Spaces:
Sleeping
Sleeping
havinashpatil commited on
Commit ·
82e39c9
1
Parent(s): 9204c04
fix: removed invalid openenv-py package from notebook install cell
Browse files- train_grpo.ipynb +157 -8
train_grpo.ipynb
CHANGED
|
@@ -12,20 +12,169 @@
|
|
| 12 |
},
|
| 13 |
{
|
| 14 |
"cell_type": "code",
|
| 15 |
-
"execution_count":
|
| 16 |
"metadata": {},
|
| 17 |
-
"outputs": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
"source": [
|
| 19 |
-
"!pip install trl transformers datasets
|
| 20 |
"!git clone https://github.com/havinashpatil/meta.git\n",
|
| 21 |
"!cd meta && pip install -r requirements.txt"
|
| 22 |
]
|
| 23 |
},
|
| 24 |
{
|
| 25 |
"cell_type": "code",
|
| 26 |
-
"execution_count":
|
| 27 |
"metadata": {},
|
| 28 |
-
"outputs": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"source": [
|
| 30 |
"import torch\n",
|
| 31 |
"from datasets import load_dataset\n",
|
|
@@ -128,7 +277,7 @@
|
|
| 128 |
],
|
| 129 |
"metadata": {
|
| 130 |
"kernelspec": {
|
| 131 |
-
"display_name": "
|
| 132 |
"language": "python",
|
| 133 |
"name": "python3"
|
| 134 |
},
|
|
@@ -142,9 +291,9 @@
|
|
| 142 |
"name": "python",
|
| 143 |
"nbconvert_exporter": "python",
|
| 144 |
"pygments_lexer": "ipython3",
|
| 145 |
-
"version": "3.
|
| 146 |
}
|
| 147 |
},
|
| 148 |
"nbformat": 4,
|
| 149 |
"nbformat_minor": 4
|
| 150 |
-
}
|
|
|
|
| 12 |
},
|
| 13 |
{
|
| 14 |
"cell_type": "code",
|
| 15 |
+
"execution_count": 2,
|
| 16 |
"metadata": {},
|
| 17 |
+
"outputs": [
|
| 18 |
+
{
|
| 19 |
+
"name": "stdout",
|
| 20 |
+
"output_type": "stream",
|
| 21 |
+
"text": [
|
| 22 |
+
"Requirement already satisfied: trl in .\\venv\\lib\\site-packages (1.2.0)\n",
|
| 23 |
+
"Requirement already satisfied: transformers in .\\venv\\lib\\site-packages (5.6.2)\n",
|
| 24 |
+
"Requirement already satisfied: datasets in .\\venv\\lib\\site-packages (4.8.4)\n"
|
| 25 |
+
]
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"name": "stderr",
|
| 29 |
+
"output_type": "stream",
|
| 30 |
+
"text": [
|
| 31 |
+
"ERROR: Could not find a version that satisfies the requirement openenv-py (from versions: none)\n",
|
| 32 |
+
"\n",
|
| 33 |
+
"[notice] A new release of pip is available: 25.2 -> 26.0.1\n",
|
| 34 |
+
"[notice] To update, run: python.exe -m pip install --upgrade pip\n",
|
| 35 |
+
"ERROR: No matching distribution found for openenv-py\n",
|
| 36 |
+
"Cloning into 'meta'...\n"
|
| 37 |
+
]
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"name": "stdout",
|
| 41 |
+
"output_type": "stream",
|
| 42 |
+
"text": [
|
| 43 |
+
"Requirement already satisfied: fastapi>=0.100.0 in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 1)) (0.136.0)\n",
|
| 44 |
+
"Requirement already satisfied: uvicorn>=0.23.0 in e:\\meta\\venv\\lib\\site-packages (from uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (0.45.0)\n",
|
| 45 |
+
"Requirement already satisfied: pydantic>=2.0.0 in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 3)) (2.13.3)\n",
|
| 46 |
+
"Requirement already satisfied: openai>=1.0.0 in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 4)) (2.32.0)\n",
|
| 47 |
+
"Requirement already satisfied: httpx>=0.24.1 in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 5)) (0.28.1)\n",
|
| 48 |
+
"Requirement already satisfied: pandas in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 6)) (3.0.2)\n",
|
| 49 |
+
"Requirement already satisfied: matplotlib in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 7)) (3.10.9)\n",
|
| 50 |
+
"Requirement already satisfied: transformers in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 8)) (5.6.2)\n",
|
| 51 |
+
"Requirement already satisfied: torch in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 9)) (2.11.0)\n",
|
| 52 |
+
"Requirement already satisfied: datasets in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 10)) (4.8.4)\n",
|
| 53 |
+
"Requirement already satisfied: trl in e:\\meta\\venv\\lib\\site-packages (from -r requirements.txt (line 11)) (1.2.0)\n",
|
| 54 |
+
"Requirement already satisfied: starlette>=0.46.0 in e:\\meta\\venv\\lib\\site-packages (from fastapi>=0.100.0->-r requirements.txt (line 1)) (1.0.0)\n",
|
| 55 |
+
"Requirement already satisfied: typing-extensions>=4.8.0 in e:\\meta\\venv\\lib\\site-packages (from fastapi>=0.100.0->-r requirements.txt (line 1)) (4.15.0)\n",
|
| 56 |
+
"Requirement already satisfied: typing-inspection>=0.4.2 in e:\\meta\\venv\\lib\\site-packages (from fastapi>=0.100.0->-r requirements.txt (line 1)) (0.4.2)\n",
|
| 57 |
+
"Requirement already satisfied: annotated-doc>=0.0.2 in e:\\meta\\venv\\lib\\site-packages (from fastapi>=0.100.0->-r requirements.txt (line 1)) (0.0.4)\n",
|
| 58 |
+
"Requirement already satisfied: click>=7.0 in e:\\meta\\venv\\lib\\site-packages (from uvicorn>=0.23.0->uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (8.3.2)\n",
|
| 59 |
+
"Requirement already satisfied: h11>=0.8 in e:\\meta\\venv\\lib\\site-packages (from uvicorn>=0.23.0->uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (0.16.0)\n",
|
| 60 |
+
"Requirement already satisfied: annotated-types>=0.6.0 in e:\\meta\\venv\\lib\\site-packages (from pydantic>=2.0.0->-r requirements.txt (line 3)) (0.7.0)\n",
|
| 61 |
+
"Requirement already satisfied: pydantic-core==2.46.3 in e:\\meta\\venv\\lib\\site-packages (from pydantic>=2.0.0->-r requirements.txt (line 3)) (2.46.3)\n",
|
| 62 |
+
"Requirement already satisfied: anyio<5,>=3.5.0 in e:\\meta\\venv\\lib\\site-packages (from openai>=1.0.0->-r requirements.txt (line 4)) (4.13.0)\n",
|
| 63 |
+
"Requirement already satisfied: distro<2,>=1.7.0 in e:\\meta\\venv\\lib\\site-packages (from openai>=1.0.0->-r requirements.txt (line 4)) (1.9.0)\n",
|
| 64 |
+
"Requirement already satisfied: jiter<1,>=0.10.0 in e:\\meta\\venv\\lib\\site-packages (from openai>=1.0.0->-r requirements.txt (line 4)) (0.14.0)\n",
|
| 65 |
+
"Requirement already satisfied: sniffio in e:\\meta\\venv\\lib\\site-packages (from openai>=1.0.0->-r requirements.txt (line 4)) (1.3.1)\n",
|
| 66 |
+
"Requirement already satisfied: tqdm>4 in e:\\meta\\venv\\lib\\site-packages (from openai>=1.0.0->-r requirements.txt (line 4)) (4.67.3)\n",
|
| 67 |
+
"Requirement already satisfied: certifi in e:\\meta\\venv\\lib\\site-packages (from httpx>=0.24.1->-r requirements.txt (line 5)) (2026.4.22)\n",
|
| 68 |
+
"Requirement already satisfied: httpcore==1.* in e:\\meta\\venv\\lib\\site-packages (from httpx>=0.24.1->-r requirements.txt (line 5)) (1.0.9)\n",
|
| 69 |
+
"Requirement already satisfied: idna in e:\\meta\\venv\\lib\\site-packages (from httpx>=0.24.1->-r requirements.txt (line 5)) (3.12)\n",
|
| 70 |
+
"Requirement already satisfied: numpy>=1.26.0 in e:\\meta\\venv\\lib\\site-packages (from pandas->-r requirements.txt (line 6)) (2.4.4)\n",
|
| 71 |
+
"Requirement already satisfied: python-dateutil>=2.8.2 in e:\\meta\\venv\\lib\\site-packages (from pandas->-r requirements.txt (line 6)) (2.9.0.post0)\n",
|
| 72 |
+
"Requirement already satisfied: tzdata in e:\\meta\\venv\\lib\\site-packages (from pandas->-r requirements.txt (line 6)) (2026.2)\n",
|
| 73 |
+
"Requirement already satisfied: contourpy>=1.0.1 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (1.3.3)\n",
|
| 74 |
+
"Requirement already satisfied: cycler>=0.10 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (0.12.1)\n",
|
| 75 |
+
"Requirement already satisfied: fonttools>=4.22.0 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (4.62.1)\n",
|
| 76 |
+
"Requirement already satisfied: kiwisolver>=1.3.1 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (1.5.0)\n",
|
| 77 |
+
"Requirement already satisfied: packaging>=20.0 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (26.2)\n",
|
| 78 |
+
"Requirement already satisfied: pillow>=8 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (12.2.0)\n",
|
| 79 |
+
"Requirement already satisfied: pyparsing>=3 in e:\\meta\\venv\\lib\\site-packages (from matplotlib->-r requirements.txt (line 7)) (3.3.2)\n",
|
| 80 |
+
"Requirement already satisfied: huggingface-hub<2.0,>=1.5.0 in e:\\meta\\venv\\lib\\site-packages (from transformers->-r requirements.txt (line 8)) (1.12.0)\n",
|
| 81 |
+
"Requirement already satisfied: pyyaml>=5.1 in e:\\meta\\venv\\lib\\site-packages (from transformers->-r requirements.txt (line 8)) (6.0.3)\n",
|
| 82 |
+
"Requirement already satisfied: regex>=2025.10.22 in e:\\meta\\venv\\lib\\site-packages (from transformers->-r requirements.txt (line 8)) (2026.4.4)\n",
|
| 83 |
+
"Requirement already satisfied: tokenizers<=0.23.0,>=0.22.0 in e:\\meta\\venv\\lib\\site-packages (from transformers->-r requirements.txt (line 8)) (0.22.2)\n",
|
| 84 |
+
"Requirement already satisfied: typer in e:\\meta\\venv\\lib\\site-packages (from transformers->-r requirements.txt (line 8)) (0.24.2)\n",
|
| 85 |
+
"Requirement already satisfied: safetensors>=0.4.3 in e:\\meta\\venv\\lib\\site-packages (from transformers->-r requirements.txt (line 8)) (0.7.0)\n",
|
| 86 |
+
"Requirement already satisfied: filelock>=3.10.0 in e:\\meta\\venv\\lib\\site-packages (from huggingface-hub<2.0,>=1.5.0->transformers->-r requirements.txt (line 8)) (3.29.0)\n",
|
| 87 |
+
"Requirement already satisfied: fsspec>=2023.5.0 in e:\\meta\\venv\\lib\\site-packages (from huggingface-hub<2.0,>=1.5.0->transformers->-r requirements.txt (line 8)) (2026.2.0)\n",
|
| 88 |
+
"Requirement already satisfied: hf-xet<2.0.0,>=1.4.3 in e:\\meta\\venv\\lib\\site-packages (from huggingface-hub<2.0,>=1.5.0->transformers->-r requirements.txt (line 8)) (1.4.3)\n",
|
| 89 |
+
"Requirement already satisfied: setuptools<82 in e:\\meta\\venv\\lib\\site-packages (from torch->-r requirements.txt (line 9)) (81.0.0)\n",
|
| 90 |
+
"Requirement already satisfied: sympy>=1.13.3 in e:\\meta\\venv\\lib\\site-packages (from torch->-r requirements.txt (line 9)) (1.14.0)\n",
|
| 91 |
+
"Requirement already satisfied: networkx>=2.5.1 in e:\\meta\\venv\\lib\\site-packages (from torch->-r requirements.txt (line 9)) (3.6.1)\n",
|
| 92 |
+
"Requirement already satisfied: jinja2 in e:\\meta\\venv\\lib\\site-packages (from torch->-r requirements.txt (line 9)) (3.1.6)\n",
|
| 93 |
+
"Requirement already satisfied: pyarrow>=21.0.0 in e:\\meta\\venv\\lib\\site-packages (from datasets->-r requirements.txt (line 10)) (24.0.0)\n",
|
| 94 |
+
"Requirement already satisfied: dill<0.4.2,>=0.3.0 in e:\\meta\\venv\\lib\\site-packages (from datasets->-r requirements.txt (line 10)) (0.4.1)\n",
|
| 95 |
+
"Requirement already satisfied: requests>=2.32.2 in e:\\meta\\venv\\lib\\site-packages (from datasets->-r requirements.txt (line 10)) (2.33.1)\n",
|
| 96 |
+
"Requirement already satisfied: xxhash in e:\\meta\\venv\\lib\\site-packages (from datasets->-r requirements.txt (line 10)) (3.6.0)\n",
|
| 97 |
+
"Requirement already satisfied: multiprocess<0.70.20 in e:\\meta\\venv\\lib\\site-packages (from datasets->-r requirements.txt (line 10)) (0.70.19)\n",
|
| 98 |
+
"Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in e:\\meta\\venv\\lib\\site-packages (from fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (3.13.5)\n",
|
| 99 |
+
"Requirement already satisfied: accelerate>=1.4.0 in e:\\meta\\venv\\lib\\site-packages (from trl->-r requirements.txt (line 11)) (1.13.0)\n",
|
| 100 |
+
"Requirement already satisfied: psutil in e:\\meta\\venv\\lib\\site-packages (from accelerate>=1.4.0->trl->-r requirements.txt (line 11)) (7.2.2)\n",
|
| 101 |
+
"Requirement already satisfied: aiohappyeyeballs>=2.5.0 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (2.6.1)\n",
|
| 102 |
+
"Requirement already satisfied: aiosignal>=1.4.0 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (1.4.0)\n",
|
| 103 |
+
"Requirement already satisfied: attrs>=17.3.0 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (26.1.0)\n",
|
| 104 |
+
"Requirement already satisfied: frozenlist>=1.1.1 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (1.8.0)\n",
|
| 105 |
+
"Requirement already satisfied: multidict<7.0,>=4.5 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (6.7.1)\n",
|
| 106 |
+
"Requirement already satisfied: propcache>=0.2.0 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (0.4.1)\n",
|
| 107 |
+
"Requirement already satisfied: yarl<2.0,>=1.17.0 in e:\\meta\\venv\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets->-r requirements.txt (line 10)) (1.23.0)\n",
|
| 108 |
+
"Requirement already satisfied: colorama in e:\\meta\\venv\\lib\\site-packages (from click>=7.0->uvicorn>=0.23.0->uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (0.4.6)\n",
|
| 109 |
+
"Requirement already satisfied: six>=1.5 in e:\\meta\\venv\\lib\\site-packages (from python-dateutil>=2.8.2->pandas->-r requirements.txt (line 6)) (1.17.0)\n",
|
| 110 |
+
"Requirement already satisfied: charset_normalizer<4,>=2 in e:\\meta\\venv\\lib\\site-packages (from requests>=2.32.2->datasets->-r requirements.txt (line 10)) (3.4.7)\n",
|
| 111 |
+
"Requirement already satisfied: urllib3<3,>=1.26 in e:\\meta\\venv\\lib\\site-packages (from requests>=2.32.2->datasets->-r requirements.txt (line 10)) (2.6.3)\n",
|
| 112 |
+
"Requirement already satisfied: mpmath<1.4,>=1.1.0 in e:\\meta\\venv\\lib\\site-packages (from sympy>=1.13.3->torch->-r requirements.txt (line 9)) (1.3.0)\n",
|
| 113 |
+
"Requirement already satisfied: httptools>=0.6.3 in e:\\meta\\venv\\lib\\site-packages (from uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (0.7.1)\n",
|
| 114 |
+
"Requirement already satisfied: python-dotenv>=0.13 in e:\\meta\\venv\\lib\\site-packages (from uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (1.2.2)\n",
|
| 115 |
+
"Requirement already satisfied: watchfiles>=0.20 in e:\\meta\\venv\\lib\\site-packages (from uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (1.1.1)\n",
|
| 116 |
+
"Requirement already satisfied: websockets>=10.4 in e:\\meta\\venv\\lib\\site-packages (from uvicorn[standard]>=0.23.0->-r requirements.txt (line 2)) (16.0)\n",
|
| 117 |
+
"Requirement already satisfied: MarkupSafe>=2.0 in e:\\meta\\venv\\lib\\site-packages (from jinja2->torch->-r requirements.txt (line 9)) (3.0.3)\n",
|
| 118 |
+
"Requirement already satisfied: shellingham>=1.3.0 in e:\\meta\\venv\\lib\\site-packages (from typer->transformers->-r requirements.txt (line 8)) (1.5.4)\n",
|
| 119 |
+
"Requirement already satisfied: rich>=12.3.0 in e:\\meta\\venv\\lib\\site-packages (from typer->transformers->-r requirements.txt (line 8)) (15.0.0)\n",
|
| 120 |
+
"Requirement already satisfied: markdown-it-py>=2.2.0 in e:\\meta\\venv\\lib\\site-packages (from rich>=12.3.0->typer->transformers->-r requirements.txt (line 8)) (4.0.0)\n",
|
| 121 |
+
"Requirement already satisfied: pygments<3.0.0,>=2.13.0 in e:\\meta\\venv\\lib\\site-packages (from rich>=12.3.0->typer->transformers->-r requirements.txt (line 8)) (2.20.0)\n",
|
| 122 |
+
"Requirement already satisfied: mdurl~=0.1 in e:\\meta\\venv\\lib\\site-packages (from markdown-it-py>=2.2.0->rich>=12.3.0->typer->transformers->-r requirements.txt (line 8)) (0.1.2)\n"
|
| 123 |
+
]
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"name": "stderr",
|
| 127 |
+
"output_type": "stream",
|
| 128 |
+
"text": [
|
| 129 |
+
"\n",
|
| 130 |
+
"[notice] A new release of pip is available: 25.2 -> 26.0.1\n",
|
| 131 |
+
"[notice] To update, run: python.exe -m pip install --upgrade pip\n"
|
| 132 |
+
]
|
| 133 |
+
}
|
| 134 |
+
],
|
| 135 |
"source": [
|
| 136 |
+
"!pip install trl transformers datasets httpx fastapi uvicorn pydantic openai\n",
|
| 137 |
"!git clone https://github.com/havinashpatil/meta.git\n",
|
| 138 |
"!cd meta && pip install -r requirements.txt"
|
| 139 |
]
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"cell_type": "code",
|
| 143 |
+
"execution_count": 3,
|
| 144 |
"metadata": {},
|
| 145 |
+
"outputs": [
|
| 146 |
+
{
|
| 147 |
+
"ename": "RuntimeError",
|
| 148 |
+
"evalue": "Failed to import trl.trainer.grpo_trainer because of the following error (look up to see its traceback):\n'charmap' codec can't decode byte 0x81 in position 932: character maps to <undefined>",
|
| 149 |
+
"output_type": "error",
|
| 150 |
+
"traceback": [
|
| 151 |
+
"\u001b[31m---------------------------------------------------------------------------\u001b[39m",
|
| 152 |
+
"\u001b[31mUnicodeDecodeError\u001b[39m Traceback (most recent call last)",
|
| 153 |
+
"\u001b[36mFile \u001b[39m\u001b[32me:\\meta\\venv\\Lib\\site-packages\\trl\\_lazy_module.py:71\u001b[39m, in \u001b[36m_LazyModule._get_module\u001b[39m\u001b[34m(self, module_name)\u001b[39m\n\u001b[32m 70\u001b[39m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[32m---> \u001b[39m\u001b[32m71\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[30;43mimportlib\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mimport_module\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43m\"\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43m\"\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43m+\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mmodule_name\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mself\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43m__name__\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 72\u001b[39m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n",
|
| 154 |
+
"\u001b[36mFile \u001b[39m\u001b[32m~\\AppData\\Local\\Programs\\Python\\Python313\\Lib\\importlib\\__init__.py:88\u001b[39m, in \u001b[36mimport_module\u001b[39m\u001b[34m(name, package)\u001b[39m\n\u001b[32m 87\u001b[39m level += \u001b[32m1\u001b[39m\n\u001b[32m---> \u001b[39m\u001b[32m88\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[30;43m_bootstrap\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43m_gcd_import\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43mname\u001b[39;49m\u001b[30;43m[\u001b[39;49m\u001b[30;43mlevel\u001b[39;49m\u001b[30;43m:\u001b[39;49m\u001b[30;43m]\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mpackage\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mlevel\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n",
|
| 155 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap>:1387\u001b[39m, in \u001b[36m_gcd_import\u001b[39m\u001b[34m(name, package, level)\u001b[39m\n",
|
| 156 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap>:1360\u001b[39m, in \u001b[36m_find_and_load\u001b[39m\u001b[34m(name, import_)\u001b[39m\n",
|
| 157 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap>:1334\u001b[39m, in \u001b[36m_find_and_load_unlocked\u001b[39m\u001b[34m(name, import_)\u001b[39m\n",
|
| 158 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap>:950\u001b[39m, in \u001b[36m_load_unlocked\u001b[39m\u001b[34m(spec)\u001b[39m\n",
|
| 159 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap_external>:1026\u001b[39m, in \u001b[36m_LoaderBasics.exec_module\u001b[39m\u001b[34m(self, module)\u001b[39m\n",
|
| 160 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap>:488\u001b[39m, in \u001b[36m_call_with_frames_removed\u001b[39m\u001b[34m(f, *args, **kwds)\u001b[39m\n",
|
| 161 |
+
"\u001b[36mFile \u001b[39m\u001b[32me:\\meta\\venv\\Lib\\site-packages\\trl\\trainer\\grpo_trainer.py:59\u001b[39m\n\u001b[32m 57\u001b[39m \u001b[38;5;28;01mfrom\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mtransformers\u001b[39;00m\u001b[34;01m.\u001b[39;00m\u001b[34;01mutils\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mimport\u001b[39;00m is_peft_available, is_rich_available\n\u001b[32m---> \u001b[39m\u001b[32m59\u001b[39m \u001b[38;5;28;01mfrom\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01m.\u001b[39;00m\u001b[34;01m.\u001b[39;00m\u001b[34;01mchat_template_utils\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[32m 60\u001b[39m add_response_schema,\n\u001b[32m 61\u001b[39m get_training_chat_template,\n\u001b[32m 62\u001b[39m is_chat_template_prefix_preserving,\n\u001b[32m 63\u001b[39m parse_response,\n\u001b[32m 64\u001b[39m supports_tool_calling,\n\u001b[32m 65\u001b[39m )\n\u001b[32m 66\u001b[39m \u001b[38;5;28;01mfrom\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01m.\u001b[39;00m\u001b[34;01m.\u001b[39;00m\u001b[34;01mdata_utils\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mimport\u001b[39;00m apply_chat_template, is_conversational, prepare_multimodal_messages\n",
|
| 162 |
+
"\u001b[36mFile \u001b[39m\u001b[32me:\\meta\\venv\\Lib\\site-packages\\trl\\chat_template_utils.py:309\u001b[39m\n\u001b[32m 273\u001b[39m qwen3_5_schema = {\n\u001b[32m 274\u001b[39m \u001b[33m\"\u001b[39m\u001b[33mx-regex\u001b[39m\u001b[33m\"\u001b[39m: \u001b[33mr\u001b[39m\u001b[33m\"\u001b[39m\u001b[33m^(?:(?:<think>\u001b[39m\u001b[33m\\\u001b[39m\u001b[33mn?)?(?:(?P<reasoning_content>.*?\u001b[39m\u001b[33m\\\u001b[39m\u001b[33mS.*?)\u001b[39m\u001b[33m\\\u001b[39m\u001b[33mn?|[\u001b[39m\u001b[33m\\\u001b[39m\u001b[33ms]*)</think>\u001b[39m\u001b[33m\\\u001b[39m\u001b[33ms*)?(?P<content>.*?)(?:\u001b[39m\u001b[33m\\\u001b[39m\u001b[33mn+(?=<tool_call>))?(?=(?:<tool_call>|<\u001b[39m\u001b[33m\\\u001b[39m\u001b[33m|im_end\u001b[39m\u001b[33m\\\u001b[39m\u001b[33m|>|$))(?P<tool_calls>(?:<tool_call>.+?</tool_call>\u001b[39m\u001b[33m\\\u001b[39m\u001b[33ms*)+)?\u001b[39m\u001b[33m\\\u001b[39m\u001b[33ms*(?:<\u001b[39m\u001b[33m\\\u001b[39m\u001b[33m|im_end\u001b[39m\u001b[33m\\\u001b[39m\u001b[33m|>|$)\u001b[39m\u001b[33m\"\u001b[39m,\n\u001b[32m 275\u001b[39m \u001b[33m\"\u001b[39m\u001b[33mtype\u001b[39m\u001b[33m\"\u001b[39m: \u001b[33m\"\u001b[39m\u001b[33mobject\u001b[39m\u001b[33m\"\u001b[39m,\n\u001b[32m (...)\u001b[39m\u001b[32m 305\u001b[39m },\n\u001b[32m 306\u001b[39m }\n\u001b[32m--> \u001b[39m\u001b[32m309\u001b[39m deepseekv3_chat_template = \u001b[30;43m(\u001b[39;49m\u001b[30;43m_CHAT_TEMPLATES_DIR\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43m/\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43m\"\u001b[39;49m\u001b[30;43mdeepseekv3.jinja\u001b[39;49m\u001b[30;43m\"\u001b[39;49m\u001b[30;43m)\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mread_text\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 311\u001b[39m glm4moe_chat_template = (_CHAT_TEMPLATES_DIR / \u001b[33m\"\u001b[39m\u001b[33mglm4moe.jinja\u001b[39m\u001b[33m\"\u001b[39m).read_text()\n",
|
| 163 |
+
"\u001b[36mFile \u001b[39m\u001b[32m~\\AppData\\Local\\Programs\\Python\\Python313\\Lib\\pathlib\\_local.py:546\u001b[39m, in \u001b[36mPath.read_text\u001b[39m\u001b[34m(self, encoding, errors, newline)\u001b[39m\n\u001b[32m 545\u001b[39m encoding = io.text_encoding(encoding)\n\u001b[32m--> \u001b[39m\u001b[32m546\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[30;43mPathBase\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mread_text\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43mself\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mencoding\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43merrors\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mnewline\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n",
|
| 164 |
+
"\u001b[36mFile \u001b[39m\u001b[32m~\\AppData\\Local\\Programs\\Python\\Python313\\Lib\\pathlib\\_abc.py:633\u001b[39m, in \u001b[36mPathBase.read_text\u001b[39m\u001b[34m(self, encoding, errors, newline)\u001b[39m\n\u001b[32m 632\u001b[39m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m.open(mode=\u001b[33m'\u001b[39m\u001b[33mr\u001b[39m\u001b[33m'\u001b[39m, encoding=encoding, errors=errors, newline=newline) \u001b[38;5;28;01mas\u001b[39;00m f:\n\u001b[32m--> \u001b[39m\u001b[32m633\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[30;43mf\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mread\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n",
|
| 165 |
+
"\u001b[36mFile \u001b[39m\u001b[32m~\\AppData\\Local\\Programs\\Python\\Python313\\Lib\\encodings\\cp1252.py:23\u001b[39m, in \u001b[36mIncrementalDecoder.decode\u001b[39m\u001b[34m(self, input, final)\u001b[39m\n\u001b[32m 22\u001b[39m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34mdecode\u001b[39m(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;28minput\u001b[39m, final=\u001b[38;5;28;01mFalse\u001b[39;00m):\n\u001b[32m---> \u001b[39m\u001b[32m23\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[30;43mcodecs\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43mcharmap_decode\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43minput\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43mself\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43merrors\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43mdecoding_table\u001b[39;49m\u001b[30;43m)\u001b[39;49m[\u001b[32m0\u001b[39m]\n",
|
| 166 |
+
"\u001b[31mUnicodeDecodeError\u001b[39m: 'charmap' codec can't decode byte 0x81 in position 932: character maps to <undefined>",
|
| 167 |
+
"\nThe above exception was the direct cause of the following exception:\n",
|
| 168 |
+
"\u001b[31mRuntimeError\u001b[39m Traceback (most recent call last)",
|
| 169 |
+
"\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[3]\u001b[39m\u001b[32m, line 4\u001b[39m\n\u001b[32m 1\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m torch\n\u001b[32m 2\u001b[39m \u001b[38;5;28;01mfrom\u001b[39;00m datasets \u001b[38;5;28;01mimport\u001b[39;00m load_dataset\n\u001b[32m 3\u001b[39m \u001b[38;5;28;01mfrom\u001b[39;00m transformers \u001b[38;5;28;01mimport\u001b[39;00m AutoModelForCausalLM, AutoTokenizer\n\u001b[32m----> \u001b[39m\u001b[32m4\u001b[39m \u001b[38;5;28;01mfrom\u001b[39;00m trl \u001b[38;5;28;01mimport\u001b[39;00m GRPOConfig, GRPOTrainer\n\u001b[32m 5\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m httpx\n\u001b[32m 6\u001b[39m \n\u001b[32m 7\u001b[39m \u001b[38;5;66;03m# Start the backend server in the background (Colab trick)\u001b[39;00m\n",
|
| 170 |
+
"\u001b[36mFile \u001b[39m\u001b[32m<frozen importlib._bootstrap>:1412\u001b[39m, in \u001b[36m_handle_fromlist\u001b[39m\u001b[34m(module, fromlist, import_, recursive)\u001b[39m\n",
|
| 171 |
+
"\u001b[36mFile \u001b[39m\u001b[32me:\\meta\\venv\\Lib\\site-packages\\trl\\_lazy_module.py:62\u001b[39m, in \u001b[36m_LazyModule.__getattr__\u001b[39m\u001b[34m(self, name)\u001b[39m\n\u001b[32m 60\u001b[39m \u001b[38;5;28;01melif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m._class_to_module.keys():\n\u001b[32m 61\u001b[39m module = \u001b[38;5;28mself\u001b[39m._get_module(\u001b[38;5;28mself\u001b[39m._class_to_module[name])\n\u001b[32m---> \u001b[39m\u001b[32m62\u001b[39m value = \u001b[30;43mgetattr\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43mmodule\u001b[39;49m\u001b[30;43m,\u001b[39;49m\u001b[30;43m \u001b[39;49m\u001b[30;43mname\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 63\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[32m 64\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mAttributeError\u001b[39;00m(\u001b[33mf\u001b[39m\u001b[33m\"\u001b[39m\u001b[33mmodule \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m.\u001b[34m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[33m has no attribute \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mname\u001b[38;5;132;01m}\u001b[39;00m\u001b[33m\"\u001b[39m)\n",
|
| 172 |
+
"\u001b[36mFile \u001b[39m\u001b[32me:\\meta\\venv\\Lib\\site-packages\\trl\\_lazy_module.py:61\u001b[39m, in \u001b[36m_LazyModule.__getattr__\u001b[39m\u001b[34m(self, name)\u001b[39m\n\u001b[32m 59\u001b[39m value = \u001b[38;5;28mself\u001b[39m._get_module(name)\n\u001b[32m 60\u001b[39m \u001b[38;5;28;01melif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m._class_to_module.keys():\n\u001b[32m---> \u001b[39m\u001b[32m61\u001b[39m module = \u001b[30;43mself\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43m_get_module\u001b[39;49m\u001b[30;43m(\u001b[39;49m\u001b[30;43mself\u001b[39;49m\u001b[30;43m.\u001b[39;49m\u001b[30;43m_class_to_module\u001b[39;49m\u001b[30;43m[\u001b[39;49m\u001b[30;43mname\u001b[39;49m\u001b[30;43m]\u001b[39;49m\u001b[30;43m)\u001b[39;49m\n\u001b[32m 62\u001b[39m value = \u001b[38;5;28mgetattr\u001b[39m(module, name)\n\u001b[32m 63\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m:\n",
|
| 173 |
+
"\u001b[36mFile \u001b[39m\u001b[32me:\\meta\\venv\\Lib\\site-packages\\trl\\_lazy_module.py:73\u001b[39m, in \u001b[36m_LazyModule._get_module\u001b[39m\u001b[34m(self, module_name)\u001b[39m\n\u001b[32m 71\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m importlib.import_module(\u001b[33m\"\u001b[39m\u001b[33m.\u001b[39m\u001b[33m\"\u001b[39m + module_name, \u001b[38;5;28mself\u001b[39m.\u001b[34m__name__\u001b[39m)\n\u001b[32m 72\u001b[39m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[32m---> \u001b[39m\u001b[32m73\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mRuntimeError\u001b[39;00m(\n\u001b[32m 74\u001b[39m \u001b[33mf\u001b[39m\u001b[33m\"\u001b[39m\u001b[33mFailed to import \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m.\u001b[34m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[33m.\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mmodule_name\u001b[38;5;132;01m}\u001b[39;00m\u001b[33m because of the following error (look up to see its\u001b[39m\u001b[33m\"\u001b[39m\n\u001b[32m 75\u001b[39m \u001b[33mf\u001b[39m\u001b[33m\"\u001b[39m\u001b[33m traceback):\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;132;01m{\u001b[39;00me\u001b[38;5;132;01m}\u001b[39;00m\u001b[33m\"\u001b[39m\n\u001b[32m 76\u001b[39m ) \u001b[38;5;28;01mfrom\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01me\u001b[39;00m\n",
|
| 174 |
+
"\u001b[31mRuntimeError\u001b[39m: Failed to import trl.trainer.grpo_trainer because of the following error (look up to see its traceback):\n'charmap' codec can't decode byte 0x81 in position 932: character maps to <undefined>"
|
| 175 |
+
]
|
| 176 |
+
}
|
| 177 |
+
],
|
| 178 |
"source": [
|
| 179 |
"import torch\n",
|
| 180 |
"from datasets import load_dataset\n",
|
|
|
|
| 277 |
],
|
| 278 |
"metadata": {
|
| 279 |
"kernelspec": {
|
| 280 |
+
"display_name": "venv",
|
| 281 |
"language": "python",
|
| 282 |
"name": "python3"
|
| 283 |
},
|
|
|
|
| 291 |
"name": "python",
|
| 292 |
"nbconvert_exporter": "python",
|
| 293 |
"pygments_lexer": "ipython3",
|
| 294 |
+
"version": "3.13.6"
|
| 295 |
}
|
| 296 |
},
|
| 297 |
"nbformat": 4,
|
| 298 |
"nbformat_minor": 4
|
| 299 |
+
}
|