|
186 | 186 | }, |
187 | 187 | { |
188 | 188 | "cell_type": "code", |
189 | | - "execution_count": 1, |
| 189 | + "execution_count": 2, |
190 | 190 | "metadata": { |
191 | 191 | "id": "ggh5TzhkJkD9" |
192 | 192 | }, |
|
229 | 229 | }, |
230 | 230 | { |
231 | 231 | "cell_type": "code", |
232 | | - "execution_count": 2, |
| 232 | + "execution_count": 5, |
233 | 233 | "metadata": { |
234 | 234 | "colab": { |
235 | 235 | "base_uri": "https://localhost:8080/" |
|
259 | 259 | }, |
260 | 260 | { |
261 | 261 | "cell_type": "code", |
262 | | - "execution_count": 3, |
| 262 | + "execution_count": 6, |
263 | 263 | "metadata": { |
264 | 264 | "colab": { |
265 | 265 | "base_uri": "https://localhost:8080/" |
|
268 | 268 | "outputId": "a8430acc-2e6d-45fd-fc8b-601fbbd8289b" |
269 | 269 | }, |
270 | 270 | "outputs": [ |
271 | | - { |
272 | | - "name": "stderr", |
273 | | - "output_type": "stream", |
274 | | - "text": [ |
275 | | - "/Users/robert.shelton/.pyenv/versions/3.11.9/lib/python3.11/site-packages/pypdf/_crypt_providers/_cryptography.py:32: CryptographyDeprecationWarning: ARC4 has been moved to cryptography.hazmat.decrepit.ciphers.algorithms.ARC4 and will be removed from this module in 48.0.0.\n", |
276 | | - " from cryptography.hazmat.primitives.ciphers.algorithms import AES, ARC4\n" |
277 | | - ] |
278 | | - }, |
279 | 271 | { |
280 | 272 | "name": "stdout", |
281 | 273 | "output_type": "stream", |
282 | 274 | "text": [ |
283 | | - "Done preprocessing. Created 210 chunks of the original pdf resources/nke-10k-2023.pdf\n" |
| 275 | + "Done preprocessing. Created 211 chunks of the original pdf resources/nke-10k-2023.pdf\n" |
284 | 276 | ] |
285 | 277 | } |
286 | 278 | ], |
|
312 | 304 | }, |
313 | 305 | { |
314 | 306 | "cell_type": "code", |
315 | | - "execution_count": 4, |
| 307 | + "execution_count": 7, |
316 | 308 | "metadata": { |
317 | 309 | "colab": { |
318 | 310 | "base_uri": "https://localhost:8080/", |
|
467 | 459 | "outputId": "b0f0d2c1-41dc-4932-990b-53d2912af19e" |
468 | 460 | }, |
469 | 461 | "outputs": [ |
470 | | - { |
471 | | - "name": "stderr", |
472 | | - "output_type": "stream", |
473 | | - "text": [ |
474 | | - "/Users/robert.shelton/.pyenv/versions/3.11.9/lib/python3.11/site-packages/huggingface_hub/file_download.py:1142: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n", |
475 | | - " warnings.warn(\n", |
476 | | - "/Users/robert.shelton/.pyenv/versions/3.11.9/lib/python3.11/site-packages/huggingface_hub/file_download.py:1142: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n", |
477 | | - " warnings.warn(\n" |
478 | | - ] |
479 | | - }, |
480 | 462 | { |
481 | 463 | "data": { |
482 | 464 | "text/plain": [ |
483 | 465 | "True" |
484 | 466 | ] |
485 | 467 | }, |
486 | | - "execution_count": 4, |
| 468 | + "execution_count": 7, |
487 | 469 | "metadata": {}, |
488 | 470 | "output_type": "execute_result" |
489 | 471 | } |
490 | 472 | ], |
491 | 473 | "source": [ |
492 | | - "from redisvl.utils.vectorize import HFTextVectorizer\n", |
| 474 | + "import warnings\n", |
493 | 475 | "import pandas as pd\n", |
494 | 476 | "from tqdm.auto import tqdm\n", |
| 477 | + "from redisvl.utils.vectorize import HFTextVectorizer\n", |
| 478 | + "\n", |
| 479 | + "warnings.filterwarnings(\"ignore\")\n", |
495 | 480 | "\n", |
496 | 481 | "hf = HFTextVectorizer(\"sentence-transformers/all-MiniLM-L6-v2\")\n", |
497 | 482 | "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n", |
|
0 commit comments