+      "      time     input    output\n",
+      "1   -0.499  0.060302  0.048241\n",
+      "2   -0.498  0.120603  0.108543\n",
+      "3   -0.497  0.192965  0.168844\n",
+      "4   -0.496  0.265327  0.241206\n",
+      "5   -0.495  0.337688  0.301507\n",
+      "..     ...       ...       ...\n",
+      "96  -0.404 -0.301507 -0.253266\n",
+      "97  -0.403 -0.229146 -0.192965\n",
+      "98  -0.402 -0.156784 -0.132663\n",
+      "99  -0.401 -0.084422 -0.072362\n",
+      "100 -0.400 -0.012060  0.000000\n",
+      "\n",
+      "[100 rows x 3 columns]\n",
+      "      time     input    output\n",
+      "101 -0.399  0.048241  0.048241\n",
+      "102 -0.398  0.120603  0.108543\n",
+      "103 -0.397  0.192965  0.168844\n",
+      "104 -0.396  0.265327  0.229146\n",
+      "105 -0.395  0.337688  0.301507\n",
+      "..     ...       ...       ...\n",
+      "196 -0.304 -0.313568 -0.265327\n",
+      "197 -0.303 -0.229146 -0.205025\n",
+      "198 -0.302 -0.156784 -0.132663\n",
+      "199 -0.301 -0.084422 -0.072362\n",
+      "200 -0.300 -0.024121 -0.012060\n",
+      "\n",
+      "[100 rows x 3 columns]\n"
+     ]
+    }
+   ],
+   "source": [
+    "cycles_per_file = 10\n",
+    "\n",
+    "def plot_cycle(cycle):\n",
+    "    time_list = list(cycle['time'])\n",
+    "    input_list = list(cycle['input'])\n",
+    "    output_list = list(cycle['output'])\n",
+    "    \n",
+    "    plt.figure()\n",
+    "    plt.plot(time_list, input_list, label=\"input\")\n",
+    "    plt.plot(time_list, output_list, label=\"output\")\n",
+    "\n",
+    "def split_cycles(filename):\n",
+    "    data = pd.read_csv(filename)\n",
+    "    zero_points = get_zero_points(filename)\n",
+    "    zero_points_len = len(zero_points)\n",
+    "    cycles = []\n",
+    "    for i in range(0, zero_points_len, 2):\n",
+    "        start = zero_points[i]\n",
+    "        if (i + 2 < zero_points_len):\n",
+    "            end = zero_points[i+2]\n",
+    "        else:\n",
+    "            end = len(data)\n",
+    "        if (end - start < 80):\n",
+    "            raise Error('Split Cycles Error')\n",
+    "        cycles.append(data[start : end])\n",
+    "    return cycles\n",
+    "\n",
+    "cycles = split_cycles('data/1_12_10000_12_1631620066')\n",
+    "print(len(cycles))\n",
+    "print(cycles[0])\n",
+    "print(cycles[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 237,
+   "id": "dbe00690",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/10000/1_10_10000_12_1631619853\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/10000/1_9_10000_12_1631619756\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/10000/1_11_10000_12_1631619951\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/10000/1_12_10000_12_1631620066\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/10000/1_1_10000_12_1631612785\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/100/1_1_100_12_1631612737\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/100/1_11_100_12_1631619903\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/100/1_10_100_12_1631619805\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/100/1_12_100_12_1631620018\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/100/1_9_100_12_1631619709\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/500/1_12_500_12_1631620035\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/500/1_11_500_12_1631619920\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/500/1_1_500_12_1631612754\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/500/1_9_500_12_1631619726\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/500/1_10_500_12_1631619822\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/1000/1_12_1000_12_1631620050\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/1000/1_10_1000_12_1631619838\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/1000/1_1_1000_12_1631612769\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/1000/1_9_1000_12_1631619741\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip1/1000/1_11_1000_12_1631619935\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_6_10000_12_1631610756\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_13_10000_12_1631611741\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_16_10000_12_1631612045\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_1_10000_12_1631609642\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_14_10000_12_1631611845\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_11_10000_12_1631611559\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_12_10000_12_1631611653\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_9_10000_12_1631611251\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_2_10000_12_1631612298\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_10_10000_12_1631611375\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_2_10000_12_1631610121\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_5_10000_12_1631610651\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_7_10000_12_1631610850\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_4_10000_12_1631610517\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_15_10000_12_1631611932\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_8_10000_12_1631611012\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_1_10000_12_1631612154\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/10000/2_3_10000_12_1631610401\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_4_100_12_1631610469\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_15_100_12_1631611885\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_2_100_12_1631610074\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_3_100_12_1631610350\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_8_100_12_1631610965\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_2_100_12_1631612250\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_10_100_12_1631611327\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_9_100_12_1631611203\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_1_100_12_1631609594\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_6_100_12_1631610708\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_14_100_12_1631611797\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_1_100_12_1631612106\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_13_100_12_1631611693\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_3_100_12_1631610281\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_12_100_12_1631611605\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_5_100_12_1631610603\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_16_100_12_1631611997\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_7_100_12_1631610803\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/100/2_11_100_12_1631611511\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_7_500_12_1631610820\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_5_500_12_1631610620\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_11_500_12_1631611528\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_2_500_12_1631610091\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_3_500_12_1631610370\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_1_500_12_1631612123\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_16_500_12_1631612014\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_9_500_12_1631611220\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_13_500_12_1631611710\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_2_500_12_1631612267\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_14_500_12_1631611814\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_15_500_12_1631611902\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_12_500_12_1631611622\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_8_500_12_1631610982\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_10_500_12_1631611344\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_1_500_12_1631609611\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_4_500_12_1631610487\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/500/2_6_500_12_1631610726\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_15_1000_12_1631611917\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_6_1000_12_1631610741\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_10_1000_12_1631611360\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_5_1000_12_1631610635\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_11_1000_12_1631611544\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_1_1000_12_1631609627\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_2_1000_12_1631610106\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_13_1000_12_1631611726\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_9_1000_12_1631611236\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_8_1000_12_1631610997\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_16_1000_12_1631612030\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_1_1000_12_1631612139\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_4_1000_12_1631610502\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_2_1000_12_1631612282\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_12_1000_12_1631611637\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_14_1000_12_1631611830\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_3_1000_12_1631610386\n",
+      "/Users/lwh/Downloads/test_data/KNOWM_Memristors/Chip2/1000/2_7_1000_12_1631610835\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os\n",
+    "directory_str = '/Users/lwh/Downloads/test_data/'\n",
+    "directory = os.fsencode(directory_str)\n",
+    "\n",
+    "a = 0\n",
+    "for root, subdirs, files in os.walk(directory):\n",
+    "    for file in files:\n",
+    "        if (file == b'.DS_Store' or file == b'meta.xml'):\n",
+    "            continue\n",
+    "        filepath = os.path.join(root, file).decode('utf-8')\n",
+    "        print(filepath)\n",
+    "        cycles = split_cycles(filepath)\n",
+    "#         print(filepath, len(cycles))"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 337,
+   "execution_count": 126,
    "id": "e1dd2cf0",
    "metadata": {},
    "outputs": [],
@@ -58,7 +324,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 338,
+   "execution_count": 233,
    "id": "bb2471c1",
    "metadata": {
     "scrolled": true
@@ -103,56 +369,30 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 339,
-   "id": "4cf8068c",
-   "metadata": {
-    "scrolled": false
-   },
-   "outputs": [],
-   "source": [
-    "import math\n",
-    "\n",
-    "def get_zero_points(filename):\n",
-    "    data = pd.read_csv(filename)\n",
-    "    input_data = data['input']\n",
-    "    is_positive = input_data[0] > 0\n",
-    "    zero_points = []\n",
-    "    for i in range(len(data)):\n",
-    "        new_is_positive = input_data[i] > 0\n",
-    "        if (is_positive != new_is_positive):\n",
-    "            # met a zero points\n",
-    "            zero_points.append(i)\n",
-    "            is_positive = new_is_positive\n",
-    "    return zero_points\n",
-    "\n",
-    "cycles_per_file = 10\n",
-    "\n",
-    "def split_cycles(filename):\n",
-    "    data = pd.read_csv(filename)\n",
-    "    cycles = []\n",
-    "#     cycles.append([list(data['input']), list(data['output'])])\n",
-    "    data_len = len(data['input'])\n",
-    "    points_per_cycle = math.ceil(data_len / cycles_per_file)\n",
-    "    for i in range(0, data_len, points_per_cycle):\n",
-    "        input_data = data['input'][i:i+points_per_cycle]\n",
-    "        output_data = data['output'][i:i+points_per_cycle]\n",
-    "        if (len(input_data) < 90):\n",
-    "            continue\n",
-    "        cycles.append([list(input_data), list(output_data)])\n",
-    "    return cycles"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 341,
+   "execution_count": 234,
    "id": "9af9415a",
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "7000\n"
+     "ename": "KeyError",
+     "evalue": "0",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
+      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/pandas/core/indexes/base.py\u001b[0m in \u001b[0;36mget_loc\u001b[0;34m(self, key, method, tolerance)\u001b[0m\n\u001b[1;32m   3360\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3361\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcasted_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3362\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/pandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/pandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
+      "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
+      "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
+      "\u001b[0;31mKeyError\u001b[0m: 0",
+      "\nThe above exception was the direct cause of the following exception:\n",
+      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
+      "\u001b[0;32m/var/folders/wl/82lk3rwd689bcgys_4j_jmch0000gn/T/ipykernel_89314/695677664.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     44\u001b[0m         \u001b[0;32mcontinue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     45\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mcycle\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcycles\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 46\u001b[0;31m         \u001b[0mtrim_cycle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcycle\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     47\u001b[0m         \u001b[0mlabels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mget_label\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcell_number\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     48\u001b[0m         \u001b[0mtrain_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcycle\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/var/folders/wl/82lk3rwd689bcgys_4j_jmch0000gn/T/ipykernel_89314/695677664.py\u001b[0m in \u001b[0;36mtrim_cycle\u001b[0;34m(cycle)\u001b[0m\n\u001b[1;32m     19\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mtrim_cycle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcycle\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 21\u001b[0;31m     \u001b[0;32mwhile\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcycle\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0mdata_len\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     22\u001b[0m         \u001b[0mpos\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrandint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcycle\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     23\u001b[0m         \u001b[0mcycle\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpos\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/pandas/core/frame.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   3453\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnlevels\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3454\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_getitem_multilevel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3455\u001b[0;31m             \u001b[0mindexer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3456\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mis_integer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindexer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3457\u001b[0m                 \u001b[0mindexer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mindexer\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m/usr/local/lib/python3.9/site-packages/pandas/core/indexes/base.py\u001b[0m in \u001b[0;36mget_loc\u001b[0;34m(self, key, method, tolerance)\u001b[0m\n\u001b[1;32m   3361\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcasted_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3362\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3363\u001b[0;31m                 \u001b[0;32mraise\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0merr\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3364\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3365\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mis_scalar\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0misna\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhasnans\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mKeyError\u001b[0m: 0"
      ]
     }
    ],
@@ -192,11 +432,11 @@
     "    voltage = float(dic['voltage'])\n",
     "    frequency = int(dic['frequency'])\n",
     "    cycles = split_cycles(file)\n",
-    "    if (len(cycles) == 0):\n",
-    "        # measurement errors;\n",
-    "        # nomarlly no more than 10 cycles in a test file;\n",
-    "        # drop these error files\n",
-    "        continue\n",
+    "#     if (len(cycles) == 0):\n",
+    "#         # measurement errors;\n",
+    "#         # nomarlly no more than 10 cycles in a test file;\n",
+    "#         # drop these error files\n",
+    "#         continue\n",
     "    if (cell_number > cells):\n",
     "        continue\n",
     "    if (voltage != t_voltage or frequency != t_frequency):\n",
@@ -211,7 +451,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 342,
+   "execution_count": 8,
    "id": "36f8817b",
    "metadata": {},
    "outputs": [],
@@ -232,71 +472,80 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 343,
+   "execution_count": 9,
    "id": "acbfab48",
    "metadata": {},
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2021-09-14 08:53:10.782103: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
+      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+      "2021-09-14 08:53:10.934090: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:176] None of the MLIR Optimization Passes are enabled (registered 2)\n"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "tf.Tensor([7000    2   90], shape=(3,), dtype=int32)\n",
-      "Model: \"sequential_37\"\n",
+      "Model: \"sequential\"\n",
       "_________________________________________________________________\n",
       "Layer (type)                 Output Shape              Param #   \n",
       "=================================================================\n",
-      "flatten_35 (Flatten)         (None, 180)               0         \n",
+      "flatten (Flatten)            (None, 180)               0         \n",
       "_________________________________________________________________\n",
-      "dense_70 (Dense)             (None, 50)                9050      \n",
+      "dense (Dense)                (None, 50)                9050      \n",
       "_________________________________________________________________\n",
-      "dense_71 (Dense)             (None, 7)                 357       \n",
+      "dense_1 (Dense)              (None, 7)                 357       \n",
       "=================================================================\n",
       "Total params: 9,407\n",
       "Trainable params: 9,407\n",
       "Non-trainable params: 0\n",
       "_________________________________________________________________\n",
       "Epoch 1/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 1.2995 - accuracy: 0.4867 - val_loss: 0.8948 - val_accuracy: 0.7316\n",
+      "784/784 [==============================] - 1s 1ms/step - loss: 1.3074 - accuracy: 0.4974 - val_loss: 0.9445 - val_accuracy: 0.6765\n",
       "Epoch 2/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.7477 - accuracy: 0.7235 - val_loss: 0.6288 - val_accuracy: 0.8061\n",
+      "784/784 [==============================] - 1s 881us/step - loss: 0.7541 - accuracy: 0.7329 - val_loss: 0.6126 - val_accuracy: 0.7969\n",
       "Epoch 3/20\n",
-      "784/784 [==============================] - 1s 971us/step - loss: 0.5289 - accuracy: 0.8342 - val_loss: 0.4552 - val_accuracy: 0.8816\n",
+      "784/784 [==============================] - 1s 883us/step - loss: 0.5276 - accuracy: 0.8436 - val_loss: 0.4232 - val_accuracy: 0.8969\n",
       "Epoch 4/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.3912 - accuracy: 0.8982 - val_loss: 0.3623 - val_accuracy: 0.8939\n",
+      "784/784 [==============================] - 1s 873us/step - loss: 0.3984 - accuracy: 0.9000 - val_loss: 0.3318 - val_accuracy: 0.9367\n",
       "Epoch 5/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.3068 - accuracy: 0.9268 - val_loss: 0.2659 - val_accuracy: 0.9367\n",
+      "784/784 [==============================] - 1s 896us/step - loss: 0.3093 - accuracy: 0.9224 - val_loss: 0.2865 - val_accuracy: 0.9082\n",
       "Epoch 6/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.2583 - accuracy: 0.9372 - val_loss: 0.2147 - val_accuracy: 0.9541\n",
+      "784/784 [==============================] - 1s 895us/step - loss: 0.2422 - accuracy: 0.9452 - val_loss: 0.2058 - val_accuracy: 0.9663\n",
       "Epoch 7/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.2003 - accuracy: 0.9559 - val_loss: 0.1910 - val_accuracy: 0.9571\n",
+      "784/784 [==============================] - 1s 895us/step - loss: 0.2040 - accuracy: 0.9528 - val_loss: 0.1744 - val_accuracy: 0.9622\n",
       "Epoch 8/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.1744 - accuracy: 0.9594 - val_loss: 0.2513 - val_accuracy: 0.8959\n",
+      "784/784 [==============================] - 1s 882us/step - loss: 0.1654 - accuracy: 0.9638 - val_loss: 0.1535 - val_accuracy: 0.9551\n",
       "Epoch 9/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.1610 - accuracy: 0.9597 - val_loss: 0.1285 - val_accuracy: 0.9765\n",
+      "784/784 [==============================] - 1s 877us/step - loss: 0.1411 - accuracy: 0.9694 - val_loss: 0.1178 - val_accuracy: 0.9816\n",
       "Epoch 10/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.1329 - accuracy: 0.9653 - val_loss: 0.1351 - val_accuracy: 0.9704\n",
+      "784/784 [==============================] - 1s 906us/step - loss: 0.1272 - accuracy: 0.9681 - val_loss: 0.1060 - val_accuracy: 0.9776\n",
       "Epoch 11/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.1222 - accuracy: 0.9686 - val_loss: 0.1027 - val_accuracy: 0.9704\n",
+      "784/784 [==============================] - 1s 878us/step - loss: 0.1164 - accuracy: 0.9742 - val_loss: 0.0926 - val_accuracy: 0.9908\n",
       "Epoch 12/20\n",
-      "784/784 [==============================] - 1s 931us/step - loss: 0.1161 - accuracy: 0.9658 - val_loss: 0.1295 - val_accuracy: 0.9612\n",
+      "784/784 [==============================] - 1s 925us/step - loss: 0.0976 - accuracy: 0.9753 - val_loss: 0.0845 - val_accuracy: 0.9796\n",
       "Epoch 13/20\n",
-      "784/784 [==============================] - 1s 960us/step - loss: 0.1114 - accuracy: 0.9684 - val_loss: 0.1075 - val_accuracy: 0.9714\n",
+      "784/784 [==============================] - 1s 919us/step - loss: 0.1011 - accuracy: 0.9735 - val_loss: 0.0637 - val_accuracy: 0.9888\n",
       "Epoch 14/20\n",
-      "784/784 [==============================] - 1s 944us/step - loss: 0.0919 - accuracy: 0.9747 - val_loss: 0.0787 - val_accuracy: 0.9765\n",
+      "784/784 [==============================] - 1s 888us/step - loss: 0.0853 - accuracy: 0.9763 - val_loss: 0.0659 - val_accuracy: 0.9878\n",
       "Epoch 15/20\n",
-      "784/784 [==============================] - 1s 938us/step - loss: 0.0905 - accuracy: 0.9730 - val_loss: 0.1007 - val_accuracy: 0.9684\n",
+      "784/784 [==============================] - 1s 858us/step - loss: 0.0906 - accuracy: 0.9755 - val_loss: 0.0817 - val_accuracy: 0.9847\n",
       "Epoch 16/20\n",
-      "784/784 [==============================] - 1s 935us/step - loss: 0.0874 - accuracy: 0.9747 - val_loss: 0.0885 - val_accuracy: 0.9704\n",
+      "784/784 [==============================] - 1s 877us/step - loss: 0.0768 - accuracy: 0.9788 - val_loss: 0.0534 - val_accuracy: 0.9867\n",
       "Epoch 17/20\n",
-      "784/784 [==============================] - 1s 931us/step - loss: 0.0916 - accuracy: 0.9709 - val_loss: 0.0785 - val_accuracy: 0.9704\n",
+      "784/784 [==============================] - 1s 890us/step - loss: 0.0774 - accuracy: 0.9758 - val_loss: 0.0776 - val_accuracy: 0.9816\n",
       "Epoch 18/20\n",
-      "784/784 [==============================] - 1s 940us/step - loss: 0.0748 - accuracy: 0.9798 - val_loss: 0.2254 - val_accuracy: 0.8898\n",
+      "784/784 [==============================] - 1s 884us/step - loss: 0.0697 - accuracy: 0.9788 - val_loss: 0.0600 - val_accuracy: 0.9857\n",
       "Epoch 19/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.0786 - accuracy: 0.9765 - val_loss: 0.0917 - val_accuracy: 0.9694\n",
+      "784/784 [==============================] - 1s 882us/step - loss: 0.0674 - accuracy: 0.9804 - val_loss: 0.0666 - val_accuracy: 0.9837\n",
       "Epoch 20/20\n",
-      "784/784 [==============================] - 1s 1ms/step - loss: 0.0691 - accuracy: 0.9806 - val_loss: 0.0531 - val_accuracy: 0.9816\n",
-      "66/66 - 0s - loss: 0.0572 - accuracy: 0.9848\n",
-      "0.057217005640268326 0.9847618937492371\n"
+      "784/784 [==============================] - 1s 881us/step - loss: 0.0679 - accuracy: 0.9798 - val_loss: 0.0665 - val_accuracy: 0.9786\n",
+      "66/66 - 0s - loss: 0.0922 - accuracy: 0.9662\n",
+      "0.09215494245290756 0.9661904573440552\n"
      ]
     }
    ],
@@ -336,23 +585,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 231,
+   "execution_count": 10,
    "id": "824ce269",
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.legend.Legend at 0x1605bf6a0>"
+       "<matplotlib.legend.Legend at 0x15f4e8160>"
       ]
      },
-     "execution_count": 231,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -364,7 +613,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]