Omar Solano commited on
Commit
84bd9c0
Β·
1 Parent(s): 9e9355f

load .env variables for vscode debugging

Browse files
notebooks/03-RAG_with_LlamaIndex.ipynb CHANGED
@@ -18,7 +18,7 @@
18
  },
19
  {
20
  "cell_type": "code",
21
- "execution_count": 1,
22
  "metadata": {
23
  "colab": {
24
  "base_uri": "https://localhost:8080/"
@@ -33,19 +33,27 @@
33
  },
34
  {
35
  "cell_type": "code",
36
- "execution_count": 2,
37
  "metadata": {
38
  "id": "CWholrWlt2OQ"
39
  },
40
  "outputs": [],
41
  "source": [
42
  "import os\n",
 
43
  "\n",
44
- "# Set your \"OPENAI_API_KEY\" environment variable\n",
45
- "os.environ[\"OPENAI_API_KEY\"] = \"<YOUR_OPENAI_KEY>\"\n",
 
 
 
 
 
46
  "\n",
47
  "# Get your GOOGLE_API_KEY from https://aistudio.google.com/app/apikey\n",
48
- "os.environ[\"GOOGLE_API_KEY\"] = \"<YOUR_GOOGLE_KEY>\""
 
 
49
  ]
50
  },
51
  {
@@ -77,7 +85,7 @@
77
  },
78
  {
79
  "cell_type": "code",
80
- "execution_count": 3,
81
  "metadata": {
82
  "colab": {
83
  "base_uri": "https://localhost:8080/"
@@ -92,7 +100,7 @@
92
  "text": [
93
  " % Total % Received % Xferd Average Speed Time Time Time Current\n",
94
  " Dload Upload Total Spent Left Speed\n",
95
- "100 169k 100 169k 0 0 1817k 0 --:--:-- --:--:-- --:--:-- 1823k\n"
96
  ]
97
  }
98
  ],
@@ -111,7 +119,7 @@
111
  },
112
  {
113
  "cell_type": "code",
114
- "execution_count": 4,
115
  "metadata": {
116
  "colab": {
117
  "base_uri": "https://localhost:8080/"
@@ -135,14 +143,16 @@
135
  "\n",
136
  "# Load the CSV file\n",
137
  "with open(\"./mini-dataset.csv\", mode=\"r\", encoding=\"utf-8\") as file:\n",
138
- " csv_reader = csv.reader(file)\n",
139
  "\n",
140
- " for idx, row in enumerate( csv_reader ):\n",
141
- " if idx == 0: continue; # Skip header row\n",
142
- " rows.append( row )\n",
 
 
143
  "\n",
144
  "# The number of characters in the dataset.\n",
145
- "print( \"number of articles:\", len( rows ) )"
146
  ]
147
  },
148
  {
@@ -156,7 +166,7 @@
156
  },
157
  {
158
  "cell_type": "code",
159
- "execution_count": 5,
160
  "metadata": {
161
  "id": "iXrr5-tnEfm9"
162
  },
@@ -170,7 +180,7 @@
170
  },
171
  {
172
  "cell_type": "code",
173
- "execution_count": 6,
174
  "metadata": {
175
  "colab": {
176
  "base_uri": "https://localhost:8080/",
@@ -210,8 +220,10 @@
210
  "text": [
211
  "/Users/omar/Documents/ai_repos/ai-tutor-rag-system/env/lib/python3.12/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
212
  " from .autonotebook import tqdm as notebook_tqdm\n",
213
- "Parsing nodes: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 14/14 [00:00<00:00, 252.38it/s]\n",
214
- "Generating embeddings: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 56/56 [00:01<00:00, 41.05it/s]\n"
 
 
215
  ]
216
  }
217
  ],
@@ -241,7 +253,7 @@
241
  },
242
  {
243
  "cell_type": "code",
244
- "execution_count": 7,
245
  "metadata": {
246
  "id": "bUaNH97dEfh9"
247
  },
@@ -259,7 +271,7 @@
259
  },
260
  {
261
  "cell_type": "code",
262
- "execution_count": 8,
263
  "metadata": {
264
  "colab": {
265
  "base_uri": "https://localhost:8080/"
@@ -272,21 +284,19 @@
272
  "name": "stdout",
273
  "output_type": "stream",
274
  "text": [
275
- "LLaMA 2 is available in four different sizes: 7 billion, 13 billion, 34 billion, and 70 billion parameters. \n",
276
  "\n"
277
  ]
278
  }
279
  ],
280
  "source": [
281
- "response = query_engine.query(\n",
282
- " \"How many parameters LLaMA2 model has?\"\n",
283
- ")\n",
284
  "print(response)"
285
  ]
286
  },
287
  {
288
  "cell_type": "code",
289
- "execution_count": 9,
290
  "metadata": {
291
  "colab": {
292
  "base_uri": "https://localhost:8080/"
@@ -299,15 +309,13 @@
299
  "name": "stdout",
300
  "output_type": "stream",
301
  "text": [
302
- "The context does not provide information about the release date of Llama 3. \n",
303
  "\n"
304
  ]
305
  }
306
  ],
307
  "source": [
308
- "response = query_engine.query(\n",
309
- " \"When will Llama3 will be released?\"\n",
310
- ")\n",
311
  "print(response)"
312
  ]
313
  }
 
18
  },
19
  {
20
  "cell_type": "code",
21
+ "execution_count": null,
22
  "metadata": {
23
  "colab": {
24
  "base_uri": "https://localhost:8080/"
 
33
  },
34
  {
35
  "cell_type": "code",
36
+ "execution_count": 1,
37
  "metadata": {
38
  "id": "CWholrWlt2OQ"
39
  },
40
  "outputs": [],
41
  "source": [
42
  "import os\n",
43
+ "from dotenv import load_dotenv\n",
44
  "\n",
45
+ "load_dotenv(\".env\")\n",
46
+ "\n",
47
+ "# Here we look for the OPENAI_API_KEY in the environment variables\n",
48
+ "OPENAI_API_KEY = os.getenv(\"OPENAI_API_KEY\")\n",
49
+ "if not OPENAI_API_KEY:\n",
50
+ " # If it's not found, you can set it manually\n",
51
+ " os.environ[\"OPENAI_API_KEY\"] = \"<YOUR_OPENAI_KEY>\"\n",
52
  "\n",
53
  "# Get your GOOGLE_API_KEY from https://aistudio.google.com/app/apikey\n",
54
+ "GOOGLE_API_KEY = os.getenv(\"GOOGLE_API_KEY\")\n",
55
+ "if not GOOGLE_API_KEY:\n",
56
+ " os.environ[\"GOOGLE_API_KEY\"] = \"<YOUR_GOOGLE_KEY>\""
57
  ]
58
  },
59
  {
 
85
  },
86
  {
87
  "cell_type": "code",
88
+ "execution_count": 2,
89
  "metadata": {
90
  "colab": {
91
  "base_uri": "https://localhost:8080/"
 
100
  "text": [
101
  " % Total % Received % Xferd Average Speed Time Time Time Current\n",
102
  " Dload Upload Total Spent Left Speed\n",
103
+ "100 169k 100 169k 0 0 772k 0 --:--:-- --:--:-- --:--:-- 774k\n"
104
  ]
105
  }
106
  ],
 
119
  },
120
  {
121
  "cell_type": "code",
122
+ "execution_count": 3,
123
  "metadata": {
124
  "colab": {
125
  "base_uri": "https://localhost:8080/"
 
143
  "\n",
144
  "# Load the CSV file\n",
145
  "with open(\"./mini-dataset.csv\", mode=\"r\", encoding=\"utf-8\") as file:\n",
146
+ " csv_reader = csv.reader(file)\n",
147
  "\n",
148
+ " for idx, row in enumerate(csv_reader):\n",
149
+ " if idx == 0:\n",
150
+ " continue\n",
151
+ " # Skip header row\n",
152
+ " rows.append(row)\n",
153
  "\n",
154
  "# The number of characters in the dataset.\n",
155
+ "print(\"number of articles:\", len(rows))"
156
  ]
157
  },
158
  {
 
166
  },
167
  {
168
  "cell_type": "code",
169
+ "execution_count": 4,
170
  "metadata": {
171
  "id": "iXrr5-tnEfm9"
172
  },
 
180
  },
181
  {
182
  "cell_type": "code",
183
+ "execution_count": 5,
184
  "metadata": {
185
  "colab": {
186
  "base_uri": "https://localhost:8080/",
 
220
  "text": [
221
  "/Users/omar/Documents/ai_repos/ai-tutor-rag-system/env/lib/python3.12/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
222
  " from .autonotebook import tqdm as notebook_tqdm\n",
223
+ "Parsing nodes: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 14/14 [00:00<00:00, 247.39it/s]\n",
224
+ "/Users/omar/Documents/ai_repos/ai-tutor-rag-system/env/lib/python3.12/site-packages/langchain/agents/json_chat/base.py:22: SyntaxWarning: invalid escape sequence '\\ '\n",
225
+ " \"\"\"Create an agent that uses JSON to format its logic, build for Chat Models.\n",
226
+ "Generating embeddings: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 56/56 [00:01<00:00, 43.08it/s]\n"
227
  ]
228
  }
229
  ],
 
253
  },
254
  {
255
  "cell_type": "code",
256
+ "execution_count": 6,
257
  "metadata": {
258
  "id": "bUaNH97dEfh9"
259
  },
 
271
  },
272
  {
273
  "cell_type": "code",
274
+ "execution_count": 7,
275
  "metadata": {
276
  "colab": {
277
  "base_uri": "https://localhost:8080/"
 
284
  "name": "stdout",
285
  "output_type": "stream",
286
  "text": [
287
+ "LLaMA 2 comes in four different sizes: 7 billion, 13 billion, 34 billion, and 70 billion parameters. \n",
288
  "\n"
289
  ]
290
  }
291
  ],
292
  "source": [
293
+ "response = query_engine.query(\"How many parameters LLaMA2 model has?\")\n",
 
 
294
  "print(response)"
295
  ]
296
  },
297
  {
298
  "cell_type": "code",
299
+ "execution_count": 8,
300
  "metadata": {
301
  "colab": {
302
  "base_uri": "https://localhost:8080/"
 
309
  "name": "stdout",
310
  "output_type": "stream",
311
  "text": [
312
+ "The context does not provide information about the release of Llama 3. \n",
313
  "\n"
314
  ]
315
  }
316
  ],
317
  "source": [
318
+ "response = query_engine.query(\"When will Llama3 will be released?\")\n",
 
 
319
  "print(response)"
320
  ]
321
  }