From c3f8dabe3235b07e0315122c43ed897051a4c807 Mon Sep 17 00:00:00 2001
From: Gregor von Laszewski <laszewski@gmail.com>
Date: Fri, 6 Oct 2023 15:26:01 -0400
Subject: [PATCH] improve an script

---
 .../cloudmask/target/greene_v0.5/an.ipynb     | 1156 +++++++++--------
 1 file changed, 647 insertions(+), 509 deletions(-)

diff --git a/benchmarks/cloudmask/target/greene_v0.5/an.ipynb b/benchmarks/cloudmask/target/greene_v0.5/an.ipynb
index 96c5175c..27ade59a 100644
--- a/benchmarks/cloudmask/target/greene_v0.5/an.ipynb
+++ b/benchmarks/cloudmask/target/greene_v0.5/an.ipynb
@@ -2,13 +2,12 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 138,
+   "execution_count": null,
    "id": "initial_id",
    "metadata": {
     "collapsed": true,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:10:18.551635207Z",
-     "start_time": "2023-10-06T14:10:17.353106833Z"
+     "start_time": "2023-10-06T17:09:24.520860982Z"
     }
    },
    "outputs": [],
@@ -40,6 +39,10 @@
     "                try:\n",
     "                    result_line = [line.replace(\":::MLLOG\", \"\").strip() for line in lines if line.startswith(\":::MLLOG\") and '\"result\"' in line][0]\n",
     "                    result_line = eval(result_line)\n",
+    "                    \n",
+    "                    for k in ['event_type', 'key', 'metadata', 'namespace', 'time_ms']:\n",
+    "                        del result_line[k]\n",
+    "                    \n",
     "                except:\n",
     "                    result_line = None\n",
     "                csv_dict = {}\n",
@@ -83,445 +86,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 139,
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "{'batch_size': 32,\n",
-      " 'card_name': 'v100',\n",
-      " 'clip_offset': 15,\n",
-      " 'cpu_num': 1,\n",
-      " 'directive': 'v100',\n",
-      " 'early_stoppage': 'False',\n",
-      " 'early_stoppage_patience': '25',\n",
-      " 'epoch': '70',\n",
-      " 'gpu': 1,\n",
-      " 'gpu_count': '1',\n",
-      " 'inference': 144.679,\n",
-      " 'learning_rate': 0.001,\n",
-      " 'loaddata': 4.118,\n",
-      " 'mem': '64GB',\n",
-      " 'name': 'result-thf2bn-53854471.out',\n",
-      " 'no_cache': False,\n",
-      " 'nodes': 1,\n",
-      " 'repeat': '5',\n",
-      " 'result': {'event_type': 'POINT_IN_TIME',\n",
-      "            'key': 'result',\n",
-      "            'metadata': {'file': 'cloudmask_v0.5.py', 'lineno': 461},\n",
-      "            'namespace': '',\n",
-      "            'time_ms': 1696595324526,\n",
-      "            'value': {'inference': {'accuracy': [0.9233438888888889,\n",
-      "                                                 0.9345094444444444,\n",
-      "                                                 0.8940772222222222,\n",
-      "                                                 0.9105144444444444,\n",
-      "                                                 0.8391288888888889,\n",
-      "                                                 0.95098,\n",
-      "                                                 0.8876861111111111,\n",
-      "                                                 0.7732322222222222,\n",
-      "                                                 0.8472777777777778,\n",
-      "                                                 0.9639461111111111,\n",
-      "                                                 0.9149733333333333,\n",
-      "                                                 0.8905033333333333,\n",
-      "                                                 0.6264644444444445,\n",
-      "                                                 0.9434594444444444,\n",
-      "                                                 0.9502511111111112,\n",
-      "                                                 0.8874255555555556,\n",
-      "                                                 0.9090383333333333,\n",
-      "                                                 0.783135,\n",
-      "                                                 0.8810644444444444,\n",
-      "                                                 0.8861566666666667,\n",
-      "                                                 0.9271861111111112,\n",
-      "                                                 0.7000683333333333,\n",
-      "                                                 0.9493627777777778,\n",
-      "                                                 0.8812888888888889,\n",
-      "                                                 0.7280755555555556,\n",
-      "                                                 0.9413794444444444,\n",
-      "                                                 0.6406661111111112,\n",
-      "                                                 0.9485711111111111,\n",
-      "                                                 0.8190166666666666,\n",
-      "                                                 0.8649538888888889,\n",
-      "                                                 0.9245216666666667,\n",
-      "                                                 0.5231455555555555,\n",
-      "                                                 0.8900266666666666,\n",
-      "                                                 0.9201622222222222,\n",
-      "                                                 0.8942827777777778,\n",
-      "                                                 0.9398694444444444,\n",
-      "                                                 0.9456305555555555,\n",
-      "                                                 0.8512122222222223,\n",
-      "                                                 0.9236327777777777,\n",
-      "                                                 0.9262188888888889,\n",
-      "                                                 0.8431988888888889,\n",
-      "                                                 0.7898244444444444,\n",
-      "                                                 0.9462755555555555,\n",
-      "                                                 0.9299205555555555,\n",
-      "                                                 0.9505644444444444,\n",
-      "                                                 0.8782683333333333,\n",
-      "                                                 0.8345605555555555,\n",
-      "                                                 0.9333472222222222,\n",
-      "                                                 0.8361561111111111,\n",
-      "                                                 0.8908183333333334,\n",
-      "                                                 0.90968,\n",
-      "                                                 0.9309866666666666,\n",
-      "                                                 0.7933927777777777,\n",
-      "                                                 0.9122016666666667,\n",
-      "                                                 0.7026127777777778,\n",
-      "                                                 0.8870366666666667,\n",
-      "                                                 0.8692122222222223,\n",
-      "                                                 0.9070755555555555,\n",
-      "                                                 0.930505,\n",
-      "                                                 0.9060172222222223,\n",
-      "                                                 0.9090455555555556,\n",
-      "                                                 0.9134227777777778,\n",
-      "                                                 0.8983916666666667,\n",
-      "                                                 0.8190194444444444,\n",
-      "                                                 0.7275561111111111,\n",
-      "                                                 0.8331155555555556,\n",
-      "                                                 0.9206822222222222,\n",
-      "                                                 0.867085,\n",
-      "                                                 0.85498,\n",
-      "                                                 0.8969744444444444,\n",
-      "                                                 0.8837505555555556,\n",
-      "                                                 0.9042238888888889,\n",
-      "                                                 0.9291705555555556,\n",
-      "                                                 0.8913488888888889,\n",
-      "                                                 0.8909755555555555,\n",
-      "                                                 0.9325277777777777,\n",
-      "                                                 0.8544688888888888,\n",
-      "                                                 0.909215,\n",
-      "                                                 0.9091288888888889,\n",
-      "                                                 0.873935,\n",
-      "                                                 0.9154433333333334,\n",
-      "                                                 0.9015288888888889,\n",
-      "                                                 0.8813138888888888,\n",
-      "                                                 0.8597738888888888,\n",
-      "                                                 0.9020194444444445,\n",
-      "                                                 0.9287488888888888,\n",
-      "                                                 0.8618527777777778,\n",
-      "                                                 0.9171522222222223,\n",
-      "                                                 0.8757566666666666,\n",
-      "                                                 0.9054572222222222,\n",
-      "                                                 0.9153788888888889,\n",
-      "                                                 0.8185816666666667,\n",
-      "                                                 0.9260683333333334,\n",
-      "                                                 0.8471155555555555,\n",
-      "                                                 0.9500361111111111,\n",
-      "                                                 0.83239,\n",
-      "                                                 0.8784616666666667,\n",
-      "                                                 0.863745,\n",
-      "                                                 0.9020511111111111,\n",
-      "                                                 0.85144],\n",
-      "                                    'avg_accuracy': 0.8767142777777777},\n",
-      "                      'inference_analyze': {'bs': 32,\n",
-      "                                            'gpus': 1,\n",
-      "                                            'nodes': 1,\n",
-      "                                            'number': 100,\n",
-      "                                            'time_per_inference': '1.45'},\n",
-      "                      'name': 'cloudmask',\n",
-      "                      'training': {'accuracy': 0.8849480152130127,\n",
-      "                                   'batch_size': 32,\n",
-      "                                   'crop_size': 80,\n",
-      "                                   'history': {'accuracy': [0.7971559762954712,\n",
-      "                                                            0.8036167621612549,\n",
-      "                                                            0.8068755269050598,\n",
-      "                                                            0.8164012432098389,\n",
-      "                                                            0.8104918003082275,\n",
-      "                                                            0.8150579929351807,\n",
-      "                                                            0.809596598148346,\n",
-      "                                                            0.8198840618133545,\n",
-      "                                                            0.8231695294380188,\n",
-      "                                                            0.8217014074325562,\n",
-      "                                                            0.8314782977104187,\n",
-      "                                                            0.8419565558433533,\n",
-      "                                                            0.8467669486999512,\n",
-      "                                                            0.851389467716217,\n",
-      "                                                            0.8547239303588867,\n",
-      "                                                            0.8569662570953369,\n",
-      "                                                            0.8529560565948486,\n",
-      "                                                            0.8562770485877991,\n",
-      "                                                            0.8571555018424988,\n",
-      "                                                            0.8612533211708069,\n",
-      "                                                            0.86093670129776,\n",
-      "                                                            0.8633123636245728,\n",
-      "                                                            0.8648520112037659,\n",
-      "                                                            0.8672714233398438,\n",
-      "                                                            0.86800217628479,\n",
-      "                                                            0.8684883713722229,\n",
-      "                                                            0.8689928650856018,\n",
-      "                                                            0.8708153367042542,\n",
-      "                                                            0.8728460073471069,\n",
-      "                                                            0.8720263242721558,\n",
-      "                                                            0.8743911981582642,\n",
-      "                                                            0.8713085055351257,\n",
-      "                                                            0.8731305599212646,\n",
-      "                                                            0.875930905342102,\n",
-      "                                                            0.8755276203155518,\n",
-      "                                                            0.8750144243240356,\n",
-      "                                                            0.8759661912918091,\n",
-      "                                                            0.8746057748794556,\n",
-      "                                                            0.8717871308326721,\n",
-      "                                                            0.8780733942985535,\n",
-      "                                                            0.8778254985809326,\n",
-      "                                                            0.8779628872871399,\n",
-      "                                                            0.8771510124206543,\n",
-      "                                                            0.8768048882484436,\n",
-      "                                                            0.8762545585632324,\n",
-      "                                                            0.8780935406684875,\n",
-      "                                                            0.878029465675354,\n",
-      "                                                            0.8791071176528931,\n",
-      "                                                            0.8806102275848389,\n",
-      "                                                            0.8794829845428467,\n",
-      "                                                            0.8801596760749817,\n",
-      "                                                            0.8821283578872681,\n",
-      "                                                            0.8824746608734131,\n",
-      "                                                            0.8801916837692261,\n",
-      "                                                            0.8809465169906616,\n",
-      "                                                            0.882117748260498,\n",
-      "                                                            0.8807238340377808,\n",
-      "                                                            0.8821170926094055,\n",
-      "                                                            0.8823124766349792,\n",
-      "                                                            0.8817462921142578,\n",
-      "                                                            0.882713258266449,\n",
-      "                                                            0.8839292526245117,\n",
-      "                                                            0.8827219605445862,\n",
-      "                                                            0.8825761675834656,\n",
-      "                                                            0.8849943280220032,\n",
-      "                                                            0.8832786083221436,\n",
-      "                                                            0.8848116397857666,\n",
-      "                                                            0.8849825263023376,\n",
-      "                                                            0.8855748772621155,\n",
-      "                                                            0.8849480152130127],\n",
-      "                                               'loss': [0.43057680130004883,\n",
-      "                                                        0.3868241310119629,\n",
-      "                                                        0.37653210759162903,\n",
-      "                                                        0.37191805243492126,\n",
-      "                                                        0.37259817123413086,\n",
-      "                                                        0.3699168264865875,\n",
-      "                                                        0.3650084137916565,\n",
-      "                                                        0.3572779893875122,\n",
-      "                                                        0.3539367914199829,\n",
-      "                                                        0.3640691936016083,\n",
-      "                                                        0.34810689091682434,\n",
-      "                                                        0.3341286778450012,\n",
-      "                                                        0.3274889886379242,\n",
-      "                                                        0.3194959759712219,\n",
-      "                                                        0.31373706459999084,\n",
-      "                                                        0.3107760548591614,\n",
-      "                                                        0.3144351541996002,\n",
-      "                                                        0.31273719668388367,\n",
-      "                                                        0.30824530124664307,\n",
-      "                                                        0.3051598370075226,\n",
-      "                                                        0.30479755997657776,\n",
-      "                                                        0.3008135259151459,\n",
-      "                                                        0.2963641881942749,\n",
-      "                                                        0.2948143780231476,\n",
-      "                                                        0.2933521568775177,\n",
-      "                                                        0.29491209983825684,\n",
-      "                                                        0.2907404899597168,\n",
-      "                                                        0.288029283285141,\n",
-      "                                                        0.2854553163051605,\n",
-      "                                                        0.28605392575263977,\n",
-      "                                                        0.2832964360713959,\n",
-      "                                                        0.2861807644367218,\n",
-      "                                                        0.2820378541946411,\n",
-      "                                                        0.2798045575618744,\n",
-      "                                                        0.28097864985466003,\n",
-      "                                                        0.2803114056587219,\n",
-      "                                                        0.27963513135910034,\n",
-      "                                                        0.28156983852386475,\n",
-      "                                                        0.2849934697151184,\n",
-      "                                                        0.27691885828971863,\n",
-      "                                                        0.2761290371417999,\n",
-      "                                                        0.27344226837158203,\n",
-      "                                                        0.2756351828575134,\n",
-      "                                                        0.2761900722980499,\n",
-      "                                                        0.2755855321884155,\n",
-      "                                                        0.2724992036819458,\n",
-      "                                                        0.2715945243835449,\n",
-      "                                                        0.2726806700229645,\n",
-      "                                                        0.2702755928039551,\n",
-      "                                                        0.2707327604293823,\n",
-      "                                                        0.26914530992507935,\n",
-      "                                                        0.2672998607158661,\n",
-      "                                                        0.2653893530368805,\n",
-      "                                                        0.2707858979701996,\n",
-      "                                                        0.26888710260391235,\n",
-      "                                                        0.26429009437561035,\n",
-      "                                                        0.26705700159072876,\n",
-      "                                                        0.26433271169662476,\n",
-      "                                                        0.2644907832145691,\n",
-      "                                                        0.2654668688774109,\n",
-      "                                                        0.2637845575809479,\n",
-      "                                                        0.2615320086479187,\n",
-      "                                                        0.2629801630973816,\n",
-      "                                                        0.2634650468826294,\n",
-      "                                                        0.25998246669769287,\n",
-      "                                                        0.2621432840824127,\n",
-      "                                                        0.2603616416454315,\n",
-      "                                                        0.260457843542099,\n",
-      "                                                        0.25872427225112915,\n",
-      "                                                        0.2601945102214813],\n",
-      "                                               'val_accuracy': [0.7767935395240784,\n",
-      "                                                                0.775540292263031,\n",
-      "                                                                0.8011550903320312,\n",
-      "                                                                0.7952496409416199,\n",
-      "                                                                0.8039715886116028,\n",
-      "                                                                0.7966544032096863,\n",
-      "                                                                0.7960512042045593,\n",
-      "                                                                0.8099052309989929,\n",
-      "                                                                0.8374516367912292,\n",
-      "                                                                0.8269608020782471,\n",
-      "                                                                0.8236475586891174,\n",
-      "                                                                0.8257650136947632,\n",
-      "                                                                0.8420281410217285,\n",
-      "                                                                0.8540118336677551,\n",
-      "                                                                0.8504691123962402,\n",
-      "                                                                0.8509190082550049,\n",
-      "                                                                0.8481219410896301,\n",
-      "                                                                0.8540127873420715,\n",
-      "                                                                0.8348009586334229,\n",
-      "                                                                0.8534428477287292,\n",
-      "                                                                0.8578217029571533,\n",
-      "                                                                0.8559457063674927,\n",
-      "                                                                0.8581292033195496,\n",
-      "                                                                0.8632622361183167,\n",
-      "                                                                0.8676828145980835,\n",
-      "                                                                0.8639155626296997,\n",
-      "                                                                0.8643791675567627,\n",
-      "                                                                0.8679339289665222,\n",
-      "                                                                0.8617703318595886,\n",
-      "                                                                0.863847553730011,\n",
-      "                                                                0.8682820796966553,\n",
-      "                                                                0.8685378432273865,\n",
-      "                                                                0.8699041604995728,\n",
-      "                                                                0.8732900619506836,\n",
-      "                                                                0.8731300830841064,\n",
-      "                                                                0.8718216419219971,\n",
-      "                                                                0.8756145238876343,\n",
-      "                                                                0.8728165030479431,\n",
-      "                                                                0.8722469210624695,\n",
-      "                                                                0.8747484683990479,\n",
-      "                                                                0.8692426681518555,\n",
-      "                                                                0.8694444894790649,\n",
-      "                                                                0.873407244682312,\n",
-      "                                                                0.8717463612556458,\n",
-      "                                                                0.8729797601699829,\n",
-      "                                                                0.8757378458976746,\n",
-      "                                                                0.8733677268028259,\n",
-      "                                                                0.8770347237586975,\n",
-      "                                                                0.8732863664627075,\n",
-      "                                                                0.872306227684021,\n",
-      "                                                                0.8756421208381653,\n",
-      "                                                                0.8771953582763672,\n",
-      "                                                                0.8779804110527039,\n",
-      "                                                                0.8791961073875427,\n",
-      "                                                                0.8724019527435303,\n",
-      "                                                                0.8737383484840393,\n",
-      "                                                                0.8651539087295532,\n",
-      "                                                                0.8782283663749695,\n",
-      "                                                                0.8751494288444519,\n",
-      "                                                                0.8749458193778992,\n",
-      "                                                                0.8758198618888855,\n",
-      "                                                                0.8772931098937988,\n",
-      "                                                                0.8728049397468567,\n",
-      "                                                                0.8781270384788513,\n",
-      "                                                                0.8808680176734924,\n",
-      "                                                                0.8775757551193237,\n",
-      "                                                                0.8762949109077454,\n",
-      "                                                                0.8789771795272827,\n",
-      "                                                                0.8788212537765503,\n",
-      "                                                                0.8799592852592468],\n",
-      "                                               'val_loss': [0.45389050245285034,\n",
-      "                                                            0.43061962723731995,\n",
-      "                                                            0.4013422429561615,\n",
-      "                                                            0.42080938816070557,\n",
-      "                                                            0.395188570022583,\n",
-      "                                                            0.39979249238967896,\n",
-      "                                                            0.39877310395240784,\n",
-      "                                                            0.4145936369895935,\n",
-      "                                                            0.3784829080104828,\n",
-      "                                                            0.3892180621623993,\n",
-      "                                                            0.3713390827178955,\n",
-      "                                                            0.36597907543182373,\n",
-      "                                                            0.34918883442878723,\n",
-      "                                                            0.3361833393573761,\n",
-      "                                                            0.33424654603004456,\n",
-      "                                                            0.34345462918281555,\n",
-      "                                                            0.3452184796333313,\n",
-      "                                                            0.33300939202308655,\n",
-      "                                                            0.3561151325702667,\n",
-      "                                                            0.33258056640625,\n",
-      "                                                            0.3100963532924652,\n",
-      "                                                            0.32722580432891846,\n",
-      "                                                            0.31929418444633484,\n",
-      "                                                            0.3067832589149475,\n",
-      "                                                            0.30295196175575256,\n",
-      "                                                            0.3147284686565399,\n",
-      "                                                            0.3125918507575989,\n",
-      "                                                            0.3050999939441681,\n",
-      "                                                            0.31659048795700073,\n",
-      "                                                            0.302707701921463,\n",
-      "                                                            0.30327436327934265,\n",
-      "                                                            0.3065930902957916,\n",
-      "                                                            0.29703590273857117,\n",
-      "                                                            0.2960212826728821,\n",
-      "                                                            0.29190030694007874,\n",
-      "                                                            0.2948874533176422,\n",
-      "                                                            0.2921930253505707,\n",
-      "                                                            0.2942402958869934,\n",
-      "                                                            0.2906084656715393,\n",
-      "                                                            0.2935509979724884,\n",
-      "                                                            0.2984870672225952,\n",
-      "                                                            0.2999739646911621,\n",
-      "                                                            0.29295942187309265,\n",
-      "                                                            0.2930569648742676,\n",
-      "                                                            0.290778249502182,\n",
-      "                                                            0.28762853145599365,\n",
-      "                                                            0.2904091477394104,\n",
-      "                                                            0.2816406786441803,\n",
-      "                                                            0.29552775621414185,\n",
-      "                                                            0.2868202030658722,\n",
-      "                                                            0.28450778126716614,\n",
-      "                                                            0.28301164507865906,\n",
-      "                                                            0.2807203531265259,\n",
-      "                                                            0.28039175271987915,\n",
-      "                                                            0.2878774106502533,\n",
-      "                                                            0.28569889068603516,\n",
-      "                                                            0.3039056658744812,\n",
-      "                                                            0.2813010513782501,\n",
-      "                                                            0.28331518173217773,\n",
-      "                                                            0.2812541127204895,\n",
-      "                                                            0.2832031548023224,\n",
-      "                                                            0.2804718613624573,\n",
-      "                                                            0.28866803646087646,\n",
-      "                                                            0.2779247462749481,\n",
-      "                                                            0.27656784653663635,\n",
-      "                                                            0.2857021391391754,\n",
-      "                                                            0.28377673029899597,\n",
-      "                                                            0.27749380469322205,\n",
-      "                                                            0.2795368432998657,\n",
-      "                                                            0.2793876826763153]},\n",
-      "                                   'learning_rate': 0.001,\n",
-      "                                   'loss': 0.2601945102214813,\n",
-      "                                   'samples': 970,\n",
-      "                                   'val_accuracy': 0.8799592852592468,\n",
-      "                                   'val_loss': 0.2793876826763153}}},\n",
-      " 'seed': 1234,\n",
-      " 'total': 10739.745,\n",
-      " 'train_split': 0.8,\n",
-      " 'training': 10590.6}\n"
-     ]
-    }
-   ],
+   "execution_count": null,
+   "outputs": [],
    "source": [
-    "pprint(result[0])\n"
+    "pprint(result[0])\n",
+    "\n",
+    "result_data = result.copy()\n"
    ],
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:10:19.536013631Z",
-     "start_time": "2023-10-06T14:10:19.272824364Z"
+     "end_time": "2023-10-06T17:09:24.520962174Z",
+     "start_time": "2023-10-06T17:09:24.520948183Z"
     }
    },
    "id": "2e365fc1c81f7e14"
@@ -536,30 +112,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 140,
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Index(['name', 'total', 'training', 'loaddata', 'inference', 'directive',\n",
-      "       'gpu_count', 'cpu_num', 'mem', 'repeat', 'epoch', 'seed',\n",
-      "       'learning_rate', 'batch_size', 'train_split', 'clip_offset', 'no_cache',\n",
-      "       'nodes', 'gpu', 'early_stoppage_patience', 'early_stoppage',\n",
-      "       'card_name', 'result', 'training_on_mutiple_GPU'],\n",
-      "      dtype='object')\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": "        total   training  loaddata  inference directive  repeat  epoch  \\\n0   10739.745  10590.600     4.118    144.679      v100       5     70   \n1    7699.156   7554.603     2.069    142.281      v100      10     50   \n2    1378.010   1232.831     2.607    142.262  a100-dgx       7     10   \n3    2641.984   2436.345     3.069    202.318      v100       9     10   \n4     659.622    514.509     2.524    142.282  a100-dgx       7      2   \n..        ...        ...       ...        ...       ...     ...    ...   \n79    827.305    682.822     1.940    142.251      v100       8      2   \n80   2103.479   1897.323     3.527    202.387      v100       6     10   \n81   2773.257   2568.251     2.362    202.304      v100      10     10   \n82   5248.492   5041.956     3.780    202.388      v100       6     30   \n83    593.296    448.220     2.499    142.264  a100-dgx       8      1   \n\n                                               result  \n0   {'namespace': '', 'time_ms': 1696595324526, 'e...  \n1   {'namespace': '', 'time_ms': 1696597201377, 'e...  \n2   {'namespace': '', 'time_ms': 1696598288235, 'e...  \n3   {'namespace': '', 'time_ms': 1696589294025, 'e...  \n4   {'namespace': '', 'time_ms': 1696596884493, 'e...  \n..                                                ...  \n79  {'namespace': '', 'time_ms': 1696585438781, 'e...  \n80  {'namespace': '', 'time_ms': 1696586688202, 'e...  \n81  {'namespace': '', 'time_ms': 1696592095380, 'e...  \n82  {'namespace': '', 'time_ms': 1696589833357, 'e...  \n83  {'namespace': '', 'time_ms': 1696599376624, 'e...  \n\n[84 rows x 8 columns]",
-      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>total</th>\n      <th>training</th>\n      <th>loaddata</th>\n      <th>inference</th>\n      <th>directive</th>\n      <th>repeat</th>\n      <th>epoch</th>\n      <th>result</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>10739.745</td>\n      <td>10590.600</td>\n      <td>4.118</td>\n      <td>144.679</td>\n      <td>v100</td>\n      <td>5</td>\n      <td>70</td>\n      <td>{'namespace': '', 'time_ms': 1696595324526, 'e...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>7699.156</td>\n      <td>7554.603</td>\n      <td>2.069</td>\n      <td>142.281</td>\n      <td>v100</td>\n      <td>10</td>\n      <td>50</td>\n      <td>{'namespace': '', 'time_ms': 1696597201377, 'e...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>1378.010</td>\n      <td>1232.831</td>\n      <td>2.607</td>\n      <td>142.262</td>\n      <td>a100-dgx</td>\n      <td>7</td>\n      <td>10</td>\n      <td>{'namespace': '', 'time_ms': 1696598288235, 'e...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2641.984</td>\n      <td>2436.345</td>\n      <td>3.069</td>\n      <td>202.318</td>\n      <td>v100</td>\n      <td>9</td>\n      <td>10</td>\n      <td>{'namespace': '', 'time_ms': 1696589294025, 'e...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>659.622</td>\n      <td>514.509</td>\n      <td>2.524</td>\n      <td>142.282</td>\n      <td>a100-dgx</td>\n      <td>7</td>\n      <td>2</td>\n      <td>{'namespace': '', 'time_ms': 1696596884493, 'e...</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>79</th>\n      <td>827.305</td>\n      <td>682.822</td>\n      <td>1.940</td>\n      <td>142.251</td>\n      <td>v100</td>\n      <td>8</td>\n      <td>2</td>\n      <td>{'namespace': '', 'time_ms': 1696585438781, 'e...</td>\n    </tr>\n    <tr>\n      <th>80</th>\n      <td>2103.479</td>\n      <td>1897.323</td>\n      <td>3.527</td>\n      <td>202.387</td>\n      <td>v100</td>\n      <td>6</td>\n      <td>10</td>\n      <td>{'namespace': '', 'time_ms': 1696586688202, 'e...</td>\n    </tr>\n    <tr>\n      <th>81</th>\n      <td>2773.257</td>\n      <td>2568.251</td>\n      <td>2.362</td>\n      <td>202.304</td>\n      <td>v100</td>\n      <td>10</td>\n      <td>10</td>\n      <td>{'namespace': '', 'time_ms': 1696592095380, 'e...</td>\n    </tr>\n    <tr>\n      <th>82</th>\n      <td>5248.492</td>\n      <td>5041.956</td>\n      <td>3.780</td>\n      <td>202.388</td>\n      <td>v100</td>\n      <td>6</td>\n      <td>30</td>\n      <td>{'namespace': '', 'time_ms': 1696589833357, 'e...</td>\n    </tr>\n    <tr>\n      <th>83</th>\n      <td>593.296</td>\n      <td>448.220</td>\n      <td>2.499</td>\n      <td>142.264</td>\n      <td>a100-dgx</td>\n      <td>8</td>\n      <td>1</td>\n      <td>{'namespace': '', 'time_ms': 1696599376624, 'e...</td>\n    </tr>\n  </tbody>\n</table>\n<p>84 rows × 8 columns</p>\n</div>"
-     },
-     "execution_count": 140,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": null,
+   "outputs": [],
    "source": [
     "df = pd.DataFrame(result)\n",
     "print(df.columns)\n",
@@ -575,15 +129,15 @@
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:11:06.901371352Z",
-     "start_time": "2023-10-06T14:11:06.860167353Z"
+     "end_time": "2023-10-06T17:09:24.521125617Z",
+     "start_time": "2023-10-06T17:09:24.521055635Z"
     }
    },
    "id": "c4b709779e248008"
   },
   {
    "cell_type": "code",
-   "execution_count": 141,
+   "execution_count": null,
    "outputs": [],
    "source": [
     "def save_plot_to_multiple_formats(plot, filename_without_extension):\n",
@@ -616,25 +170,15 @@
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:11:12.966208316Z",
-     "start_time": "2023-10-06T14:11:12.950017915Z"
+     "start_time": "2023-10-06T17:09:24.521088356Z"
     }
    },
    "id": "c6f2b393d28ea51"
   },
   {
    "cell_type": "code",
-   "execution_count": 142,
-   "outputs": [
-    {
-     "data": {
-      "text/plain": "<Figure size 1000x600 with 1 Axes>",
-      "image/png": "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\n"
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": null,
+   "outputs": [],
    "source": [
     "sns.set(style=\"whitegrid\")\n",
     "\n",
@@ -653,25 +197,15 @@
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:11:14.251775968Z",
-     "start_time": "2023-10-06T14:11:13.934435281Z"
+     "start_time": "2023-10-06T17:09:24.521114947Z"
     }
    },
    "id": "fd537299d421c3a2"
   },
   {
    "cell_type": "code",
-   "execution_count": 143,
-   "outputs": [
-    {
-     "data": {
-      "text/plain": "<Figure size 1000x600 with 1 Axes>",
-      "image/png": "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\n"
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": null,
+   "outputs": [],
    "source": [
     "\n",
     "# Create a line plot using Seaborn\n",
@@ -694,25 +228,16 @@
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:11:14.840796794Z",
-     "start_time": "2023-10-06T14:11:14.567232097Z"
+     "end_time": "2023-10-06T17:09:24.521251719Z",
+     "start_time": "2023-10-06T17:09:24.521160877Z"
     }
    },
    "id": "79834fd2012b6d7d"
   },
   {
    "cell_type": "code",
-   "execution_count": 144,
-   "outputs": [
-    {
-     "data": {
-      "text/plain": "<Figure size 1000x600 with 1 Axes>",
-      "image/png": "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\n"
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": null,
+   "outputs": [],
    "source": [
     "\n",
     "# Set the style of the plot\n",
@@ -733,26 +258,639 @@
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:11:15.389931328Z",
-     "start_time": "2023-10-06T14:11:15.259522846Z"
+     "start_time": "2023-10-06T17:09:24.521236449Z"
     }
    },
    "id": "331740827b3dd78a"
   },
   {
    "cell_type": "code",
-   "execution_count": 144,
+   "execution_count": null,
    "outputs": [],
    "source": [],
    "metadata": {
     "collapsed": false,
     "ExecuteTime": {
-     "end_time": "2023-10-06T14:11:16.208068597Z",
-     "start_time": "2023-10-06T14:11:16.204620709Z"
+     "end_time": "2023-10-06T17:09:24.521998123Z",
+     "start_time": "2023-10-06T17:09:24.521284010Z"
     }
    },
    "id": "b61c85fd0a4b6771"
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "\n",
+    "\n",
+    "# Filter rows with valid directive names\n",
+    "df = df[df['directive'].notna()]\n",
+    "\n",
+    "# Specify the directive you want to select\n",
+    "selected_directive = 'a100-dgx'  # Replace with your desired directive\n",
+    "selected_epoch = 100\n",
+    "\n",
+    "# Filter the DataFrame for the selected directive and epoch\n",
+    "filtered_df = df[(df['directive'] == selected_directive) & (df['result'].apply(lambda x: len(x['value']['training']['history']['accuracy'])) >= selected_epoch)]\n",
+    "\n",
+    "# Create an empty list to store accuracy histories\n",
+    "accuracy_histories = []\n",
+    "\n",
+    "# Iterate through the filtered DataFrame and extract accuracy histories\n",
+    "for index, row in filtered_df.iterrows():\n",
+    "    result = row['result']\n",
+    "    accuracy_history = result['value']['training']['history']['accuracy']\n",
+    "    accuracy_histories.append(accuracy_history[:selected_epoch])\n",
+    "\n",
+    "# Create a Seaborn line plot for all matching directives\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "sns.set_style(\"darkgrid\")\n",
+    "\n",
+    "for idx, accuracy_history in enumerate(accuracy_histories):\n",
+    "    sns.lineplot(x=range(1, selected_epoch + 1), y=accuracy_history, label=f'Result {idx + 1}')\n",
+    "\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.title(f'Accuracy History for Directive: {selected_directive} at Epoch {selected_epoch}')\n",
+    "plt.legend(title='Results')\n",
+    "plt.show()\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521310280Z"
+    }
+   },
+   "id": "32f8b5abdceb8cd7"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# Filter rows with valid directive names\n",
+    "df = df[df['directive'].notna()]\n",
+    "\n",
+    "# Specify the directive you want to plot\n",
+    "selected_directive = 'v100'  # Replace with your desired directive\n",
+    "\n",
+    "# Filter the DataFrame for the selected directive\n",
+    "directive_df = df[df['directive'] == selected_directive]\n",
+    "\n",
+    "# Extract all accuracy histories for the selected directive\n",
+    "accuracy_histories = directive_df['result'].apply(lambda x: x['value']['training']['history']['accuracy'])\n",
+    "\n",
+    "# Create a Seaborn line plot for all accuracy histories\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "sns.set(style=\"whitegrid\")\n",
+    "\n",
+    "for idx, accuracy_history in enumerate(accuracy_histories):\n",
+    "    x_values = list(range(len(accuracy_history)))\n",
+    "    sns.lineplot(x=x_values, y=accuracy_history, label=f'Result {idx + 1}')\n",
+    "\n",
+    "plt.xlabel('Epoch (Index)')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.title(f'Accuracy History for Directive: {selected_directive}')\n",
+    "plt.legend(title='Results')\n",
+    "plt.show()\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521337441Z"
+    }
+   },
+   "id": "507603c8d8759e8f"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "\n",
+    "\n",
+    "# Filter rows with valid directive names\n",
+    "df = df[df['directive'].notna()]\n",
+    "\n",
+    "# Specify the directive you want to select\n",
+    "selected_directive = 'v100'  # Replace with your desired directive\n",
+    "selected_epoch = 100\n",
+    "\n",
+    "# Filter the DataFrame for the selected directive and epoch\n",
+    "filtered_df = df[(df['directive'] == selected_directive) & (df['result'].apply(lambda x: len(x['value']['training']['history']['accuracy'])) >= selected_epoch)]\n",
+    "\n",
+    "# Create an empty list to store accuracy histories\n",
+    "accuracy_histories = []\n",
+    "\n",
+    "# Iterate through the filtered DataFrame and extract accuracy histories\n",
+    "for index, row in filtered_df.iterrows():\n",
+    "    result = row['result']\n",
+    "    accuracy_history = result['value']['training']['history']['accuracy']\n",
+    "    accuracy_histories.append(accuracy_history[:selected_epoch])\n",
+    "\n",
+    "# Create a Seaborn line plot for all matching directives\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "sns.set(style=\"whitegrid\")\n",
+    "\n",
+    "for idx, accuracy_history in enumerate(accuracy_histories):\n",
+    "    sns.lineplot(x=range(1, selected_epoch + 1), y=accuracy_history, label=f'Result {idx + 1}')\n",
+    "\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.title(f'Accuracy History for Directive: {selected_directive} at Epoch {selected_epoch}')\n",
+    "plt.legend(title='Results')\n",
+    "plt.show()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521355671Z"
+    }
+   },
+   "id": "e09acd37de6e1387"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# Filter rows with valid directive names\n",
+    "df = df[df['directive'].notna()]\n",
+    "\n",
+    "# Specify the directive you want to select\n",
+    "selected_directive = 'a100-dgx'  # Replace with your desired directive\n",
+    "selected_epoch = 100\n",
+    "\n",
+    "# Filter the DataFrame for the selected directive and epoch\n",
+    "filtered_df = df[(df['directive'] == selected_directive) & (df['result'].apply(lambda x: len(x['value']['training']['history']['accuracy'])) >= selected_epoch)]\n",
+    "\n",
+    "# Create an empty list to store accuracy histories\n",
+    "accuracy_histories = []\n",
+    "\n",
+    "# Iterate through the filtered DataFrame and extract accuracy histories\n",
+    "for index, row in filtered_df.iterrows():\n",
+    "    result = row['result']\n",
+    "    accuracy_history = result['value']['training']['history']['accuracy']\n",
+    "    accuracy_histories.append(accuracy_history[:selected_epoch])\n",
+    "\n",
+    "# Create a Seaborn line plot for all matching directives without a legend\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "sns.set(style=\"whitegrid\")\n",
+    "\n",
+    "for idx, accuracy_history in enumerate(accuracy_histories):\n",
+    "    sns.lineplot(x=range(1, selected_epoch + 1), y=accuracy_history, label=f'Result {idx + 1}')\n",
+    "\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.title(f'Accuracy History for Directive: {selected_directive} at Epoch {selected_epoch}')\n",
+    "plt.show()\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521373032Z"
+    }
+   },
+   "id": "df31aa6725841098"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "def plot_history(data, epoch=100, directives=[\"a100-dgx\",\"v100\"], value=\"accuracy\"):\n",
+    "    df = data[data['directive'].notna()].copy()\n",
+    "    \n",
+    "    # Specify the list of directives you want to select\n",
+    "    selected_directives = directives # ['a100-dgx', 'v100']  # Replace with your desired directives\n",
+    "    selected_epoch = epoch\n",
+    "    \n",
+    "    # Create an empty list to store accuracy histories for each directive\n",
+    "    accuracy_histories_dict = {directive: [] for directive in selected_directives}\n",
+    "    \n",
+    "    \n",
+    "    \n",
+    "    # Iterate through the DataFrame and extract accuracy histories for selected directives\n",
+    "    for directive in selected_directives:\n",
+    "        filtered_df = df[(df['directive'] == directive) & (df['result'].apply(lambda x: len(x['value']['training']['history'][value])) >= selected_epoch)]\n",
+    "        \n",
+    "        for index, row in filtered_df.iterrows():\n",
+    "            result = row['result']\n",
+    "            accuracy_history = result['value']['training']['history'][value]\n",
+    "            accuracy_histories_dict[directive].append(accuracy_history[:selected_epoch])\n",
+    "    \n",
+    "    # Create Seaborn line plots for each directive with a legend\n",
+    "    plt.figure(figsize=(12, 6))\n",
+    "    sns.set(style=\"whitegrid\")\n",
+    "    \n",
+    "    for directive, accuracy_histories in accuracy_histories_dict.items():\n",
+    "        for idx, accuracy_history in enumerate(accuracy_histories):\n",
+    "            sns.lineplot(x=range(1, selected_epoch + 1), y=accuracy_history, label=f'{directive} - Result {idx + 1}')\n",
+    "    \n",
+    "    plt.xlabel('Epoch')\n",
+    "    plt.ylabel(value.capitalize())\n",
+    "    plt.title(f'Accuracy History for Directives at Epoch {selected_epoch}')\n",
+    "    plt.legend()\n",
+    "    plt.show()\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521392182Z"
+    }
+   },
+   "id": "e6ae2cfcf32b1e21"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "\n",
+    "value = 'accuracy'\n",
+    "# Filter rows with valid directive names\n",
+    "df = df[df['directive'].notna()]\n",
+    "\n",
+    "plot_history(df, directives=['a100-dgx', 'v100'], epoch=100, value=\"accuracy\")\n",
+    "\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521409922Z"
+    }
+   },
+   "id": "cac8fff44eb01905"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "for k in result['value']['training']['history'].keys():\n",
+    "    plot_history(df, directives=['a100-dgx', 'v100'], epoch=100, value=k)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521427083Z"
+    }
+   },
+   "id": "90e0869b186fbeaf"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "result['value']['training']['history'].keys()\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521464853Z"
+    }
+   },
+   "id": "36d19ff42b3c4de9"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "\n",
+    "# Filter rows with valid directive names\n",
+    "df = df[df['directive'].notna()]\n",
+    "\n",
+    "# Specify the list of directives you want to select\n",
+    "selected_directives = ['a100-dgx', 'v100']  # Replace with your desired directives\n",
+    "selected_epoch = 100\n",
+    "\n",
+    "# Create a dictionary to store accuracy histories and best accuracies for each directive\n",
+    "directive_data = {}\n",
+    "\n",
+    "# Iterate through the DataFrame and extract accuracy histories for selected directives\n",
+    "for directive in selected_directives:\n",
+    "    filtered_df = df[(df['directive'] == directive) & (df['result'].apply(lambda x: len(x['value']['training']['history']['accuracy'])) >= selected_epoch)]\n",
+    "    \n",
+    "    directive_accuracies = []\n",
+    "    \n",
+    "    for index, row in filtered_df.iterrows():\n",
+    "        result = row['result']\n",
+    "        accuracy_history = result['value']['training']['history']['accuracy']\n",
+    "        directive_accuracies.append(accuracy_history[:selected_epoch])\n",
+    "    \n",
+    "    # Calculate the best accuracy for this directive\n",
+    "    best_accuracy = max([accuracy[-1] for accuracy in directive_accuracies])\n",
+    "    \n",
+    "    directive_data[directive] = {\n",
+    "        'accuracies': directive_accuracies,\n",
+    "        'best_accuracy': best_accuracy\n",
+    "    }\n",
+    "\n",
+    "# Sort directives by best accuracy in descending order\n",
+    "sorted_directives = sorted(directive_data.keys(), key=lambda x: directive_data[x]['best_accuracy'], reverse=True)\n",
+    "\n",
+    "# Create Seaborn line plots for each directive with a legend\n",
+    "plt.figure(figsize=(12, 6))\n",
+    "sns.set_style(\"darkgrid\")\n",
+    "\n",
+    "for directive in sorted_directives:\n",
+    "    for idx, accuracy_history in enumerate(directive_data[directive]['accuracies']):\n",
+    "        best_accuracy = directive_data[directive]['best_accuracy']\n",
+    "        sns.lineplot(x=range(1, selected_epoch + 1), y=accuracy_history, label=f'{directive} - Result {idx + 1} (Best: {best_accuracy:.4f})')\n",
+    "\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.title(f'Accuracy History for Directives at Epoch {selected_epoch}')\n",
+    "plt.legend()\n",
+    "plt.show()\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-10-06T17:09:24.521490194Z"
+    }
+   },
+   "id": "3f607b99bc95f595"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 290,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{'name': 'result-thf2bn-53854471.out', 'total': 10739.745, 'training': 10590.6, 'loaddata': 4.118, 'inference': 144.679, 'directive': 'v100', 'gpu_count': '1', 'cpu_num': 1, 'mem': '64GB', 'repeat': '5', 'epoch': '70', 'seed': 1234, 'learning_rate': 0.001, 'batch_size': 32, 'train_split': 0.8, 'clip_offset': 15, 'no_cache': False, 'nodes': 1, 'gpu': 1, 'early_stoppage_patience': '25', 'early_stoppage': 'False', 'card_name': 'v100', 'result': {'value': {'name': 'cloudmask', 'training': {'samples': 970, 'accuracy': 0.8849480152130127, 'loss': 0.2601945102214813, 'val_loss': 0.2793876826763153, 'val_accuracy': 0.8799592852592468, 'history': {'accuracy': [0.7971559762954712, 0.8036167621612549, 0.8068755269050598, 0.8164012432098389, 0.8104918003082275, 0.8150579929351807, 0.809596598148346, 0.8198840618133545, 0.8231695294380188, 0.8217014074325562, 0.8314782977104187, 0.8419565558433533, 0.8467669486999512, 0.851389467716217, 0.8547239303588867, 0.8569662570953369, 0.8529560565948486, 0.8562770485877991, 0.8571555018424988, 0.8612533211708069, 0.86093670129776, 0.8633123636245728, 0.8648520112037659, 0.8672714233398438, 0.86800217628479, 0.8684883713722229, 0.8689928650856018, 0.8708153367042542, 0.8728460073471069, 0.8720263242721558, 0.8743911981582642, 0.8713085055351257, 0.8731305599212646, 0.875930905342102, 0.8755276203155518, 0.8750144243240356, 0.8759661912918091, 0.8746057748794556, 0.8717871308326721, 0.8780733942985535, 0.8778254985809326, 0.8779628872871399, 0.8771510124206543, 0.8768048882484436, 0.8762545585632324, 0.8780935406684875, 0.878029465675354, 0.8791071176528931, 0.8806102275848389, 0.8794829845428467, 0.8801596760749817, 0.8821283578872681, 0.8824746608734131, 0.8801916837692261, 0.8809465169906616, 0.882117748260498, 0.8807238340377808, 0.8821170926094055, 0.8823124766349792, 0.8817462921142578, 0.882713258266449, 0.8839292526245117, 0.8827219605445862, 0.8825761675834656, 0.8849943280220032, 0.8832786083221436, 0.8848116397857666, 0.8849825263023376, 0.8855748772621155, 0.8849480152130127], 'loss': [0.43057680130004883, 0.3868241310119629, 0.37653210759162903, 0.37191805243492126, 0.37259817123413086, 0.3699168264865875, 0.3650084137916565, 0.3572779893875122, 0.3539367914199829, 0.3640691936016083, 0.34810689091682434, 0.3341286778450012, 0.3274889886379242, 0.3194959759712219, 0.31373706459999084, 0.3107760548591614, 0.3144351541996002, 0.31273719668388367, 0.30824530124664307, 0.3051598370075226, 0.30479755997657776, 0.3008135259151459, 0.2963641881942749, 0.2948143780231476, 0.2933521568775177, 0.29491209983825684, 0.2907404899597168, 0.288029283285141, 0.2854553163051605, 0.28605392575263977, 0.2832964360713959, 0.2861807644367218, 0.2820378541946411, 0.2798045575618744, 0.28097864985466003, 0.2803114056587219, 0.27963513135910034, 0.28156983852386475, 0.2849934697151184, 0.27691885828971863, 0.2761290371417999, 0.27344226837158203, 0.2756351828575134, 0.2761900722980499, 0.2755855321884155, 0.2724992036819458, 0.2715945243835449, 0.2726806700229645, 0.2702755928039551, 0.2707327604293823, 0.26914530992507935, 0.2672998607158661, 0.2653893530368805, 0.2707858979701996, 0.26888710260391235, 0.26429009437561035, 0.26705700159072876, 0.26433271169662476, 0.2644907832145691, 0.2654668688774109, 0.2637845575809479, 0.2615320086479187, 0.2629801630973816, 0.2634650468826294, 0.25998246669769287, 0.2621432840824127, 0.2603616416454315, 0.260457843542099, 0.25872427225112915, 0.2601945102214813], 'val_loss': [0.45389050245285034, 0.43061962723731995, 0.4013422429561615, 0.42080938816070557, 0.395188570022583, 0.39979249238967896, 0.39877310395240784, 0.4145936369895935, 0.3784829080104828, 0.3892180621623993, 0.3713390827178955, 0.36597907543182373, 0.34918883442878723, 0.3361833393573761, 0.33424654603004456, 0.34345462918281555, 0.3452184796333313, 0.33300939202308655, 0.3561151325702667, 0.33258056640625, 0.3100963532924652, 0.32722580432891846, 0.31929418444633484, 0.3067832589149475, 0.30295196175575256, 0.3147284686565399, 0.3125918507575989, 0.3050999939441681, 0.31659048795700073, 0.302707701921463, 0.30327436327934265, 0.3065930902957916, 0.29703590273857117, 0.2960212826728821, 0.29190030694007874, 0.2948874533176422, 0.2921930253505707, 0.2942402958869934, 0.2906084656715393, 0.2935509979724884, 0.2984870672225952, 0.2999739646911621, 0.29295942187309265, 0.2930569648742676, 0.290778249502182, 0.28762853145599365, 0.2904091477394104, 0.2816406786441803, 0.29552775621414185, 0.2868202030658722, 0.28450778126716614, 0.28301164507865906, 0.2807203531265259, 0.28039175271987915, 0.2878774106502533, 0.28569889068603516, 0.3039056658744812, 0.2813010513782501, 0.28331518173217773, 0.2812541127204895, 0.2832031548023224, 0.2804718613624573, 0.28866803646087646, 0.2779247462749481, 0.27656784653663635, 0.2857021391391754, 0.28377673029899597, 0.27749380469322205, 0.2795368432998657, 0.2793876826763153], 'val_accuracy': [0.7767935395240784, 0.775540292263031, 0.8011550903320312, 0.7952496409416199, 0.8039715886116028, 0.7966544032096863, 0.7960512042045593, 0.8099052309989929, 0.8374516367912292, 0.8269608020782471, 0.8236475586891174, 0.8257650136947632, 0.8420281410217285, 0.8540118336677551, 0.8504691123962402, 0.8509190082550049, 0.8481219410896301, 0.8540127873420715, 0.8348009586334229, 0.8534428477287292, 0.8578217029571533, 0.8559457063674927, 0.8581292033195496, 0.8632622361183167, 0.8676828145980835, 0.8639155626296997, 0.8643791675567627, 0.8679339289665222, 0.8617703318595886, 0.863847553730011, 0.8682820796966553, 0.8685378432273865, 0.8699041604995728, 0.8732900619506836, 0.8731300830841064, 0.8718216419219971, 0.8756145238876343, 0.8728165030479431, 0.8722469210624695, 0.8747484683990479, 0.8692426681518555, 0.8694444894790649, 0.873407244682312, 0.8717463612556458, 0.8729797601699829, 0.8757378458976746, 0.8733677268028259, 0.8770347237586975, 0.8732863664627075, 0.872306227684021, 0.8756421208381653, 0.8771953582763672, 0.8779804110527039, 0.8791961073875427, 0.8724019527435303, 0.8737383484840393, 0.8651539087295532, 0.8782283663749695, 0.8751494288444519, 0.8749458193778992, 0.8758198618888855, 0.8772931098937988, 0.8728049397468567, 0.8781270384788513, 0.8808680176734924, 0.8775757551193237, 0.8762949109077454, 0.8789771795272827, 0.8788212537765503, 0.8799592852592468]}, 'batch_size': 32, 'crop_size': 80, 'learning_rate': 0.001}, 'inference': {'avg_accuracy': 0.8767142777777777, 'accuracy': [0.9233438888888889, 0.9345094444444444, 0.8940772222222222, 0.9105144444444444, 0.8391288888888889, 0.95098, 0.8876861111111111, 0.7732322222222222, 0.8472777777777778, 0.9639461111111111, 0.9149733333333333, 0.8905033333333333, 0.6264644444444445, 0.9434594444444444, 0.9502511111111112, 0.8874255555555556, 0.9090383333333333, 0.783135, 0.8810644444444444, 0.8861566666666667, 0.9271861111111112, 0.7000683333333333, 0.9493627777777778, 0.8812888888888889, 0.7280755555555556, 0.9413794444444444, 0.6406661111111112, 0.9485711111111111, 0.8190166666666666, 0.8649538888888889, 0.9245216666666667, 0.5231455555555555, 0.8900266666666666, 0.9201622222222222, 0.8942827777777778, 0.9398694444444444, 0.9456305555555555, 0.8512122222222223, 0.9236327777777777, 0.9262188888888889, 0.8431988888888889, 0.7898244444444444, 0.9462755555555555, 0.9299205555555555, 0.9505644444444444, 0.8782683333333333, 0.8345605555555555, 0.9333472222222222, 0.8361561111111111, 0.8908183333333334, 0.90968, 0.9309866666666666, 0.7933927777777777, 0.9122016666666667, 0.7026127777777778, 0.8870366666666667, 0.8692122222222223, 0.9070755555555555, 0.930505, 0.9060172222222223, 0.9090455555555556, 0.9134227777777778, 0.8983916666666667, 0.8190194444444444, 0.7275561111111111, 0.8331155555555556, 0.9206822222222222, 0.867085, 0.85498, 0.8969744444444444, 0.8837505555555556, 0.9042238888888889, 0.9291705555555556, 0.8913488888888889, 0.8909755555555555, 0.9325277777777777, 0.8544688888888888, 0.909215, 0.9091288888888889, 0.873935, 0.9154433333333334, 0.9015288888888889, 0.8813138888888888, 0.8597738888888888, 0.9020194444444445, 0.9287488888888888, 0.8618527777777778, 0.9171522222222223, 0.8757566666666666, 0.9054572222222222, 0.9153788888888889, 0.8185816666666667, 0.9260683333333334, 0.8471155555555555, 0.9500361111111111, 0.83239, 0.8784616666666667, 0.863745, 0.9020511111111111, 0.85144]}, 'inference_analyze': {'number': 100, 'bs': 32, 'nodes': 1, 'gpus': 1, 'time_per_inference': '1.45'}}}}\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(result_data[0])"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:09:39.014578210Z",
+     "start_time": "2023-10-06T17:09:38.954683058Z"
+    }
+   },
+   "id": "46df1e8d2333481f"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 296,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "OOOO\n",
+      "10500\n",
+      "3378\n",
+      "3378\n",
+      "3378\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": "          Loss  Val_Accuracy  Val_Loss  Index\n0     0.430577      0.776794  0.453891      1\n1     0.386824      0.775540  0.430620      2\n2     0.376532      0.801155  0.401342      3\n3     0.371918      0.795250  0.420809      4\n4     0.372598      0.803972  0.395189      5\n...        ...           ...       ...    ...\n3373  0.293370      0.866613  0.312281   3374\n3374  0.292041      0.866306  0.307600   3375\n3375  0.292379      0.863489  0.318265   3376\n3376  0.288750      0.862671  0.312229   3377\n3377  0.429758      0.771661  0.435519   3378\n\n[3378 rows x 4 columns]",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Loss</th>\n      <th>Val_Accuracy</th>\n      <th>Val_Loss</th>\n      <th>Index</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.430577</td>\n      <td>0.776794</td>\n      <td>0.453891</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0.386824</td>\n      <td>0.775540</td>\n      <td>0.430620</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0.376532</td>\n      <td>0.801155</td>\n      <td>0.401342</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0.371918</td>\n      <td>0.795250</td>\n      <td>0.420809</td>\n      <td>4</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0.372598</td>\n      <td>0.803972</td>\n      <td>0.395189</td>\n      <td>5</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>3373</th>\n      <td>0.293370</td>\n      <td>0.866613</td>\n      <td>0.312281</td>\n      <td>3374</td>\n    </tr>\n    <tr>\n      <th>3374</th>\n      <td>0.292041</td>\n      <td>0.866306</td>\n      <td>0.307600</td>\n      <td>3375</td>\n    </tr>\n    <tr>\n      <th>3375</th>\n      <td>0.292379</td>\n      <td>0.863489</td>\n      <td>0.318265</td>\n      <td>3376</td>\n    </tr>\n    <tr>\n      <th>3376</th>\n      <td>0.288750</td>\n      <td>0.862671</td>\n      <td>0.312229</td>\n      <td>3377</td>\n    </tr>\n    <tr>\n      <th>3377</th>\n      <td>0.429758</td>\n      <td>0.771661</td>\n      <td>0.435519</td>\n      <td>3378</td>\n    </tr>\n  </tbody>\n</table>\n<p>3378 rows × 4 columns</p>\n</div>"
+     },
+     "execution_count": 296,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "accuracy_list = []\n",
+    "loss_list = []\n",
+    "val_accuracy_list = []\n",
+    "val_loss_list = []\n",
+    "\n",
+    "# Iterate through the list of dictionaries and extract the values\n",
+    "for entry in result_data:\n",
+    "    accuracy_list.extend(entry['result']['value']['inference']['accuracy'])\n",
+    "    loss_list.extend(entry['result']['value']['training']['history']['loss'])\n",
+    "    val_accuracy_list.extend(entry['result']['value']['training']['history']['val_accuracy'])\n",
+    "    val_loss_list.extend(entry['result']['value']['training']['history']['val_loss'])\n",
+    "    \n",
+    "print (\"OOOO\")\n",
+    "print (len(accuracy_list))\n",
+    "print (len(loss_list))\n",
+    "print (len(val_accuracy_list))\n",
+    "print (len(val_loss_list))\n",
+    "\n",
+    "# Create a DataFrame with the extracted values\n",
+    "df = pd.DataFrame({\n",
+    "#    'Accuracy': accuracy_list,\n",
+    "    'Loss': loss_list,\n",
+    "    'Val_Accuracy': val_accuracy_list,\n",
+    "    'Val_Loss': val_loss_list\n",
+    "})\n",
+    "df['Index'] = range(1, len(df) + 1)\n",
+    "# Display the DataFrame\n",
+    "df"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:13:36.954000094Z",
+     "start_time": "2023-10-06T17:13:36.911286595Z"
+    }
+   },
+   "id": "b6892b3966f04fcb"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 297,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<bound method NDFrame.head of           Loss  Val_Accuracy  Val_Loss  Index\n",
+      "0     0.430577      0.776794  0.453891      1\n",
+      "1     0.386824      0.775540  0.430620      2\n",
+      "2     0.376532      0.801155  0.401342      3\n",
+      "3     0.371918      0.795250  0.420809      4\n",
+      "4     0.372598      0.803972  0.395189      5\n",
+      "...        ...           ...       ...    ...\n",
+      "3373  0.293370      0.866613  0.312281   3374\n",
+      "3374  0.292041      0.866306  0.307600   3375\n",
+      "3375  0.292379      0.863489  0.318265   3376\n",
+      "3376  0.288750      0.862671  0.312229   3377\n",
+      "3377  0.429758      0.771661  0.435519   3378\n",
+      "\n",
+      "[3378 rows x 4 columns]>\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(df.head)\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:18:41.323295185Z",
+     "start_time": "2023-10-06T17:18:41.258573736Z"
+    }
+   },
+   "id": "fcb9d9b9d82a3df4"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 301,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "          Loss  Val_Accuracy  Val_Loss  Index\n1083  0.213068      0.898055  0.243567   1084\n1081  0.215439      0.899954  0.247064   1082\n1080  0.216453      0.897066  0.255360   1081\n1079  0.217210      0.899515  0.247711   1080\n1082  0.218885      0.899921  0.241348   1083\n1076  0.219385      0.896986  0.254129   1077\n1075  0.220052      0.899768  0.246114   1076\n547   0.220269      0.888677  0.254200    548\n1078  0.221557      0.900421  0.240277   1079\n543   0.221967      0.893634  0.251591    544",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Loss</th>\n      <th>Val_Accuracy</th>\n      <th>Val_Loss</th>\n      <th>Index</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>1083</th>\n      <td>0.213068</td>\n      <td>0.898055</td>\n      <td>0.243567</td>\n      <td>1084</td>\n    </tr>\n    <tr>\n      <th>1081</th>\n      <td>0.215439</td>\n      <td>0.899954</td>\n      <td>0.247064</td>\n      <td>1082</td>\n    </tr>\n    <tr>\n      <th>1080</th>\n      <td>0.216453</td>\n      <td>0.897066</td>\n      <td>0.255360</td>\n      <td>1081</td>\n    </tr>\n    <tr>\n      <th>1079</th>\n      <td>0.217210</td>\n      <td>0.899515</td>\n      <td>0.247711</td>\n      <td>1080</td>\n    </tr>\n    <tr>\n      <th>1082</th>\n      <td>0.218885</td>\n      <td>0.899921</td>\n      <td>0.241348</td>\n      <td>1083</td>\n    </tr>\n    <tr>\n      <th>1076</th>\n      <td>0.219385</td>\n      <td>0.896986</td>\n      <td>0.254129</td>\n      <td>1077</td>\n    </tr>\n    <tr>\n      <th>1075</th>\n      <td>0.220052</td>\n      <td>0.899768</td>\n      <td>0.246114</td>\n      <td>1076</td>\n    </tr>\n    <tr>\n      <th>547</th>\n      <td>0.220269</td>\n      <td>0.888677</td>\n      <td>0.254200</td>\n      <td>548</td>\n    </tr>\n    <tr>\n      <th>1078</th>\n      <td>0.221557</td>\n      <td>0.900421</td>\n      <td>0.240277</td>\n      <td>1079</td>\n    </tr>\n    <tr>\n      <th>543</th>\n      <td>0.221967</td>\n      <td>0.893634</td>\n      <td>0.251591</td>\n      <td>544</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 301,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "n = 10\n",
+    "sorted_loss_df = df.sort_values(by='Loss', ascending=True)\n",
+    "sorted_loss_df.head(n)\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:22:53.472835949Z",
+     "start_time": "2023-10-06T17:22:53.459234535Z"
+    }
+   },
+   "id": "120eb23bf5e6701f"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 302,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "          Loss  Val_Accuracy  Val_Loss  Index\n1078  0.221557      0.900421  0.240277   1079\n1081  0.215439      0.899954  0.247064   1082\n1082  0.218885      0.899921  0.241348   1083\n1075  0.220052      0.899768  0.246114   1076\n1079  0.217210      0.899515  0.247711   1080\n545   0.222495      0.898925  0.245414    546\n1083  0.213068      0.898055  0.243567   1084\n1074  0.226875      0.897595  0.244554   1075\n1069  0.229369      0.897237  0.246573   1070\n536   0.225548      0.897211  0.245603    537",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Loss</th>\n      <th>Val_Accuracy</th>\n      <th>Val_Loss</th>\n      <th>Index</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>1078</th>\n      <td>0.221557</td>\n      <td>0.900421</td>\n      <td>0.240277</td>\n      <td>1079</td>\n    </tr>\n    <tr>\n      <th>1081</th>\n      <td>0.215439</td>\n      <td>0.899954</td>\n      <td>0.247064</td>\n      <td>1082</td>\n    </tr>\n    <tr>\n      <th>1082</th>\n      <td>0.218885</td>\n      <td>0.899921</td>\n      <td>0.241348</td>\n      <td>1083</td>\n    </tr>\n    <tr>\n      <th>1075</th>\n      <td>0.220052</td>\n      <td>0.899768</td>\n      <td>0.246114</td>\n      <td>1076</td>\n    </tr>\n    <tr>\n      <th>1079</th>\n      <td>0.217210</td>\n      <td>0.899515</td>\n      <td>0.247711</td>\n      <td>1080</td>\n    </tr>\n    <tr>\n      <th>545</th>\n      <td>0.222495</td>\n      <td>0.898925</td>\n      <td>0.245414</td>\n      <td>546</td>\n    </tr>\n    <tr>\n      <th>1083</th>\n      <td>0.213068</td>\n      <td>0.898055</td>\n      <td>0.243567</td>\n      <td>1084</td>\n    </tr>\n    <tr>\n      <th>1074</th>\n      <td>0.226875</td>\n      <td>0.897595</td>\n      <td>0.244554</td>\n      <td>1075</td>\n    </tr>\n    <tr>\n      <th>1069</th>\n      <td>0.229369</td>\n      <td>0.897237</td>\n      <td>0.246573</td>\n      <td>1070</td>\n    </tr>\n    <tr>\n      <th>536</th>\n      <td>0.225548</td>\n      <td>0.897211</td>\n      <td>0.245603</td>\n      <td>537</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 302,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "n = 10\n",
+    "sorted_loss_df = df.sort_values(by='Val_Accuracy', ascending=False)\n",
+    "sorted_loss_df.head(n)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:23:30.143329734Z",
+     "start_time": "2023-10-06T17:23:30.089134704Z"
+    }
+   },
+   "id": "f408b0a3ed8c889d"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 303,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "          Loss  Val_Accuracy  Val_Loss  Index\n1078  0.221557      0.900421  0.240277   1079\n1082  0.218885      0.899921  0.241348   1083\n1083  0.213068      0.898055  0.243567   1084\n1074  0.226875      0.897595  0.244554   1075\n545   0.222495      0.898925  0.245414    546\n536   0.225548      0.897211  0.245603    537\n1075  0.220052      0.899768  0.246114   1076\n1069  0.229369      0.897237  0.246573   1070\n539   0.223255      0.895780  0.246660    540\n1081  0.215439      0.899954  0.247064   1082",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Loss</th>\n      <th>Val_Accuracy</th>\n      <th>Val_Loss</th>\n      <th>Index</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>1078</th>\n      <td>0.221557</td>\n      <td>0.900421</td>\n      <td>0.240277</td>\n      <td>1079</td>\n    </tr>\n    <tr>\n      <th>1082</th>\n      <td>0.218885</td>\n      <td>0.899921</td>\n      <td>0.241348</td>\n      <td>1083</td>\n    </tr>\n    <tr>\n      <th>1083</th>\n      <td>0.213068</td>\n      <td>0.898055</td>\n      <td>0.243567</td>\n      <td>1084</td>\n    </tr>\n    <tr>\n      <th>1074</th>\n      <td>0.226875</td>\n      <td>0.897595</td>\n      <td>0.244554</td>\n      <td>1075</td>\n    </tr>\n    <tr>\n      <th>545</th>\n      <td>0.222495</td>\n      <td>0.898925</td>\n      <td>0.245414</td>\n      <td>546</td>\n    </tr>\n    <tr>\n      <th>536</th>\n      <td>0.225548</td>\n      <td>0.897211</td>\n      <td>0.245603</td>\n      <td>537</td>\n    </tr>\n    <tr>\n      <th>1075</th>\n      <td>0.220052</td>\n      <td>0.899768</td>\n      <td>0.246114</td>\n      <td>1076</td>\n    </tr>\n    <tr>\n      <th>1069</th>\n      <td>0.229369</td>\n      <td>0.897237</td>\n      <td>0.246573</td>\n      <td>1070</td>\n    </tr>\n    <tr>\n      <th>539</th>\n      <td>0.223255</td>\n      <td>0.895780</td>\n      <td>0.246660</td>\n      <td>540</td>\n    </tr>\n    <tr>\n      <th>1081</th>\n      <td>0.215439</td>\n      <td>0.899954</td>\n      <td>0.247064</td>\n      <td>1082</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 303,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "n = 10\n",
+    "sorted_loss_df = df.sort_values(by='Val_Loss', ascending=True)\n",
+    "sorted_loss_df.head(n)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:24:21.772498806Z",
+     "start_time": "2023-10-06T17:24:21.713136937Z"
+    }
+   },
+   "id": "6bd469e8177bfdaa"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 304,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "       Accuracy  Index\n0      0.923344      1\n1      0.934509      2\n2      0.894077      3\n3      0.910514      4\n4      0.839129      5\n...         ...    ...\n10495  0.751298  10496\n10496  0.867493  10497\n10497  0.718101  10498\n10498  0.826334  10499\n10499  0.837769  10500\n\n[10500 rows x 2 columns]",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Accuracy</th>\n      <th>Index</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.923344</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0.934509</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0.894077</td>\n      <td>3</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0.910514</td>\n      <td>4</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0.839129</td>\n      <td>5</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>10495</th>\n      <td>0.751298</td>\n      <td>10496</td>\n    </tr>\n    <tr>\n      <th>10496</th>\n      <td>0.867493</td>\n      <td>10497</td>\n    </tr>\n    <tr>\n      <th>10497</th>\n      <td>0.718101</td>\n      <td>10498</td>\n    </tr>\n    <tr>\n      <th>10498</th>\n      <td>0.826334</td>\n      <td>10499</td>\n    </tr>\n    <tr>\n      <th>10499</th>\n      <td>0.837769</td>\n      <td>10500</td>\n    </tr>\n  </tbody>\n</table>\n<p>10500 rows × 2 columns</p>\n</div>"
+     },
+     "execution_count": 304,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df = pd.DataFrame({\n",
+    "    'Accuracy': accuracy_list,\n",
+    "})\n",
+    "df['Index'] = range(1, len(df) + 1)\n",
+    "# Display the DataFrame\n",
+    "df"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:25:08.097115129Z",
+     "start_time": "2023-10-06T17:25:08.041006334Z"
+    }
+   },
+   "id": "c3a753ccc3e662d2"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 305,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "      Accuracy  Index\n3009  0.989144   3010\n3022  0.988317   3023\n3005  0.984737   3006\n3057  0.980435   3058\n3094  0.969111   3095\n3092  0.967100   3093\n1709  0.964410   1710\n2209  0.964246   2210\n1409  0.964241   1410\n1109  0.964023   1110",
+      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Accuracy</th>\n      <th>Index</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>3009</th>\n      <td>0.989144</td>\n      <td>3010</td>\n    </tr>\n    <tr>\n      <th>3022</th>\n      <td>0.988317</td>\n      <td>3023</td>\n    </tr>\n    <tr>\n      <th>3005</th>\n      <td>0.984737</td>\n      <td>3006</td>\n    </tr>\n    <tr>\n      <th>3057</th>\n      <td>0.980435</td>\n      <td>3058</td>\n    </tr>\n    <tr>\n      <th>3094</th>\n      <td>0.969111</td>\n      <td>3095</td>\n    </tr>\n    <tr>\n      <th>3092</th>\n      <td>0.967100</td>\n      <td>3093</td>\n    </tr>\n    <tr>\n      <th>1709</th>\n      <td>0.964410</td>\n      <td>1710</td>\n    </tr>\n    <tr>\n      <th>2209</th>\n      <td>0.964246</td>\n      <td>2210</td>\n    </tr>\n    <tr>\n      <th>1409</th>\n      <td>0.964241</td>\n      <td>1410</td>\n    </tr>\n    <tr>\n      <th>1109</th>\n      <td>0.964023</td>\n      <td>1110</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
+     },
+     "execution_count": 305,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "n = 10\n",
+    "sorted_accuracy_loss = df.sort_values(by='Accuracy', ascending=False)\n",
+    "sorted_accuracy_loss.head(n)\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2023-10-06T17:26:28.517987815Z",
+     "start_time": "2023-10-06T17:26:28.453832257Z"
+    }
+   },
+   "id": "73fdae48fd5a8078"
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -761,7 +899,7 @@
    "metadata": {
     "collapsed": false
    },
-   "id": "507603c8d8759e8f"
+   "id": "416e59f43a6a1ac8"
   }
  ],
  "metadata": {