Hugging Face's logo

osiria
/
deberta-italian-question-answering

Question Answering
Transformers
PyTorch
Safetensors
Italian
deberta-v2
Eval Results
Inference Endpoints
Model card Files Community
1
osiria commited on
Commit
7f35c24
•
1 Parent(s): 2b164a0

Upload osiria_deberta_italian_qa_evaluation.ipynb

Browse files
Files changed (1) hide show
  1. osiria_deberta_italian_qa_evaluation.ipynb +890 -0
osiria_deberta_italian_qa_evaluation.ipynb ADDED
@@ -0,0 +1,890 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "nbformat": 4,
3
+ "nbformat_minor": 0,
4
+ "metadata": {
5
+ "colab": {
6
+ "provenance": [],
7
+ "gpuType": "T4"
8
+ },
9
+ "kernelspec": {
10
+ "name": "python3",
11
+ "display_name": "Python 3"
12
+ },
13
+ "language_info": {
14
+ "name": "python"
15
+ },
16
+ "accelerator": "GPU"
17
+ },
18
+ "cells": [
19
+ {
20
+ "cell_type": "markdown",
21
+ "source": [
22
+ "# osiria/deberta-italian-question-answering\n",
23
+ "## test set metrics"
24
+ ],
25
+ "metadata": {
26
+ "id": "oub2ir8ZDkrH"
27
+ }
28
+ },
29
+ {
30
+ "cell_type": "code",
31
+ "execution_count": 1,
32
+ "metadata": {
33
+ "colab": {
34
+ "base_uri": "https://localhost:8080/"
35
+ },
36
+ "id": "02K9xO6Gvto1",
37
+ "outputId": "29e1d40b-749c-42b4-e182-4b63ebcc9418"
38
+ },
39
+ "outputs": [
40
+ {
41
+ "output_type": "stream",
42
+ "name": "stdout",
43
+ "text": [
44
+ "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
45
+ "Collecting git+https://github.com/huggingface/transformers\n",
46
+ " Cloning https://github.com/huggingface/transformers to /tmp/pip-req-build-k2caxkht\n",
47
+ " Running command git clone --filter=blob:none --quiet https://github.com/huggingface/transformers /tmp/pip-req-build-k2caxkht\n",
48
+ " Resolved https://github.com/huggingface/transformers to commit f49a3453caa6fe606bb31c571423f72264152fce\n",
49
+ " Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
50
+ " Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
51
+ " Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
52
+ "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (3.12.0)\n",
53
+ "Collecting huggingface-hub<1.0,>=0.14.1 (from transformers==4.30.0.dev0)\n",
54
+ " Downloading huggingface_hub-0.15.1-py3-none-any.whl (236 kB)\n",
55
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m236.8/236.8 kB\u001b[0m \u001b[31m20.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
56
+ "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (1.22.4)\n",
57
+ "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (23.1)\n",
58
+ "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (6.0)\n",
59
+ "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (2022.10.31)\n",
60
+ "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (2.27.1)\n",
61
+ "Collecting tokenizers!=0.11.3,<0.14,>=0.11.1 (from transformers==4.30.0.dev0)\n",
62
+ " Downloading tokenizers-0.13.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.8 MB)\n",
63
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.8/7.8 MB\u001b[0m \u001b[31m127.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
64
+ "\u001b[?25hCollecting safetensors>=0.3.1 (from transformers==4.30.0.dev0)\n",
65
+ " Downloading safetensors-0.3.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n",
66
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m77.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
67
+ "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.10/dist-packages (from transformers==4.30.0.dev0) (4.65.0)\n",
68
+ "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0,>=0.14.1->transformers==4.30.0.dev0) (2023.4.0)\n",
69
+ "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0,>=0.14.1->transformers==4.30.0.dev0) (4.5.0)\n",
70
+ "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->transformers==4.30.0.dev0) (1.26.15)\n",
71
+ "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->transformers==4.30.0.dev0) (2022.12.7)\n",
72
+ "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->transformers==4.30.0.dev0) (2.0.12)\n",
73
+ "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->transformers==4.30.0.dev0) (3.4)\n",
74
+ "Building wheels for collected packages: transformers\n",
75
+ " Building wheel for transformers (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
76
+ " Created wheel for transformers: filename=transformers-4.30.0.dev0-py3-none-any.whl size=7138714 sha256=89a0efbe4dd8919f94cc3c42cefeba7534b015aae8caed503d8b9d9c5a1ef7dd\n",
77
+ " Stored in directory: /tmp/pip-ephem-wheel-cache-oc_3dlyv/wheels/c0/14/d6/6c9a5582d2ac191ec0a483be151a4495fe1eb2a6706ca49f1b\n",
78
+ "Successfully built transformers\n",
79
+ "Installing collected packages: tokenizers, safetensors, huggingface-hub, transformers\n",
80
+ "Successfully installed huggingface-hub-0.15.1 safetensors-0.3.1 tokenizers-0.13.3 transformers-4.30.0.dev0\n"
81
+ ]
82
+ }
83
+ ],
84
+ "source": [
85
+ "!pip install git+https://github.com/huggingface/transformers"
86
+ ]
87
+ },
88
+ {
89
+ "cell_type": "code",
90
+ "source": [
91
+ "!pip install datasets"
92
+ ],
93
+ "metadata": {
94
+ "colab": {
95
+ "base_uri": "https://localhost:8080/"
96
+ },
97
+ "id": "VKgybrUswa1q",
98
+ "outputId": "7953a664-d7ec-42f0-8de0-37034a8d77e7"
99
+ },
100
+ "execution_count": 2,
101
+ "outputs": [
102
+ {
103
+ "output_type": "stream",
104
+ "name": "stdout",
105
+ "text": [
106
+ "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
107
+ "Collecting datasets\n",
108
+ " Downloading datasets-2.12.0-py3-none-any.whl (474 kB)\n",
109
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m474.6/474.6 kB\u001b[0m \u001b[31m39.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
110
+ "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from datasets) (1.22.4)\n",
111
+ "Requirement already satisfied: pyarrow>=8.0.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (9.0.0)\n",
112
+ "Collecting dill<0.3.7,>=0.3.0 (from datasets)\n",
113
+ " Downloading dill-0.3.6-py3-none-any.whl (110 kB)\n",
114
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m110.5/110.5 kB\u001b[0m \u001b[31m14.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
115
+ "\u001b[?25hRequirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from datasets) (1.5.3)\n",
116
+ "Requirement already satisfied: requests>=2.19.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (2.27.1)\n",
117
+ "Requirement already satisfied: tqdm>=4.62.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (4.65.0)\n",
118
+ "Collecting xxhash (from datasets)\n",
119
+ " Downloading xxhash-3.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (212 kB)\n",
120
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m212.5/212.5 kB\u001b[0m \u001b[31m32.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
121
+ "\u001b[?25hCollecting multiprocess (from datasets)\n",
122
+ " Downloading multiprocess-0.70.14-py310-none-any.whl (134 kB)\n",
123
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m134.3/134.3 kB\u001b[0m \u001b[31m11.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
124
+ "\u001b[?25hRequirement already satisfied: fsspec[http]>=2021.11.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (2023.4.0)\n",
125
+ "Collecting aiohttp (from datasets)\n",
126
+ " Downloading aiohttp-3.8.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.0 MB)\n",
127
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.0/1.0 MB\u001b[0m \u001b[31m66.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
128
+ "\u001b[?25hRequirement already satisfied: huggingface-hub<1.0.0,>=0.11.0 in /usr/local/lib/python3.10/dist-packages (from datasets) (0.15.1)\n",
129
+ "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from datasets) (23.1)\n",
130
+ "Collecting responses<0.19 (from datasets)\n",
131
+ " Downloading responses-0.18.0-py3-none-any.whl (38 kB)\n",
132
+ "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from datasets) (6.0)\n",
133
+ "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (23.1.0)\n",
134
+ "Requirement already satisfied: charset-normalizer<4.0,>=2.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets) (2.0.12)\n",
135
+ "Collecting multidict<7.0,>=4.5 (from aiohttp->datasets)\n",
136
+ " Downloading multidict-6.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (114 kB)\n",
137
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m114.5/114.5 kB\u001b[0m \u001b[31m16.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
138
+ "\u001b[?25hCollecting async-timeout<5.0,>=4.0.0a3 (from aiohttp->datasets)\n",
139
+ " Downloading async_timeout-4.0.2-py3-none-any.whl (5.8 kB)\n",
140
+ "Collecting yarl<2.0,>=1.0 (from aiohttp->datasets)\n",
141
+ " Downloading yarl-1.9.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (268 kB)\n",
142
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m268.8/268.8 kB\u001b[0m \u001b[31m33.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
143
+ "\u001b[?25hCollecting frozenlist>=1.1.1 (from aiohttp->datasets)\n",
144
+ " Downloading frozenlist-1.3.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (149 kB)\n",
145
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m149.6/149.6 kB\u001b[0m \u001b[31m18.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
146
+ "\u001b[?25hCollecting aiosignal>=1.1.2 (from aiohttp->datasets)\n",
147
+ " Downloading aiosignal-1.3.1-py3-none-any.whl (7.6 kB)\n",
148
+ "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0.0,>=0.11.0->datasets) (3.12.0)\n",
149
+ "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0.0,>=0.11.0->datasets) (4.5.0)\n",
150
+ "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (1.26.15)\n",
151
+ "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (2022.12.7)\n",
152
+ "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->datasets) (3.4)\n",
153
+ "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets) (2.8.2)\n",
154
+ "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->datasets) (2022.7.1)\n",
155
+ "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->datasets) (1.16.0)\n",
156
+ "Installing collected packages: xxhash, multidict, frozenlist, dill, async-timeout, yarl, responses, multiprocess, aiosignal, aiohttp, datasets\n",
157
+ "Successfully installed aiohttp-3.8.4 aiosignal-1.3.1 async-timeout-4.0.2 datasets-2.12.0 dill-0.3.6 frozenlist-1.3.3 multidict-6.0.4 multiprocess-0.70.14 responses-0.18.0 xxhash-3.2.0 yarl-1.9.2\n"
158
+ ]
159
+ }
160
+ ]
161
+ },
162
+ {
163
+ "cell_type": "code",
164
+ "source": [
165
+ "!pip install accelerate"
166
+ ],
167
+ "metadata": {
168
+ "colab": {
169
+ "base_uri": "https://localhost:8080/"
170
+ },
171
+ "id": "xYEW0Hjqz2lo",
172
+ "outputId": "3aa2a2bc-9051-4021-8ae4-66f77b5d1ac1"
173
+ },
174
+ "execution_count": 3,
175
+ "outputs": [
176
+ {
177
+ "output_type": "stream",
178
+ "name": "stdout",
179
+ "text": [
180
+ "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
181
+ "Collecting accelerate\n",
182
+ " Downloading accelerate-0.19.0-py3-none-any.whl (219 kB)\n",
183
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m219.1/219.1 kB\u001b[0m \u001b[31m18.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
184
+ "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from accelerate) (1.22.4)\n",
185
+ "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from accelerate) (23.1)\n",
186
+ "Requirement already satisfied: psutil in /usr/local/lib/python3.10/dist-packages (from accelerate) (5.9.5)\n",
187
+ "Requirement already satisfied: pyyaml in /usr/local/lib/python3.10/dist-packages (from accelerate) (6.0)\n",
188
+ "Requirement already satisfied: torch>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from accelerate) (2.0.1+cu118)\n",
189
+ "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch>=1.6.0->accelerate) (3.12.0)\n",
190
+ "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch>=1.6.0->accelerate) (4.5.0)\n",
191
+ "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch>=1.6.0->accelerate) (1.11.1)\n",
192
+ "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.6.0->accelerate) (3.1)\n",
193
+ "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch>=1.6.0->accelerate) (3.1.2)\n",
194
+ "Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.6.0->accelerate) (2.0.0)\n",
195
+ "Requirement already satisfied: cmake in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.6.0->accelerate) (3.25.2)\n",
196
+ "Requirement already satisfied: lit in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch>=1.6.0->accelerate) (16.0.5)\n",
197
+ "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch>=1.6.0->accelerate) (2.1.2)\n",
198
+ "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch>=1.6.0->accelerate) (1.3.0)\n",
199
+ "Installing collected packages: accelerate\n",
200
+ "Successfully installed accelerate-0.19.0\n"
201
+ ]
202
+ }
203
+ ]
204
+ },
205
+ {
206
+ "cell_type": "code",
207
+ "source": [
208
+ "!pip install evaluate"
209
+ ],
210
+ "metadata": {
211
+ "colab": {
212
+ "base_uri": "https://localhost:8080/"
213
+ },
214
+ "id": "9OFz3JdR0cQ_",
215
+ "outputId": "d91b5794-3f05-4b34-dca7-0331cd94f5ec"
216
+ },
217
+ "execution_count": 4,
218
+ "outputs": [
219
+ {
220
+ "output_type": "stream",
221
+ "name": "stdout",
222
+ "text": [
223
+ "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
224
+ "Collecting evaluate\n",
225
+ " Downloading evaluate-0.4.0-py3-none-any.whl (81 kB)\n",
226
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m81.4/81.4 kB\u001b[0m \u001b[31m5.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
227
+ "\u001b[?25hRequirement already satisfied: datasets>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from evaluate) (2.12.0)\n",
228
+ "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from evaluate) (1.22.4)\n",
229
+ "Requirement already satisfied: dill in /usr/local/lib/python3.10/dist-packages (from evaluate) (0.3.6)\n",
230
+ "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from evaluate) (1.5.3)\n",
231
+ "Requirement already satisfied: requests>=2.19.0 in /usr/local/lib/python3.10/dist-packages (from evaluate) (2.27.1)\n",
232
+ "Requirement already satisfied: tqdm>=4.62.1 in /usr/local/lib/python3.10/dist-packages (from evaluate) (4.65.0)\n",
233
+ "Requirement already satisfied: xxhash in /usr/local/lib/python3.10/dist-packages (from evaluate) (3.2.0)\n",
234
+ "Requirement already satisfied: multiprocess in /usr/local/lib/python3.10/dist-packages (from evaluate) (0.70.14)\n",
235
+ "Requirement already satisfied: fsspec[http]>=2021.05.0 in /usr/local/lib/python3.10/dist-packages (from evaluate) (2023.4.0)\n",
236
+ "Requirement already satisfied: huggingface-hub>=0.7.0 in /usr/local/lib/python3.10/dist-packages (from evaluate) (0.15.1)\n",
237
+ "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from evaluate) (23.1)\n",
238
+ "Requirement already satisfied: responses<0.19 in /usr/local/lib/python3.10/dist-packages (from evaluate) (0.18.0)\n",
239
+ "Requirement already satisfied: pyarrow>=8.0.0 in /usr/local/lib/python3.10/dist-packages (from datasets>=2.0.0->evaluate) (9.0.0)\n",
240
+ "Requirement already satisfied: aiohttp in /usr/local/lib/python3.10/dist-packages (from datasets>=2.0.0->evaluate) (3.8.4)\n",
241
+ "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from datasets>=2.0.0->evaluate) (6.0)\n",
242
+ "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from huggingface-hub>=0.7.0->evaluate) (3.12.0)\n",
243
+ "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub>=0.7.0->evaluate) (4.5.0)\n",
244
+ "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->evaluate) (1.26.15)\n",
245
+ "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->evaluate) (2022.12.7)\n",
246
+ "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->evaluate) (2.0.12)\n",
247
+ "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests>=2.19.0->evaluate) (3.4)\n",
248
+ "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas->evaluate) (2.8.2)\n",
249
+ "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->evaluate) (2022.7.1)\n",
250
+ "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.0.0->evaluate) (23.1.0)\n",
251
+ "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.0.0->evaluate) (6.0.4)\n",
252
+ "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.0.0->evaluate) (4.0.2)\n",
253
+ "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.0.0->evaluate) (1.9.2)\n",
254
+ "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.0.0->evaluate) (1.3.3)\n",
255
+ "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp->datasets>=2.0.0->evaluate) (1.3.1)\n",
256
+ "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->evaluate) (1.16.0)\n",
257
+ "Installing collected packages: evaluate\n",
258
+ "Successfully installed evaluate-0.4.0\n"
259
+ ]
260
+ }
261
+ ]
262
+ },
263
+ {
264
+ "cell_type": "code",
265
+ "source": [
266
+ "!wget https://raw.githubusercontent.com/huggingface/transformers/main/examples/pytorch/question-answering/run_qa.py\n",
267
+ "!wget https://raw.githubusercontent.com/huggingface/transformers/main/examples/pytorch/question-answering/trainer_qa.py\n",
268
+ "!wget https://raw.githubusercontent.com/huggingface/transformers/main/examples/pytorch/question-answering/utils_qa.py"
269
+ ],
270
+ "metadata": {
271
+ "colab": {
272
+ "base_uri": "https://localhost:8080/"
273
+ },
274
+ "id": "JJhpBNwpxOTK",
275
+ "outputId": "4be12780-f34e-4834-8c10-9b3f4ee8a873"
276
+ },
277
+ "execution_count": 5,
278
+ "outputs": [
279
+ {
280
+ "output_type": "stream",
281
+ "name": "stdout",
282
+ "text": [
283
+ "--2023-06-01 22:15:29-- https://raw.githubusercontent.com/huggingface/transformers/main/examples/pytorch/question-answering/run_qa.py\n",
284
+ "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
285
+ "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
286
+ "HTTP request sent, awaiting response... 200 OK\n",
287
+ "Length: 31717 (31K) [text/plain]\n",
288
+ "Saving to: ‘run_qa.py’\n",
289
+ "\n",
290
+ "run_qa.py 100%[===================>] 30.97K --.-KB/s in 0s \n",
291
+ "\n",
292
+ "2023-06-01 22:15:29 (125 MB/s) - ‘run_qa.py’ saved [31717/31717]\n",
293
+ "\n",
294
+ "--2023-06-01 22:15:29-- https://raw.githubusercontent.com/huggingface/transformers/main/examples/pytorch/question-answering/trainer_qa.py\n",
295
+ "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
296
+ "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
297
+ "HTTP request sent, awaiting response... 200 OK\n",
298
+ "Length: 6253 (6.1K) [text/plain]\n",
299
+ "Saving to: ‘trainer_qa.py’\n",
300
+ "\n",
301
+ "trainer_qa.py 100%[===================>] 6.11K --.-KB/s in 0s \n",
302
+ "\n",
303
+ "2023-06-01 22:15:29 (72.9 MB/s) - ‘trainer_qa.py’ saved [6253/6253]\n",
304
+ "\n",
305
+ "--2023-06-01 22:15:29-- https://raw.githubusercontent.com/huggingface/transformers/main/examples/pytorch/question-answering/utils_qa.py\n",
306
+ "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
307
+ "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
308
+ "HTTP request sent, awaiting response... 200 OK\n",
309
+ "Length: 22777 (22K) [text/plain]\n",
310
+ "Saving to: ‘utils_qa.py’\n",
311
+ "\n",
312
+ "utils_qa.py 100%[===================>] 22.24K --.-KB/s in 0s \n",
313
+ "\n",
314
+ "2023-06-01 22:15:30 (110 MB/s) - ‘utils_qa.py’ saved [22777/22777]\n",
315
+ "\n"
316
+ ]
317
+ }
318
+ ]
319
+ },
320
+ {
321
+ "cell_type": "code",
322
+ "source": [
323
+ "!ls"
324
+ ],
325
+ "metadata": {
326
+ "colab": {
327
+ "base_uri": "https://localhost:8080/"
328
+ },
329
+ "id": "Loa6y9XqzXJ4",
330
+ "outputId": "9f4f930a-21f8-43d4-8c73-eca71c8ad4e7"
331
+ },
332
+ "execution_count": 6,
333
+ "outputs": [
334
+ {
335
+ "output_type": "stream",
336
+ "name": "stdout",
337
+ "text": [
338
+ "run_qa.py sample_data\ttrainer_qa.py utils_qa.py\n"
339
+ ]
340
+ }
341
+ ]
342
+ },
343
+ {
344
+ "cell_type": "code",
345
+ "source": [
346
+ "!mkdir eval_results"
347
+ ],
348
+ "metadata": {
349
+ "id": "QDNxV31P0reW"
350
+ },
351
+ "execution_count": 7,
352
+ "outputs": []
353
+ },
354
+ {
355
+ "cell_type": "code",
356
+ "source": [
357
+ "!python -m run_qa --model_name_or_path osiria/deberta-italian-question-answering --dataset_name squad_it --do_predict --per_device_eval_batch_size=5 --output_dir eval_results"
358
+ ],
359
+ "metadata": {
360
+ "colab": {
361
+ "base_uri": "https://localhost:8080/"
362
+ },
363
+ "id": "B5O5Lom8x_c4",
364
+ "outputId": "abcf118b-6a83-4bb2-db83-b715ac727688"
365
+ },
366
+ "execution_count": 8,
367
+ "outputs": [
368
+ {
369
+ "output_type": "stream",
370
+ "name": "stdout",
371
+ "text": [
372
+ "2023-06-01 22:15:37.369548: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
373
+ "06/01/2023 22:15:41 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False\n",
374
+ "06/01/2023 22:15:41 - INFO - __main__ - Training/evaluation parameters TrainingArguments(\n",
375
+ "_n_gpu=1,\n",
376
+ "adafactor=False,\n",
377
+ "adam_beta1=0.9,\n",
378
+ "adam_beta2=0.999,\n",
379
+ "adam_epsilon=1e-08,\n",
380
+ "auto_find_batch_size=False,\n",
381
+ "bf16=False,\n",
382
+ "bf16_full_eval=False,\n",
383
+ "data_seed=None,\n",
384
+ "dataloader_drop_last=False,\n",
385
+ "dataloader_num_workers=0,\n",
386
+ "dataloader_pin_memory=True,\n",
387
+ "ddp_backend=None,\n",
388
+ "ddp_bucket_cap_mb=None,\n",
389
+ "ddp_find_unused_parameters=None,\n",
390
+ "ddp_timeout=1800,\n",
391
+ "debug=[],\n",
392
+ "deepspeed=None,\n",
393
+ "disable_tqdm=False,\n",
394
+ "do_eval=False,\n",
395
+ "do_predict=True,\n",
396
+ "do_train=False,\n",
397
+ "eval_accumulation_steps=None,\n",
398
+ "eval_delay=0,\n",
399
+ "eval_steps=None,\n",
400
+ "evaluation_strategy=no,\n",
401
+ "fp16=False,\n",
402
+ "fp16_backend=auto,\n",
403
+ "fp16_full_eval=False,\n",
404
+ "fp16_opt_level=O1,\n",
405
+ "fsdp=[],\n",
406
+ "fsdp_config={'fsdp_min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},\n",
407
+ "fsdp_min_num_params=0,\n",
408
+ "fsdp_transformer_layer_cls_to_wrap=None,\n",
409
+ "full_determinism=False,\n",
410
+ "gradient_accumulation_steps=1,\n",
411
+ "gradient_checkpointing=False,\n",
412
+ "greater_is_better=None,\n",
413
+ "group_by_length=False,\n",
414
+ "half_precision_backend=auto,\n",
415
+ "hub_model_id=None,\n",
416
+ "hub_private_repo=False,\n",
417
+ "hub_strategy=every_save,\n",
418
+ "hub_token=<HUB_TOKEN>,\n",
419
+ "ignore_data_skip=False,\n",
420
+ "include_inputs_for_metrics=False,\n",
421
+ "jit_mode_eval=False,\n",
422
+ "label_names=None,\n",
423
+ "label_smoothing_factor=0.0,\n",
424
+ "learning_rate=5e-05,\n",
425
+ "length_column_name=length,\n",
426
+ "load_best_model_at_end=False,\n",
427
+ "local_rank=0,\n",
428
+ "log_level=passive,\n",
429
+ "log_level_replica=warning,\n",
430
+ "log_on_each_node=True,\n",
431
+ "logging_dir=eval_results/runs/Jun01_22-15-40_8f1ed0a6ad1b,\n",
432
+ "logging_first_step=False,\n",
433
+ "logging_nan_inf_filter=True,\n",
434
+ "logging_steps=500,\n",
435
+ "logging_strategy=steps,\n",
436
+ "lr_scheduler_type=linear,\n",
437
+ "max_grad_norm=1.0,\n",
438
+ "max_steps=-1,\n",
439
+ "metric_for_best_model=None,\n",
440
+ "mp_parameters=,\n",
441
+ "no_cuda=False,\n",
442
+ "num_train_epochs=3.0,\n",
443
+ "optim=adamw_hf,\n",
444
+ "optim_args=None,\n",
445
+ "output_dir=eval_results,\n",
446
+ "overwrite_output_dir=False,\n",
447
+ "past_index=-1,\n",
448
+ "per_device_eval_batch_size=5,\n",
449
+ "per_device_train_batch_size=8,\n",
450
+ "prediction_loss_only=False,\n",
451
+ "push_to_hub=False,\n",
452
+ "push_to_hub_model_id=None,\n",
453
+ "push_to_hub_organization=None,\n",
454
+ "push_to_hub_token=<PUSH_TO_HUB_TOKEN>,\n",
455
+ "ray_scope=last,\n",
456
+ "remove_unused_columns=True,\n",
457
+ "report_to=['tensorboard'],\n",
458
+ "resume_from_checkpoint=None,\n",
459
+ "run_name=eval_results,\n",
460
+ "save_on_each_node=False,\n",
461
+ "save_safetensors=False,\n",
462
+ "save_steps=500,\n",
463
+ "save_strategy=steps,\n",
464
+ "save_total_limit=None,\n",
465
+ "seed=42,\n",
466
+ "sharded_ddp=[],\n",
467
+ "skip_memory_metrics=True,\n",
468
+ "tf32=None,\n",
469
+ "torch_compile=False,\n",
470
+ "torch_compile_backend=None,\n",
471
+ "torch_compile_mode=None,\n",
472
+ "torchdynamo=None,\n",
473
+ "tpu_metrics_debug=False,\n",
474
+ "tpu_num_cores=None,\n",
475
+ "use_ipex=False,\n",
476
+ "use_legacy_prediction_loop=False,\n",
477
+ "use_mps_device=False,\n",
478
+ "warmup_ratio=0.0,\n",
479
+ "warmup_steps=0,\n",
480
+ "weight_decay=0.0,\n",
481
+ "xpu_backend=None,\n",
482
+ ")\n",
483
+ "06/01/2023 22:15:42 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/squad_it/resolve/main/squad_it.py not found in cache or force_download set to True, downloading to /root/.cache/huggingface/datasets/downloads/tmpofsetih3\n",
484
+ "Downloading builder script: 100% 4.74k/4.74k [00:00<00:00, 4.58MB/s]\n",
485
+ "06/01/2023 22:15:42 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/squad_it/resolve/main/squad_it.py in cache at /root/.cache/huggingface/datasets/downloads/2f7a246f661e6acb3a9ad072ed14241ea343265e7d863b88717897941c3ab1dd.671c3c3daf2f05e0355b498b0f5a7f2167fa94dc6de36195fe1ba0503e6dcc30.py\n",
486
+ "06/01/2023 22:15:42 - INFO - datasets.utils.file_utils - creating metadata file for /root/.cache/huggingface/datasets/downloads/2f7a246f661e6acb3a9ad072ed14241ea343265e7d863b88717897941c3ab1dd.671c3c3daf2f05e0355b498b0f5a7f2167fa94dc6de36195fe1ba0503e6dcc30.py\n",
487
+ "06/01/2023 22:15:43 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/squad_it/resolve/main/dataset_infos.json not found in cache or force_download set to True, downloading to /root/.cache/huggingface/datasets/downloads/tmpoickid2o\n",
488
+ "Downloading metadata: 100% 2.51k/2.51k [00:00<00:00, 2.36MB/s]\n",
489
+ "06/01/2023 22:15:43 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/squad_it/resolve/main/dataset_infos.json in cache at /root/.cache/huggingface/datasets/downloads/f6d7650a1e4a9f0e42b54f34ac614f7c1142ad26c1bd19a448eb606fe09f7515.b459489a7b1a4d859decbeef851034c7846ef6cce8ec2cf192cf4b4f40d0865d\n",
490
+ "06/01/2023 22:15:43 - INFO - datasets.utils.file_utils - creating metadata file for /root/.cache/huggingface/datasets/downloads/f6d7650a1e4a9f0e42b54f34ac614f7c1142ad26c1bd19a448eb606fe09f7515.b459489a7b1a4d859decbeef851034c7846ef6cce8ec2cf192cf4b4f40d0865d\n",
491
+ "06/01/2023 22:15:43 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/squad_it/resolve/main/README.md not found in cache or force_download set to True, downloading to /root/.cache/huggingface/datasets/downloads/tmp99adhx5x\n",
492
+ "Downloading readme: 100% 7.27k/7.27k [00:00<00:00, 6.54MB/s]\n",
493
+ "06/01/2023 22:15:43 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/squad_it/resolve/main/README.md in cache at /root/.cache/huggingface/datasets/downloads/b664892e5965a1ef2e173eadc7e9cbf6ad4aeb6240d7b3bb6ab9f32850363a6f.71fe7dbf5f3b574d85b57b5a722ebc49fe8ec8b03d655344891fb1b05c4634fb\n",
494
+ "06/01/2023 22:15:43 - INFO - datasets.utils.file_utils - creating metadata file for /root/.cache/huggingface/datasets/downloads/b664892e5965a1ef2e173eadc7e9cbf6ad4aeb6240d7b3bb6ab9f32850363a6f.71fe7dbf5f3b574d85b57b5a722ebc49fe8ec8b03d655344891fb1b05c4634fb\n",
495
+ "06/01/2023 22:15:43 - INFO - datasets.info - Loading Dataset Infos from /root/.cache/huggingface/modules/datasets_modules/datasets/squad_it/d442bdb4794b4bae227ab19105b76d706ed7cf2ac342e4c9da4a5c36bde19d71\n",
496
+ "06/01/2023 22:15:43 - INFO - datasets.builder - Generating dataset squad_it (/root/.cache/huggingface/datasets/squad_it/default/0.1.0/d442bdb4794b4bae227ab19105b76d706ed7cf2ac342e4c9da4a5c36bde19d71)\n",
497
+ "Downloading and preparing dataset squad_it/default to /root/.cache/huggingface/datasets/squad_it/default/0.1.0/d442bdb4794b4bae227ab19105b76d706ed7cf2ac342e4c9da4a5c36bde19d71...\n",
498
+ "06/01/2023 22:15:45 - INFO - datasets.builder - Dataset not on Hf google storage. Downloading and preparing it from source\n",
499
+ "Downloading data files: 0% 0/2 [00:00<?, ?it/s]06/01/2023 22:15:46 - INFO - datasets.utils.file_utils - https://github.com/crux82/squad-it/raw/master/SQuAD_it-train.json.gz not found in cache or force_download set to True, downloading to /root/.cache/huggingface/datasets/downloads/tmpj9esiuss\n",
500
+ "\n",
501
+ "Downloading data: 0% 0.00/7.73M [00:00<?, ?B/s]\u001b[A\n",
502
+ "Downloading data: 100% 7.73M/7.73M [00:00<00:00, 75.5MB/s]\n",
503
+ "06/01/2023 22:15:46 - INFO - datasets.utils.file_utils - storing https://github.com/crux82/squad-it/raw/master/SQuAD_it-train.json.gz in cache at /root/.cache/huggingface/datasets/downloads/44ac0548c3284061d1356ff7f5212fe42b923b0aa0854c886ce93dfeb7ab629b\n",
504
+ "06/01/2023 22:15:46 - INFO - datasets.utils.file_utils - creating metadata file for /root/.cache/huggingface/datasets/downloads/44ac0548c3284061d1356ff7f5212fe42b923b0aa0854c886ce93dfeb7ab629b\n",
505
+ "Downloading data files: 50% 1/2 [00:01<00:01, 1.35s/it]06/01/2023 22:15:47 - INFO - datasets.utils.file_utils - https://github.com/crux82/squad-it/raw/master/SQuAD_it-test.json.gz not found in cache or force_download set to True, downloading to /root/.cache/huggingface/datasets/downloads/tmp0dopuyxr\n",
506
+ "\n",
507
+ "Downloading data: 100% 1.05M/1.05M [00:00<00:00, 57.2MB/s]\n",
508
+ "06/01/2023 22:15:47 - INFO - datasets.utils.file_utils - storing https://github.com/crux82/squad-it/raw/master/SQuAD_it-test.json.gz in cache at /root/.cache/huggingface/datasets/downloads/e560a4a54eb929ce542e945627942afca80859dcbf499866a59984eb9026d31f\n",
509
+ "06/01/2023 22:15:47 - INFO - datasets.utils.file_utils - creating metadata file for /root/.cache/huggingface/datasets/downloads/e560a4a54eb929ce542e945627942afca80859dcbf499866a59984eb9026d31f\n",
510
+ "Downloading data files: 100% 2/2 [00:02<00:00, 1.06s/it]\n",
511
+ "06/01/2023 22:15:47 - INFO - datasets.download.download_manager - Downloading took 0.0 min\n",
512
+ "06/01/2023 22:15:47 - INFO - datasets.download.download_manager - Checksum Computation took 0.0 min\n",
513
+ "Extracting data files: 100% 2/2 [00:00<00:00, 8.00it/s]\n",
514
+ "06/01/2023 22:15:47 - INFO - datasets.builder - Generating train split\n",
515
+ "06/01/2023 22:15:52 - INFO - datasets.builder - Generating test split\n",
516
+ "06/01/2023 22:15:53 - INFO - datasets.utils.info_utils - All the splits matched successfully.\n",
517
+ "Dataset squad_it downloaded and prepared to /root/.cache/huggingface/datasets/squad_it/default/0.1.0/d442bdb4794b4bae227ab19105b76d706ed7cf2ac342e4c9da4a5c36bde19d71. Subsequent calls will reuse this data.\n",
518
+ "100% 2/2 [00:00<00:00, 177.35it/s]\n",
519
+ "Downloading (…)lve/main/config.json: 100% 826/826 [00:00<00:00, 3.77MB/s]\n",
520
+ "[INFO|configuration_utils.py:669] 2023-06-01 22:15:54,044 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--osiria--deberta-italian-question-answering/snapshots/2b164a0777ece87f9ea321432e558423ca862eea/config.json\n",
521
+ "[INFO|configuration_utils.py:725] 2023-06-01 22:15:54,050 >> Model config DebertaV2Config {\n",
522
+ " \"_name_or_path\": \"osiria/deberta-italian-question-answering\",\n",
523
+ " \"architectures\": [\n",
524
+ " \"DebertaV2ForQuestionAnswering\"\n",
525
+ " ],\n",
526
+ " \"attention_probs_dropout_prob\": 0.1,\n",
527
+ " \"hidden_act\": \"gelu\",\n",
528
+ " \"hidden_dropout_prob\": 0.1,\n",
529
+ " \"hidden_size\": 768,\n",
530
+ " \"initializer_range\": 0.02,\n",
531
+ " \"intermediate_size\": 3072,\n",
532
+ " \"layer_norm_eps\": 1e-07,\n",
533
+ " \"max_position_embeddings\": 512,\n",
534
+ " \"max_relative_positions\": -1,\n",
535
+ " \"model_type\": \"deberta-v2\",\n",
536
+ " \"norm_rel_ebd\": \"layer_norm\",\n",
537
+ " \"num_attention_heads\": 12,\n",
538
+ " \"num_hidden_layers\": 12,\n",
539
+ " \"pad_token_id\": 0,\n",
540
+ " \"pooler_dropout\": 0,\n",
541
+ " \"pooler_hidden_act\": \"gelu\",\n",
542
+ " \"pooler_hidden_size\": 768,\n",
543
+ " \"pos_att_type\": [\n",
544
+ " \"p2c\",\n",
545
+ " \"c2p\"\n",
546
+ " ],\n",
547
+ " \"position_biased_input\": false,\n",
548
+ " \"position_buckets\": 256,\n",
549
+ " \"relative_attention\": true,\n",
550
+ " \"share_att_key\": true,\n",
551
+ " \"torch_dtype\": \"float32\",\n",
552
+ " \"transformers_version\": \"4.30.0.dev0\",\n",
553
+ " \"type_vocab_size\": 0,\n",
554
+ " \"vocab_size\": 50256\n",
555
+ "}\n",
556
+ "\n",
557
+ "Downloading (…)okenizer_config.json: 100% 412/412 [00:00<00:00, 2.28MB/s]\n",
558
+ "Downloading (…)/main/tokenizer.json: 100% 2.02M/2.02M [00:01<00:00, 1.87MB/s]\n",
559
+ "[INFO|tokenization_utils_base.py:1818] 2023-06-01 22:15:57,482 >> loading file spm.model from cache at None\n",
560
+ "[INFO|tokenization_utils_base.py:1818] 2023-06-01 22:15:57,482 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--osiria--deberta-italian-question-answering/snapshots/2b164a0777ece87f9ea321432e558423ca862eea/tokenizer.json\n",
561
+ "[INFO|tokenization_utils_base.py:1818] 2023-06-01 22:15:57,482 >> loading file added_tokens.json from cache at None\n",
562
+ "[INFO|tokenization_utils_base.py:1818] 2023-06-01 22:15:57,482 >> loading file special_tokens_map.json from cache at None\n",
563
+ "[INFO|tokenization_utils_base.py:1818] 2023-06-01 22:15:57,482 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--osiria--deberta-italian-question-answering/snapshots/2b164a0777ece87f9ea321432e558423ca862eea/tokenizer_config.json\n",
564
+ "Downloading pytorch_model.bin: 100% 496M/496M [00:28<00:00, 17.5MB/s]\n",
565
+ "[INFO|modeling_utils.py:2574] 2023-06-01 22:16:27,690 >> loading weights file pytorch_model.bin from cache at /root/.cache/huggingface/hub/models--osiria--deberta-italian-question-answering/snapshots/2b164a0777ece87f9ea321432e558423ca862eea/pytorch_model.bin\n",
566
+ "[INFO|modeling_utils.py:3283] 2023-06-01 22:16:29,065 >> All model checkpoint weights were used when initializing DebertaV2ForQuestionAnswering.\n",
567
+ "\n",
568
+ "[INFO|modeling_utils.py:3291] 2023-06-01 22:16:29,065 >> All the weights of DebertaV2ForQuestionAnswering were initialized from the model checkpoint at osiria/deberta-italian-question-answering.\n",
569
+ "If your task is similar to the task the model of the checkpoint was trained on, you can already use DebertaV2ForQuestionAnswering for predictions without further training.\n",
570
+ "Running tokenizer on prediction dataset: 0% 0/7609 [00:00<?, ? examples/s]06/01/2023 22:16:29 - INFO - datasets.arrow_dataset - Caching processed dataset at /root/.cache/huggingface/datasets/squad_it/default/0.1.0/d442bdb4794b4bae227ab19105b76d706ed7cf2ac342e4c9da4a5c36bde19d71/cache-c70dd3f912ae21c4.arrow\n",
571
+ "Downloading builder script: 100% 4.53k/4.53k [00:00<00:00, 3.39MB/s]\n",
572
+ "Downloading extra modules: 100% 3.32k/3.32k [00:00<00:00, 3.26MB/s]\n",
573
+ "06/01/2023 22:16:44 - INFO - __main__ - *** Predict ***\n",
574
+ "[INFO|trainer.py:793] 2023-06-01 22:16:44,390 >> The following columns in the test set don't have a corresponding argument in `DebertaV2ForQuestionAnswering.forward` and have been ignored: example_id, offset_mapping. If example_id, offset_mapping are not expected by `DebertaV2ForQuestionAnswering.forward`, you can safely ignore this message.\n",
575
+ "[INFO|trainer.py:3175] 2023-06-01 22:16:44,392 >> ***** Running Prediction *****\n",
576
+ "[INFO|trainer.py:3177] 2023-06-01 22:16:44,392 >> Num examples = 8565\n",
577
+ "[INFO|trainer.py:3180] 2023-06-01 22:16:44,392 >> Batch size = 5\n",
578
+ "100% 1713/1713 [04:43<00:00, 6.06it/s]06/01/2023 22:21:41 - INFO - utils_qa - Post-processing 7609 example predictions split into 8565 features.\n",
579
+ "\n",
580
+ " 0% 0/7609 [00:00<?, ?it/s]\u001b[A\n",
581
+ " 0% 17/7609 [00:00<00:47, 159.40it/s]\u001b[A\n",
582
+ " 0% 33/7609 [00:00<00:47, 158.19it/s]\u001b[A\n",
583
+ " 1% 53/7609 [00:00<00:42, 176.81it/s]\u001b[A\n",
584
+ " 1% 71/7609 [00:00<00:45, 165.29it/s]\u001b[A\n",
585
+ " 1% 90/7609 [00:00<00:43, 171.27it/s]\u001b[A\n",
586
+ " 1% 109/7609 [00:00<00:42, 175.74it/s]\u001b[A\n",
587
+ " 2% 127/7609 [00:00<00:43, 173.80it/s]\u001b[A\n",
588
+ " 2% 146/7609 [00:00<00:42, 177.15it/s]\u001b[A\n",
589
+ " 2% 165/7609 [00:00<00:41, 179.53it/s]\u001b[A\n",
590
+ " 2% 184/7609 [00:01<00:40, 181.50it/s]\u001b[A\n",
591
+ " 3% 203/7609 [00:01<00:40, 180.75it/s]\u001b[A\n",
592
+ " 3% 224/7609 [00:01<00:39, 186.81it/s]\u001b[A\n",
593
+ " 3% 243/7609 [00:01<00:40, 183.61it/s]\u001b[A\n",
594
+ " 3% 262/7609 [00:01<00:39, 184.42it/s]\u001b[A\n",
595
+ " 4% 281/7609 [00:01<00:39, 184.43it/s]\u001b[A\n",
596
+ " 4% 300/7609 [00:01<00:41, 174.25it/s]\u001b[A\n",
597
+ " 4% 318/7609 [00:01<00:44, 165.66it/s]\u001b[A\n",
598
+ " 4% 341/7609 [00:01<00:39, 181.94it/s]\u001b[A\n",
599
+ " 5% 368/7609 [00:02<00:35, 205.82it/s]\u001b[A\n",
600
+ " 5% 402/7609 [00:02<00:29, 242.50it/s]\u001b[A\n",
601
+ " 6% 431/7609 [00:02<00:28, 254.96it/s]\u001b[A\n",
602
+ " 6% 464/7609 [00:02<00:25, 275.38it/s]\u001b[A\n",
603
+ " 7% 497/7609 [00:02<00:24, 288.13it/s]\u001b[A\n",
604
+ " 7% 526/7609 [00:02<00:27, 255.52it/s]\u001b[A\n",
605
+ " 7% 556/7609 [00:02<00:26, 266.39it/s]\u001b[A\n",
606
+ " 8% 584/7609 [00:02<00:27, 260.00it/s]\u001b[A\n",
607
+ " 8% 616/7609 [00:02<00:25, 274.49it/s]\u001b[A\n",
608
+ " 9% 651/7609 [00:02<00:23, 294.48it/s]\u001b[A\n",
609
+ " 9% 685/7609 [00:03<00:22, 306.17it/s]\u001b[A\n",
610
+ " 9% 716/7609 [00:03<00:22, 306.03it/s]\u001b[A\n",
611
+ " 10% 748/7609 [00:03<00:22, 308.88it/s]\u001b[A\n",
612
+ " 10% 781/7609 [00:03<00:22, 309.86it/s]\u001b[A\n",
613
+ " 11% 813/7609 [00:03<00:23, 288.11it/s]\u001b[A\n",
614
+ " 11% 843/7609 [00:03<00:23, 285.63it/s]\u001b[A\n",
615
+ " 11% 875/7609 [00:03<00:23, 292.71it/s]\u001b[A\n",
616
+ " 12% 908/7609 [00:03<00:22, 302.47it/s]\u001b[A\n",
617
+ " 12% 940/7609 [00:03<00:21, 305.39it/s]\u001b[A\n",
618
+ " 13% 971/7609 [00:04<00:21, 306.39it/s]\u001b[A\n",
619
+ " 13% 1003/7609 [00:04<00:21, 308.09it/s]\u001b[A\n",
620
+ " 14% 1036/7609 [00:04<00:20, 313.79it/s]\u001b[A\n",
621
+ " 14% 1068/7609 [00:04<00:20, 314.74it/s]\u001b[A\n",
622
+ " 14% 1100/7609 [00:04<00:22, 288.69it/s]\u001b[A\n",
623
+ " 15% 1130/7609 [00:04<00:22, 283.51it/s]\u001b[A\n",
624
+ " 15% 1159/7609 [00:04<00:23, 275.71it/s]\u001b[A\n",
625
+ " 16% 1187/7609 [00:04<00:23, 270.13it/s]\u001b[A\n",
626
+ " 16% 1219/7609 [00:04<00:22, 284.03it/s]\u001b[A\n",
627
+ " 16% 1249/7609 [00:05<00:22, 288.00it/s]\u001b[A\n",
628
+ " 17% 1282/7609 [00:05<00:21, 299.12it/s]\u001b[A\n",
629
+ " 17% 1313/7609 [00:05<00:20, 301.98it/s]\u001b[A\n",
630
+ " 18% 1344/7609 [00:05<00:20, 300.23it/s]\u001b[A\n",
631
+ " 18% 1375/7609 [00:05<00:20, 298.52it/s]\u001b[A\n",
632
+ " 18% 1405/7609 [00:05<00:24, 250.07it/s]\u001b[A\n",
633
+ " 19% 1434/7609 [00:05<00:23, 258.63it/s]\u001b[A\n",
634
+ " 19% 1461/7609 [00:05<00:24, 254.66it/s]\u001b[A\n",
635
+ " 20% 1488/7609 [00:05<00:25, 238.54it/s]\u001b[A\n",
636
+ " 20% 1517/7609 [00:06<00:24, 248.67it/s]\u001b[A\n",
637
+ " 20% 1543/7609 [00:06<00:24, 249.56it/s]\u001b[A\n",
638
+ " 21% 1575/7609 [00:06<00:22, 267.06it/s]\u001b[A\n",
639
+ " 21% 1603/7609 [00:06<00:23, 252.92it/s]\u001b[A\n",
640
+ " 21% 1629/7609 [00:06<00:23, 253.28it/s]\u001b[A\n",
641
+ " 22% 1655/7609 [00:06<00:25, 237.38it/s]\u001b[A\n",
642
+ " 22% 1683/7609 [00:06<00:23, 246.97it/s]\u001b[A\n",
643
+ " 22% 1709/7609 [00:06<00:23, 248.09it/s]\u001b[A\n",
644
+ " 23% 1741/7609 [00:06<00:22, 266.58it/s]\u001b[A\n",
645
+ " 23% 1768/7609 [00:07<00:23, 253.74it/s]\u001b[A\n",
646
+ " 24% 1794/7609 [00:07<00:23, 251.59it/s]\u001b[A\n",
647
+ " 24% 1823/7609 [00:07<00:22, 255.77it/s]\u001b[A\n",
648
+ " 24% 1849/7609 [00:07<00:22, 253.87it/s]\u001b[A\n",
649
+ " 25% 1879/7609 [00:07<00:21, 262.85it/s]\u001b[A\n",
650
+ " 25% 1906/7609 [00:07<00:22, 251.41it/s]\u001b[A\n",
651
+ " 25% 1932/7609 [00:07<00:22, 253.44it/s]\u001b[A\n",
652
+ " 26% 1958/7609 [00:07<00:23, 243.18it/s]\u001b[A\n",
653
+ " 26% 1983/7609 [00:07<00:25, 217.06it/s]\u001b[A\n",
654
+ " 26% 2006/7609 [00:08<00:28, 195.02it/s]\u001b[A\n",
655
+ " 27% 2027/7609 [00:08<00:33, 168.09it/s]\u001b[A\n",
656
+ " 27% 2049/7609 [00:08<00:30, 179.97it/s]\u001b[A\n",
657
+ " 27% 2069/7609 [00:08<00:31, 175.25it/s]\u001b[A\n",
658
+ " 27% 2088/7609 [00:08<00:41, 132.97it/s]\u001b[A\n",
659
+ " 28% 2104/7609 [00:08<00:44, 122.73it/s]\u001b[A\n",
660
+ " 28% 2136/7609 [00:08<00:33, 164.02it/s]\u001b[A\n",
661
+ " 29% 2170/7609 [00:09<00:26, 203.87it/s]\u001b[A\n",
662
+ " 29% 2194/7609 [00:09<00:25, 211.26it/s]\u001b[A\n",
663
+ " 29% 2218/7609 [00:09<00:24, 218.07it/s]\u001b[A\n",
664
+ " 29% 2244/7609 [00:09<00:23, 228.15it/s]\u001b[A\n",
665
+ " 30% 2270/7609 [00:09<00:22, 235.60it/s]\u001b[A\n",
666
+ " 30% 2301/7609 [00:09<00:20, 254.68it/s]\u001b[A\n",
667
+ " 31% 2329/7609 [00:09<00:20, 260.56it/s]\u001b[A\n",
668
+ " 31% 2363/7609 [00:09<00:18, 281.33it/s]\u001b[A\n",
669
+ " 31% 2395/7609 [00:09<00:17, 291.27it/s]\u001b[A\n",
670
+ " 32% 2429/7609 [00:10<00:17, 303.90it/s]\u001b[A\n",
671
+ " 32% 2460/7609 [00:10<00:17, 289.95it/s]\u001b[A\n",
672
+ " 33% 2490/7609 [00:10<00:19, 267.73it/s]\u001b[A\n",
673
+ " 33% 2523/7609 [00:10<00:17, 283.94it/s]\u001b[A\n",
674
+ " 34% 2553/7609 [00:10<00:17, 286.19it/s]\u001b[A\n",
675
+ " 34% 2582/7609 [00:10<00:17, 281.98it/s]\u001b[A\n",
676
+ " 34% 2611/7609 [00:10<00:18, 265.35it/s]\u001b[A\n",
677
+ " 35% 2639/7609 [00:10<00:18, 268.49it/s]\u001b[A\n",
678
+ " 35% 2671/7609 [00:10<00:17, 280.79it/s]\u001b[A\n",
679
+ " 36% 2702/7609 [00:11<00:17, 288.04it/s]\u001b[A\n",
680
+ " 36% 2732/7609 [00:11<00:16, 290.66it/s]\u001b[A\n",
681
+ " 36% 2762/7609 [00:11<00:17, 284.93it/s]\u001b[A\n",
682
+ " 37% 2795/7609 [00:11<00:16, 297.24it/s]\u001b[A\n",
683
+ " 37% 2825/7609 [00:11<00:16, 297.64it/s]\u001b[A\n",
684
+ " 38% 2855/7609 [00:11<00:16, 289.95it/s]\u001b[A\n",
685
+ " 38% 2886/7609 [00:11<00:16, 292.29it/s]\u001b[A\n",
686
+ " 38% 2916/7609 [00:11<00:15, 293.43it/s]\u001b[A\n",
687
+ " 39% 2947/7609 [00:11<00:15, 296.51it/s]\u001b[A\n",
688
+ " 39% 2977/7609 [00:11<00:16, 285.57it/s]\u001b[A\n",
689
+ " 40% 3006/7609 [00:12<00:20, 225.91it/s]\u001b[A\n",
690
+ " 40% 3031/7609 [00:12<00:23, 196.79it/s]\u001b[A\n",
691
+ " 40% 3053/7609 [00:12<00:23, 192.50it/s]\u001b[A\n",
692
+ " 40% 3074/7609 [00:12<00:24, 186.74it/s]\u001b[A\n",
693
+ " 41% 3094/7609 [00:12<00:26, 169.89it/s]\u001b[A\n",
694
+ " 41% 3112/7609 [00:12<00:27, 162.10it/s]\u001b[A\n",
695
+ " 41% 3129/7609 [00:12<00:28, 156.46it/s]\u001b[A\n",
696
+ " 41% 3146/7609 [00:13<00:28, 158.18it/s]\u001b[A\n",
697
+ " 42% 3164/7609 [00:13<00:27, 163.03it/s]\u001b[A\n",
698
+ " 42% 3181/7609 [00:13<00:29, 148.02it/s]\u001b[A\n",
699
+ " 42% 3197/7609 [00:13<00:31, 138.03it/s]\u001b[A\n",
700
+ " 42% 3213/7609 [00:13<00:30, 142.24it/s]\u001b[A\n",
701
+ " 42% 3228/7609 [00:13<00:32, 135.79it/s]\u001b[A\n",
702
+ " 43% 3247/7609 [00:13<00:29, 148.69it/s]\u001b[A\n",
703
+ " 43% 3263/7609 [00:13<00:28, 150.77it/s]\u001b[A\n",
704
+ " 43% 3279/7609 [00:13<00:28, 150.85it/s]\u001b[A\n",
705
+ " 43% 3298/7609 [00:14<00:27, 159.49it/s]\u001b[A\n",
706
+ " 44% 3316/7609 [00:14<00:26, 163.96it/s]\u001b[A\n",
707
+ " 44% 3333/7609 [00:14<00:28, 148.56it/s]\u001b[A\n",
708
+ " 44% 3349/7609 [00:14<00:28, 151.24it/s]\u001b[A\n",
709
+ " 44% 3368/7609 [00:14<00:26, 161.74it/s]\u001b[A\n",
710
+ " 45% 3388/7609 [00:14<00:24, 170.84it/s]\u001b[A\n",
711
+ " 45% 3406/7609 [00:14<00:24, 172.08it/s]\u001b[A\n",
712
+ " 45% 3424/7609 [00:14<00:24, 170.93it/s]\u001b[A\n",
713
+ " 45% 3444/7609 [00:14<00:23, 176.91it/s]\u001b[A\n",
714
+ " 46% 3464/7609 [00:15<00:22, 183.10it/s]\u001b[A\n",
715
+ " 46% 3484/7609 [00:15<00:22, 185.87it/s]\u001b[A\n",
716
+ " 46% 3505/7609 [00:15<00:21, 191.49it/s]\u001b[A\n",
717
+ " 46% 3525/7609 [00:15<00:21, 189.64it/s]\u001b[A\n",
718
+ " 47% 3545/7609 [00:15<00:22, 177.47it/s]\u001b[A\n",
719
+ " 47% 3564/7609 [00:15<00:22, 180.37it/s]\u001b[A\n",
720
+ " 47% 3584/7609 [00:15<00:21, 183.68it/s]\u001b[A\n",
721
+ " 47% 3603/7609 [00:15<00:23, 166.99it/s]\u001b[A\n",
722
+ " 48% 3621/7609 [00:15<00:24, 163.33it/s]\u001b[A\n",
723
+ " 48% 3638/7609 [00:16<00:24, 160.40it/s]\u001b[A\n",
724
+ " 48% 3659/7609 [00:16<00:22, 172.66it/s]\u001b[A\n",
725
+ " 49% 3691/7609 [00:16<00:18, 212.19it/s]\u001b[A\n",
726
+ " 49% 3722/7609 [00:16<00:16, 237.97it/s]\u001b[A\n",
727
+ " 49% 3755/7609 [00:16<00:14, 262.59it/s]\u001b[A\n",
728
+ " 50% 3782/7609 [00:16<00:14, 261.90it/s]\u001b[A\n",
729
+ " 50% 3810/7609 [00:16<00:14, 265.74it/s]\u001b[A\n",
730
+ " 51% 3847/7609 [00:16<00:12, 294.39it/s]\u001b[A\n",
731
+ " 51% 3877/7609 [00:16<00:12, 294.43it/s]\u001b[A\n",
732
+ " 51% 3912/7609 [00:16<00:11, 309.91it/s]\u001b[A\n",
733
+ " 52% 3946/7609 [00:17<00:11, 318.01it/s]\u001b[A\n",
734
+ " 52% 3978/7609 [00:17<00:11, 303.10it/s]\u001b[A\n",
735
+ " 53% 4010/7609 [00:17<00:11, 305.08it/s]\u001b[A\n",
736
+ " 53% 4041/7609 [00:17<00:11, 304.66it/s]\u001b[A\n",
737
+ " 54% 4072/7609 [00:17<00:12, 287.62it/s]\u001b[A\n",
738
+ " 54% 4102/7609 [00:17<00:13, 262.87it/s]\u001b[A\n",
739
+ " 54% 4130/7609 [00:17<00:13, 267.01it/s]\u001b[A\n",
740
+ " 55% 4160/7609 [00:17<00:12, 275.11it/s]\u001b[A\n",
741
+ " 55% 4188/7609 [00:17<00:12, 275.05it/s]\u001b[A\n",
742
+ " 55% 4216/7609 [00:18<00:12, 274.27it/s]\u001b[A\n",
743
+ " 56% 4246/7609 [00:18<00:12, 278.88it/s]\u001b[A\n",
744
+ " 56% 4278/7609 [00:18<00:11, 290.37it/s]\u001b[A\n",
745
+ " 57% 4311/7609 [00:18<00:11, 299.60it/s]\u001b[A\n",
746
+ " 57% 4344/7609 [00:18<00:10, 305.95it/s]\u001b[A\n",
747
+ " 58% 4376/7609 [00:18<00:10, 308.56it/s]\u001b[A\n",
748
+ " 58% 4409/7609 [00:18<00:10, 312.28it/s]\u001b[A\n",
749
+ " 58% 4441/7609 [00:18<00:10, 299.20it/s]\u001b[A\n",
750
+ " 59% 4473/7609 [00:18<00:10, 303.12it/s]\u001b[A\n",
751
+ " 59% 4507/7609 [00:19<00:09, 312.51it/s]\u001b[A\n",
752
+ " 60% 4539/7609 [00:19<00:10, 280.99it/s]\u001b[A\n",
753
+ " 60% 4572/7609 [00:19<00:10, 292.89it/s]\u001b[A\n",
754
+ " 60% 4603/7609 [00:19<00:10, 297.20it/s]\u001b[A\n",
755
+ " 61% 4635/7609 [00:19<00:09, 302.90it/s]\u001b[A\n",
756
+ " 61% 4668/7609 [00:19<00:09, 308.96it/s]\u001b[A\n",
757
+ " 62% 4701/7609 [00:19<00:09, 315.05it/s]\u001b[A\n",
758
+ " 62% 4735/7609 [00:19<00:08, 322.22it/s]\u001b[A\n",
759
+ " 63% 4768/7609 [00:19<00:08, 316.34it/s]\u001b[A\n",
760
+ " 63% 4800/7609 [00:19<00:09, 290.19it/s]\u001b[A\n",
761
+ " 64% 4832/7609 [00:20<00:09, 297.73it/s]\u001b[A\n",
762
+ " 64% 4863/7609 [00:20<00:09, 296.36it/s]\u001b[A\n",
763
+ " 64% 4895/7609 [00:20<00:08, 302.81it/s]\u001b[A\n",
764
+ " 65% 4926/7609 [00:20<00:08, 301.93it/s]\u001b[A\n",
765
+ " 65% 4957/7609 [00:20<00:09, 282.39it/s]\u001b[A\n",
766
+ " 66% 4990/7609 [00:20<00:08, 295.33it/s]\u001b[A\n",
767
+ " 66% 5022/7609 [00:20<00:08, 301.40it/s]\u001b[A\n",
768
+ " 66% 5055/7609 [00:20<00:08, 308.96it/s]\u001b[A\n",
769
+ " 67% 5087/7609 [00:20<00:08, 303.63it/s]\u001b[A\n",
770
+ " 67% 5120/7609 [00:21<00:08, 309.00it/s]\u001b[A\n",
771
+ " 68% 5152/7609 [00:21<00:08, 304.81it/s]\u001b[A\n",
772
+ " 68% 5184/7609 [00:21<00:07, 307.53it/s]\u001b[A\n",
773
+ " 69% 5219/7609 [00:21<00:07, 318.71it/s]\u001b[A\n",
774
+ " 69% 5253/7609 [00:21<00:07, 323.27it/s]\u001b[A\n",
775
+ " 69% 5288/7609 [00:21<00:07, 330.41it/s]\u001b[A\n",
776
+ " 70% 5323/7609 [00:21<00:06, 334.92it/s]\u001b[A\n",
777
+ " 70% 5358/7609 [00:21<00:06, 336.91it/s]\u001b[A\n",
778
+ " 71% 5392/7609 [00:21<00:06, 327.19it/s]\u001b[A\n",
779
+ " 71% 5425/7609 [00:21<00:06, 315.70it/s]\u001b[A\n",
780
+ " 72% 5457/7609 [00:22<00:06, 314.59it/s]\u001b[A\n",
781
+ " 72% 5489/7609 [00:22<00:06, 313.63it/s]\u001b[A\n",
782
+ " 73% 5522/7609 [00:22<00:06, 317.64it/s]\u001b[A\n",
783
+ " 73% 5554/7609 [00:22<00:06, 312.75it/s]\u001b[A\n",
784
+ " 73% 5586/7609 [00:22<00:06, 314.14it/s]\u001b[A\n",
785
+ " 74% 5619/7609 [00:22<00:06, 317.95it/s]\u001b[A\n",
786
+ " 74% 5651/7609 [00:22<00:06, 294.98it/s]\u001b[A\n",
787
+ " 75% 5683/7609 [00:22<00:06, 300.49it/s]\u001b[A\n",
788
+ " 75% 5716/7609 [00:22<00:06, 307.75it/s]\u001b[A\n",
789
+ " 76% 5747/7609 [00:23<00:06, 297.03it/s]\u001b[A\n",
790
+ " 76% 5781/7609 [00:23<00:05, 308.96it/s]\u001b[A\n",
791
+ " 76% 5817/7609 [00:23<00:05, 323.26it/s]\u001b[A\n",
792
+ " 77% 5850/7609 [00:23<00:05, 305.63it/s]\u001b[A\n",
793
+ " 77% 5884/7609 [00:23<00:05, 314.96it/s]\u001b[A\n",
794
+ " 78% 5918/7609 [00:23<00:05, 321.21it/s]\u001b[A\n",
795
+ " 78% 5951/7609 [00:23<00:05, 299.26it/s]\u001b[A\n",
796
+ " 79% 5983/7609 [00:23<00:05, 303.73it/s]\u001b[A\n",
797
+ " 79% 6014/7609 [00:23<00:05, 301.59it/s]\u001b[A\n",
798
+ " 79% 6045/7609 [00:24<00:06, 259.58it/s]\u001b[A\n",
799
+ " 80% 6079/7609 [00:24<00:05, 278.08it/s]\u001b[A\n",
800
+ " 80% 6108/7609 [00:24<00:05, 271.65it/s]\u001b[A\n",
801
+ " 81% 6136/7609 [00:24<00:06, 219.93it/s]\u001b[A\n",
802
+ " 81% 6160/7609 [00:24<00:07, 203.67it/s]\u001b[A\n",
803
+ " 81% 6191/7609 [00:24<00:06, 227.18it/s]\u001b[A\n",
804
+ " 82% 6225/7609 [00:24<00:05, 254.98it/s]\u001b[A\n",
805
+ " 82% 6260/7609 [00:24<00:04, 277.45it/s]\u001b[A\n",
806
+ " 83% 6290/7609 [00:25<00:04, 283.26it/s]\u001b[A\n",
807
+ " 83% 6325/7609 [00:25<00:04, 299.94it/s]\u001b[A\n",
808
+ " 84% 6358/7609 [00:25<00:04, 306.40it/s]\u001b[A\n",
809
+ " 84% 6395/7609 [00:25<00:03, 322.63it/s]\u001b[A\n",
810
+ " 85% 6432/7609 [00:25<00:03, 334.50it/s]\u001b[A\n",
811
+ " 85% 6466/7609 [00:25<00:03, 323.57it/s]\u001b[A\n",
812
+ " 85% 6499/7609 [00:25<00:03, 297.71it/s]\u001b[A\n",
813
+ " 86% 6530/7609 [00:25<00:03, 299.54it/s]\u001b[A\n",
814
+ " 86% 6564/7609 [00:25<00:03, 310.64it/s]\u001b[A\n",
815
+ " 87% 6599/7609 [00:25<00:03, 321.81it/s]\u001b[A\n",
816
+ " 87% 6632/7609 [00:26<00:03, 284.04it/s]\u001b[A\n",
817
+ " 88% 6662/7609 [00:26<00:03, 246.42it/s]\u001b[A\n",
818
+ " 88% 6689/7609 [00:26<00:04, 206.91it/s]\u001b[A\n",
819
+ " 88% 6712/7609 [00:26<00:04, 196.18it/s]\u001b[A\n",
820
+ " 88% 6733/7609 [00:26<00:04, 191.25it/s]\u001b[A\n",
821
+ " 89% 6753/7609 [00:26<00:04, 180.30it/s]\u001b[A\n",
822
+ " 89% 6772/7609 [00:26<00:04, 169.16it/s]\u001b[A\n",
823
+ " 89% 6790/7609 [00:27<00:04, 167.71it/s]\u001b[A\n",
824
+ " 89% 6807/7609 [00:27<00:05, 154.04it/s]\u001b[A\n",
825
+ " 90% 6823/7609 [00:27<00:05, 147.19it/s]\u001b[A\n",
826
+ " 90% 6841/7609 [00:27<00:04, 154.53it/s]\u001b[A\n",
827
+ " 90% 6859/7609 [00:27<00:04, 160.38it/s]\u001b[A\n",
828
+ " 90% 6876/7609 [00:27<00:05, 146.05it/s]\u001b[A\n",
829
+ " 91% 6893/7609 [00:27<00:04, 150.70it/s]\u001b[A\n",
830
+ " 91% 6912/7609 [00:27<00:04, 159.53it/s]\u001b[A\n",
831
+ " 91% 6929/7609 [00:28<00:04, 159.56it/s]\u001b[A\n",
832
+ " 91% 6947/7609 [00:28<00:04, 163.48it/s]\u001b[A\n",
833
+ " 92% 6965/7609 [00:28<00:03, 165.79it/s]\u001b[A\n",
834
+ " 92% 6982/7609 [00:28<00:03, 165.51it/s]\u001b[A\n",
835
+ " 92% 7001/7609 [00:28<00:03, 170.32it/s]\u001b[A\n",
836
+ " 92% 7021/7609 [00:28<00:03, 177.92it/s]\u001b[A\n",
837
+ " 93% 7039/7609 [00:28<00:03, 175.03it/s]\u001b[A\n",
838
+ " 93% 7057/7609 [00:28<00:03, 163.32it/s]\u001b[A\n",
839
+ " 93% 7074/7609 [00:28<00:03, 137.57it/s]\u001b[A\n",
840
+ " 93% 7089/7609 [00:29<00:03, 135.50it/s]\u001b[A\n",
841
+ " 93% 7104/7609 [00:29<00:03, 134.23it/s]\u001b[A\n",
842
+ " 94% 7118/7609 [00:29<00:03, 132.23it/s]\u001b[A\n",
843
+ " 94% 7136/7609 [00:29<00:03, 142.38it/s]\u001b[A\n",
844
+ " 94% 7151/7609 [00:29<00:03, 140.73it/s]\u001b[A\n",
845
+ " 94% 7167/7609 [00:29<00:03, 145.65it/s]\u001b[A\n",
846
+ " 94% 7186/7609 [00:29<00:02, 156.49it/s]\u001b[A\n",
847
+ " 95% 7202/7609 [00:29<00:02, 142.59it/s]\u001b[A\n",
848
+ " 95% 7222/7609 [00:29<00:02, 155.98it/s]\u001b[A\n",
849
+ " 95% 7241/7609 [00:30<00:02, 163.57it/s]\u001b[A\n",
850
+ " 95% 7258/7609 [00:30<00:02, 163.20it/s]\u001b[A\n",
851
+ " 96% 7276/7609 [00:30<00:01, 167.01it/s]\u001b[A\n",
852
+ " 96% 7296/7609 [00:30<00:01, 175.67it/s]\u001b[A\n",
853
+ " 96% 7317/7609 [00:30<00:01, 183.09it/s]\u001b[A\n",
854
+ " 97% 7345/7609 [00:30<00:01, 210.09it/s]\u001b[A\n",
855
+ " 97% 7377/7609 [00:30<00:00, 242.02it/s]\u001b[A\n",
856
+ " 97% 7407/7609 [00:30<00:00, 252.08it/s]\u001b[A\n",
857
+ " 98% 7434/7609 [00:30<00:00, 256.70it/s]\u001b[A\n",
858
+ " 98% 7460/7609 [00:30<00:00, 248.40it/s]\u001b[A\n",
859
+ " 98% 7488/7609 [00:31<00:00, 252.82it/s]\u001b[A\n",
860
+ " 99% 7514/7609 [00:31<00:00, 248.94it/s]\u001b[A\n",
861
+ " 99% 7539/7609 [00:31<00:00, 231.82it/s]\u001b[A\n",
862
+ " 99% 7563/7609 [00:31<00:00, 218.37it/s]\u001b[A\n",
863
+ "100% 7609/7609 [00:31<00:00, 240.54it/s]\n",
864
+ "06/01/2023 22:22:13 - INFO - utils_qa - Saving predictions to eval_results/predict_predictions.json.\n",
865
+ "06/01/2023 22:22:13 - INFO - utils_qa - Saving nbest_preds to eval_results/predict_nbest_predictions.json.\n",
866
+ "***** predict metrics *****\n",
867
+ " predict_samples = 8565\n",
868
+ " test_exact_match = 68.8001\n",
869
+ " test_f1 = 80.0818\n",
870
+ " test_runtime = 0:04:46.57\n",
871
+ " test_samples_per_second = 29.887\n",
872
+ " test_steps_per_second = 5.977\n",
873
+ "[INFO|modelcard.py:451] 2023-06-01 22:22:16,979 >> Dropping the following result as it does not have all the necessary fields:\n",
874
+ "{'task': {'name': 'Question Answering', 'type': 'question-answering'}, 'dataset': {'name': 'squad_it', 'type': 'squad_it'}}\n",
875
+ "100% 1713/1713 [05:29<00:00, 5.19it/s]\n"
876
+ ]
877
+ }
878
+ ]
879
+ },
880
+ {
881
+ "cell_type": "code",
882
+ "source": [],
883
+ "metadata": {
884
+ "id": "ij8YkY0xyNXI"
885
+ },
886
+ "execution_count": null,
887
+ "outputs": []
888
+ }
889
+ ]
890
+ }