at least one filter in each category has to be true
Browse files- ai_config_faiss.py +12 -11
- cache/db_metadata.json +70 -70
- db/faiss_index.faiss/index.pkl +2 -2
- ttv_web_scraper.py +2 -2
ai_config_faiss.py
CHANGED
|
@@ -62,17 +62,18 @@ class FAISSAIAssistant:
|
|
| 62 |
if not filters:
|
| 63 |
return True
|
| 64 |
for key, values in filters.items():
|
| 65 |
-
if key in doc_metadata:
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
|
|
|
| 76 |
|
| 77 |
filtered_docs = [
|
| 78 |
doc for doc in all_docs
|
|
|
|
| 62 |
if not filters:
|
| 63 |
return True
|
| 64 |
for key, values in filters.items():
|
| 65 |
+
if key not in doc_metadata:
|
| 66 |
+
return False
|
| 67 |
+
doc_value = doc_metadata[key]
|
| 68 |
+
if isinstance(doc_value, list):
|
| 69 |
+
# If doc_value is a list, check if any item in doc_value is in values
|
| 70 |
+
if not any(item in values for item in doc_value):
|
| 71 |
+
return False
|
| 72 |
+
else:
|
| 73 |
+
# If doc_value is a single string, check if it's in values
|
| 74 |
+
if doc_value not in values:
|
| 75 |
+
return False
|
| 76 |
+
return True
|
| 77 |
|
| 78 |
filtered_docs = [
|
| 79 |
doc for doc in all_docs
|
cache/db_metadata.json
CHANGED
|
@@ -1,73 +1,73 @@
|
|
| 1 |
{
|
| 2 |
"content_hashes": [
|
| 3 |
"4974b044047d2523c747b79d938915c9",
|
| 4 |
-
"5754ba35c4f9f27e3e1d5b4d9bb972f2",
|
| 5 |
"9ae73679959943c591be3d1c81b7c26c",
|
| 6 |
-
"6286818c51fc82ffc065ba12d3c48c19",
|
| 7 |
"e068f68ad0aed4134d075210f871ea95",
|
| 8 |
"f8f43b2e1413f709038506c3a2dfd7b9",
|
| 9 |
-
"
|
|
|
|
|
|
|
| 10 |
],
|
| 11 |
"speakers": [
|
| 12 |
-
"Sadayuki Abeta",
|
| 13 |
-
"Juan Manuel Caro",
|
| 14 |
-
"Franz Seiser",
|
| 15 |
"Hasan Jafri",
|
| 16 |
-
"
|
| 17 |
-
"
|
| 18 |
-
"
|
| 19 |
-
"
|
| 20 |
"Alfredo Musitani",
|
| 21 |
-
"
|
| 22 |
-
"Gabriela Styf Sj\u00f6man",
|
| 23 |
"Alex Foster",
|
|
|
|
|
|
|
|
|
|
| 24 |
"Vishal Mathur",
|
| 25 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
"Alexandra Foster",
|
| 27 |
-
"
|
| 28 |
-
"
|
| 29 |
-
"Laura Murphy",
|
| 30 |
-
"Ahmed Hafez",
|
| 31 |
"Jose Antonio Martin Martinez",
|
| 32 |
-
"
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
"Anita D\u00f6hler",
|
| 37 |
-
"Geoff Hollingworth",
|
| 38 |
-
"Abdu Mudesir",
|
| 39 |
-
"Akira Tada",
|
| 40 |
"Matthias Fridstrom",
|
| 41 |
-
"
|
|
|
|
| 42 |
"Guy Daniels",
|
| 43 |
-
"
|
| 44 |
-
"
|
|
|
|
| 45 |
"Luis Velarde Tazon",
|
| 46 |
"Colin Bannon",
|
| 47 |
-
"Dennis Hoffman",
|
| 48 |
"Michele Campriani",
|
| 49 |
-
"
|
| 50 |
"Ray Le Maistre",
|
| 51 |
-
"
|
|
|
|
|
|
|
|
|
|
| 52 |
"Philippe Ensarguet",
|
| 53 |
-
"
|
| 54 |
-
"
|
| 55 |
-
"
|
| 56 |
-
"Mark Gilmour",
|
| 57 |
-
"Francesca Serravalle",
|
| 58 |
"Robert Curran",
|
| 59 |
-
"Faiq Khan",
|
| 60 |
-
"Chivas Nambiar",
|
| 61 |
-
"Mirko Voltolini",
|
| 62 |
"Komal Aggarwal",
|
| 63 |
-
"
|
| 64 |
-
"Vivek Chadha",
|
| 65 |
-
"Yago Tenorio"
|
| 66 |
],
|
| 67 |
"companies": {
|
| 68 |
"TelecomTV": [
|
| 69 |
-
"
|
| 70 |
-
"
|
| 71 |
],
|
| 72 |
"Connectivitree": [
|
| 73 |
"Mark Gilmour"
|
|
@@ -77,15 +77,15 @@
|
|
| 77 |
"Terje Jensen"
|
| 78 |
],
|
| 79 |
"Vodafone": [
|
| 80 |
-
"
|
| 81 |
-
"
|
| 82 |
],
|
| 83 |
"BT": [
|
| 84 |
-
"Sarwar Khan",
|
| 85 |
"Alex Foster",
|
| 86 |
-
"
|
|
|
|
| 87 |
"Mark Henry",
|
| 88 |
-
"
|
| 89 |
],
|
| 90 |
"American Tower": [
|
| 91 |
"Susan James"
|
|
@@ -101,15 +101,15 @@
|
|
| 101 |
"Sandeep Raithatha"
|
| 102 |
],
|
| 103 |
"Deutsche Telekom": [
|
| 104 |
-
"
|
| 105 |
-
"
|
| 106 |
],
|
| 107 |
"Telefonica": [
|
| 108 |
-
"Juan Manuel Caro",
|
| 109 |
-
"Enrique Blanco",
|
| 110 |
"Jose Antonio Martin Martinez",
|
| 111 |
-
"
|
| 112 |
-
"
|
|
|
|
|
|
|
| 113 |
],
|
| 114 |
"SoftBank": [
|
| 115 |
"Akira Tada"
|
|
@@ -138,8 +138,8 @@
|
|
| 138 |
"Sushil Rawat"
|
| 139 |
],
|
| 140 |
"Orange": [
|
| 141 |
-
"
|
| 142 |
-
"
|
| 143 |
],
|
| 144 |
"BT Business": [
|
| 145 |
"Colin Bannon"
|
|
@@ -163,9 +163,9 @@
|
|
| 163 |
"Dean Dennis"
|
| 164 |
],
|
| 165 |
"Rakuten Symphony": [
|
| 166 |
-
"Faiq Khan",
|
| 167 |
"Geoff Hollingworth",
|
| 168 |
-
"Vivek Chadha"
|
|
|
|
| 169 |
],
|
| 170 |
"Cambridge Management Consulting": [
|
| 171 |
"Tom Burton"
|
|
@@ -194,16 +194,16 @@
|
|
| 194 |
},
|
| 195 |
"sentiments": [],
|
| 196 |
"subjects": [
|
| 197 |
-
"
|
| 198 |
-
"
|
| 199 |
-
"
|
| 200 |
-
"
|
| 201 |
-
"
|
| 202 |
-
"
|
| 203 |
-
"
|
| 204 |
-
"
|
| 205 |
-
"
|
| 206 |
-
"
|
| 207 |
-
"
|
| 208 |
]
|
| 209 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"content_hashes": [
|
| 3 |
"4974b044047d2523c747b79d938915c9",
|
|
|
|
| 4 |
"9ae73679959943c591be3d1c81b7c26c",
|
|
|
|
| 5 |
"e068f68ad0aed4134d075210f871ea95",
|
| 6 |
"f8f43b2e1413f709038506c3a2dfd7b9",
|
| 7 |
+
"5754ba35c4f9f27e3e1d5b4d9bb972f2",
|
| 8 |
+
"d81ba6e90c2c42d82c4003c4d158d3e3",
|
| 9 |
+
"6286818c51fc82ffc065ba12d3c48c19"
|
| 10 |
],
|
| 11 |
"speakers": [
|
|
|
|
|
|
|
|
|
|
| 12 |
"Hasan Jafri",
|
| 13 |
+
"Atoosa Hatefi",
|
| 14 |
+
"Chris Lewis",
|
| 15 |
+
"Laura Murphy",
|
| 16 |
+
"Tom Burton",
|
| 17 |
"Alfredo Musitani",
|
| 18 |
+
"Amith Maharaj",
|
|
|
|
| 19 |
"Alex Foster",
|
| 20 |
+
"Mirko Voltolini",
|
| 21 |
+
"Juan Manuel Caro",
|
| 22 |
+
"Mark Gilmour",
|
| 23 |
"Vishal Mathur",
|
| 24 |
+
"Abdu Mudesir",
|
| 25 |
+
"Anita D\u00f6hler",
|
| 26 |
+
"Chivas Nambiar",
|
| 27 |
+
"Manish Singh",
|
| 28 |
+
"Vivek Chadha",
|
| 29 |
+
"Enrique Blanco",
|
| 30 |
+
"Francesca Serravalle",
|
| 31 |
+
"Sushil Rawat",
|
| 32 |
+
"Dean Dennis",
|
| 33 |
+
"Mallik Rao",
|
| 34 |
+
"Sadayuki Abeta",
|
| 35 |
"Alexandra Foster",
|
| 36 |
+
"Terje Jensen",
|
| 37 |
+
"Yago Tenorio",
|
|
|
|
|
|
|
| 38 |
"Jose Antonio Martin Martinez",
|
| 39 |
+
"Madhukiran Medithe",
|
| 40 |
+
"Susan James",
|
| 41 |
+
"Sarwar Khan",
|
| 42 |
+
"Faiq Khan",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
"Matthias Fridstrom",
|
| 44 |
+
"Mojdeh Amani",
|
| 45 |
+
"Francis Haysom",
|
| 46 |
"Guy Daniels",
|
| 47 |
+
"Dennis Hoffman",
|
| 48 |
+
"Amol Phadke",
|
| 49 |
+
"Akira Tada",
|
| 50 |
"Luis Velarde Tazon",
|
| 51 |
"Colin Bannon",
|
|
|
|
| 52 |
"Michele Campriani",
|
| 53 |
+
"Harkirit Singh",
|
| 54 |
"Ray Le Maistre",
|
| 55 |
+
"Franz Seiser",
|
| 56 |
+
"Andrew Coward",
|
| 57 |
+
"Nik Willets",
|
| 58 |
+
"Sandeep Raithatha",
|
| 59 |
"Philippe Ensarguet",
|
| 60 |
+
"Geoff Hollingworth",
|
| 61 |
+
"Ahmed Hafez",
|
| 62 |
+
"Mark Henry",
|
|
|
|
|
|
|
| 63 |
"Robert Curran",
|
|
|
|
|
|
|
|
|
|
| 64 |
"Komal Aggarwal",
|
| 65 |
+
"Gabriela Styf Sj\u00f6man"
|
|
|
|
|
|
|
| 66 |
],
|
| 67 |
"companies": {
|
| 68 |
"TelecomTV": [
|
| 69 |
+
"Ray Le Maistre",
|
| 70 |
+
"Guy Daniels"
|
| 71 |
],
|
| 72 |
"Connectivitree": [
|
| 73 |
"Mark Gilmour"
|
|
|
|
| 77 |
"Terje Jensen"
|
| 78 |
],
|
| 79 |
"Vodafone": [
|
| 80 |
+
"Yago Tenorio",
|
| 81 |
+
"Komal Aggarwal"
|
| 82 |
],
|
| 83 |
"BT": [
|
|
|
|
| 84 |
"Alex Foster",
|
| 85 |
+
"Sarwar Khan",
|
| 86 |
+
"Mojdeh Amani",
|
| 87 |
"Mark Henry",
|
| 88 |
+
"Gabriela Styf Sj\u00f6man"
|
| 89 |
],
|
| 90 |
"American Tower": [
|
| 91 |
"Susan James"
|
|
|
|
| 101 |
"Sandeep Raithatha"
|
| 102 |
],
|
| 103 |
"Deutsche Telekom": [
|
| 104 |
+
"Ahmed Hafez",
|
| 105 |
+
"Abdu Mudesir"
|
| 106 |
],
|
| 107 |
"Telefonica": [
|
|
|
|
|
|
|
| 108 |
"Jose Antonio Martin Martinez",
|
| 109 |
+
"Enrique Blanco",
|
| 110 |
+
"Juan Manuel Caro",
|
| 111 |
+
"Mallik Rao",
|
| 112 |
+
"Luis Velarde Tazon"
|
| 113 |
],
|
| 114 |
"SoftBank": [
|
| 115 |
"Akira Tada"
|
|
|
|
| 138 |
"Sushil Rawat"
|
| 139 |
],
|
| 140 |
"Orange": [
|
| 141 |
+
"Philippe Ensarguet",
|
| 142 |
+
"Atoosa Hatefi"
|
| 143 |
],
|
| 144 |
"BT Business": [
|
| 145 |
"Colin Bannon"
|
|
|
|
| 163 |
"Dean Dennis"
|
| 164 |
],
|
| 165 |
"Rakuten Symphony": [
|
|
|
|
| 166 |
"Geoff Hollingworth",
|
| 167 |
+
"Vivek Chadha",
|
| 168 |
+
"Faiq Khan"
|
| 169 |
],
|
| 170 |
"Cambridge Management Consulting": [
|
| 171 |
"Tom Burton"
|
|
|
|
| 194 |
},
|
| 195 |
"sentiments": [],
|
| 196 |
"subjects": [
|
| 197 |
+
" AI ",
|
| 198 |
+
" B2B ",
|
| 199 |
+
" Network ",
|
| 200 |
+
" API ",
|
| 201 |
+
" Infrastructure ",
|
| 202 |
+
" Enterprise ",
|
| 203 |
+
" Innovation ",
|
| 204 |
+
" TechCo ",
|
| 205 |
+
" Connectivity ",
|
| 206 |
+
" Open RAN ",
|
| 207 |
+
" 5G "
|
| 208 |
]
|
| 209 |
}
|
db/faiss_index.faiss/index.pkl
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e22f1ebeec963e6b3a901c8810aafc0cca37ca0e06006b737f7cc7dab74f47f
|
| 3 |
+
size 573721
|
ttv_web_scraper.py
CHANGED
|
@@ -15,8 +15,8 @@ if not os.path.exists(CACHE_DIR):
|
|
| 15 |
|
| 16 |
DB_METADATA_FILE = os.path.join(CACHE_DIR, "db_metadata.json")
|
| 17 |
SUBJECTS = [
|
| 18 |
-
"5G",
|
| 19 |
-
"TechCo",
|
| 20 |
]
|
| 21 |
|
| 22 |
|
|
|
|
| 15 |
|
| 16 |
DB_METADATA_FILE = os.path.join(CACHE_DIR, "db_metadata.json")
|
| 17 |
SUBJECTS = [
|
| 18 |
+
" 5G ", " AI ", " Innovation ", " Network ", " Enterprise ", " Open RAN ",
|
| 19 |
+
" TechCo ", " B2B ", " API ", " Infrastructure ", " Connectivity "
|
| 20 |
]
|
| 21 |
|
| 22 |
|