Newer
Older
" </tr>\n",
" <tr>\n",
" <th>145</th>\n",
" <td>CORRECT</td>\n",
" <td>_</td>\n",
" <td>_</td>\n",
" </tr>\n",
" <tr>\n",
" <th>146</th>\n",
" <td>CORRECT</td>\n",
" </tr>\n",
" <tr>\n",
" <th>147</th>\n",
" <td>CORRECT</td>\n",
" </tr>\n",
" <tr>\n",
" <th>148</th>\n",
" <td>CORRECT</td>\n",
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
" <td>_</td>\n",
" <td>_</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" step_type reference_word_text hypothesis_word_text\n",
"0 DELETION _ \n",
"1 DELETION _ \n",
"2 DELETION _ \n",
"3 DELETION _ \n",
"4 DELETION _ \n",
"5 CORRECT _ _\n",
"6 CORRECT _ _\n",
"7 CORRECT _ _\n",
"8 CORRECT _ _\n",
"9 CORRECT _ _\n",
"10 CORRECT _ _\n",
"11 CORRECT _ _\n",
"12 CORRECT _ _\n",
"13 CORRECT _ _\n",
"14 CORRECT _ _\n",
"15 CORRECT _ _\n",
"16 CORRECT _ _\n",
"17 CORRECT _ _\n",
"18 DELETION geogName \n",
"19 DELETION geogName \n",
"20 DELETION geogName \n",
"21 CORRECT _ _\n",
"22 CORRECT _ _\n",
"23 DELETION placeName \n",
"24 CORRECT _ _\n",
"25 CORRECT _ _\n",
"26 CORRECT _ _\n",
"27 CORRECT _ _\n",
"28 CORRECT _ _\n",
"29 CORRECT _ _\n",
"30 CORRECT _ _\n",
"31 CORRECT _ _\n",
"32 DELETION orgName \n",
"33 CORRECT _ _\n",
"34 CORRECT _ _\n",
"35 CORRECT _ _\n",
"36 CORRECT _ _\n",
"37 CORRECT _ _\n",
"38 CORRECT _ _\n",
"39 CORRECT _ _\n",
"40 CORRECT _ _\n",
"41 CORRECT _ _\n",
"42 CORRECT _ _\n",
"43 DELETION geogName \n",
"44 DELETION geogName \n",
"45 CORRECT _ _\n",
"46 CORRECT placeName placeName\n",
"47 CORRECT _ _\n",
"48 CORRECT _ _\n",
"49 CORRECT _ _\n",
"50 DELETION placeName \n",
"51 CORRECT _ _\n",
"52 CORRECT _ _\n",
"53 CORRECT _ _\n",
"54 CORRECT _ _\n",
"55 CORRECT _ _\n",
"56 CORRECT _ _\n",
"57 DELETION geogName \n",
"58 CORRECT _ _\n",
"59 CORRECT _ _\n",
"60 CORRECT _ _\n",
"61 CORRECT _ _\n",
"62 CORRECT _ _\n",
"63 CORRECT _ _\n",
"64 DELETION persName \n",
"65 CORRECT _ _\n",
"66 CORRECT _ _\n",
"67 CORRECT _ _\n",
"68 CORRECT _ _\n",
"69 CORRECT _ _\n",
"70 CORRECT _ _\n",
"71 CORRECT _ _\n",
"72 CORRECT _ _\n",
"73 DELETION geogName \n",
"74 DELETION geogName \n",
"75 CORRECT _ _\n",
"76 CORRECT _ _\n",
"77 CORRECT _ _\n",
"78 CORRECT placeName placeName\n",
"79 DELETION _ \n",
"80 DELETION _ \n",
"81 DELETION _ \n",
"82 DELETION _ \n",
"83 CORRECT _ _\n",
"84 CORRECT _ _\n",
"85 CORRECT _ _\n",
"86 SUBSTITUTION _ placeName\n",
"87 CORRECT _ _\n",
"88 CORRECT _ _\n",
"89 CORRECT _ _\n",
"90 DELETION geogName \n",
"91 DELETION geogName \n",
"92 CORRECT _ _\n",
"93 CORRECT _ _\n",
"94 CORRECT _ _\n",
"95 CORRECT _ _\n",
"96 DELETION orgName \n",
"97 DELETION orgName \n",
"98 CORRECT _ _\n",
"99 CORRECT _ _\n",
"100 CORRECT _ _\n",
"101 CORRECT _ _\n",
"102 CORRECT _ _\n",
"103 CORRECT _ _\n",
"104 CORRECT _ _\n",
"105 CORRECT _ _\n",
"106 CORRECT _ _\n",
"107 CORRECT _ _\n",
"108 CORRECT _ _\n",
"109 CORRECT _ _\n",
"110 CORRECT _ _\n",
"111 CORRECT _ _\n",
"112 CORRECT _ _\n",
"113 CORRECT _ _\n",
"114 CORRECT _ _\n",
"115 CORRECT _ _\n",
"116 CORRECT _ _\n",
"117 DELETION placeName \n",
"118 DELETION placeName \n",
"119 CORRECT _ _\n",
"120 CORRECT _ _\n",
"121 CORRECT _ _\n",
"122 CORRECT _ _\n",
"123 CORRECT _ _\n",
"124 CORRECT _ _\n",
"125 CORRECT _ _\n",
"126 CORRECT _ _\n",
"127 CORRECT _ _\n",
"128 CORRECT _ _\n",
"129 DELETION geogName \n",
"130 CORRECT _ _\n",
"131 CORRECT _ _\n",
"132 CORRECT _ _\n",
"133 CORRECT _ _\n",
"134 CORRECT _ _\n",
"135 CORRECT time time\n",
"136 CORRECT time time\n",
"137 CORRECT time time\n",
"138 CORRECT time time\n",
"139 DELETION _ \n",
"140 DELETION _ \n",
"141 DELETION _ \n",
"142 DELETION _ \n",
"143 SUBSTITUTION _ time\n",
"144 CORRECT _ _\n",
"145 CORRECT _ _\n",
"146 CORRECT _ _\n",
"147 CORRECT _ _\n",
"148 CORRECT _ _"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"--------------------------------------------------------------\n",
"--------------------------------------------------------------\n",
"--------------------------------------------------------------\n"
]
}
],
"source": [
"for it in list(luna_record_provider.get_all_records())[:1]:\n",
" show_report_for(it, luna_experiment_repository, techmo_connections_config)"
]
},
{
"cell_type": "markdown",
"id": "a4b4f1bc-f0b3-4d23-bf1d-97f75715f332",
"metadata": {},
"source": [
"# Report for LUNA and AJN ASR"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "27797bad-c92a-45a6-87f2-071645c18150",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"record_id: LUNA.PL__PRZYSTANKI__DOBRAJAKOSC__M__1_2007-03-30_14_37_38__1_2007-03-30_14_37_38\n",
"properties_confiog: AJN ASR\n",
"\n",
"gold transcript: xxx dzień dobry dzień dobry proszę pana ja mam o której będzie sto czterdzieści trzy będzie najbliższy z Alei Alei Witosa w stronę Gocławia z tego przystanku vis-a-vis ZUSu tak ? no nie nie tam vis-a-vis Panoramy przez Siekierkowski tam w stronę Gocławia ale to nie z tego przystanku Aleja Witosa ? no to jest czy Bartoszka na żądanie wie pan gdzie jest się zaczyna Most Siekierkowski tam w stronę Gocławia gdzie ma on przystanek pierwszy sto czterdzieści trzy znaczy tak naprawdę Most Siekierkowski to się zaczyna za Siekierki Sanktuarium no właśnie przed tym przed tym przystanek o której jest sto czterdzieści trzy najbliższy aha czyli to wtedy przystanek Małe Siekierki ale to nie jest tak tak tak tak na wysokości Panoramy aha dobrze najbliższy już chwileczkę czternasta czterdzieści sześć piętnasta zero jeden aha no dobrze dziękuję do widzenia do widzenia\n",
"\n",
"asr transcript: dobre dyrektor proszę wybaczyć tryb numer jeden to teresie trzeba będzie najbliższy ale i tu aleje pić od roztropnego działania swego przystankom vis-a-vis zus-u to jak mnie nie <unk> widać <unk> panoramę przez siekierko trwałego tonę antoniego przystanki kolejowe to było coś czy bartoszka na żądanie <unk> <unk> <unk> wielbłądy się toczy na most siekierkowski to trud i czytanek pieszczoty się trzy płaczesz tak naprawdę może kierkowski to zaczną siekierki sanktuarium w lourdes się tym przejęty przystanku który torresie czne edition - czyli to wspólny przystanek małe siekierki ale to nie do końca dotrzymać dziś są czternastej czterdzieści sześć piętnasto żeruje a nowy gen. \n",
"\n",
"word wer {'classic_wer': 0.8297872340425532}\n",
"\n",
"pos wer {'classic_wer': 0.6778523489932886}\n",
"\n",
"tag wer {'classic_wer': 0.8456375838926175}\n",
"\n",
"ner wer {'classic_wer': 0.22818791946308725}\n",
"\n"
]
},
{
"data": {
"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>step_type</th>\n",
" <th>reference_word_text</th>\n",
" <th>hypothesis_word_text</th>\n",
" <th>2</th>\n",
" <td>DELETION</td>\n",
" <td>dobry</td>\n",
" <td></td>\n",
" <th>3</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>dzień</td>\n",
" <td>dobre</td>\n",
" <th>4</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>dobry</td>\n",
" <td>dyrektor</td>\n",
" <th>6</th>\n",
" <td>DELETION</td>\n",
" <td>pana</td>\n",
" <td></td>\n",
" <th>7</th>\n",
" <td>DELETION</td>\n",
" <td>ja</td>\n",
" <td></td>\n",
" <th>8</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>mam</td>\n",
" <td>wybaczyć</td>\n",
" <th>9</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>o</td>\n",
" <td>tryb</td>\n",
" <th>10</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>której</td>\n",
" <td>numer</td>\n",
" <th>11</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>będzie</td>\n",
" <td>jeden</td>\n",
" <th>12</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>sto</td>\n",
" <td>to</td>\n",
" <td>czterdzieści</td>\n",
" <td>teresie</td>\n",
" <th>14</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>trzy</td>\n",
" <td>trzeba</td>\n",
" <td>najbliższy</td>\n",
" <td>najbliższy</td>\n",
" <th>18</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Alei</td>\n",
" <td>i</td>\n",
" <th>19</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Alei</td>\n",
" <td>tu</td>\n",
" <th>20</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Witosa</td>\n",
" <td>aleje</td>\n",
" <th>21</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>w</td>\n",
" <td>pić</td>\n",
" <th>23</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Gocławia</td>\n",
" <td>roztropnego</td>\n",
" <th>24</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>z</td>\n",
" <td>działania</td>\n",
" <th>25</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>tego</td>\n",
" <td>swego</td>\n",
" <th>26</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>przystanku</td>\n",
" <td>przystankom</td>\n",
" <td>vis-a-vis</td>\n",
" <td>vis-a-vis</td>\n",
" <th>32</th>\n",
" <td>CORRECT</td>\n",
" <td>nie</td>\n",
" <td>nie</td>\n",
" <th>33</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>nie</td>\n",
" <td><unk></td>\n",
" <th>35</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>vis-a-vis</td>\n",
" <td><unk></td>\n",
" <th>36</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Panoramy</td>\n",
" <td>panoramę</td>\n",
" <th>38</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Siekierkowski</td>\n",
" <td>siekierko</td>\n",
" <th>39</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>tam</td>\n",
" <td>trwałego</td>\n",
" <th>40</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>w</td>\n",
" <td>tonę</td>\n",
" <td>stronę</td>\n",
" <td>antoniego</td>\n",
" <td>Gocławia</td>\n",
" <td>przystanki</td>\n",
" <th>45</th>\n",
" <td>DELETION</td>\n",
" <td>nie</td>\n",
" <td></td>\n",
" <th>46</th>\n",
" <td>DELETION</td>\n",
" <td>z</td>\n",
" <td></td>\n",
" <th>47</th>\n",
" <td>DELETION</td>\n",
" <td>tego</td>\n",
" <td></td>\n",
" <th>48</th>\n",
" <td>DELETION</td>\n",
" <td>przystanku</td>\n",
" <td></td>\n",
" <th>49</th>\n",
" <td>DELETION</td>\n",
" <td>Aleja</td>\n",
" <td></td>\n",
" <th>52</th>\n",
" <td>DELETION</td>\n",
" <td>no</td>\n",
" <td></td>\n",
" <th>54</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>jest</td>\n",
" <td>coś</td>\n",
" <th>56</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Bartoszka</td>\n",
" <td>bartoszka</td>\n",
" <th>59</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>wie</td>\n",
" <td><unk></td>\n",
" <td>gdzie</td>\n",
" <td><unk></td>\n",
" <tr>\n",
" <th>62</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>jest</td>\n",
" <td>wielbłądy</td>\n",
" <th>63</th>\n",
" <td>CORRECT</td>\n",
" <td>się</td>\n",
" <td>się</td>\n",
" <td>gdzie</td>\n",
" <td>siekierkowski</td>\n",
" <th>75</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>pierwszy</td>\n",
" <td>czytanek</td>\n",
" <td>czterdzieści</td>\n",
" <td>się</td>\n",
" <th>78</th>\n",
" <td>CORRECT</td>\n",
" <td>trzy</td>\n",
" <td>trzy</td>\n",
" <th>81</th>\n",
" <td>CORRECT</td>\n",
" <td>naprawdę</td>\n",
" <td>naprawdę</td>\n",
" <td>Siekierkowski</td>\n",
" <td>kierkowski</td>\n",
" <td>Sanktuarium</td>\n",
" <td>zaczną</td>\n",
" <td>właśnie</td>\n",
" <td>sanktuarium</td>\n",
" <th>93</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>tym</td>\n",
" <td>lourdes</td>\n",
" <th>94</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>przed</td>\n",
" <td>się</td>\n",
" <th>95</th>\n",
" <td>CORRECT</td>\n",
" <td>tym</td>\n",
" <td>tym</td>\n",
" </tr>\n",
" <tr>\n",
" <th>96</th>\n",
" <td>czterdzieści</td>\n",
" <td>torresie</td>\n",
" <td>najbliższy</td>\n",
" <td>edition</td>\n",
" <th>105</th>\n",
" <td>CORRECT</td>\n",
" <td>czyli</td>\n",
" <td>czyli</td>\n",
" </tr>\n",
" <tr>\n",
" <th>106</th>\n",
" <td>CORRECT</td>\n",
" <td>to</td>\n",
" <td>przystanek</td>\n",
" <td>przystanek</td>\n",
" <th>109</th>\n",
" <td>SUBSTITUTION</td>\n",
" <td>Małe</td>\n",
" <td>małe</td>\n",
" <td>Siekierki</td>\n",
" <td>siekierki</td>\n",
" <th>112</th>\n",
" <td>CORRECT</td>\n",
" <td>to</td>\n",
" <td>to</td>\n",
" </tr>\n",
" <tr>\n",
" <th>113</th>\n",
" <td>CORRECT</td>\n",
" <td>nie</td>\n",
" <td>nie</td>\n",
" </tr>\n",
" <tr>\n",
" <th>114</th>\n",
" <th>116</th>\n",
" <td>DELETION</td>\n",
" <td>tak</td>\n",
" <td></td>\n",
" <th>117</th>\n",
" <td>DELETION</td>\n",
" <td>tak</td>\n",
" <td></td>\n",
" <th>119</th>\n",
" <td>DELETION</td>\n",
" <td>na</td>\n",
" <td></td>\n",