Commit
•
489dc47
1
Parent(s):
81dd58a
Training in progress, epoch 0
Browse files- .ipynb_checkpoints/data-checkpoint.csv +0 -0
- .ipynb_checkpoints/fine-tuned-phi2-checkpoint.ipynb +6 -0
- .ipynb_checkpoints/phi-2-custom-checkpoint.ipynb +697 -0
- adapter_config.json +28 -0
- adapter_model.safetensors +3 -0
- fine-tuned-phi2.ipynb +33 -0
- phi-2-custom.ipynb +1045 -0
- training_args.bin +3 -0
.ipynb_checkpoints/data-checkpoint.csv
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
.ipynb_checkpoints/fine-tuned-phi2-checkpoint.ipynb
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [],
|
| 3 |
+
"metadata": {},
|
| 4 |
+
"nbformat": 4,
|
| 5 |
+
"nbformat_minor": 5
|
| 6 |
+
}
|
.ipynb_checkpoints/phi-2-custom-checkpoint.ipynb
ADDED
|
@@ -0,0 +1,697 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [
|
| 3 |
+
{
|
| 4 |
+
"cell_type": "code",
|
| 5 |
+
"execution_count": 1,
|
| 6 |
+
"id": "b3b92bc7-d105-405f-970d-804d298b9976",
|
| 7 |
+
"metadata": {},
|
| 8 |
+
"outputs": [
|
| 9 |
+
{
|
| 10 |
+
"name": "stdout",
|
| 11 |
+
"output_type": "stream",
|
| 12 |
+
"text": [
|
| 13 |
+
"Requirement already satisfied: accelerate in /usr/local/lib/python3.11/site-packages (0.27.2)\n",
|
| 14 |
+
"Requirement already satisfied: transformers in /usr/local/lib/python3.11/site-packages (4.37.2)\n",
|
| 15 |
+
"Requirement already satisfied: einops in /usr/local/lib/python3.11/site-packages (0.7.0)\n",
|
| 16 |
+
"Requirement already satisfied: datasets in /usr/local/lib/python3.11/site-packages (2.17.0)\n",
|
| 17 |
+
"Requirement already satisfied: peft in /usr/local/lib/python3.11/site-packages (0.8.2)\n",
|
| 18 |
+
"Requirement already satisfied: bitsandbytes in /usr/local/lib/python3.11/site-packages (0.42.0)\n",
|
| 19 |
+
"Requirement already satisfied: torch in /usr/local/lib/python3.11/site-packages (2.2.0)\n",
|
| 20 |
+
"Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.11/site-packages (from accelerate) (1.26.4)\n",
|
| 21 |
+
"Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.11/site-packages (from accelerate) (23.2)\n",
|
| 22 |
+
"Requirement already satisfied: psutil in /usr/local/lib/python3.11/site-packages (from accelerate) (5.9.8)\n",
|
| 23 |
+
"Requirement already satisfied: pyyaml in /usr/local/lib/python3.11/site-packages (from accelerate) (6.0.1)\n",
|
| 24 |
+
"Requirement already satisfied: huggingface-hub in /usr/local/lib/python3.11/site-packages (from accelerate) (0.20.3)\n",
|
| 25 |
+
"Requirement already satisfied: safetensors>=0.3.1 in /usr/local/lib/python3.11/site-packages (from accelerate) (0.4.2)\n",
|
| 26 |
+
"Requirement already satisfied: filelock in /usr/local/lib/python3.11/site-packages (from transformers) (3.13.1)\n",
|
| 27 |
+
"Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.11/site-packages (from transformers) (2023.12.25)\n",
|
| 28 |
+
"Requirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from transformers) (2.31.0)\n",
|
| 29 |
+
"Requirement already satisfied: tokenizers<0.19,>=0.14 in /usr/local/lib/python3.11/site-packages (from transformers) (0.15.2)\n",
|
| 30 |
+
"Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.11/site-packages (from transformers) (4.66.2)\n",
|
| 31 |
+
"Requirement already satisfied: pyarrow>=12.0.0 in /usr/local/lib/python3.11/site-packages (from datasets) (15.0.0)\n",
|
| 32 |
+
"Requirement already satisfied: pyarrow-hotfix in /usr/local/lib/python3.11/site-packages (from datasets) (0.6)\n",
|
| 33 |
+
"Requirement already satisfied: dill<0.3.9,>=0.3.0 in /usr/local/lib/python3.11/site-packages (from datasets) (0.3.8)\n",
|
| 34 |
+
"Requirement already satisfied: pandas in /usr/local/lib/python3.11/site-packages (from datasets) (2.2.0)\n",
|
| 35 |
+
"Requirement already satisfied: xxhash in /usr/local/lib/python3.11/site-packages (from datasets) (3.4.1)\n",
|
| 36 |
+
"Requirement already satisfied: multiprocess in /usr/local/lib/python3.11/site-packages (from datasets) (0.70.16)\n",
|
| 37 |
+
"Requirement already satisfied: fsspec<=2023.10.0,>=2023.1.0 in /usr/local/lib/python3.11/site-packages (from fsspec[http]<=2023.10.0,>=2023.1.0->datasets) (2023.10.0)\n",
|
| 38 |
+
"Requirement already satisfied: aiohttp in /usr/local/lib/python3.11/site-packages (from datasets) (3.8.3)\n",
|
| 39 |
+
"Requirement already satisfied: scipy in /usr/local/lib/python3.11/site-packages (from bitsandbytes) (1.12.0)\n",
|
| 40 |
+
"Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.11/site-packages (from torch) (4.9.0)\n",
|
| 41 |
+
"Requirement already satisfied: sympy in /usr/local/lib/python3.11/site-packages (from torch) (1.12)\n",
|
| 42 |
+
"Requirement already satisfied: networkx in /usr/local/lib/python3.11/site-packages (from torch) (3.2.1)\n",
|
| 43 |
+
"Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/site-packages (from torch) (3.1.3)\n",
|
| 44 |
+
"Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch) (12.1.105)\n",
|
| 45 |
+
"Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch) (12.1.105)\n",
|
| 46 |
+
"Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch) (12.1.105)\n",
|
| 47 |
+
"Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /usr/local/lib/python3.11/site-packages (from torch) (8.9.2.26)\n",
|
| 48 |
+
"Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /usr/local/lib/python3.11/site-packages (from torch) (12.1.3.1)\n",
|
| 49 |
+
"Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /usr/local/lib/python3.11/site-packages (from torch) (11.0.2.54)\n",
|
| 50 |
+
"Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /usr/local/lib/python3.11/site-packages (from torch) (10.3.2.106)\n",
|
| 51 |
+
"Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /usr/local/lib/python3.11/site-packages (from torch) (11.4.5.107)\n",
|
| 52 |
+
"Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /usr/local/lib/python3.11/site-packages (from torch) (12.1.0.106)\n",
|
| 53 |
+
"Requirement already satisfied: nvidia-nccl-cu12==2.19.3 in /usr/local/lib/python3.11/site-packages (from torch) (2.19.3)\n",
|
| 54 |
+
"Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /usr/local/lib/python3.11/site-packages (from torch) (12.1.105)\n",
|
| 55 |
+
"Requirement already satisfied: triton==2.2.0 in /usr/local/lib/python3.11/site-packages (from torch) (2.2.0)\n",
|
| 56 |
+
"Requirement already satisfied: nvidia-nvjitlink-cu12 in /usr/local/lib/python3.11/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch) (12.3.101)\n",
|
| 57 |
+
"Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (23.2.0)\n",
|
| 58 |
+
"Requirement already satisfied: charset-normalizer<3.0,>=2.0 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (2.1.1)\n",
|
| 59 |
+
"Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (6.0.5)\n",
|
| 60 |
+
"Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (4.0.3)\n",
|
| 61 |
+
"Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (1.9.4)\n",
|
| 62 |
+
"Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (1.4.1)\n",
|
| 63 |
+
"Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (1.3.1)\n",
|
| 64 |
+
"Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (3.6)\n",
|
| 65 |
+
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (2.2.0)\n",
|
| 66 |
+
"Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (2024.2.2)\n",
|
| 67 |
+
"Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2->torch) (2.1.5)\n",
|
| 68 |
+
"Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from pandas->datasets) (2.8.2)\n",
|
| 69 |
+
"Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.11/site-packages (from pandas->datasets) (2024.1)\n",
|
| 70 |
+
"Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.11/site-packages (from pandas->datasets) (2024.1)\n",
|
| 71 |
+
"Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.11/site-packages (from sympy->torch) (1.3.0)\n",
|
| 72 |
+
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.11/site-packages (from python-dateutil>=2.8.2->pandas->datasets) (1.16.0)\n"
|
| 73 |
+
]
|
| 74 |
+
}
|
| 75 |
+
],
|
| 76 |
+
"source": [
|
| 77 |
+
"!pip install accelerate transformers einops datasets peft bitsandbytes torch"
|
| 78 |
+
]
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"cell_type": "code",
|
| 82 |
+
"execution_count": 2,
|
| 83 |
+
"id": "46a53303-b585-4b02-956f-4af173410e25",
|
| 84 |
+
"metadata": {},
|
| 85 |
+
"outputs": [
|
| 86 |
+
{
|
| 87 |
+
"name": "stderr",
|
| 88 |
+
"output_type": "stream",
|
| 89 |
+
"text": [
|
| 90 |
+
"/usr/local/lib/python3.11/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",
|
| 91 |
+
" from .autonotebook import tqdm as notebook_tqdm\n"
|
| 92 |
+
]
|
| 93 |
+
}
|
| 94 |
+
],
|
| 95 |
+
"source": [
|
| 96 |
+
"import torch\n",
|
| 97 |
+
"from datasets import load_dataset, Dataset\n",
|
| 98 |
+
"from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments, Trainer, DataCollatorForLanguageModeling, BitsAndBytesConfig\n",
|
| 99 |
+
"from peft import LoraConfig, get_peft_model\n",
|
| 100 |
+
"import os"
|
| 101 |
+
]
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"cell_type": "code",
|
| 105 |
+
"execution_count": 3,
|
| 106 |
+
"id": "ffa3ee11-3929-4032-8758-322cda3a912e",
|
| 107 |
+
"metadata": {},
|
| 108 |
+
"outputs": [
|
| 109 |
+
{
|
| 110 |
+
"name": "stderr",
|
| 111 |
+
"output_type": "stream",
|
| 112 |
+
"text": [
|
| 113 |
+
"Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
|
| 114 |
+
]
|
| 115 |
+
}
|
| 116 |
+
],
|
| 117 |
+
"source": [
|
| 118 |
+
"tokenizer = AutoTokenizer.from_pretrained(\"microsoft/phi-2\", trust_remote_code=True)\n",
|
| 119 |
+
"tokenizer.pad_token = tokenizer.eos_token"
|
| 120 |
+
]
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"cell_type": "code",
|
| 124 |
+
"execution_count": 4,
|
| 125 |
+
"id": "b17096e6-c633-4d88-8cfc-fb738c1e4ca0",
|
| 126 |
+
"metadata": {
|
| 127 |
+
"scrolled": true
|
| 128 |
+
},
|
| 129 |
+
"outputs": [
|
| 130 |
+
{
|
| 131 |
+
"name": "stderr",
|
| 132 |
+
"output_type": "stream",
|
| 133 |
+
"text": [
|
| 134 |
+
"Loading checkpoint shards: 100%|██████████| 2/2 [00:06<00:00, 3.08s/it]\n"
|
| 135 |
+
]
|
| 136 |
+
}
|
| 137 |
+
],
|
| 138 |
+
"source": [
|
| 139 |
+
"bnb_config = BitsAndBytesConfig(\n",
|
| 140 |
+
" load_in_4bit=True,\n",
|
| 141 |
+
" bnb_4bit_use_double_quant=True,\n",
|
| 142 |
+
" bnb_4bit_quant_type=\"nf4\",\n",
|
| 143 |
+
" bnb_4bit_compute_dtype=torch.float16\n",
|
| 144 |
+
")\n",
|
| 145 |
+
"\n",
|
| 146 |
+
"model = AutoModelForCausalLM.from_pretrained(\n",
|
| 147 |
+
" \"microsoft/phi-2\",\n",
|
| 148 |
+
" device_map={\"\":0},\n",
|
| 149 |
+
" trust_remote_code=True,\n",
|
| 150 |
+
" quantization_config=bnb_config\n",
|
| 151 |
+
")"
|
| 152 |
+
]
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"cell_type": "code",
|
| 156 |
+
"execution_count": 5,
|
| 157 |
+
"id": "e15aa794-e17c-4b09-a64a-c60377259218",
|
| 158 |
+
"metadata": {},
|
| 159 |
+
"outputs": [
|
| 160 |
+
{
|
| 161 |
+
"name": "stdout",
|
| 162 |
+
"output_type": "stream",
|
| 163 |
+
"text": [
|
| 164 |
+
"PhiForCausalLM(\n",
|
| 165 |
+
" (model): PhiModel(\n",
|
| 166 |
+
" (embed_tokens): Embedding(51200, 2560)\n",
|
| 167 |
+
" (embed_dropout): Dropout(p=0.0, inplace=False)\n",
|
| 168 |
+
" (layers): ModuleList(\n",
|
| 169 |
+
" (0-31): 32 x PhiDecoderLayer(\n",
|
| 170 |
+
" (self_attn): PhiAttention(\n",
|
| 171 |
+
" (q_proj): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 172 |
+
" (k_proj): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 173 |
+
" (v_proj): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 174 |
+
" (dense): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 175 |
+
" (rotary_emb): PhiRotaryEmbedding()\n",
|
| 176 |
+
" )\n",
|
| 177 |
+
" (mlp): PhiMLP(\n",
|
| 178 |
+
" (activation_fn): NewGELUActivation()\n",
|
| 179 |
+
" (fc1): Linear4bit(in_features=2560, out_features=10240, bias=True)\n",
|
| 180 |
+
" (fc2): Linear4bit(in_features=10240, out_features=2560, bias=True)\n",
|
| 181 |
+
" )\n",
|
| 182 |
+
" (input_layernorm): LayerNorm((2560,), eps=1e-05, elementwise_affine=True)\n",
|
| 183 |
+
" (resid_dropout): Dropout(p=0.1, inplace=False)\n",
|
| 184 |
+
" )\n",
|
| 185 |
+
" )\n",
|
| 186 |
+
" (final_layernorm): LayerNorm((2560,), eps=1e-05, elementwise_affine=True)\n",
|
| 187 |
+
" )\n",
|
| 188 |
+
" (lm_head): Linear(in_features=2560, out_features=51200, bias=True)\n",
|
| 189 |
+
")\n"
|
| 190 |
+
]
|
| 191 |
+
}
|
| 192 |
+
],
|
| 193 |
+
"source": [
|
| 194 |
+
"print(model)"
|
| 195 |
+
]
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"cell_type": "code",
|
| 199 |
+
"execution_count": 6,
|
| 200 |
+
"id": "18d5599f-992d-4d8e-a90c-4d43774be473",
|
| 201 |
+
"metadata": {},
|
| 202 |
+
"outputs": [
|
| 203 |
+
{
|
| 204 |
+
"name": "stdout",
|
| 205 |
+
"output_type": "stream",
|
| 206 |
+
"text": [
|
| 207 |
+
"trainable params: 17,039,360 || all params: 2,796,723,200 || trainable%: 0.6092615815537269\n"
|
| 208 |
+
]
|
| 209 |
+
}
|
| 210 |
+
],
|
| 211 |
+
"source": [
|
| 212 |
+
"config = LoraConfig(\n",
|
| 213 |
+
" r=16,\n",
|
| 214 |
+
" lora_alpha=16,\n",
|
| 215 |
+
" target_modules=[\"q_proj\",\"k_proj\",\"v_proj\"],\n",
|
| 216 |
+
" lora_dropout=0.05,\n",
|
| 217 |
+
" bias=\"none\",\n",
|
| 218 |
+
" task_type=\"CAUSAL_LM\"\n",
|
| 219 |
+
")\n",
|
| 220 |
+
"\n",
|
| 221 |
+
"model = get_peft_model(model, config)\n",
|
| 222 |
+
"model.print_trainable_parameters()"
|
| 223 |
+
]
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"cell_type": "code",
|
| 227 |
+
"execution_count": 7,
|
| 228 |
+
"id": "baeee903-3dce-48b2-93c3-7a697d8c6daf",
|
| 229 |
+
"metadata": {},
|
| 230 |
+
"outputs": [],
|
| 231 |
+
"source": [
|
| 232 |
+
"def tokenize(sample):\n",
|
| 233 |
+
" model_inps = tokenizer(sample[\"text\"], padding=True, truncation=True, max_length=512)\n",
|
| 234 |
+
" return model_inps"
|
| 235 |
+
]
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"cell_type": "code",
|
| 239 |
+
"execution_count": 8,
|
| 240 |
+
"id": "28a9b24a-a822-4fcb-96b3-d77b7ea30a5f",
|
| 241 |
+
"metadata": {},
|
| 242 |
+
"outputs": [
|
| 243 |
+
{
|
| 244 |
+
"name": "stdout",
|
| 245 |
+
"output_type": "stream",
|
| 246 |
+
"text": [
|
| 247 |
+
"Requirement already satisfied: scikit-learn in /usr/local/lib/python3.11/site-packages (1.4.0)\n",
|
| 248 |
+
"Requirement already satisfied: numpy<2.0,>=1.19.5 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (1.26.4)\n",
|
| 249 |
+
"Requirement already satisfied: scipy>=1.6.0 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (1.12.0)\n",
|
| 250 |
+
"Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (1.3.2)\n",
|
| 251 |
+
"Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (3.3.0)\n"
|
| 252 |
+
]
|
| 253 |
+
}
|
| 254 |
+
],
|
| 255 |
+
"source": [
|
| 256 |
+
"!pip install scikit-learn"
|
| 257 |
+
]
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"cell_type": "code",
|
| 261 |
+
"execution_count": 9,
|
| 262 |
+
"id": "1ee7fd2a-38e4-4f23-a978-0bdeeda64d8b",
|
| 263 |
+
"metadata": {},
|
| 264 |
+
"outputs": [],
|
| 265 |
+
"source": [
|
| 266 |
+
"import pandas as pd\n",
|
| 267 |
+
"\n",
|
| 268 |
+
"from sklearn.model_selection import train_test_split\n",
|
| 269 |
+
"dataset_name='data.csv'\n",
|
| 270 |
+
"df = pd.read_csv(dataset_name)\n",
|
| 271 |
+
"#train, test = train_test_split(df, test_size=0.2)"
|
| 272 |
+
]
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"cell_type": "code",
|
| 276 |
+
"execution_count": 10,
|
| 277 |
+
"id": "e84c29e2-843e-42c2-8c0f-324d392e671c",
|
| 278 |
+
"metadata": {},
|
| 279 |
+
"outputs": [
|
| 280 |
+
{
|
| 281 |
+
"name": "stderr",
|
| 282 |
+
"output_type": "stream",
|
| 283 |
+
"text": [
|
| 284 |
+
"Tokenizing data: 100%|██████████| 16412/16412 [00:05<00:00, 2902.91 examples/s]\n"
|
| 285 |
+
]
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"data": {
|
| 289 |
+
"text/plain": [
|
| 290 |
+
"Dataset({\n",
|
| 291 |
+
" features: ['input_ids', 'attention_mask'],\n",
|
| 292 |
+
" num_rows: 16412\n",
|
| 293 |
+
"})"
|
| 294 |
+
]
|
| 295 |
+
},
|
| 296 |
+
"execution_count": 10,
|
| 297 |
+
"metadata": {},
|
| 298 |
+
"output_type": "execute_result"
|
| 299 |
+
}
|
| 300 |
+
],
|
| 301 |
+
"source": [
|
| 302 |
+
"data_df = df\n",
|
| 303 |
+
"data_df[\"text\"] = data_df[[\"user\", \"assistant\"]].apply(lambda x: \"question: \" + str(x[\"user\"]) + \" answer: \" + str(x[\"assistant\"]), axis=1)\n",
|
| 304 |
+
"data = Dataset.from_pandas(data_df)\n",
|
| 305 |
+
"tokenized_data = data.map(tokenize, batched=True, desc=\"Tokenizing data\", remove_columns=data.column_names)\n",
|
| 306 |
+
"tokenized_data"
|
| 307 |
+
]
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"cell_type": "code",
|
| 311 |
+
"execution_count": 11,
|
| 312 |
+
"id": "ac968254-5338-49df-950d-222b82647407",
|
| 313 |
+
"metadata": {},
|
| 314 |
+
"outputs": [],
|
| 315 |
+
"source": [
|
| 316 |
+
"training_arguments = TrainingArguments(\n",
|
| 317 |
+
" output_dir=\".\",\n",
|
| 318 |
+
" per_device_train_batch_size=4,\n",
|
| 319 |
+
" gradient_accumulation_steps=1,\n",
|
| 320 |
+
" learning_rate=2e-4,\n",
|
| 321 |
+
" lr_scheduler_type=\"cosine\",\n",
|
| 322 |
+
" save_strategy=\"epoch\",\n",
|
| 323 |
+
" logging_steps=100,\n",
|
| 324 |
+
" max_steps=1100,\n",
|
| 325 |
+
" num_train_epochs=3,\n",
|
| 326 |
+
" push_to_hub=True\n",
|
| 327 |
+
" )"
|
| 328 |
+
]
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"cell_type": "code",
|
| 332 |
+
"execution_count": 12,
|
| 333 |
+
"id": "729df2d0-0890-4ac4-adf3-c167a6e9669d",
|
| 334 |
+
"metadata": {},
|
| 335 |
+
"outputs": [
|
| 336 |
+
{
|
| 337 |
+
"name": "stderr",
|
| 338 |
+
"output_type": "stream",
|
| 339 |
+
"text": [
|
| 340 |
+
"huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
|
| 341 |
+
"To disable this warning, you can either:\n",
|
| 342 |
+
"\t- Avoid using `tokenizers` before the fork if possible\n",
|
| 343 |
+
"\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
|
| 344 |
+
]
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"name": "stdout",
|
| 348 |
+
"output_type": "stream",
|
| 349 |
+
"text": [
|
| 350 |
+
"Requirement already satisfied: huggingface_hub in /usr/local/lib/python3.11/site-packages (0.20.3)\n",
|
| 351 |
+
"Requirement already satisfied: filelock in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (3.13.1)\n",
|
| 352 |
+
"Requirement already satisfied: fsspec>=2023.5.0 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (2023.10.0)\n",
|
| 353 |
+
"Requirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (2.31.0)\n",
|
| 354 |
+
"Requirement already satisfied: tqdm>=4.42.1 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (4.66.2)\n",
|
| 355 |
+
"Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (6.0.1)\n",
|
| 356 |
+
"Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (4.9.0)\n",
|
| 357 |
+
"Requirement already satisfied: packaging>=20.9 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (23.2)\n",
|
| 358 |
+
"Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (2.1.1)\n",
|
| 359 |
+
"Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (3.6)\n",
|
| 360 |
+
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (2.2.0)\n",
|
| 361 |
+
"Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (2024.2.2)\n"
|
| 362 |
+
]
|
| 363 |
+
}
|
| 364 |
+
],
|
| 365 |
+
"source": [
|
| 366 |
+
"!pip install huggingface_hub"
|
| 367 |
+
]
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"cell_type": "code",
|
| 371 |
+
"execution_count": 13,
|
| 372 |
+
"id": "525f81a4-eb92-466e-bb9c-cd63122231ab",
|
| 373 |
+
"metadata": {},
|
| 374 |
+
"outputs": [
|
| 375 |
+
{
|
| 376 |
+
"name": "stderr",
|
| 377 |
+
"output_type": "stream",
|
| 378 |
+
"text": [
|
| 379 |
+
"huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
|
| 380 |
+
"To disable this warning, you can either:\n",
|
| 381 |
+
"\t- Avoid using `tokenizers` before the fork if possible\n",
|
| 382 |
+
"\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
|
| 383 |
+
]
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"name": "stdout",
|
| 387 |
+
"output_type": "stream",
|
| 388 |
+
"text": [
|
| 389 |
+
"Requirement already satisfied: jupyterlab in /usr/local/lib/python3.11/site-packages (4.1.0)\n",
|
| 390 |
+
"Collecting jupyterlab\n",
|
| 391 |
+
" Downloading jupyterlab-4.1.1-py3-none-any.whl.metadata (15 kB)\n",
|
| 392 |
+
"Collecting jupyterlab-git\n",
|
| 393 |
+
" Downloading jupyterlab_git-0.50.0-py3-none-any.whl.metadata (31 kB)\n",
|
| 394 |
+
"Requirement already satisfied: async-lru>=1.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.0.4)\n",
|
| 395 |
+
"Requirement already satisfied: httpx>=0.25.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (0.26.0)\n",
|
| 396 |
+
"Requirement already satisfied: ipykernel in /usr/local/lib/python3.11/site-packages (from jupyterlab) (6.29.1)\n",
|
| 397 |
+
"Requirement already satisfied: jinja2>=3.0.3 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (3.1.3)\n",
|
| 398 |
+
"Requirement already satisfied: jupyter-core in /usr/local/lib/python3.11/site-packages (from jupyterlab) (5.7.1)\n",
|
| 399 |
+
"Requirement already satisfied: jupyter-lsp>=2.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.2.2)\n",
|
| 400 |
+
"Requirement already satisfied: jupyter-server<3,>=2.4.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.12.5)\n",
|
| 401 |
+
"Requirement already satisfied: jupyterlab-server<3,>=2.19.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.25.2)\n",
|
| 402 |
+
"Requirement already satisfied: notebook-shim>=0.2 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (0.2.3)\n",
|
| 403 |
+
"Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from jupyterlab) (23.2)\n",
|
| 404 |
+
"Requirement already satisfied: tornado>=6.2.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (6.4)\n",
|
| 405 |
+
"Requirement already satisfied: traitlets in /usr/local/lib/python3.11/site-packages (from jupyterlab) (5.14.1)\n",
|
| 406 |
+
"Collecting nbdime~=4.0.1 (from jupyterlab-git)\n",
|
| 407 |
+
" Downloading nbdime-4.0.1-py3-none-any.whl.metadata (9.5 kB)\n",
|
| 408 |
+
"Requirement already satisfied: nbformat in /usr/local/lib/python3.11/site-packages (from jupyterlab-git) (5.9.2)\n",
|
| 409 |
+
"Requirement already satisfied: pexpect in /usr/local/lib/python3.11/site-packages (from jupyterlab-git) (4.9.0)\n",
|
| 410 |
+
"Requirement already satisfied: anyio in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (4.2.0)\n",
|
| 411 |
+
"Requirement already satisfied: certifi in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (2024.2.2)\n",
|
| 412 |
+
"Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (1.0.2)\n",
|
| 413 |
+
"Requirement already satisfied: idna in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (3.6)\n",
|
| 414 |
+
"Requirement already satisfied: sniffio in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (1.3.0)\n",
|
| 415 |
+
"Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.11/site-packages (from httpcore==1.*->httpx>=0.25.0->jupyterlab) (0.14.0)\n",
|
| 416 |
+
"Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2>=3.0.3->jupyterlab) (2.1.5)\n",
|
| 417 |
+
"Requirement already satisfied: argon2-cffi in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (23.1.0)\n",
|
| 418 |
+
"Requirement already satisfied: jupyter-client>=7.4.4 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (8.6.0)\n",
|
| 419 |
+
"Requirement already satisfied: jupyter-events>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.9.0)\n",
|
| 420 |
+
"Requirement already satisfied: jupyter-server-terminals in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.5.2)\n",
|
| 421 |
+
"Requirement already satisfied: nbconvert>=6.4.4 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (7.15.0)\n",
|
| 422 |
+
"Requirement already satisfied: overrides in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (7.7.0)\n",
|
| 423 |
+
"Requirement already satisfied: prometheus-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.19.0)\n",
|
| 424 |
+
"Requirement already satisfied: pyzmq>=24 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (25.1.2)\n",
|
| 425 |
+
"Requirement already satisfied: send2trash>=1.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (1.8.2)\n",
|
| 426 |
+
"Requirement already satisfied: terminado>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.18.0)\n",
|
| 427 |
+
"Requirement already satisfied: websocket-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (1.7.0)\n",
|
| 428 |
+
"Requirement already satisfied: platformdirs>=2.5 in /usr/local/lib/python3.11/site-packages (from jupyter-core->jupyterlab) (4.2.0)\n",
|
| 429 |
+
"Requirement already satisfied: babel>=2.10 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (2.14.0)\n",
|
| 430 |
+
"Requirement already satisfied: json5>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (0.9.14)\n",
|
| 431 |
+
"Requirement already satisfied: jsonschema>=4.18.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (4.21.1)\n",
|
| 432 |
+
"Requirement already satisfied: requests>=2.31 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (2.31.0)\n",
|
| 433 |
+
"Collecting colorama (from nbdime~=4.0.1->jupyterlab-git)\n",
|
| 434 |
+
" Downloading colorama-0.4.6-py2.py3-none-any.whl (25 kB)\n",
|
| 435 |
+
"Collecting gitpython!=2.1.4,!=2.1.5,!=2.1.6 (from nbdime~=4.0.1->jupyterlab-git)\n",
|
| 436 |
+
" Downloading GitPython-3.1.41-py3-none-any.whl.metadata (14 kB)\n",
|
| 437 |
+
"Collecting jupyter-server-mathjax>=0.2.2 (from nbdime~=4.0.1->jupyterlab-git)\n",
|
| 438 |
+
" Downloading jupyter_server_mathjax-0.2.6-py3-none-any.whl (3.1 MB)\n",
|
| 439 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.1/3.1 MB\u001b[0m \u001b[31m97.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 440 |
+
"\u001b[?25hRequirement already satisfied: pygments in /usr/local/lib/python3.11/site-packages (from nbdime~=4.0.1->jupyterlab-git) (2.17.2)\n",
|
| 441 |
+
"Requirement already satisfied: fastjsonschema in /usr/local/lib/python3.11/site-packages (from nbformat->jupyterlab-git) (2.19.1)\n",
|
| 442 |
+
"Requirement already satisfied: comm>=0.1.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (0.2.1)\n",
|
| 443 |
+
"Requirement already satisfied: debugpy>=1.6.5 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (1.8.0)\n",
|
| 444 |
+
"Requirement already satisfied: ipython>=7.23.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (8.21.0)\n",
|
| 445 |
+
"Requirement already satisfied: matplotlib-inline>=0.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (0.1.6)\n",
|
| 446 |
+
"Requirement already satisfied: nest-asyncio in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (1.6.0)\n",
|
| 447 |
+
"Requirement already satisfied: psutil in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (5.9.8)\n",
|
| 448 |
+
"Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.11/site-packages (from pexpect->jupyterlab-git) (0.7.0)\n",
|
| 449 |
+
"Collecting gitdb<5,>=4.0.1 (from gitpython!=2.1.4,!=2.1.5,!=2.1.6->nbdime~=4.0.1->jupyterlab-git)\n",
|
| 450 |
+
" Downloading gitdb-4.0.11-py3-none-any.whl.metadata (1.2 kB)\n",
|
| 451 |
+
"Requirement already satisfied: decorator in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (5.1.1)\n",
|
| 452 |
+
"Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (0.19.1)\n",
|
| 453 |
+
"Requirement already satisfied: prompt-toolkit<3.1.0,>=3.0.41 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (3.0.43)\n",
|
| 454 |
+
"Requirement already satisfied: stack-data in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (0.6.3)\n",
|
| 455 |
+
"Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (23.2.0)\n",
|
| 456 |
+
"Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (2023.12.1)\n",
|
| 457 |
+
"Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (0.33.0)\n",
|
| 458 |
+
"Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (0.17.1)\n",
|
| 459 |
+
"Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-client>=7.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (2.8.2)\n",
|
| 460 |
+
"Requirement already satisfied: python-json-logger>=2.0.4 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.0.7)\n",
|
| 461 |
+
"Requirement already satisfied: pyyaml>=5.3 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (6.0.1)\n",
|
| 462 |
+
"Requirement already satisfied: rfc3339-validator in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (0.1.4)\n",
|
| 463 |
+
"Requirement already satisfied: rfc3986-validator>=0.1.1 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (0.1.1)\n",
|
| 464 |
+
"Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (4.12.3)\n",
|
| 465 |
+
"Requirement already satisfied: bleach!=5.0.0 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (6.1.0)\n",
|
| 466 |
+
"Requirement already satisfied: defusedxml in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.7.1)\n",
|
| 467 |
+
"Requirement already satisfied: jupyterlab-pygments in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.3.0)\n",
|
| 468 |
+
"Requirement already satisfied: mistune<4,>=2.0.3 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (3.0.2)\n",
|
| 469 |
+
"Requirement already satisfied: nbclient>=0.5.0 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.9.0)\n",
|
| 470 |
+
"Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.5.1)\n",
|
| 471 |
+
"Requirement already satisfied: tinycss2 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.2.1)\n",
|
| 472 |
+
"Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.19.0->jupyterlab) (2.1.1)\n",
|
| 473 |
+
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.19.0->jupyterlab) (2.2.0)\n",
|
| 474 |
+
"Requirement already satisfied: argon2-cffi-bindings in /usr/local/lib/python3.11/site-packages (from argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (21.2.0)\n",
|
| 475 |
+
"Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.16.0)\n",
|
| 476 |
+
"Requirement already satisfied: webencodings in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.5.1)\n",
|
| 477 |
+
"Collecting smmap<6,>=3.0.1 (from gitdb<5,>=4.0.1->gitpython!=2.1.4,!=2.1.5,!=2.1.6->nbdime~=4.0.1->jupyterlab-git)\n",
|
| 478 |
+
" Downloading smmap-5.0.1-py3-none-any.whl.metadata (4.3 kB)\n",
|
| 479 |
+
"Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jedi>=0.16->ipython>=7.23.1->ipykernel->jupyterlab) (0.8.3)\n",
|
| 480 |
+
"Requirement already satisfied: fqdn in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.5.1)\n",
|
| 481 |
+
"Requirement already satisfied: isoduration in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (20.11.0)\n",
|
| 482 |
+
"Requirement already satisfied: jsonpointer>1.13 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.4)\n",
|
| 483 |
+
"Requirement already satisfied: uri-template in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.3.0)\n",
|
| 484 |
+
"Requirement already satisfied: webcolors>=1.11 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.13)\n",
|
| 485 |
+
"Requirement already satisfied: wcwidth in /usr/local/lib/python3.11/site-packages (from prompt-toolkit<3.1.0,>=3.0.41->ipython>=7.23.1->ipykernel->jupyterlab) (0.2.13)\n",
|
| 486 |
+
"Requirement already satisfied: cffi>=1.0.1 in /usr/local/lib/python3.11/site-packages (from argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (1.16.0)\n",
|
| 487 |
+
"Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.11/site-packages (from beautifulsoup4->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (2.5)\n",
|
| 488 |
+
"Requirement already satisfied: executing>=1.2.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (2.0.1)\n",
|
| 489 |
+
"Requirement already satisfied: asttokens>=2.1.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (2.4.1)\n",
|
| 490 |
+
"Requirement already satisfied: pure-eval in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (0.2.2)\n",
|
| 491 |
+
"Requirement already satisfied: pycparser in /usr/local/lib/python3.11/site-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (2.21)\n",
|
| 492 |
+
"Requirement already satisfied: arrow>=0.15.0 in /usr/local/lib/python3.11/site-packages (from isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.3.0)\n",
|
| 493 |
+
"Requirement already satisfied: types-python-dateutil>=2.8.10 in /usr/local/lib/python3.11/site-packages (from arrow>=0.15.0->isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.8.19.20240106)\n",
|
| 494 |
+
"Downloading jupyterlab-4.1.1-py3-none-any.whl (11.4 MB)\n",
|
| 495 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11.4/11.4 MB\u001b[0m \u001b[31m150.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 496 |
+
"\u001b[?25hDownloading jupyterlab_git-0.50.0-py3-none-any.whl (1.2 MB)\n",
|
| 497 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.2/1.2 MB\u001b[0m \u001b[31m165.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 498 |
+
"\u001b[?25hDownloading nbdime-4.0.1-py3-none-any.whl (5.9 MB)\n",
|
| 499 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━��━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.9/5.9 MB\u001b[0m \u001b[31m140.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 500 |
+
"\u001b[?25hDownloading GitPython-3.1.41-py3-none-any.whl (196 kB)\n",
|
| 501 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m196.4/196.4 kB\u001b[0m \u001b[31m94.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 502 |
+
"\u001b[?25hDownloading gitdb-4.0.11-py3-none-any.whl (62 kB)\n",
|
| 503 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m62.7/62.7 kB\u001b[0m \u001b[31m156.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 504 |
+
"\u001b[?25hDownloading smmap-5.0.1-py3-none-any.whl (24 kB)\n",
|
| 505 |
+
"Installing collected packages: smmap, colorama, gitdb, gitpython, jupyter-server-mathjax, nbdime, jupyterlab, jupyterlab-git\n",
|
| 506 |
+
" Attempting uninstall: jupyterlab\n",
|
| 507 |
+
" Found existing installation: jupyterlab 4.1.0\n",
|
| 508 |
+
" Uninstalling jupyterlab-4.1.0:\n",
|
| 509 |
+
" Successfully uninstalled jupyterlab-4.1.0\n",
|
| 510 |
+
"Successfully installed colorama-0.4.6 gitdb-4.0.11 gitpython-3.1.41 jupyter-server-mathjax-0.2.6 jupyterlab-4.1.1 jupyterlab-git-0.50.0 nbdime-4.0.1 smmap-5.0.1\n",
|
| 511 |
+
"Note: you may need to restart the kernel to use updated packages.\n"
|
| 512 |
+
]
|
| 513 |
+
}
|
| 514 |
+
],
|
| 515 |
+
"source": [
|
| 516 |
+
"pip install --upgrade jupyterlab jupyterlab-git"
|
| 517 |
+
]
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"cell_type": "code",
|
| 521 |
+
"execution_count": 14,
|
| 522 |
+
"id": "05d58512-a9e2-4319-88bf-9331c6a0584c",
|
| 523 |
+
"metadata": {},
|
| 524 |
+
"outputs": [
|
| 525 |
+
{
|
| 526 |
+
"name": "stdout",
|
| 527 |
+
"output_type": "stream",
|
| 528 |
+
"text": [
|
| 529 |
+
"\n",
|
| 530 |
+
" _| _| _| _| _|_|_| _|_|_| _|_|_| _| _| _|_|_| _|_|_|_| _|_| _|_|_| _|_|_|_|\n",
|
| 531 |
+
" _| _| _| _| _| _| _| _|_| _| _| _| _| _| _| _|\n",
|
| 532 |
+
" _|_|_|_| _| _| _| _|_| _| _|_| _| _| _| _| _| _|_| _|_|_| _|_|_|_| _| _|_|_|\n",
|
| 533 |
+
" _| _| _| _| _| _| _| _| _| _| _|_| _| _| _| _| _| _| _|\n",
|
| 534 |
+
" _| _| _|_| _|_|_| _|_|_| _|_|_| _| _| _|_|_| _| _| _| _|_|_| _|_|_|_|\n",
|
| 535 |
+
"\n",
|
| 536 |
+
" To login, `huggingface_hub` requires a token generated from https://huggingface.co/settings/tokens .\n"
|
| 537 |
+
]
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"name": "stdin",
|
| 541 |
+
"output_type": "stream",
|
| 542 |
+
"text": [
|
| 543 |
+
"Token: ········\n",
|
| 544 |
+
"Add token as git credential? (Y/n) n\n"
|
| 545 |
+
]
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"name": "stdout",
|
| 549 |
+
"output_type": "stream",
|
| 550 |
+
"text": [
|
| 551 |
+
"Token is valid (permission: write).\n",
|
| 552 |
+
"Your token has been saved to /root/.cache/huggingface/token\n",
|
| 553 |
+
"Login successful\n"
|
| 554 |
+
]
|
| 555 |
+
}
|
| 556 |
+
],
|
| 557 |
+
"source": [
|
| 558 |
+
"from huggingface_hub import interpreter_login\n",
|
| 559 |
+
"interpreter_login()"
|
| 560 |
+
]
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"cell_type": "code",
|
| 564 |
+
"execution_count": 15,
|
| 565 |
+
"id": "3bf553b6-b26c-49c3-9407-74c8d53a395e",
|
| 566 |
+
"metadata": {},
|
| 567 |
+
"outputs": [
|
| 568 |
+
{
|
| 569 |
+
"name": "stderr",
|
| 570 |
+
"output_type": "stream",
|
| 571 |
+
"text": [
|
| 572 |
+
"Detected kernel version 4.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.\n"
|
| 573 |
+
]
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"data": {
|
| 577 |
+
"text/html": [
|
| 578 |
+
"\n",
|
| 579 |
+
" <div>\n",
|
| 580 |
+
" \n",
|
| 581 |
+
" <progress value='1100' max='1100' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
|
| 582 |
+
" [1100/1100 07:32, Epoch 0/1]\n",
|
| 583 |
+
" </div>\n",
|
| 584 |
+
" <table border=\"1\" class=\"dataframe\">\n",
|
| 585 |
+
" <thead>\n",
|
| 586 |
+
" <tr style=\"text-align: left;\">\n",
|
| 587 |
+
" <th>Step</th>\n",
|
| 588 |
+
" <th>Training Loss</th>\n",
|
| 589 |
+
" </tr>\n",
|
| 590 |
+
" </thead>\n",
|
| 591 |
+
" <tbody>\n",
|
| 592 |
+
" <tr>\n",
|
| 593 |
+
" <td>100</td>\n",
|
| 594 |
+
" <td>1.414100</td>\n",
|
| 595 |
+
" </tr>\n",
|
| 596 |
+
" <tr>\n",
|
| 597 |
+
" <td>200</td>\n",
|
| 598 |
+
" <td>1.213500</td>\n",
|
| 599 |
+
" </tr>\n",
|
| 600 |
+
" <tr>\n",
|
| 601 |
+
" <td>300</td>\n",
|
| 602 |
+
" <td>1.205500</td>\n",
|
| 603 |
+
" </tr>\n",
|
| 604 |
+
" <tr>\n",
|
| 605 |
+
" <td>400</td>\n",
|
| 606 |
+
" <td>1.142100</td>\n",
|
| 607 |
+
" </tr>\n",
|
| 608 |
+
" <tr>\n",
|
| 609 |
+
" <td>500</td>\n",
|
| 610 |
+
" <td>1.158900</td>\n",
|
| 611 |
+
" </tr>\n",
|
| 612 |
+
" <tr>\n",
|
| 613 |
+
" <td>600</td>\n",
|
| 614 |
+
" <td>1.131100</td>\n",
|
| 615 |
+
" </tr>\n",
|
| 616 |
+
" <tr>\n",
|
| 617 |
+
" <td>700</td>\n",
|
| 618 |
+
" <td>1.160300</td>\n",
|
| 619 |
+
" </tr>\n",
|
| 620 |
+
" <tr>\n",
|
| 621 |
+
" <td>800</td>\n",
|
| 622 |
+
" <td>1.148900</td>\n",
|
| 623 |
+
" </tr>\n",
|
| 624 |
+
" <tr>\n",
|
| 625 |
+
" <td>900</td>\n",
|
| 626 |
+
" <td>1.115100</td>\n",
|
| 627 |
+
" </tr>\n",
|
| 628 |
+
" <tr>\n",
|
| 629 |
+
" <td>1000</td>\n",
|
| 630 |
+
" <td>1.123700</td>\n",
|
| 631 |
+
" </tr>\n",
|
| 632 |
+
" <tr>\n",
|
| 633 |
+
" <td>1100</td>\n",
|
| 634 |
+
" <td>1.143100</td>\n",
|
| 635 |
+
" </tr>\n",
|
| 636 |
+
" </tbody>\n",
|
| 637 |
+
"</table><p>"
|
| 638 |
+
],
|
| 639 |
+
"text/plain": [
|
| 640 |
+
"<IPython.core.display.HTML object>"
|
| 641 |
+
]
|
| 642 |
+
},
|
| 643 |
+
"metadata": {},
|
| 644 |
+
"output_type": "display_data"
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"data": {
|
| 648 |
+
"text/plain": [
|
| 649 |
+
"TrainOutput(global_step=1100, training_loss=1.177838169444691, metrics={'train_runtime': 454.4095, 'train_samples_per_second': 9.683, 'train_steps_per_second': 2.421, 'total_flos': 3.603107414016e+16, 'train_loss': 1.177838169444691, 'epoch': 0.27})"
|
| 650 |
+
]
|
| 651 |
+
},
|
| 652 |
+
"execution_count": 15,
|
| 653 |
+
"metadata": {},
|
| 654 |
+
"output_type": "execute_result"
|
| 655 |
+
}
|
| 656 |
+
],
|
| 657 |
+
"source": [
|
| 658 |
+
"trainer = Trainer(\n",
|
| 659 |
+
" model=model,\n",
|
| 660 |
+
" train_dataset=tokenized_data,\n",
|
| 661 |
+
" args=training_arguments,\n",
|
| 662 |
+
" data_collator=DataCollatorForLanguageModeling(tokenizer, mlm=False)\n",
|
| 663 |
+
")\n",
|
| 664 |
+
"trainer.train()"
|
| 665 |
+
]
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"cell_type": "code",
|
| 669 |
+
"execution_count": null,
|
| 670 |
+
"id": "263cc15e-8e9d-4bd8-9708-ec1638bc1165",
|
| 671 |
+
"metadata": {},
|
| 672 |
+
"outputs": [],
|
| 673 |
+
"source": []
|
| 674 |
+
}
|
| 675 |
+
],
|
| 676 |
+
"metadata": {
|
| 677 |
+
"kernelspec": {
|
| 678 |
+
"display_name": "Python 3 (ipykernel)",
|
| 679 |
+
"language": "python",
|
| 680 |
+
"name": "python3"
|
| 681 |
+
},
|
| 682 |
+
"language_info": {
|
| 683 |
+
"codemirror_mode": {
|
| 684 |
+
"name": "ipython",
|
| 685 |
+
"version": 3
|
| 686 |
+
},
|
| 687 |
+
"file_extension": ".py",
|
| 688 |
+
"mimetype": "text/x-python",
|
| 689 |
+
"name": "python",
|
| 690 |
+
"nbconvert_exporter": "python",
|
| 691 |
+
"pygments_lexer": "ipython3",
|
| 692 |
+
"version": "3.11.5"
|
| 693 |
+
}
|
| 694 |
+
},
|
| 695 |
+
"nbformat": 4,
|
| 696 |
+
"nbformat_minor": 5
|
| 697 |
+
}
|
adapter_config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": null,
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 16,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 16,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"k_proj",
|
| 24 |
+
"q_proj"
|
| 25 |
+
],
|
| 26 |
+
"task_type": "CAUSAL_LM",
|
| 27 |
+
"use_rslora": false
|
| 28 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9791595178e2c022ea24fe5360aa7af0060a60c77d5b2f35583ba9045871ca6
|
| 3 |
+
size 31486304
|
fine-tuned-phi2.ipynb
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [
|
| 3 |
+
{
|
| 4 |
+
"cell_type": "code",
|
| 5 |
+
"execution_count": null,
|
| 6 |
+
"id": "6cb144df-3103-4fbe-96cf-5bc7e7e119c6",
|
| 7 |
+
"metadata": {},
|
| 8 |
+
"outputs": [],
|
| 9 |
+
"source": []
|
| 10 |
+
}
|
| 11 |
+
],
|
| 12 |
+
"metadata": {
|
| 13 |
+
"kernelspec": {
|
| 14 |
+
"display_name": "Python 3 (ipykernel)",
|
| 15 |
+
"language": "python",
|
| 16 |
+
"name": "python3"
|
| 17 |
+
},
|
| 18 |
+
"language_info": {
|
| 19 |
+
"codemirror_mode": {
|
| 20 |
+
"name": "ipython",
|
| 21 |
+
"version": 3
|
| 22 |
+
},
|
| 23 |
+
"file_extension": ".py",
|
| 24 |
+
"mimetype": "text/x-python",
|
| 25 |
+
"name": "python",
|
| 26 |
+
"nbconvert_exporter": "python",
|
| 27 |
+
"pygments_lexer": "ipython3",
|
| 28 |
+
"version": "3.11.5"
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"nbformat": 4,
|
| 32 |
+
"nbformat_minor": 5
|
| 33 |
+
}
|
phi-2-custom.ipynb
ADDED
|
@@ -0,0 +1,1045 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [
|
| 3 |
+
{
|
| 4 |
+
"cell_type": "code",
|
| 5 |
+
"execution_count": 1,
|
| 6 |
+
"id": "b3b92bc7-d105-405f-970d-804d298b9976",
|
| 7 |
+
"metadata": {
|
| 8 |
+
"scrolled": true
|
| 9 |
+
},
|
| 10 |
+
"outputs": [
|
| 11 |
+
{
|
| 12 |
+
"name": "stdout",
|
| 13 |
+
"output_type": "stream",
|
| 14 |
+
"text": [
|
| 15 |
+
"Collecting accelerate\n",
|
| 16 |
+
" Downloading accelerate-0.27.2-py3-none-any.whl.metadata (18 kB)\n",
|
| 17 |
+
"Collecting transformers\n",
|
| 18 |
+
" Downloading transformers-4.37.2-py3-none-any.whl.metadata (129 kB)\n",
|
| 19 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m129.4/129.4 kB\u001b[0m \u001b[31m80.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 20 |
+
"\u001b[?25hCollecting einops\n",
|
| 21 |
+
" Downloading einops-0.7.0-py3-none-any.whl.metadata (13 kB)\n",
|
| 22 |
+
"Collecting datasets\n",
|
| 23 |
+
" Downloading datasets-2.17.0-py3-none-any.whl.metadata (20 kB)\n",
|
| 24 |
+
"Collecting peft\n",
|
| 25 |
+
" Downloading peft-0.8.2-py3-none-any.whl.metadata (25 kB)\n",
|
| 26 |
+
"Collecting bitsandbytes\n",
|
| 27 |
+
" Downloading bitsandbytes-0.42.0-py3-none-any.whl.metadata (9.9 kB)\n",
|
| 28 |
+
"Collecting torch\n",
|
| 29 |
+
" Downloading torch-2.2.0-cp311-cp311-manylinux1_x86_64.whl.metadata (25 kB)\n",
|
| 30 |
+
"Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.11/site-packages (from accelerate) (1.26.4)\n",
|
| 31 |
+
"Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.11/site-packages (from accelerate) (23.2)\n",
|
| 32 |
+
"Requirement already satisfied: psutil in /usr/local/lib/python3.11/site-packages (from accelerate) (5.9.8)\n",
|
| 33 |
+
"Requirement already satisfied: pyyaml in /usr/local/lib/python3.11/site-packages (from accelerate) (6.0.1)\n",
|
| 34 |
+
"Collecting huggingface-hub (from accelerate)\n",
|
| 35 |
+
" Downloading huggingface_hub-0.20.3-py3-none-any.whl.metadata (12 kB)\n",
|
| 36 |
+
"Collecting safetensors>=0.3.1 (from accelerate)\n",
|
| 37 |
+
" Downloading safetensors-0.4.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (3.8 kB)\n",
|
| 38 |
+
"Collecting filelock (from transformers)\n",
|
| 39 |
+
" Downloading filelock-3.13.1-py3-none-any.whl.metadata (2.8 kB)\n",
|
| 40 |
+
"Collecting regex!=2019.12.17 (from transformers)\n",
|
| 41 |
+
" Downloading regex-2023.12.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (40 kB)\n",
|
| 42 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m40.9/40.9 kB\u001b[0m \u001b[31m416.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 43 |
+
"\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from transformers) (2.31.0)\n",
|
| 44 |
+
"Collecting tokenizers<0.19,>=0.14 (from transformers)\n",
|
| 45 |
+
" Downloading tokenizers-0.15.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.7 kB)\n",
|
| 46 |
+
"Collecting tqdm>=4.27 (from transformers)\n",
|
| 47 |
+
" Downloading tqdm-4.66.2-py3-none-any.whl.metadata (57 kB)\n",
|
| 48 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m57.6/57.6 kB\u001b[0m \u001b[31m156.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 49 |
+
"\u001b[?25hCollecting pyarrow>=12.0.0 (from datasets)\n",
|
| 50 |
+
" Downloading pyarrow-15.0.0-cp311-cp311-manylinux_2_28_x86_64.whl.metadata (3.0 kB)\n",
|
| 51 |
+
"Collecting pyarrow-hotfix (from datasets)\n",
|
| 52 |
+
" Downloading pyarrow_hotfix-0.6-py3-none-any.whl.metadata (3.6 kB)\n",
|
| 53 |
+
"Collecting dill<0.3.9,>=0.3.0 (from datasets)\n",
|
| 54 |
+
" Downloading dill-0.3.8-py3-none-any.whl.metadata (10 kB)\n",
|
| 55 |
+
"Collecting pandas (from datasets)\n",
|
| 56 |
+
" Downloading pandas-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (19 kB)\n",
|
| 57 |
+
"Collecting xxhash (from datasets)\n",
|
| 58 |
+
" Downloading xxhash-3.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB)\n",
|
| 59 |
+
"Collecting multiprocess (from datasets)\n",
|
| 60 |
+
" Downloading multiprocess-0.70.16-py311-none-any.whl.metadata (7.2 kB)\n",
|
| 61 |
+
"Collecting fsspec<=2023.10.0,>=2023.1.0 (from fsspec[http]<=2023.10.0,>=2023.1.0->datasets)\n",
|
| 62 |
+
" Downloading fsspec-2023.10.0-py3-none-any.whl.metadata (6.8 kB)\n",
|
| 63 |
+
"Requirement already satisfied: aiohttp in /usr/local/lib/python3.11/site-packages (from datasets) (3.8.3)\n",
|
| 64 |
+
"Collecting scipy (from bitsandbytes)\n",
|
| 65 |
+
" Downloading scipy-1.12.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB)\n",
|
| 66 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m60.4/60.4 kB\u001b[0m \u001b[31m432.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 67 |
+
"\u001b[?25hRequirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.11/site-packages (from torch) (4.9.0)\n",
|
| 68 |
+
"Collecting sympy (from torch)\n",
|
| 69 |
+
" Downloading sympy-1.12-py3-none-any.whl.metadata (12 kB)\n",
|
| 70 |
+
"Collecting networkx (from torch)\n",
|
| 71 |
+
" Downloading networkx-3.2.1-py3-none-any.whl.metadata (5.2 kB)\n",
|
| 72 |
+
"Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/site-packages (from torch) (3.1.3)\n",
|
| 73 |
+
"Collecting nvidia-cuda-nvrtc-cu12==12.1.105 (from torch)\n",
|
| 74 |
+
" Downloading nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\n",
|
| 75 |
+
"Collecting nvidia-cuda-runtime-cu12==12.1.105 (from torch)\n",
|
| 76 |
+
" Downloading nvidia_cuda_runtime_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\n",
|
| 77 |
+
"Collecting nvidia-cuda-cupti-cu12==12.1.105 (from torch)\n",
|
| 78 |
+
" Downloading nvidia_cuda_cupti_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\n",
|
| 79 |
+
"Collecting nvidia-cudnn-cu12==8.9.2.26 (from torch)\n",
|
| 80 |
+
" Downloading nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\n",
|
| 81 |
+
"Collecting nvidia-cublas-cu12==12.1.3.1 (from torch)\n",
|
| 82 |
+
" Downloading nvidia_cublas_cu12-12.1.3.1-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\n",
|
| 83 |
+
"Collecting nvidia-cufft-cu12==11.0.2.54 (from torch)\n",
|
| 84 |
+
" Downloading nvidia_cufft_cu12-11.0.2.54-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\n",
|
| 85 |
+
"Collecting nvidia-curand-cu12==10.3.2.106 (from torch)\n",
|
| 86 |
+
" Downloading nvidia_curand_cu12-10.3.2.106-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\n",
|
| 87 |
+
"Collecting nvidia-cusolver-cu12==11.4.5.107 (from torch)\n",
|
| 88 |
+
" Downloading nvidia_cusolver_cu12-11.4.5.107-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\n",
|
| 89 |
+
"Collecting nvidia-cusparse-cu12==12.1.0.106 (from torch)\n",
|
| 90 |
+
" Downloading nvidia_cusparse_cu12-12.1.0.106-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\n",
|
| 91 |
+
"Collecting nvidia-nccl-cu12==2.19.3 (from torch)\n",
|
| 92 |
+
" Downloading nvidia_nccl_cu12-2.19.3-py3-none-manylinux1_x86_64.whl.metadata (1.8 kB)\n",
|
| 93 |
+
"Collecting nvidia-nvtx-cu12==12.1.105 (from torch)\n",
|
| 94 |
+
" Downloading nvidia_nvtx_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.7 kB)\n",
|
| 95 |
+
"Collecting triton==2.2.0 (from torch)\n",
|
| 96 |
+
" Downloading triton-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (1.4 kB)\n",
|
| 97 |
+
"Collecting nvidia-nvjitlink-cu12 (from nvidia-cusolver-cu12==11.4.5.107->torch)\n",
|
| 98 |
+
" Downloading nvidia_nvjitlink_cu12-12.3.101-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\n",
|
| 99 |
+
"Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (23.2.0)\n",
|
| 100 |
+
"Requirement already satisfied: charset-normalizer<3.0,>=2.0 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (2.1.1)\n",
|
| 101 |
+
"Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (6.0.5)\n",
|
| 102 |
+
"Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (4.0.3)\n",
|
| 103 |
+
"Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (1.9.4)\n",
|
| 104 |
+
"Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (1.4.1)\n",
|
| 105 |
+
"Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.11/site-packages (from aiohttp->datasets) (1.3.1)\n",
|
| 106 |
+
"Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (3.6)\n",
|
| 107 |
+
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (2.2.0)\n",
|
| 108 |
+
"Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->transformers) (2024.2.2)\n",
|
| 109 |
+
"Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2->torch) (2.1.5)\n",
|
| 110 |
+
"Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from pandas->datasets) (2.8.2)\n",
|
| 111 |
+
"Collecting pytz>=2020.1 (from pandas->datasets)\n",
|
| 112 |
+
" Downloading pytz-2024.1-py2.py3-none-any.whl.metadata (22 kB)\n",
|
| 113 |
+
"Collecting tzdata>=2022.7 (from pandas->datasets)\n",
|
| 114 |
+
" Downloading tzdata-2024.1-py2.py3-none-any.whl.metadata (1.4 kB)\n",
|
| 115 |
+
"Collecting mpmath>=0.19 (from sympy->torch)\n",
|
| 116 |
+
" Downloading mpmath-1.3.0-py3-none-any.whl (536 kB)\n",
|
| 117 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m536.2/536.2 kB\u001b[0m \u001b[31m344.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 118 |
+
"\u001b[?25hRequirement already satisfied: six>=1.5 in /usr/local/lib/python3.11/site-packages (from python-dateutil>=2.8.2->pandas->datasets) (1.16.0)\n",
|
| 119 |
+
"Downloading accelerate-0.27.2-py3-none-any.whl (279 kB)\n",
|
| 120 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m280.0/280.0 kB\u001b[0m \u001b[31m352.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 121 |
+
"\u001b[?25hDownloading transformers-4.37.2-py3-none-any.whl (8.4 MB)\n",
|
| 122 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m8.4/8.4 MB\u001b[0m \u001b[31m235.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 123 |
+
"\u001b[?25hDownloading einops-0.7.0-py3-none-any.whl (44 kB)\n",
|
| 124 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m44.6/44.6 kB\u001b[0m \u001b[31m358.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 125 |
+
"\u001b[?25hDownloading datasets-2.17.0-py3-none-any.whl (536 kB)\n",
|
| 126 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m536.6/536.6 kB\u001b[0m \u001b[31m461.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 127 |
+
"\u001b[?25hDownloading peft-0.8.2-py3-none-any.whl (183 kB)\n",
|
| 128 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m183.4/183.4 kB\u001b[0m \u001b[31m386.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 129 |
+
"\u001b[?25hDownloading bitsandbytes-0.42.0-py3-none-any.whl (105.0 MB)\n",
|
| 130 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m105.0/105.0 MB\u001b[0m \u001b[31m164.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 131 |
+
"\u001b[?25hDownloading torch-2.2.0-cp311-cp311-manylinux1_x86_64.whl (755.5 MB)\n",
|
| 132 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m755.5/755.5 MB\u001b[0m \u001b[31m257.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
|
| 133 |
+
"\u001b[?25hDownloading nvidia_cublas_cu12-12.1.3.1-py3-none-manylinux1_x86_64.whl (410.6 MB)\n",
|
| 134 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m410.6/410.6 MB\u001b[0m \u001b[31m272.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
|
| 135 |
+
"\u001b[?25hDownloading nvidia_cuda_cupti_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (14.1 MB)\n",
|
| 136 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m14.1/14.1 MB\u001b[0m \u001b[31m191.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m \u001b[36m0:00:01\u001b[0m\n",
|
| 137 |
+
"\u001b[?25hDownloading nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (23.7 MB)\n",
|
| 138 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m23.7/23.7 MB\u001b[0m \u001b[31m262.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 139 |
+
"\u001b[?25hDownloading nvidia_cuda_runtime_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (823 kB)\n",
|
| 140 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m823.6/823.6 kB\u001b[0m \u001b[31m553.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 141 |
+
"\u001b[?25hDownloading nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl (731.7 MB)\n",
|
| 142 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m731.7/731.7 MB\u001b[0m \u001b[31m319.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
|
| 143 |
+
"\u001b[?25hDownloading nvidia_cufft_cu12-11.0.2.54-py3-none-manylinux1_x86_64.whl (121.6 MB)\n",
|
| 144 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m121.6/121.6 MB\u001b[0m \u001b[31m244.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 145 |
+
"\u001b[?25hDownloading nvidia_curand_cu12-10.3.2.106-py3-none-manylinux1_x86_64.whl (56.5 MB)\n",
|
| 146 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m56.5/56.5 MB\u001b[0m \u001b[31m294.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 147 |
+
"\u001b[?25hDownloading nvidia_cusolver_cu12-11.4.5.107-py3-none-manylinux1_x86_64.whl (124.2 MB)\n",
|
| 148 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m124.2/124.2 MB\u001b[0m \u001b[31m276.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 149 |
+
"\u001b[?25hDownloading nvidia_cusparse_cu12-12.1.0.106-py3-none-manylinux1_x86_64.whl (196.0 MB)\n",
|
| 150 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m196.0/196.0 MB\u001b[0m \u001b[31m163.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
|
| 151 |
+
"\u001b[?25hDownloading nvidia_nccl_cu12-2.19.3-py3-none-manylinux1_x86_64.whl (166.0 MB)\n",
|
| 152 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m166.0/166.0 MB\u001b[0m \u001b[31m207.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
|
| 153 |
+
"\u001b[?25hDownloading nvidia_nvtx_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (99 kB)\n",
|
| 154 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m99.1/99.1 kB\u001b[0m \u001b[31m338.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 155 |
+
"\u001b[?25hDownloading triton-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (167.9 MB)\n",
|
| 156 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m167.9/167.9 MB\u001b[0m \u001b[31m217.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 157 |
+
"\u001b[?25hDownloading dill-0.3.8-py3-none-any.whl (116 kB)\n",
|
| 158 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m116.3/116.3 kB\u001b[0m \u001b[31m340.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 159 |
+
"\u001b[?25hDownloading fsspec-2023.10.0-py3-none-any.whl (166 kB)\n",
|
| 160 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m166.4/166.4 kB\u001b[0m \u001b[31m579.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 161 |
+
"\u001b[?25hDownloading huggingface_hub-0.20.3-py3-none-any.whl (330 kB)\n",
|
| 162 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m330.1/330.1 kB\u001b[0m \u001b[31m454.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 163 |
+
"\u001b[?25hDownloading pyarrow-15.0.0-cp311-cp311-manylinux_2_28_x86_64.whl (38.3 MB)\n",
|
| 164 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m38.3/38.3 MB\u001b[0m \u001b[31m283.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 165 |
+
"\u001b[?25hDownloading regex-2023.12.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (785 kB)\n",
|
| 166 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m785.1/785.1 kB\u001b[0m \u001b[31m547.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 167 |
+
"\u001b[?25hDownloading safetensors-0.4.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n",
|
| 168 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m538.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 169 |
+
"\u001b[?25hDownloading tokenizers-0.15.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.6 MB)\n",
|
| 170 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.6/3.6 MB\u001b[0m \u001b[31m80.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0mta \u001b[36m0:00:01\u001b[0m\n",
|
| 171 |
+
"\u001b[?25hDownloading tqdm-4.66.2-py3-none-any.whl (78 kB)\n",
|
| 172 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m78.3/78.3 kB\u001b[0m \u001b[31m97.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 173 |
+
"\u001b[?25hDownloading filelock-3.13.1-py3-none-any.whl (11 kB)\n",
|
| 174 |
+
"Downloading multiprocess-0.70.16-py311-none-any.whl (143 kB)\n",
|
| 175 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m143.5/143.5 kB\u001b[0m \u001b[31m396.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 176 |
+
"\u001b[?25hDownloading networkx-3.2.1-py3-none-any.whl (1.6 MB)\n",
|
| 177 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.6/1.6 MB\u001b[0m \u001b[31m464.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 178 |
+
"\u001b[?25hDownloading pandas-2.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.0 MB)\n",
|
| 179 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m13.0/13.0 MB\u001b[0m \u001b[31m256.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m \u001b[36m0:00:01\u001b[0m\n",
|
| 180 |
+
"\u001b[?25hDownloading pyarrow_hotfix-0.6-py3-none-any.whl (7.9 kB)\n",
|
| 181 |
+
"Downloading scipy-1.12.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (38.4 MB)\n",
|
| 182 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m38.4/38.4 MB\u001b[0m \u001b[31m282.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 183 |
+
"\u001b[?25hDownloading sympy-1.12-py3-none-any.whl (5.7 MB)\n",
|
| 184 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.7/5.7 MB\u001b[0m \u001b[31m180.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 185 |
+
"\u001b[?25hDownloading xxhash-3.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (194 kB)\n",
|
| 186 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m194.8/194.8 kB\u001b[0m \u001b[31m540.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 187 |
+
"\u001b[?25hDownloading pytz-2024.1-py2.py3-none-any.whl (505 kB)\n",
|
| 188 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m505.5/505.5 kB\u001b[0m \u001b[31m586.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 189 |
+
"\u001b[?25hDownloading tzdata-2024.1-py2.py3-none-any.whl (345 kB)\n",
|
| 190 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m345.4/345.4 kB\u001b[0m \u001b[31m585.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 191 |
+
"\u001b[?25hDownloading nvidia_nvjitlink_cu12-12.3.101-py3-none-manylinux1_x86_64.whl (20.5 MB)\n",
|
| 192 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m20.5/20.5 MB\u001b[0m \u001b[31m199.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
|
| 193 |
+
"\u001b[?25hInstalling collected packages: pytz, mpmath, xxhash, tzdata, tqdm, sympy, scipy, safetensors, regex, pyarrow-hotfix, pyarrow, nvidia-nvtx-cu12, nvidia-nvjitlink-cu12, nvidia-nccl-cu12, nvidia-curand-cu12, nvidia-cufft-cu12, nvidia-cuda-runtime-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-cupti-cu12, nvidia-cublas-cu12, networkx, fsspec, filelock, einops, dill, triton, pandas, nvidia-cusparse-cu12, nvidia-cudnn-cu12, multiprocess, huggingface-hub, bitsandbytes, tokenizers, nvidia-cusolver-cu12, transformers, torch, datasets, accelerate, peft\n",
|
| 194 |
+
"Successfully installed accelerate-0.27.2 bitsandbytes-0.42.0 datasets-2.17.0 dill-0.3.8 einops-0.7.0 filelock-3.13.1 fsspec-2023.10.0 huggingface-hub-0.20.3 mpmath-1.3.0 multiprocess-0.70.16 networkx-3.2.1 nvidia-cublas-cu12-12.1.3.1 nvidia-cuda-cupti-cu12-12.1.105 nvidia-cuda-nvrtc-cu12-12.1.105 nvidia-cuda-runtime-cu12-12.1.105 nvidia-cudnn-cu12-8.9.2.26 nvidia-cufft-cu12-11.0.2.54 nvidia-curand-cu12-10.3.2.106 nvidia-cusolver-cu12-11.4.5.107 nvidia-cusparse-cu12-12.1.0.106 nvidia-nccl-cu12-2.19.3 nvidia-nvjitlink-cu12-12.3.101 nvidia-nvtx-cu12-12.1.105 pandas-2.2.0 peft-0.8.2 pyarrow-15.0.0 pyarrow-hotfix-0.6 pytz-2024.1 regex-2023.12.25 safetensors-0.4.2 scipy-1.12.0 sympy-1.12 tokenizers-0.15.2 torch-2.2.0 tqdm-4.66.2 transformers-4.37.2 triton-2.2.0 tzdata-2024.1 xxhash-3.4.1\n"
|
| 195 |
+
]
|
| 196 |
+
}
|
| 197 |
+
],
|
| 198 |
+
"source": [
|
| 199 |
+
"!pip install accelerate transformers einops datasets peft bitsandbytes torch"
|
| 200 |
+
]
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"cell_type": "code",
|
| 204 |
+
"execution_count": 2,
|
| 205 |
+
"id": "46a53303-b585-4b02-956f-4af173410e25",
|
| 206 |
+
"metadata": {},
|
| 207 |
+
"outputs": [
|
| 208 |
+
{
|
| 209 |
+
"name": "stderr",
|
| 210 |
+
"output_type": "stream",
|
| 211 |
+
"text": [
|
| 212 |
+
"/usr/local/lib/python3.11/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",
|
| 213 |
+
" from .autonotebook import tqdm as notebook_tqdm\n"
|
| 214 |
+
]
|
| 215 |
+
}
|
| 216 |
+
],
|
| 217 |
+
"source": [
|
| 218 |
+
"import torch\n",
|
| 219 |
+
"from datasets import load_dataset, Dataset\n",
|
| 220 |
+
"from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments, Trainer, DataCollatorForLanguageModeling, BitsAndBytesConfig\n",
|
| 221 |
+
"from peft import LoraConfig, get_peft_model\n",
|
| 222 |
+
"import os"
|
| 223 |
+
]
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"cell_type": "code",
|
| 227 |
+
"execution_count": 3,
|
| 228 |
+
"id": "ffa3ee11-3929-4032-8758-322cda3a912e",
|
| 229 |
+
"metadata": {},
|
| 230 |
+
"outputs": [
|
| 231 |
+
{
|
| 232 |
+
"name": "stderr",
|
| 233 |
+
"output_type": "stream",
|
| 234 |
+
"text": [
|
| 235 |
+
"tokenizer_config.json: 100%|██████████| 7.34k/7.34k [00:00<00:00, 22.1MB/s]\n",
|
| 236 |
+
"vocab.json: 100%|██████████| 798k/798k [00:00<00:00, 3.38MB/s]\n",
|
| 237 |
+
"merges.txt: 100%|██████████| 456k/456k [00:00<00:00, 50.2MB/s]\n",
|
| 238 |
+
"tokenizer.json: 100%|██████████| 2.11M/2.11M [00:00<00:00, 45.9MB/s]\n",
|
| 239 |
+
"added_tokens.json: 100%|██████████| 1.08k/1.08k [00:00<00:00, 8.97MB/s]\n",
|
| 240 |
+
"special_tokens_map.json: 100%|██████████| 99.0/99.0 [00:00<00:00, 779kB/s]\n",
|
| 241 |
+
"Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
|
| 242 |
+
]
|
| 243 |
+
}
|
| 244 |
+
],
|
| 245 |
+
"source": [
|
| 246 |
+
"tokenizer = AutoTokenizer.from_pretrained(\"microsoft/phi-2\", trust_remote_code=True)\n",
|
| 247 |
+
"tokenizer.pad_token = tokenizer.eos_token"
|
| 248 |
+
]
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"cell_type": "code",
|
| 252 |
+
"execution_count": 4,
|
| 253 |
+
"id": "b17096e6-c633-4d88-8cfc-fb738c1e4ca0",
|
| 254 |
+
"metadata": {
|
| 255 |
+
"scrolled": true
|
| 256 |
+
},
|
| 257 |
+
"outputs": [
|
| 258 |
+
{
|
| 259 |
+
"name": "stderr",
|
| 260 |
+
"output_type": "stream",
|
| 261 |
+
"text": [
|
| 262 |
+
"config.json: 100%|██████████| 863/863 [00:00<00:00, 6.70MB/s]\n",
|
| 263 |
+
"configuration_phi.py: 100%|██████████| 9.26k/9.26k [00:00<00:00, 46.0MB/s]\n",
|
| 264 |
+
"A new version of the following files was downloaded from https://huggingface.co/microsoft/phi-2:\n",
|
| 265 |
+
"- configuration_phi.py\n",
|
| 266 |
+
". Make sure to double-check they do not contain any added malicious code. To avoid downloading new versions of the code file, you can pin a revision.\n",
|
| 267 |
+
"modeling_phi.py: 100%|██████████| 62.7k/62.7k [00:00<00:00, 18.8MB/s]\n",
|
| 268 |
+
"A new version of the following files was downloaded from https://huggingface.co/microsoft/phi-2:\n",
|
| 269 |
+
"- modeling_phi.py\n",
|
| 270 |
+
". Make sure to double-check they do not contain any added malicious code. To avoid downloading new versions of the code file, you can pin a revision.\n",
|
| 271 |
+
"model.safetensors.index.json: 100%|██████████| 35.7k/35.7k [00:00<00:00, 126MB/s]\n",
|
| 272 |
+
"Downloading shards: 0%| | 0/2 [00:00<?, ?it/s]\n",
|
| 273 |
+
"model-00001-of-00002.safetensors: 0%| | 0.00/5.00G [00:00<?, ?B/s]\u001b[A\n",
|
| 274 |
+
"model-00001-of-00002.safetensors: 1%| | 31.5M/5.00G [00:00<00:19, 254MB/s]\u001b[A\n",
|
| 275 |
+
"model-00001-of-00002.safetensors: 1%|▏ | 62.9M/5.00G [00:00<00:21, 233MB/s]\u001b[A\n",
|
| 276 |
+
"model-00001-of-00002.safetensors: 2%|▏ | 94.4M/5.00G [00:00<00:21, 227MB/s]\u001b[A\n",
|
| 277 |
+
"model-00001-of-00002.safetensors: 3%|▎ | 126M/5.00G [00:00<00:19, 247MB/s] \u001b[A\n",
|
| 278 |
+
"model-00001-of-00002.safetensors: 3%|▎ | 157M/5.00G [00:00<00:18, 263MB/s]\u001b[A\n",
|
| 279 |
+
"model-00001-of-00002.safetensors: 4%|▍ | 189M/5.00G [00:00<00:18, 262MB/s]\u001b[A\n",
|
| 280 |
+
"model-00001-of-00002.safetensors: 4%|▍ | 220M/5.00G [00:00<00:20, 232MB/s]\u001b[A\n",
|
| 281 |
+
"model-00001-of-00002.safetensors: 5%|▌ | 252M/5.00G [00:01<00:21, 223MB/s]\u001b[A\n",
|
| 282 |
+
"model-00001-of-00002.safetensors: 6%|▌ | 283M/5.00G [00:01<00:22, 207MB/s]\u001b[A\n",
|
| 283 |
+
"model-00001-of-00002.safetensors: 6%|▋ | 315M/5.00G [00:01<00:21, 221MB/s]\u001b[A\n",
|
| 284 |
+
"model-00001-of-00002.safetensors: 7%|▋ | 346M/5.00G [00:01<00:21, 217MB/s]\u001b[A\n",
|
| 285 |
+
"model-00001-of-00002.safetensors: 8%|▊ | 377M/5.00G [00:01<00:22, 205MB/s]\u001b[A\n",
|
| 286 |
+
"model-00001-of-00002.safetensors: 8%|▊ | 409M/5.00G [00:02<00:57, 79.3MB/s]\u001b[A\n",
|
| 287 |
+
"model-00001-of-00002.safetensors: 9%|▊ | 430M/5.00G [00:02<00:49, 91.5MB/s]\u001b[A\n",
|
| 288 |
+
"model-00001-of-00002.safetensors: 9%|▉ | 451M/5.00G [00:02<00:45, 101MB/s] \u001b[A\n",
|
| 289 |
+
"model-00001-of-00002.safetensors: 9%|▉ | 472M/5.00G [00:02<00:39, 115MB/s]\u001b[A\n",
|
| 290 |
+
"model-00001-of-00002.safetensors: 10%|█ | 514M/5.00G [00:03<00:28, 155MB/s]\u001b[A\n",
|
| 291 |
+
"model-00001-of-00002.safetensors: 11%|█ | 545M/5.00G [00:03<00:25, 176MB/s]\u001b[A\n",
|
| 292 |
+
"model-00001-of-00002.safetensors: 12%|█▏ | 577M/5.00G [00:03<00:25, 174MB/s]\u001b[A\n",
|
| 293 |
+
"model-00001-of-00002.safetensors: 12%|█▏ | 598M/5.00G [00:03<00:28, 153MB/s]\u001b[A\n",
|
| 294 |
+
"model-00001-of-00002.safetensors: 13%|█▎ | 629M/5.00G [00:03<00:24, 177MB/s]\u001b[A\n",
|
| 295 |
+
"model-00001-of-00002.safetensors: 13%|█▎ | 661M/5.00G [00:03<00:22, 192MB/s]\u001b[A\n",
|
| 296 |
+
"model-00001-of-00002.safetensors: 14%|█▍ | 692M/5.00G [00:04<00:20, 206MB/s]\u001b[A\n",
|
| 297 |
+
"model-00001-of-00002.safetensors: 14%|█▍ | 724M/5.00G [00:04<00:21, 201MB/s]\u001b[A\n",
|
| 298 |
+
"model-00001-of-00002.safetensors: 15%|█▌ | 755M/5.00G [00:04<00:19, 214MB/s]\u001b[A\n",
|
| 299 |
+
"model-00001-of-00002.safetensors: 16%|█▌ | 786M/5.00G [00:04<00:20, 203MB/s]\u001b[A\n",
|
| 300 |
+
"model-00001-of-00002.safetensors: 16%|█▋ | 818M/5.00G [00:04<00:21, 198MB/s]\u001b[A\n",
|
| 301 |
+
"model-00001-of-00002.safetensors: 17%|█▋ | 839M/5.00G [00:04<00:21, 198MB/s]\u001b[A\n",
|
| 302 |
+
"model-00001-of-00002.safetensors: 17%|█▋ | 870M/5.00G [00:04<00:18, 221MB/s]\u001b[A\n",
|
| 303 |
+
"model-00001-of-00002.safetensors: 18%|█▊ | 902M/5.00G [00:04<00:17, 237MB/s]\u001b[A\n",
|
| 304 |
+
"model-00001-of-00002.safetensors: 19%|█▊ | 933M/5.00G [00:05<00:47, 86.2MB/s]\u001b[A\n",
|
| 305 |
+
"model-00001-of-00002.safetensors: 19%|█▉ | 954M/5.00G [00:05<00:40, 99.0MB/s]\u001b[A\n",
|
| 306 |
+
"model-00001-of-00002.safetensors: 20%|█▉ | 986M/5.00G [00:07<01:16, 52.4MB/s]\u001b[A\n",
|
| 307 |
+
"model-00001-of-00002.safetensors: 20%|██ | 1.01G/5.00G [00:07<01:03, 62.8MB/s]\u001b[A\n",
|
| 308 |
+
"model-00001-of-00002.safetensors: 21%|██ | 1.03G/5.00G [00:07<00:52, 75.3MB/s]\u001b[A\n",
|
| 309 |
+
"model-00001-of-00002.safetensors: 21%|██ | 1.05G/5.00G [00:07<00:44, 89.7MB/s]\u001b[A\n",
|
| 310 |
+
"model-00001-of-00002.safetensors: 22%|██▏ | 1.08G/5.00G [00:07<00:33, 118MB/s] \u001b[A\n",
|
| 311 |
+
"model-00001-of-00002.safetensors: 22%|██▏ | 1.10G/5.00G [00:07<00:29, 130MB/s]\u001b[A\n",
|
| 312 |
+
"model-00001-of-00002.safetensors: 22%|██▏ | 1.12G/5.00G [00:07<00:27, 140MB/s]\u001b[A\n",
|
| 313 |
+
"model-00001-of-00002.safetensors: 23%|██▎ | 1.14G/5.00G [00:08<01:16, 50.1MB/s]\u001b[A\n",
|
| 314 |
+
"model-00001-of-00002.safetensors: 24%|██▎ | 1.17G/5.00G [00:09<00:54, 70.6MB/s]\u001b[A\n",
|
| 315 |
+
"model-00001-of-00002.safetensors: 24%|██▍ | 1.20G/5.00G [00:09<00:44, 84.8MB/s]\u001b[A\n",
|
| 316 |
+
"model-00001-of-00002.safetensors: 24%|██▍ | 1.22G/5.00G [00:09<00:37, 99.8MB/s]\u001b[A\n",
|
| 317 |
+
"model-00001-of-00002.safetensors: 25%|██▍ | 1.24G/5.00G [00:09<00:33, 113MB/s] \u001b[A\n",
|
| 318 |
+
"model-00001-of-00002.safetensors: 25%|██▌ | 1.26G/5.00G [00:09<00:30, 124MB/s]\u001b[A\n",
|
| 319 |
+
"model-00001-of-00002.safetensors: 26%|██▌ | 1.28G/5.00G [00:09<00:29, 126MB/s]\u001b[A\n",
|
| 320 |
+
"model-00001-of-00002.safetensors: 26%|██▌ | 1.31G/5.00G [00:09<00:24, 148MB/s]\u001b[A\n",
|
| 321 |
+
"model-00001-of-00002.safetensors: 27%|██▋ | 1.33G/5.00G [00:10<00:25, 142MB/s]\u001b[A\n",
|
| 322 |
+
"model-00001-of-00002.safetensors: 27%|██▋ | 1.35G/5.00G [00:10<00:26, 136MB/s]\u001b[A\n",
|
| 323 |
+
"model-00001-of-00002.safetensors: 27%|██▋ | 1.37G/5.00G [00:11<01:13, 49.0MB/s]\u001b[A\n",
|
| 324 |
+
"model-00001-of-00002.safetensors: 28%|██▊ | 1.39G/5.00G [00:11<00:59, 60.7MB/s]\u001b[A\n",
|
| 325 |
+
"model-00001-of-00002.safetensors: 28%|██▊ | 1.42G/5.00G [00:11<00:52, 68.8MB/s]\u001b[A\n",
|
| 326 |
+
"model-00001-of-00002.safetensors: 29%|██▉ | 1.45G/5.00G [00:11<00:38, 91.4MB/s]\u001b[A\n",
|
| 327 |
+
"model-00001-of-00002.safetensors: 29%|██▉ | 1.47G/5.00G [00:11<00:34, 103MB/s] \u001b[A\n",
|
| 328 |
+
"model-00001-of-00002.safetensors: 30%|██▉ | 1.49G/5.00G [00:12<00:40, 86.3MB/s]\u001b[A\n",
|
| 329 |
+
"model-00001-of-00002.safetensors: 30%|███ | 1.51G/5.00G [00:12<00:33, 103MB/s] \u001b[A\n",
|
| 330 |
+
"model-00001-of-00002.safetensors: 31%|███ | 1.53G/5.00G [00:13<01:17, 44.5MB/s]\u001b[A\n",
|
| 331 |
+
"model-00001-of-00002.safetensors: 31%|███▏ | 1.56G/5.00G [00:13<00:54, 63.6MB/s]\u001b[A\n",
|
| 332 |
+
"model-00001-of-00002.safetensors: 32%|███▏ | 1.58G/5.00G [00:13<00:47, 71.3MB/s]\u001b[A\n",
|
| 333 |
+
"model-00001-of-00002.safetensors: 32%|███▏ | 1.60G/5.00G [00:14<00:45, 75.2MB/s]\u001b[A\n",
|
| 334 |
+
"model-00001-of-00002.safetensors: 33%|███▎ | 1.63G/5.00G [00:14<00:38, 87.8MB/s]\u001b[A\n",
|
| 335 |
+
"model-00001-of-00002.safetensors: 33%|███▎ | 1.65G/5.00G [00:14<00:32, 102MB/s] \u001b[A\n",
|
| 336 |
+
"model-00001-of-00002.safetensors: 34%|███▎ | 1.68G/5.00G [00:14<00:30, 110MB/s]\u001b[A\n",
|
| 337 |
+
"model-00001-of-00002.safetensors: 34%|███▍ | 1.72G/5.00G [00:14<00:22, 148MB/s]\u001b[A\n",
|
| 338 |
+
"model-00001-of-00002.safetensors: 35%|███▌ | 1.75G/5.00G [00:14<00:19, 171MB/s]\u001b[A\n",
|
| 339 |
+
"model-00001-of-00002.safetensors: 35%|███▌ | 1.77G/5.00G [00:15<00:18, 176MB/s]\u001b[A\n",
|
| 340 |
+
"model-00001-of-00002.safetensors: 36%|███▌ | 1.79G/5.00G [00:15<00:18, 170MB/s]\u001b[A\n",
|
| 341 |
+
"model-00001-of-00002.safetensors: 36%|███▋ | 1.81G/5.00G [00:15<00:19, 165MB/s]\u001b[A\n",
|
| 342 |
+
"model-00001-of-00002.safetensors: 37%|███▋ | 1.85G/5.00G [00:15<00:17, 184MB/s]\u001b[A\n",
|
| 343 |
+
"model-00001-of-00002.safetensors: 38%|███▊ | 1.88G/5.00G [00:15<00:14, 212MB/s]\u001b[A\n",
|
| 344 |
+
"model-00001-of-00002.safetensors: 38%|███▊ | 1.91G/5.00G [00:15<00:14, 216MB/s]\u001b[A\n",
|
| 345 |
+
"model-00001-of-00002.safetensors: 39%|███▉ | 1.94G/5.00G [00:15<00:14, 206MB/s]\u001b[A\n",
|
| 346 |
+
"model-00001-of-00002.safetensors: 39%|███▉ | 1.97G/5.00G [00:16<00:14, 215MB/s]\u001b[A\n",
|
| 347 |
+
"model-00001-of-00002.safetensors: 40%|████ | 2.00G/5.00G [00:16<00:14, 207MB/s]\u001b[A\n",
|
| 348 |
+
"model-00001-of-00002.safetensors: 41%|████ | 2.03G/5.00G [00:16<00:13, 220MB/s]\u001b[A\n",
|
| 349 |
+
"model-00001-of-00002.safetensors: 41%|████▏ | 2.07G/5.00G [00:16<00:13, 222MB/s]\u001b[A\n",
|
| 350 |
+
"model-00001-of-00002.safetensors: 42%|████▏ | 2.10G/5.00G [00:16<00:12, 239MB/s]\u001b[A\n",
|
| 351 |
+
"model-00001-of-00002.safetensors: 43%|████▎ | 2.13G/5.00G [00:16<00:12, 223MB/s]\u001b[A\n",
|
| 352 |
+
"model-00001-of-00002.safetensors: 43%|████▎ | 2.16G/5.00G [00:16<00:13, 207MB/s]\u001b[A\n",
|
| 353 |
+
"model-00001-of-00002.safetensors: 44%|████▍ | 2.19G/5.00G [00:17<00:16, 167MB/s]\u001b[A\n",
|
| 354 |
+
"model-00001-of-00002.safetensors: 44%|████▍ | 2.21G/5.00G [00:17<00:17, 163MB/s]\u001b[A\n",
|
| 355 |
+
"model-00001-of-00002.safetensors: 45%|████▍ | 2.23G/5.00G [00:17<00:16, 165MB/s]\u001b[A\n",
|
| 356 |
+
"model-00001-of-00002.safetensors: 45%|████▌ | 2.25G/5.00G [00:17<00:18, 147MB/s]\u001b[A\n",
|
| 357 |
+
"model-00001-of-00002.safetensors: 46%|████▌ | 2.28G/5.00G [00:17<00:17, 154MB/s]\u001b[A\n",
|
| 358 |
+
"model-00001-of-00002.safetensors: 46%|████▌ | 2.30G/5.00G [00:17<00:16, 159MB/s]\u001b[A\n",
|
| 359 |
+
"model-00001-of-00002.safetensors: 46%|████▋ | 2.32G/5.00G [00:17<00:16, 161MB/s]\u001b[A\n",
|
| 360 |
+
"model-00001-of-00002.safetensors: 47%|████▋ | 2.34G/5.00G [00:18<00:16, 160MB/s]\u001b[A\n",
|
| 361 |
+
"model-00001-of-00002.safetensors: 47%|████▋ | 2.36G/5.00G [00:18<00:16, 163MB/s]\u001b[A\n",
|
| 362 |
+
"model-00001-of-00002.safetensors: 48%|████▊ | 2.38G/5.00G [00:18<00:15, 167MB/s]\u001b[A\n",
|
| 363 |
+
"model-00001-of-00002.safetensors: 48%|████▊ | 2.41G/5.00G [00:18<00:13, 186MB/s]\u001b[A\n",
|
| 364 |
+
"model-00001-of-00002.safetensors: 49%|████▊ | 2.43G/5.00G [00:18<00:14, 182MB/s]\u001b[A\n",
|
| 365 |
+
"model-00001-of-00002.safetensors: 49%|████▉ | 2.45G/5.00G [00:18<00:14, 178MB/s]\u001b[A\n",
|
| 366 |
+
"model-00001-of-00002.safetensors: 50%|████▉ | 2.47G/5.00G [00:18<00:13, 184MB/s]\u001b[A\n",
|
| 367 |
+
"model-00001-of-00002.safetensors: 50%|████▉ | 2.50G/5.00G [00:18<00:15, 166MB/s]\u001b[A\n",
|
| 368 |
+
"model-00001-of-00002.safetensors: 50%|█████ | 2.52G/5.00G [00:19<00:15, 164MB/s]\u001b[A\n",
|
| 369 |
+
"model-00001-of-00002.safetensors: 51%|█████ | 2.54G/5.00G [00:19<00:15, 164MB/s]\u001b[A\n",
|
| 370 |
+
"model-00001-of-00002.safetensors: 51%|█████ | 2.56G/5.00G [00:19<00:14, 166MB/s]\u001b[A\n",
|
| 371 |
+
"model-00001-of-00002.safetensors: 52%|█████▏ | 2.59G/5.00G [00:19<00:12, 192MB/s]\u001b[A\n",
|
| 372 |
+
"model-00001-of-00002.safetensors: 52%|█████▏ | 2.61G/5.00G [00:19<00:13, 180MB/s]\u001b[A\n",
|
| 373 |
+
"model-00001-of-00002.safetensors: 53%|█████▎ | 2.63G/5.00G [00:19<00:14, 167MB/s]\u001b[A\n",
|
| 374 |
+
"model-00001-of-00002.safetensors: 53%|█████▎ | 2.66G/5.00G [00:19<00:12, 194MB/s]\u001b[A\n",
|
| 375 |
+
"model-00001-of-00002.safetensors: 54%|█████▍ | 2.69G/5.00G [00:20<00:10, 212MB/s]\u001b[A\n",
|
| 376 |
+
"model-00001-of-00002.safetensors: 55%|█████▍ | 2.73G/5.00G [00:20<00:11, 200MB/s]\u001b[A\n",
|
| 377 |
+
"model-00001-of-00002.safetensors: 55%|█████▌ | 2.76G/5.00G [00:20<00:10, 211MB/s]\u001b[A\n",
|
| 378 |
+
"model-00001-of-00002.safetensors: 56%|█████▌ | 2.79G/5.00G [00:20<00:10, 205MB/s]\u001b[A\n",
|
| 379 |
+
"model-00001-of-00002.safetensors: 56%|█████▋ | 2.81G/5.00G [00:20<00:11, 192MB/s]\u001b[A\n",
|
| 380 |
+
"model-00001-of-00002.safetensors: 57%|█████▋ | 2.83G/5.00G [00:20<00:12, 178MB/s]\u001b[A\n",
|
| 381 |
+
"model-00001-of-00002.safetensors: 57%|█████▋ | 2.85G/5.00G [00:20<00:11, 180MB/s]\u001b[A\n",
|
| 382 |
+
"model-00001-of-00002.safetensors: 58%|█████▊ | 2.88G/5.00G [00:20<00:10, 201MB/s]\u001b[A\n",
|
| 383 |
+
"model-00001-of-00002.safetensors: 58%|█████▊ | 2.92G/5.00G [00:21<00:09, 214MB/s]\u001b[A\n",
|
| 384 |
+
"model-00001-of-00002.safetensors: 59%|█████▉ | 2.95G/5.00G [00:21<00:11, 174MB/s]\u001b[A\n",
|
| 385 |
+
"model-00001-of-00002.safetensors: 59%|█████▉ | 2.97G/5.00G [00:21<00:13, 154MB/s]\u001b[A\n",
|
| 386 |
+
"model-00001-of-00002.safetensors: 60%|█████▉ | 2.99G/5.00G [00:21<00:13, 154MB/s]\u001b[A\n",
|
| 387 |
+
"model-00001-of-00002.safetensors: 60%|██████ | 3.01G/5.00G [00:21<00:13, 152MB/s]\u001b[A\n",
|
| 388 |
+
"model-00001-of-00002.safetensors: 61%|██████ | 3.03G/5.00G [00:21<00:12, 160MB/s]\u001b[A\n",
|
| 389 |
+
"model-00001-of-00002.safetensors: 61%|██████▏ | 3.06G/5.00G [00:22<00:10, 188MB/s]\u001b[A\n",
|
| 390 |
+
"model-00001-of-00002.safetensors: 62%|██████▏ | 3.08G/5.00G [00:22<00:09, 192MB/s]\u001b[A\n",
|
| 391 |
+
"model-00001-of-00002.safetensors: 62%|██████▏ | 3.10G/5.00G [00:22<00:10, 182MB/s]\u001b[A\n",
|
| 392 |
+
"model-00001-of-00002.safetensors: 63%|██████▎ | 3.14G/5.00G [00:22<00:09, 196MB/s]\u001b[A\n",
|
| 393 |
+
"model-00001-of-00002.safetensors: 63%|██████▎ | 3.17G/5.00G [00:22<00:08, 216MB/s]\u001b[A\n",
|
| 394 |
+
"model-00001-of-00002.safetensors: 64%|██████▍ | 3.21G/5.00G [00:22<00:07, 239MB/s]\u001b[A\n",
|
| 395 |
+
"model-00001-of-00002.safetensors: 65%|██████▍ | 3.24G/5.00G [00:22<00:08, 218MB/s]\u001b[A\n",
|
| 396 |
+
"model-00001-of-00002.safetensors: 65%|██████▌ | 3.27G/5.00G [00:23<00:08, 194MB/s]\u001b[A\n",
|
| 397 |
+
"model-00001-of-00002.safetensors: 66%|██████▌ | 3.30G/5.00G [00:23<00:08, 206MB/s]\u001b[A\n",
|
| 398 |
+
"model-00001-of-00002.safetensors: 67%|██████▋ | 3.33G/5.00G [00:23<00:09, 184MB/s]\u001b[A\n",
|
| 399 |
+
"model-00001-of-00002.safetensors: 67%|██████▋ | 3.36G/5.00G [00:23<00:09, 164MB/s]\u001b[A\n",
|
| 400 |
+
"model-00001-of-00002.safetensors: 68%|██████▊ | 3.38G/5.00G [00:23<00:12, 132MB/s]\u001b[A\n",
|
| 401 |
+
"model-00001-of-00002.safetensors: 68%|██████▊ | 3.40G/5.00G [00:24<00:13, 122MB/s]\u001b[A\n",
|
| 402 |
+
"model-00001-of-00002.safetensors: 68%|██████▊ | 3.42G/5.00G [00:24<00:13, 120MB/s]\u001b[A\n",
|
| 403 |
+
"model-00001-of-00002.safetensors: 69%|██████▉ | 3.44G/5.00G [00:24<00:14, 110MB/s]\u001b[A\n",
|
| 404 |
+
"model-00001-of-00002.safetensors: 69%|██████▉ | 3.46G/5.00G [00:24<00:14, 104MB/s]\u001b[A\n",
|
| 405 |
+
"model-00001-of-00002.safetensors: 70%|██████▉ | 3.48G/5.00G [00:24<00:15, 95.3MB/s]\u001b[A\n",
|
| 406 |
+
"model-00001-of-00002.safetensors: 70%|██████▉ | 3.49G/5.00G [00:25<00:16, 93.6MB/s]\u001b[A\n",
|
| 407 |
+
"model-00001-of-00002.safetensors: 70%|███████ | 3.51G/5.00G [00:25<00:15, 96.6MB/s]\u001b[A\n",
|
| 408 |
+
"model-00001-of-00002.safetensors: 71%|███████ | 3.53G/5.00G [00:25<00:14, 97.9MB/s]\u001b[A\n",
|
| 409 |
+
"model-00001-of-00002.safetensors: 71%|███████ | 3.54G/5.00G [00:25<00:14, 99.1MB/s]\u001b[A\n",
|
| 410 |
+
"model-00001-of-00002.safetensors: 71%|███████ | 3.55G/5.00G [00:25<00:14, 99.1MB/s]\u001b[A\n",
|
| 411 |
+
"model-00001-of-00002.safetensors: 72%|███████▏ | 3.58G/5.00G [00:25<00:11, 122MB/s] \u001b[A\n",
|
| 412 |
+
"model-00001-of-00002.safetensors: 72%|███████▏ | 3.60G/5.00G [00:26<00:11, 119MB/s]\u001b[A\n",
|
| 413 |
+
"model-00001-of-00002.safetensors: 73%|███████▎ | 3.63G/5.00G [00:26<00:08, 154MB/s]\u001b[A\n",
|
| 414 |
+
"model-00001-of-00002.safetensors: 73%|███████▎ | 3.65G/5.00G [00:26<00:10, 128MB/s]\u001b[A\n",
|
| 415 |
+
"model-00001-of-00002.safetensors: 73%|███████▎ | 3.67G/5.00G [00:26<00:11, 115MB/s]\u001b[A\n",
|
| 416 |
+
"model-00001-of-00002.safetensors: 74%|███████▍ | 3.69G/5.00G [00:26<00:12, 101MB/s]\u001b[A\n",
|
| 417 |
+
"model-00001-of-00002.safetensors: 74%|███████▍ | 3.71G/5.00G [00:27<00:11, 110MB/s]\u001b[A\n",
|
| 418 |
+
"model-00001-of-00002.safetensors: 75%|███████▍ | 3.74G/5.00G [00:27<00:09, 127MB/s]\u001b[A\n",
|
| 419 |
+
"model-00001-of-00002.safetensors: 75%|███████▌ | 3.76G/5.00G [00:27<00:09, 128MB/s]\u001b[A\n",
|
| 420 |
+
"model-00001-of-00002.safetensors: 76%|███████▌ | 3.79G/5.00G [00:27<00:08, 142MB/s]\u001b[A\n",
|
| 421 |
+
"model-00001-of-00002.safetensors: 76%|███████▌ | 3.81G/5.00G [00:27<00:10, 112MB/s]\u001b[A\n",
|
| 422 |
+
"model-00001-of-00002.safetensors: 77%|███████▋ | 3.83G/5.00G [00:27<00:11, 105MB/s]\u001b[A\n",
|
| 423 |
+
"model-00001-of-00002.safetensors: 77%|███████▋ | 3.85G/5.00G [00:28<00:11, 98.9MB/s]\u001b[A\n",
|
| 424 |
+
"model-00001-of-00002.safetensors: 77%|███████▋ | 3.87G/5.00G [00:28<00:10, 106MB/s] \u001b[A\n",
|
| 425 |
+
"model-00001-of-00002.safetensors: 78%|███████▊ | 3.89G/5.00G [00:28<00:10, 106MB/s]\u001b[A\n",
|
| 426 |
+
"model-00001-of-00002.safetensors: 78%|███████▊ | 3.91G/5.00G [00:28<00:09, 119MB/s]\u001b[A\n",
|
| 427 |
+
"model-00001-of-00002.safetensors: 79%|███████▊ | 3.93G/5.00G [00:28<00:08, 129MB/s]\u001b[A\n",
|
| 428 |
+
"model-00001-of-00002.safetensors: 79%|███████▉ | 3.96G/5.00G [00:28<00:06, 163MB/s]\u001b[A\n",
|
| 429 |
+
"model-00001-of-00002.safetensors: 80%|███████▉ | 4.00G/5.00G [00:29<00:05, 182MB/s]\u001b[A\n",
|
| 430 |
+
"model-00001-of-00002.safetensors: 80%|████████ | 4.02G/5.00G [00:29<00:05, 183MB/s]\u001b[A\n",
|
| 431 |
+
"model-00001-of-00002.safetensors: 81%|████████ | 4.05G/5.00G [00:29<00:04, 210MB/s]\u001b[A\n",
|
| 432 |
+
"model-00001-of-00002.safetensors: 82%|████████▏ | 4.08G/5.00G [00:29<00:04, 204MB/s]\u001b[A\n",
|
| 433 |
+
"model-00001-of-00002.safetensors: 82%|████████▏ | 4.11G/5.00G [00:29<00:04, 205MB/s]\u001b[A\n",
|
| 434 |
+
"model-00001-of-00002.safetensors: 83%|████████▎ | 4.14G/5.00G [00:29<00:03, 220MB/s]\u001b[A\n",
|
| 435 |
+
"model-00001-of-00002.safetensors: 84%|████████▍ | 4.18G/5.00G [00:29<00:03, 250MB/s]\u001b[A\n",
|
| 436 |
+
"model-00001-of-00002.safetensors: 84%|████████▍ | 4.22G/5.00G [00:30<00:03, 247MB/s]\u001b[A\n",
|
| 437 |
+
"model-00001-of-00002.safetensors: 85%|████████▌ | 4.25G/5.00G [00:30<00:03, 223MB/s]\u001b[A\n",
|
| 438 |
+
"model-00001-of-00002.safetensors: 86%|████████▌ | 4.28G/5.00G [00:30<00:04, 165MB/s]\u001b[A\n",
|
| 439 |
+
"model-00001-of-00002.safetensors: 86%|████████▌ | 4.30G/5.00G [00:30<00:04, 148MB/s]\u001b[A\n",
|
| 440 |
+
"model-00001-of-00002.safetensors: 87%|████████▋ | 4.33G/5.00G [00:30<00:04, 164MB/s]\u001b[A\n",
|
| 441 |
+
"model-00001-of-00002.safetensors: 87%|████████▋ | 4.36G/5.00G [00:31<00:03, 168MB/s]\u001b[A\n",
|
| 442 |
+
"model-00001-of-00002.safetensors: 88%|████████▊ | 4.38G/5.00G [00:31<00:03, 157MB/s]\u001b[A\n",
|
| 443 |
+
"model-00001-of-00002.safetensors: 88%|████████▊ | 4.40G/5.00G [00:31<00:05, 116MB/s]\u001b[A\n",
|
| 444 |
+
"model-00001-of-00002.safetensors: 89%|████████▊ | 4.42G/5.00G [00:31<00:04, 130MB/s]\u001b[A\n",
|
| 445 |
+
"model-00001-of-00002.safetensors: 89%|████████▉ | 4.45G/5.00G [00:31<00:03, 142MB/s]\u001b[A\n",
|
| 446 |
+
"model-00001-of-00002.safetensors: 90%|████████▉ | 4.49G/5.00G [00:31<00:02, 189MB/s]\u001b[A\n",
|
| 447 |
+
"model-00001-of-00002.safetensors: 90%|█████████ | 4.52G/5.00G [00:31<00:02, 205MB/s]\u001b[A\n",
|
| 448 |
+
"model-00001-of-00002.safetensors: 91%|█████████ | 4.55G/5.00G [00:32<00:02, 174MB/s]\u001b[A\n",
|
| 449 |
+
"model-00001-of-00002.safetensors: 92%|█████████▏| 4.57G/5.00G [00:32<00:02, 166MB/s]\u001b[A\n",
|
| 450 |
+
"model-00001-of-00002.safetensors: 92%|█████████▏| 4.60G/5.00G [00:32<00:02, 187MB/s]\u001b[A\n",
|
| 451 |
+
"model-00001-of-00002.safetensors: 93%|█████████▎| 4.63G/5.00G [00:32<00:01, 188MB/s]\u001b[A\n",
|
| 452 |
+
"model-00001-of-00002.safetensors: 93%|█████████▎| 4.67G/5.00G [00:32<00:01, 201MB/s]\u001b[A\n",
|
| 453 |
+
"model-00001-of-00002.safetensors: 94%|█████████▍| 4.70G/5.00G [00:32<00:01, 218MB/s]\u001b[A\n",
|
| 454 |
+
"model-00001-of-00002.safetensors: 95%|█████████▍| 4.73G/5.00G [00:33<00:01, 232MB/s]\u001b[A\n",
|
| 455 |
+
"model-00001-of-00002.safetensors: 95%|█████████▌| 4.76G/5.00G [00:33<00:00, 250MB/s]\u001b[A\n",
|
| 456 |
+
"model-00001-of-00002.safetensors: 96%|█████████▌| 4.79G/5.00G [00:33<00:00, 244MB/s]\u001b[A\n",
|
| 457 |
+
"model-00001-of-00002.safetensors: 97%|█████████▋| 4.82G/5.00G [00:33<00:00, 202MB/s]\u001b[A\n",
|
| 458 |
+
"model-00001-of-00002.safetensors: 97%|█████████▋| 4.85G/5.00G [00:33<00:00, 200MB/s]\u001b[A\n",
|
| 459 |
+
"model-00001-of-00002.safetensors: 98%|█████████▊| 4.89G/5.00G [00:33<00:00, 222MB/s]\u001b[A\n",
|
| 460 |
+
"model-00001-of-00002.safetensors: 98%|█████████▊| 4.92G/5.00G [00:33<00:00, 208MB/s]\u001b[A\n",
|
| 461 |
+
"model-00001-of-00002.safetensors: 99%|█████████▉| 4.95G/5.00G [00:34<00:00, 209MB/s]\u001b[A\n",
|
| 462 |
+
"model-00001-of-00002.safetensors: 100%|██████████| 5.00G/5.00G [00:34<00:00, 146MB/s]\u001b[A\n",
|
| 463 |
+
"Downloading shards: 50%|█████ | 1/2 [00:34<00:34, 34.54s/it]\n",
|
| 464 |
+
"model-00002-of-00002.safetensors: 0%| | 0.00/564M [00:00<?, ?B/s]\u001b[A\n",
|
| 465 |
+
"model-00002-of-00002.safetensors: 6%|▌ | 31.5M/564M [00:00<00:02, 235MB/s]\u001b[A\n",
|
| 466 |
+
"model-00002-of-00002.safetensors: 11%|█ | 62.9M/564M [00:00<00:01, 256MB/s]\u001b[A\n",
|
| 467 |
+
"model-00002-of-00002.safetensors: 17%|█▋ | 94.4M/564M [00:00<00:02, 234MB/s]\u001b[A\n",
|
| 468 |
+
"model-00002-of-00002.safetensors: 24%|██▍ | 136M/564M [00:00<00:01, 273MB/s] \u001b[A\n",
|
| 469 |
+
"model-00002-of-00002.safetensors: 30%|██▉ | 168M/564M [00:00<00:01, 222MB/s]\u001b[A\n",
|
| 470 |
+
"model-00002-of-00002.safetensors: 35%|███▌ | 199M/564M [00:01<00:02, 167MB/s]\u001b[A\n",
|
| 471 |
+
"model-00002-of-00002.safetensors: 39%|███▉ | 220M/564M [00:01<00:01, 175MB/s]\u001b[A\n",
|
| 472 |
+
"model-00002-of-00002.safetensors: 45%|████▍ | 252M/564M [00:01<00:01, 196MB/s]\u001b[A\n",
|
| 473 |
+
"model-00002-of-00002.safetensors: 50%|█████ | 283M/564M [00:01<00:01, 183MB/s]\u001b[A\n",
|
| 474 |
+
"model-00002-of-00002.safetensors: 54%|█████▍ | 304M/564M [00:01<00:01, 187MB/s]\u001b[A\n",
|
| 475 |
+
"model-00002-of-00002.safetensors: 60%|█████▉ | 336M/564M [00:01<00:01, 204MB/s]\u001b[A\n",
|
| 476 |
+
"model-00002-of-00002.safetensors: 67%|██████▋ | 377M/564M [00:01<00:00, 239MB/s]\u001b[A\n",
|
| 477 |
+
"model-00002-of-00002.safetensors: 73%|███████▎ | 409M/564M [00:01<00:00, 245MB/s]\u001b[A\n",
|
| 478 |
+
"model-00002-of-00002.safetensors: 78%|███████▊ | 440M/564M [00:02<00:00, 242MB/s]\u001b[A\n",
|
| 479 |
+
"model-00002-of-00002.safetensors: 84%|████████▎ | 472M/564M [00:02<00:00, 234MB/s]\u001b[A\n",
|
| 480 |
+
"model-00002-of-00002.safetensors: 89%|████████▉ | 503M/564M [00:02<00:00, 192MB/s]\u001b[A\n",
|
| 481 |
+
"model-00002-of-00002.safetensors: 93%|█████████▎| 524M/564M [00:02<00:00, 193MB/s]\u001b[A\n",
|
| 482 |
+
"model-00002-of-00002.safetensors: 100%|██████████| 564M/564M [00:02<00:00, 213MB/s]\u001b[A\n",
|
| 483 |
+
"Downloading shards: 100%|██████████| 2/2 [00:37<00:00, 18.66s/it]\n",
|
| 484 |
+
"Loading checkpoint shards: 100%|██████████| 2/2 [00:02<00:00, 1.40s/it]\n",
|
| 485 |
+
"generation_config.json: 100%|██████████| 124/124 [00:00<00:00, 861kB/s]\n"
|
| 486 |
+
]
|
| 487 |
+
}
|
| 488 |
+
],
|
| 489 |
+
"source": [
|
| 490 |
+
"bnb_config = BitsAndBytesConfig(\n",
|
| 491 |
+
" load_in_4bit=True,\n",
|
| 492 |
+
" bnb_4bit_use_double_quant=True,\n",
|
| 493 |
+
" bnb_4bit_quant_type=\"nf4\",\n",
|
| 494 |
+
" bnb_4bit_compute_dtype=torch.float16\n",
|
| 495 |
+
")\n",
|
| 496 |
+
"\n",
|
| 497 |
+
"model = AutoModelForCausalLM.from_pretrained(\n",
|
| 498 |
+
" \"microsoft/phi-2\",\n",
|
| 499 |
+
" device_map={\"\":0},\n",
|
| 500 |
+
" trust_remote_code=True,\n",
|
| 501 |
+
" quantization_config=bnb_config\n",
|
| 502 |
+
")"
|
| 503 |
+
]
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"cell_type": "code",
|
| 507 |
+
"execution_count": 5,
|
| 508 |
+
"id": "e15aa794-e17c-4b09-a64a-c60377259218",
|
| 509 |
+
"metadata": {},
|
| 510 |
+
"outputs": [
|
| 511 |
+
{
|
| 512 |
+
"name": "stdout",
|
| 513 |
+
"output_type": "stream",
|
| 514 |
+
"text": [
|
| 515 |
+
"PhiForCausalLM(\n",
|
| 516 |
+
" (model): PhiModel(\n",
|
| 517 |
+
" (embed_tokens): Embedding(51200, 2560)\n",
|
| 518 |
+
" (embed_dropout): Dropout(p=0.0, inplace=False)\n",
|
| 519 |
+
" (layers): ModuleList(\n",
|
| 520 |
+
" (0-31): 32 x PhiDecoderLayer(\n",
|
| 521 |
+
" (self_attn): PhiAttention(\n",
|
| 522 |
+
" (q_proj): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 523 |
+
" (k_proj): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 524 |
+
" (v_proj): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 525 |
+
" (dense): Linear4bit(in_features=2560, out_features=2560, bias=True)\n",
|
| 526 |
+
" (rotary_emb): PhiRotaryEmbedding()\n",
|
| 527 |
+
" )\n",
|
| 528 |
+
" (mlp): PhiMLP(\n",
|
| 529 |
+
" (activation_fn): NewGELUActivation()\n",
|
| 530 |
+
" (fc1): Linear4bit(in_features=2560, out_features=10240, bias=True)\n",
|
| 531 |
+
" (fc2): Linear4bit(in_features=10240, out_features=2560, bias=True)\n",
|
| 532 |
+
" )\n",
|
| 533 |
+
" (input_layernorm): LayerNorm((2560,), eps=1e-05, elementwise_affine=True)\n",
|
| 534 |
+
" (resid_dropout): Dropout(p=0.1, inplace=False)\n",
|
| 535 |
+
" )\n",
|
| 536 |
+
" )\n",
|
| 537 |
+
" (final_layernorm): LayerNorm((2560,), eps=1e-05, elementwise_affine=True)\n",
|
| 538 |
+
" )\n",
|
| 539 |
+
" (lm_head): Linear(in_features=2560, out_features=51200, bias=True)\n",
|
| 540 |
+
")\n"
|
| 541 |
+
]
|
| 542 |
+
}
|
| 543 |
+
],
|
| 544 |
+
"source": [
|
| 545 |
+
"print(model)"
|
| 546 |
+
]
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"cell_type": "code",
|
| 550 |
+
"execution_count": 13,
|
| 551 |
+
"id": "18d5599f-992d-4d8e-a90c-4d43774be473",
|
| 552 |
+
"metadata": {},
|
| 553 |
+
"outputs": [
|
| 554 |
+
{
|
| 555 |
+
"name": "stdout",
|
| 556 |
+
"output_type": "stream",
|
| 557 |
+
"text": [
|
| 558 |
+
"trainable params: 7,864,320 || all params: 2,787,548,160 || trainable%: 0.2821231974697076\n"
|
| 559 |
+
]
|
| 560 |
+
}
|
| 561 |
+
],
|
| 562 |
+
"source": [
|
| 563 |
+
"config = LoraConfig(\n",
|
| 564 |
+
" r=16,\n",
|
| 565 |
+
" lora_alpha=16,\n",
|
| 566 |
+
" target_modules=[\"q_proj\",\"k_proj\",\"v_proj\"],\n",
|
| 567 |
+
" lora_dropout=0.05,\n",
|
| 568 |
+
" bias=\"none\",\n",
|
| 569 |
+
" task_type=\"CAUSAL_LM\"\n",
|
| 570 |
+
")\n",
|
| 571 |
+
"\n",
|
| 572 |
+
"model = get_peft_model(model, config)\n",
|
| 573 |
+
"model.print_trainable_parameters()"
|
| 574 |
+
]
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"cell_type": "code",
|
| 578 |
+
"execution_count": 14,
|
| 579 |
+
"id": "baeee903-3dce-48b2-93c3-7a697d8c6daf",
|
| 580 |
+
"metadata": {},
|
| 581 |
+
"outputs": [],
|
| 582 |
+
"source": [
|
| 583 |
+
"def tokenize(sample):\n",
|
| 584 |
+
" model_inps = tokenizer(sample[\"text\"], padding=True, truncation=True, max_length=512)\n",
|
| 585 |
+
" return model_inps"
|
| 586 |
+
]
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"cell_type": "code",
|
| 590 |
+
"execution_count": 15,
|
| 591 |
+
"id": "28a9b24a-a822-4fcb-96b3-d77b7ea30a5f",
|
| 592 |
+
"metadata": {},
|
| 593 |
+
"outputs": [
|
| 594 |
+
{
|
| 595 |
+
"name": "stdout",
|
| 596 |
+
"output_type": "stream",
|
| 597 |
+
"text": [
|
| 598 |
+
"Requirement already satisfied: scikit-learn in /usr/local/lib/python3.11/site-packages (1.4.0)\n",
|
| 599 |
+
"Requirement already satisfied: numpy<2.0,>=1.19.5 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (1.26.4)\n",
|
| 600 |
+
"Requirement already satisfied: scipy>=1.6.0 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (1.12.0)\n",
|
| 601 |
+
"Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (1.3.2)\n",
|
| 602 |
+
"Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.11/site-packages (from scikit-learn) (3.3.0)\n"
|
| 603 |
+
]
|
| 604 |
+
}
|
| 605 |
+
],
|
| 606 |
+
"source": [
|
| 607 |
+
"!pip install scikit-learn"
|
| 608 |
+
]
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"cell_type": "code",
|
| 612 |
+
"execution_count": 16,
|
| 613 |
+
"id": "1ee7fd2a-38e4-4f23-a978-0bdeeda64d8b",
|
| 614 |
+
"metadata": {},
|
| 615 |
+
"outputs": [],
|
| 616 |
+
"source": [
|
| 617 |
+
"import pandas as pd\n",
|
| 618 |
+
"\n",
|
| 619 |
+
"from sklearn.model_selection import train_test_split\n",
|
| 620 |
+
"dataset_name='data.csv'\n",
|
| 621 |
+
"df = pd.read_csv(dataset_name)\n",
|
| 622 |
+
"#train, test = train_test_split(df, test_size=0.2)"
|
| 623 |
+
]
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"cell_type": "code",
|
| 627 |
+
"execution_count": 17,
|
| 628 |
+
"id": "e84c29e2-843e-42c2-8c0f-324d392e671c",
|
| 629 |
+
"metadata": {},
|
| 630 |
+
"outputs": [
|
| 631 |
+
{
|
| 632 |
+
"name": "stderr",
|
| 633 |
+
"output_type": "stream",
|
| 634 |
+
"text": [
|
| 635 |
+
"Tokenizing data: 100%|██████████| 16412/16412 [00:02<00:00, 6502.58 examples/s]\n"
|
| 636 |
+
]
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"data": {
|
| 640 |
+
"text/plain": [
|
| 641 |
+
"Dataset({\n",
|
| 642 |
+
" features: ['input_ids', 'attention_mask'],\n",
|
| 643 |
+
" num_rows: 16412\n",
|
| 644 |
+
"})"
|
| 645 |
+
]
|
| 646 |
+
},
|
| 647 |
+
"execution_count": 17,
|
| 648 |
+
"metadata": {},
|
| 649 |
+
"output_type": "execute_result"
|
| 650 |
+
}
|
| 651 |
+
],
|
| 652 |
+
"source": [
|
| 653 |
+
"data_df = df\n",
|
| 654 |
+
"data_df[\"text\"] = data_df[[\"user\", \"assistant\"]].apply(lambda x: \"question: \" + str(x[\"user\"]) + \" answer: \" + str(x[\"assistant\"]), axis=1)\n",
|
| 655 |
+
"data = Dataset.from_pandas(data_df)\n",
|
| 656 |
+
"tokenized_data = data.map(tokenize, batched=True, desc=\"Tokenizing data\", remove_columns=data.column_names)\n",
|
| 657 |
+
"tokenized_data"
|
| 658 |
+
]
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"cell_type": "code",
|
| 662 |
+
"execution_count": 18,
|
| 663 |
+
"id": "ac968254-5338-49df-950d-222b82647407",
|
| 664 |
+
"metadata": {},
|
| 665 |
+
"outputs": [],
|
| 666 |
+
"source": [
|
| 667 |
+
"training_arguments = TrainingArguments(\n",
|
| 668 |
+
" output_dir=\".\",\n",
|
| 669 |
+
" per_device_train_batch_size=4,\n",
|
| 670 |
+
" gradient_accumulation_steps=1,\n",
|
| 671 |
+
" learning_rate=2e-4,\n",
|
| 672 |
+
" lr_scheduler_type=\"cosine\",\n",
|
| 673 |
+
" save_strategy=\"epoch\",\n",
|
| 674 |
+
" logging_steps=100,\n",
|
| 675 |
+
" max_steps=1100,\n",
|
| 676 |
+
" num_train_epochs=3,\n",
|
| 677 |
+
" push_to_hub=True\n",
|
| 678 |
+
" )"
|
| 679 |
+
]
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"cell_type": "code",
|
| 683 |
+
"execution_count": 19,
|
| 684 |
+
"id": "729df2d0-0890-4ac4-adf3-c167a6e9669d",
|
| 685 |
+
"metadata": {},
|
| 686 |
+
"outputs": [
|
| 687 |
+
{
|
| 688 |
+
"name": "stderr",
|
| 689 |
+
"output_type": "stream",
|
| 690 |
+
"text": [
|
| 691 |
+
"huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
|
| 692 |
+
"To disable this warning, you can either:\n",
|
| 693 |
+
"\t- Avoid using `tokenizers` before the fork if possible\n",
|
| 694 |
+
"\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
|
| 695 |
+
]
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"name": "stdout",
|
| 699 |
+
"output_type": "stream",
|
| 700 |
+
"text": [
|
| 701 |
+
"Requirement already satisfied: huggingface_hub in /usr/local/lib/python3.11/site-packages (0.20.3)\n",
|
| 702 |
+
"Requirement already satisfied: filelock in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (3.13.1)\n",
|
| 703 |
+
"Requirement already satisfied: fsspec>=2023.5.0 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (2023.10.0)\n",
|
| 704 |
+
"Requirement already satisfied: requests in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (2.31.0)\n",
|
| 705 |
+
"Requirement already satisfied: tqdm>=4.42.1 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (4.66.2)\n",
|
| 706 |
+
"Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (6.0.1)\n",
|
| 707 |
+
"Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (4.9.0)\n",
|
| 708 |
+
"Requirement already satisfied: packaging>=20.9 in /usr/local/lib/python3.11/site-packages (from huggingface_hub) (23.2)\n",
|
| 709 |
+
"Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (2.1.1)\n",
|
| 710 |
+
"Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (3.6)\n",
|
| 711 |
+
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (2.2.0)\n",
|
| 712 |
+
"Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/site-packages (from requests->huggingface_hub) (2024.2.2)\n"
|
| 713 |
+
]
|
| 714 |
+
}
|
| 715 |
+
],
|
| 716 |
+
"source": [
|
| 717 |
+
"!pip install huggingface_hub"
|
| 718 |
+
]
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"cell_type": "code",
|
| 722 |
+
"execution_count": 20,
|
| 723 |
+
"id": "525f81a4-eb92-466e-bb9c-cd63122231ab",
|
| 724 |
+
"metadata": {},
|
| 725 |
+
"outputs": [
|
| 726 |
+
{
|
| 727 |
+
"name": "stderr",
|
| 728 |
+
"output_type": "stream",
|
| 729 |
+
"text": [
|
| 730 |
+
"huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
|
| 731 |
+
"To disable this warning, you can either:\n",
|
| 732 |
+
"\t- Avoid using `tokenizers` before the fork if possible\n",
|
| 733 |
+
"\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
|
| 734 |
+
]
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"name": "stdout",
|
| 738 |
+
"output_type": "stream",
|
| 739 |
+
"text": [
|
| 740 |
+
"Requirement already satisfied: jupyterlab in /usr/local/lib/python3.11/site-packages (4.1.0)\n",
|
| 741 |
+
"Collecting jupyterlab\n",
|
| 742 |
+
" Downloading jupyterlab-4.1.1-py3-none-any.whl.metadata (15 kB)\n",
|
| 743 |
+
"Collecting jupyterlab-git\n",
|
| 744 |
+
" Downloading jupyterlab_git-0.50.0-py3-none-any.whl.metadata (31 kB)\n",
|
| 745 |
+
"Requirement already satisfied: async-lru>=1.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.0.4)\n",
|
| 746 |
+
"Requirement already satisfied: httpx>=0.25.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (0.26.0)\n",
|
| 747 |
+
"Requirement already satisfied: ipykernel in /usr/local/lib/python3.11/site-packages (from jupyterlab) (6.29.1)\n",
|
| 748 |
+
"Requirement already satisfied: jinja2>=3.0.3 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (3.1.3)\n",
|
| 749 |
+
"Requirement already satisfied: jupyter-core in /usr/local/lib/python3.11/site-packages (from jupyterlab) (5.7.1)\n",
|
| 750 |
+
"Requirement already satisfied: jupyter-lsp>=2.0.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.2.2)\n",
|
| 751 |
+
"Requirement already satisfied: jupyter-server<3,>=2.4.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.12.5)\n",
|
| 752 |
+
"Requirement already satisfied: jupyterlab-server<3,>=2.19.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (2.25.2)\n",
|
| 753 |
+
"Requirement already satisfied: notebook-shim>=0.2 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (0.2.3)\n",
|
| 754 |
+
"Requirement already satisfied: packaging in /usr/local/lib/python3.11/site-packages (from jupyterlab) (23.2)\n",
|
| 755 |
+
"Requirement already satisfied: tornado>=6.2.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab) (6.4)\n",
|
| 756 |
+
"Requirement already satisfied: traitlets in /usr/local/lib/python3.11/site-packages (from jupyterlab) (5.14.1)\n",
|
| 757 |
+
"Collecting nbdime~=4.0.1 (from jupyterlab-git)\n",
|
| 758 |
+
" Downloading nbdime-4.0.1-py3-none-any.whl.metadata (9.5 kB)\n",
|
| 759 |
+
"Requirement already satisfied: nbformat in /usr/local/lib/python3.11/site-packages (from jupyterlab-git) (5.9.2)\n",
|
| 760 |
+
"Requirement already satisfied: pexpect in /usr/local/lib/python3.11/site-packages (from jupyterlab-git) (4.9.0)\n",
|
| 761 |
+
"Requirement already satisfied: anyio in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (4.2.0)\n",
|
| 762 |
+
"Requirement already satisfied: certifi in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (2024.2.2)\n",
|
| 763 |
+
"Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (1.0.2)\n",
|
| 764 |
+
"Requirement already satisfied: idna in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (3.6)\n",
|
| 765 |
+
"Requirement already satisfied: sniffio in /usr/local/lib/python3.11/site-packages (from httpx>=0.25.0->jupyterlab) (1.3.0)\n",
|
| 766 |
+
"Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.11/site-packages (from httpcore==1.*->httpx>=0.25.0->jupyterlab) (0.14.0)\n",
|
| 767 |
+
"Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/site-packages (from jinja2>=3.0.3->jupyterlab) (2.1.5)\n",
|
| 768 |
+
"Requirement already satisfied: argon2-cffi in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (23.1.0)\n",
|
| 769 |
+
"Requirement already satisfied: jupyter-client>=7.4.4 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (8.6.0)\n",
|
| 770 |
+
"Requirement already satisfied: jupyter-events>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.9.0)\n",
|
| 771 |
+
"Requirement already satisfied: jupyter-server-terminals in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.5.2)\n",
|
| 772 |
+
"Requirement already satisfied: nbconvert>=6.4.4 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (7.15.0)\n",
|
| 773 |
+
"Requirement already satisfied: overrides in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (7.7.0)\n",
|
| 774 |
+
"Requirement already satisfied: prometheus-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.19.0)\n",
|
| 775 |
+
"Requirement already satisfied: pyzmq>=24 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (25.1.2)\n",
|
| 776 |
+
"Requirement already satisfied: send2trash>=1.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (1.8.2)\n",
|
| 777 |
+
"Requirement already satisfied: terminado>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (0.18.0)\n",
|
| 778 |
+
"Requirement already satisfied: websocket-client in /usr/local/lib/python3.11/site-packages (from jupyter-server<3,>=2.4.0->jupyterlab) (1.7.0)\n",
|
| 779 |
+
"Requirement already satisfied: platformdirs>=2.5 in /usr/local/lib/python3.11/site-packages (from jupyter-core->jupyterlab) (4.2.0)\n",
|
| 780 |
+
"Requirement already satisfied: babel>=2.10 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (2.14.0)\n",
|
| 781 |
+
"Requirement already satisfied: json5>=0.9.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (0.9.14)\n",
|
| 782 |
+
"Requirement already satisfied: jsonschema>=4.18.0 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (4.21.1)\n",
|
| 783 |
+
"Requirement already satisfied: requests>=2.31 in /usr/local/lib/python3.11/site-packages (from jupyterlab-server<3,>=2.19.0->jupyterlab) (2.31.0)\n",
|
| 784 |
+
"Collecting colorama (from nbdime~=4.0.1->jupyterlab-git)\n",
|
| 785 |
+
" Downloading colorama-0.4.6-py2.py3-none-any.whl (25 kB)\n",
|
| 786 |
+
"Collecting gitpython!=2.1.4,!=2.1.5,!=2.1.6 (from nbdime~=4.0.1->jupyterlab-git)\n",
|
| 787 |
+
" Downloading GitPython-3.1.41-py3-none-any.whl.metadata (14 kB)\n",
|
| 788 |
+
"Collecting jupyter-server-mathjax>=0.2.2 (from nbdime~=4.0.1->jupyterlab-git)\n",
|
| 789 |
+
" Downloading jupyter_server_mathjax-0.2.6-py3-none-any.whl (3.1 MB)\n",
|
| 790 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.1/3.1 MB\u001b[0m \u001b[31m265.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 791 |
+
"\u001b[?25hRequirement already satisfied: pygments in /usr/local/lib/python3.11/site-packages (from nbdime~=4.0.1->jupyterlab-git) (2.17.2)\n",
|
| 792 |
+
"Requirement already satisfied: fastjsonschema in /usr/local/lib/python3.11/site-packages (from nbformat->jupyterlab-git) (2.19.1)\n",
|
| 793 |
+
"Requirement already satisfied: comm>=0.1.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (0.2.1)\n",
|
| 794 |
+
"Requirement already satisfied: debugpy>=1.6.5 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (1.8.0)\n",
|
| 795 |
+
"Requirement already satisfied: ipython>=7.23.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (8.21.0)\n",
|
| 796 |
+
"Requirement already satisfied: matplotlib-inline>=0.1 in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (0.1.6)\n",
|
| 797 |
+
"Requirement already satisfied: nest-asyncio in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (1.6.0)\n",
|
| 798 |
+
"Requirement already satisfied: psutil in /usr/local/lib/python3.11/site-packages (from ipykernel->jupyterlab) (5.9.8)\n",
|
| 799 |
+
"Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.11/site-packages (from pexpect->jupyterlab-git) (0.7.0)\n",
|
| 800 |
+
"Collecting gitdb<5,>=4.0.1 (from gitpython!=2.1.4,!=2.1.5,!=2.1.6->nbdime~=4.0.1->jupyterlab-git)\n",
|
| 801 |
+
" Downloading gitdb-4.0.11-py3-none-any.whl.metadata (1.2 kB)\n",
|
| 802 |
+
"Requirement already satisfied: decorator in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (5.1.1)\n",
|
| 803 |
+
"Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (0.19.1)\n",
|
| 804 |
+
"Requirement already satisfied: prompt-toolkit<3.1.0,>=3.0.41 in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (3.0.43)\n",
|
| 805 |
+
"Requirement already satisfied: stack-data in /usr/local/lib/python3.11/site-packages (from ipython>=7.23.1->ipykernel->jupyterlab) (0.6.3)\n",
|
| 806 |
+
"Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (23.2.0)\n",
|
| 807 |
+
"Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (2023.12.1)\n",
|
| 808 |
+
"Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (0.33.0)\n",
|
| 809 |
+
"Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.11/site-packages (from jsonschema>=4.18.0->jupyterlab-server<3,>=2.19.0->jupyterlab) (0.17.1)\n",
|
| 810 |
+
"Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/site-packages (from jupyter-client>=7.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (2.8.2)\n",
|
| 811 |
+
"Requirement already satisfied: python-json-logger>=2.0.4 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.0.7)\n",
|
| 812 |
+
"Requirement already satisfied: pyyaml>=5.3 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (6.0.1)\n",
|
| 813 |
+
"Requirement already satisfied: rfc3339-validator in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (0.1.4)\n",
|
| 814 |
+
"Requirement already satisfied: rfc3986-validator>=0.1.1 in /usr/local/lib/python3.11/site-packages (from jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (0.1.1)\n",
|
| 815 |
+
"Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (4.12.3)\n",
|
| 816 |
+
"Requirement already satisfied: bleach!=5.0.0 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (6.1.0)\n",
|
| 817 |
+
"Requirement already satisfied: defusedxml in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.7.1)\n",
|
| 818 |
+
"Requirement already satisfied: jupyterlab-pygments in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.3.0)\n",
|
| 819 |
+
"Requirement already satisfied: mistune<4,>=2.0.3 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (3.0.2)\n",
|
| 820 |
+
"Requirement already satisfied: nbclient>=0.5.0 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.9.0)\n",
|
| 821 |
+
"Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.5.1)\n",
|
| 822 |
+
"Requirement already satisfied: tinycss2 in /usr/local/lib/python3.11/site-packages (from nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.2.1)\n",
|
| 823 |
+
"Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.19.0->jupyterlab) (2.1.1)\n",
|
| 824 |
+
"Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/site-packages (from requests>=2.31->jupyterlab-server<3,>=2.19.0->jupyterlab) (2.2.0)\n",
|
| 825 |
+
"Requirement already satisfied: argon2-cffi-bindings in /usr/local/lib/python3.11/site-packages (from argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (21.2.0)\n",
|
| 826 |
+
"Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (1.16.0)\n",
|
| 827 |
+
"Requirement already satisfied: webencodings in /usr/local/lib/python3.11/site-packages (from bleach!=5.0.0->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (0.5.1)\n",
|
| 828 |
+
"Collecting smmap<6,>=3.0.1 (from gitdb<5,>=4.0.1->gitpython!=2.1.4,!=2.1.5,!=2.1.6->nbdime~=4.0.1->jupyterlab-git)\n",
|
| 829 |
+
" Downloading smmap-5.0.1-py3-none-any.whl.metadata (4.3 kB)\n",
|
| 830 |
+
"Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.11/site-packages (from jedi>=0.16->ipython>=7.23.1->ipykernel->jupyterlab) (0.8.3)\n",
|
| 831 |
+
"Requirement already satisfied: fqdn in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.5.1)\n",
|
| 832 |
+
"Requirement already satisfied: isoduration in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (20.11.0)\n",
|
| 833 |
+
"Requirement already satisfied: jsonpointer>1.13 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.4)\n",
|
| 834 |
+
"Requirement already satisfied: uri-template in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.3.0)\n",
|
| 835 |
+
"Requirement already satisfied: webcolors>=1.11 in /usr/local/lib/python3.11/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.13)\n",
|
| 836 |
+
"Requirement already satisfied: wcwidth in /usr/local/lib/python3.11/site-packages (from prompt-toolkit<3.1.0,>=3.0.41->ipython>=7.23.1->ipykernel->jupyterlab) (0.2.13)\n",
|
| 837 |
+
"Requirement already satisfied: cffi>=1.0.1 in /usr/local/lib/python3.11/site-packages (from argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (1.16.0)\n",
|
| 838 |
+
"Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.11/site-packages (from beautifulsoup4->nbconvert>=6.4.4->jupyter-server<3,>=2.4.0->jupyterlab) (2.5)\n",
|
| 839 |
+
"Requirement already satisfied: executing>=1.2.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (2.0.1)\n",
|
| 840 |
+
"Requirement already satisfied: asttokens>=2.1.0 in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (2.4.1)\n",
|
| 841 |
+
"Requirement already satisfied: pure-eval in /usr/local/lib/python3.11/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyterlab) (0.2.2)\n",
|
| 842 |
+
"Requirement already satisfied: pycparser in /usr/local/lib/python3.11/site-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->jupyter-server<3,>=2.4.0->jupyterlab) (2.21)\n",
|
| 843 |
+
"Requirement already satisfied: arrow>=0.15.0 in /usr/local/lib/python3.11/site-packages (from isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (1.3.0)\n",
|
| 844 |
+
"Requirement already satisfied: types-python-dateutil>=2.8.10 in /usr/local/lib/python3.11/site-packages (from arrow>=0.15.0->isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server<3,>=2.4.0->jupyterlab) (2.8.19.20240106)\n",
|
| 845 |
+
"Downloading jupyterlab-4.1.1-py3-none-any.whl (11.4 MB)\n",
|
| 846 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m11.4/11.4 MB\u001b[0m \u001b[31m434.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 847 |
+
"\u001b[?25hDownloading jupyterlab_git-0.50.0-py3-none-any.whl (1.2 MB)\n",
|
| 848 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.2/1.2 MB\u001b[0m \u001b[31m546.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 849 |
+
"\u001b[?25hDownloading nbdime-4.0.1-py3-none-any.whl (5.9 MB)\n",
|
| 850 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.9/5.9 MB\u001b[0m \u001b[31m279.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 851 |
+
"\u001b[?25hDownloading GitPython-3.1.41-py3-none-any.whl (196 kB)\n",
|
| 852 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m196.4/196.4 kB\u001b[0m \u001b[31m474.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 853 |
+
"\u001b[?25hDownloading gitdb-4.0.11-py3-none-any.whl (62 kB)\n",
|
| 854 |
+
"\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m62.7/62.7 kB\u001b[0m \u001b[31m406.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
|
| 855 |
+
"\u001b[?25hDownloading smmap-5.0.1-py3-none-any.whl (24 kB)\n",
|
| 856 |
+
"Installing collected packages: smmap, colorama, gitdb, gitpython, jupyter-server-mathjax, nbdime, jupyterlab, jupyterlab-git\n",
|
| 857 |
+
" Attempting uninstall: jupyterlab\n",
|
| 858 |
+
" Found existing installation: jupyterlab 4.1.0\n",
|
| 859 |
+
" Uninstalling jupyterlab-4.1.0:\n",
|
| 860 |
+
" Successfully uninstalled jupyterlab-4.1.0\n",
|
| 861 |
+
"Successfully installed colorama-0.4.6 gitdb-4.0.11 gitpython-3.1.41 jupyter-server-mathjax-0.2.6 jupyterlab-4.1.1 jupyterlab-git-0.50.0 nbdime-4.0.1 smmap-5.0.1\n",
|
| 862 |
+
"Note: you may need to restart the kernel to use updated packages.\n"
|
| 863 |
+
]
|
| 864 |
+
}
|
| 865 |
+
],
|
| 866 |
+
"source": [
|
| 867 |
+
"pip install --upgrade jupyterlab jupyterlab-git"
|
| 868 |
+
]
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"cell_type": "code",
|
| 872 |
+
"execution_count": 22,
|
| 873 |
+
"id": "05d58512-a9e2-4319-88bf-9331c6a0584c",
|
| 874 |
+
"metadata": {},
|
| 875 |
+
"outputs": [
|
| 876 |
+
{
|
| 877 |
+
"name": "stdout",
|
| 878 |
+
"output_type": "stream",
|
| 879 |
+
"text": [
|
| 880 |
+
"\n",
|
| 881 |
+
" _| _| _| _| _|_|_| _|_|_| _|_|_| _| _| _|_|_| _|_|_|_| _|_| _|_|_| _|_|_|_|\n",
|
| 882 |
+
" _| _| _| _| _| _| _| _|_| _| _| _| _| _| _| _|\n",
|
| 883 |
+
" _|_|_|_| _| _| _| _|_| _| _|_| _| _| _| _| _| _|_| _|_|_| _|_|_|_| _| _|_|_|\n",
|
| 884 |
+
" _| _| _| _| _| _| _| _| _| _| _|_| _| _| _| _| _| _| _|\n",
|
| 885 |
+
" _| _| _|_| _|_|_| _|_|_| _|_|_| _| _| _|_|_| _| _| _| _|_|_| _|_|_|_|\n",
|
| 886 |
+
"\n",
|
| 887 |
+
" To login, `huggingface_hub` requires a token generated from https://huggingface.co/settings/tokens .\n"
|
| 888 |
+
]
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"name": "stdin",
|
| 892 |
+
"output_type": "stream",
|
| 893 |
+
"text": [
|
| 894 |
+
"Token: ········\n",
|
| 895 |
+
"Add token as git credential? (Y/n) n\n"
|
| 896 |
+
]
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"name": "stdout",
|
| 900 |
+
"output_type": "stream",
|
| 901 |
+
"text": [
|
| 902 |
+
"Token is valid (permission: write).\n",
|
| 903 |
+
"Your token has been saved to /root/.cache/huggingface/token\n",
|
| 904 |
+
"Login successful\n"
|
| 905 |
+
]
|
| 906 |
+
}
|
| 907 |
+
],
|
| 908 |
+
"source": [
|
| 909 |
+
"from huggingface_hub import interpreter_login\n",
|
| 910 |
+
"interpreter_login()"
|
| 911 |
+
]
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"cell_type": "code",
|
| 915 |
+
"execution_count": null,
|
| 916 |
+
"id": "3bf553b6-b26c-49c3-9407-74c8d53a395e",
|
| 917 |
+
"metadata": {},
|
| 918 |
+
"outputs": [
|
| 919 |
+
{
|
| 920 |
+
"data": {
|
| 921 |
+
"text/html": [
|
| 922 |
+
"\n",
|
| 923 |
+
" <div>\n",
|
| 924 |
+
" \n",
|
| 925 |
+
" <progress value='1044' max='1100' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
|
| 926 |
+
" [1044/1100 03:16 < 00:10, 5.31 it/s, Epoch 0.25/1]\n",
|
| 927 |
+
" </div>\n",
|
| 928 |
+
" <table border=\"1\" class=\"dataframe\">\n",
|
| 929 |
+
" <thead>\n",
|
| 930 |
+
" <tr style=\"text-align: left;\">\n",
|
| 931 |
+
" <th>Step</th>\n",
|
| 932 |
+
" <th>Training Loss</th>\n",
|
| 933 |
+
" </tr>\n",
|
| 934 |
+
" </thead>\n",
|
| 935 |
+
" <tbody>\n",
|
| 936 |
+
" <tr>\n",
|
| 937 |
+
" <td>100</td>\n",
|
| 938 |
+
" <td>1.473100</td>\n",
|
| 939 |
+
" </tr>\n",
|
| 940 |
+
" <tr>\n",
|
| 941 |
+
" <td>200</td>\n",
|
| 942 |
+
" <td>1.236100</td>\n",
|
| 943 |
+
" </tr>\n",
|
| 944 |
+
" <tr>\n",
|
| 945 |
+
" <td>300</td>\n",
|
| 946 |
+
" <td>1.224800</td>\n",
|
| 947 |
+
" </tr>\n",
|
| 948 |
+
" <tr>\n",
|
| 949 |
+
" <td>400</td>\n",
|
| 950 |
+
" <td>1.157900</td>\n",
|
| 951 |
+
" </tr>\n",
|
| 952 |
+
" <tr>\n",
|
| 953 |
+
" <td>500</td>\n",
|
| 954 |
+
" <td>1.177800</td>\n",
|
| 955 |
+
" </tr>\n",
|
| 956 |
+
" <tr>\n",
|
| 957 |
+
" <td>600</td>\n",
|
| 958 |
+
" <td>1.150500</td>\n",
|
| 959 |
+
" </tr>\n",
|
| 960 |
+
" <tr>\n",
|
| 961 |
+
" <td>700</td>\n",
|
| 962 |
+
" <td>1.178100</td>\n",
|
| 963 |
+
" </tr>\n",
|
| 964 |
+
" <tr>\n",
|
| 965 |
+
" <td>800</td>\n",
|
| 966 |
+
" <td>1.166400</td>\n",
|
| 967 |
+
" </tr>\n",
|
| 968 |
+
" <tr>\n",
|
| 969 |
+
" <td>900</td>\n",
|
| 970 |
+
" <td>1.138400</td>\n",
|
| 971 |
+
" </tr>\n",
|
| 972 |
+
" <tr>\n",
|
| 973 |
+
" <td>1000</td>\n",
|
| 974 |
+
" <td>1.144000</td>\n",
|
| 975 |
+
" </tr>\n",
|
| 976 |
+
" </tbody>\n",
|
| 977 |
+
"</table><p>"
|
| 978 |
+
],
|
| 979 |
+
"text/plain": [
|
| 980 |
+
"<IPython.core.display.HTML object>"
|
| 981 |
+
]
|
| 982 |
+
},
|
| 983 |
+
"metadata": {},
|
| 984 |
+
"output_type": "display_data"
|
| 985 |
+
}
|
| 986 |
+
],
|
| 987 |
+
"source": [
|
| 988 |
+
"trainer = Trainer(\n",
|
| 989 |
+
" model=model,\n",
|
| 990 |
+
" train_dataset=tokenized_data,\n",
|
| 991 |
+
" args=training_arguments,\n",
|
| 992 |
+
" data_collator=DataCollatorForLanguageModeling(tokenizer, mlm=False)\n",
|
| 993 |
+
")\n",
|
| 994 |
+
"trainer.train()"
|
| 995 |
+
]
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"cell_type": "code",
|
| 999 |
+
"execution_count": null,
|
| 1000 |
+
"id": "263cc15e-8e9d-4bd8-9708-ec1638bc1165",
|
| 1001 |
+
"metadata": {},
|
| 1002 |
+
"outputs": [],
|
| 1003 |
+
"source": [
|
| 1004 |
+
"from peft import PeftModel\n",
|
| 1005 |
+
"from transformers import AutoModelForCausalLM\n",
|
| 1006 |
+
"import torch\n",
|
| 1007 |
+
"model = AutoModelForCausalLM.from_pretrained(\"microsoft/phi-2\", trust_remote_code=True, torch_dtype=torch.float32)\n",
|
| 1008 |
+
"peft_model = PeftModel.from_pretrained(model, \"aissatoubalde/lab\", from_transformers=True)\n",
|
| 1009 |
+
"model = peft_model.merge_and_unload()\n",
|
| 1010 |
+
"model"
|
| 1011 |
+
]
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"cell_type": "code",
|
| 1015 |
+
"execution_count": null,
|
| 1016 |
+
"id": "8eef4f4f-52da-4ba9-8a22-2b7874420562",
|
| 1017 |
+
"metadata": {},
|
| 1018 |
+
"outputs": [],
|
| 1019 |
+
"source": [
|
| 1020 |
+
"model.push_to_hub('aissatoubalde/lab')"
|
| 1021 |
+
]
|
| 1022 |
+
}
|
| 1023 |
+
],
|
| 1024 |
+
"metadata": {
|
| 1025 |
+
"kernelspec": {
|
| 1026 |
+
"display_name": "Python 3 (ipykernel)",
|
| 1027 |
+
"language": "python",
|
| 1028 |
+
"name": "python3"
|
| 1029 |
+
},
|
| 1030 |
+
"language_info": {
|
| 1031 |
+
"codemirror_mode": {
|
| 1032 |
+
"name": "ipython",
|
| 1033 |
+
"version": 3
|
| 1034 |
+
},
|
| 1035 |
+
"file_extension": ".py",
|
| 1036 |
+
"mimetype": "text/x-python",
|
| 1037 |
+
"name": "python",
|
| 1038 |
+
"nbconvert_exporter": "python",
|
| 1039 |
+
"pygments_lexer": "ipython3",
|
| 1040 |
+
"version": "3.11.5"
|
| 1041 |
+
}
|
| 1042 |
+
},
|
| 1043 |
+
"nbformat": 4,
|
| 1044 |
+
"nbformat_minor": 5
|
| 1045 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6018ee5dd9088c029dec6b7490a081b122e08b3ae81ee63f1cb6fda91dcb9fd4
|
| 3 |
+
size 4664
|