unknown commited on
Commit
42b2077
·
0 Parent(s):

Initial commit of finished project

Browse files
.gitattributes ADDED
@@ -0,0 +1 @@
 
 
1
+ *.pth filter=lfs diff=lfs merge=lfs -text
Untitled24vitsmall (1) - Copy.ipynb ADDED
@@ -0,0 +1,2154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 5,
6
+ "metadata": {
7
+ "id": "DDjtMfKwauQl"
8
+ },
9
+ "outputs": [
10
+ {
11
+ "name": "stdout",
12
+ "output_type": "stream",
13
+ "text": [
14
+ "Requirement already satisfied: timm in /venv/main/lib/python3.12/site-packages (1.0.25)\n",
15
+ "Requirement already satisfied: gdown in /venv/main/lib/python3.12/site-packages (5.2.1)\n",
16
+ "Collecting tensorboard\n",
17
+ " Downloading tensorboard-2.20.0-py3-none-any.whl.metadata (1.8 kB)\n",
18
+ "Requirement already satisfied: torch in /venv/main/lib/python3.12/site-packages (from timm) (2.10.0+cu130)\n",
19
+ "Requirement already satisfied: torchvision in /venv/main/lib/python3.12/site-packages (from timm) (0.25.0+cu130)\n",
20
+ "Requirement already satisfied: pyyaml in /venv/main/lib/python3.12/site-packages (from timm) (6.0.3)\n",
21
+ "Requirement already satisfied: huggingface_hub in /venv/main/lib/python3.12/site-packages (from timm) (1.2.3)\n",
22
+ "Requirement already satisfied: safetensors in /venv/main/lib/python3.12/site-packages (from timm) (0.7.0)\n",
23
+ "Requirement already satisfied: beautifulsoup4 in /venv/main/lib/python3.12/site-packages (from gdown) (4.14.3)\n",
24
+ "Requirement already satisfied: filelock in /venv/main/lib/python3.12/site-packages (from gdown) (3.20.1)\n",
25
+ "Requirement already satisfied: requests[socks] in /venv/main/lib/python3.12/site-packages (from gdown) (2.32.5)\n",
26
+ "Requirement already satisfied: tqdm in /venv/main/lib/python3.12/site-packages (from gdown) (4.67.1)\n",
27
+ "Collecting absl-py>=0.4 (from tensorboard)\n",
28
+ " Downloading absl_py-2.4.0-py3-none-any.whl.metadata (3.3 kB)\n",
29
+ "Collecting grpcio>=1.48.2 (from tensorboard)\n",
30
+ " Downloading grpcio-1.78.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (3.8 kB)\n",
31
+ "Collecting markdown>=2.6.8 (from tensorboard)\n",
32
+ " Downloading markdown-3.10.2-py3-none-any.whl.metadata (5.1 kB)\n",
33
+ "Requirement already satisfied: numpy>=1.12.0 in /venv/main/lib/python3.12/site-packages (from tensorboard) (2.4.1)\n",
34
+ "Requirement already satisfied: packaging in /venv/main/lib/python3.12/site-packages (from tensorboard) (25.0)\n",
35
+ "Requirement already satisfied: pillow in /venv/main/lib/python3.12/site-packages (from tensorboard) (12.1.0)\n",
36
+ "Collecting protobuf!=4.24.0,>=3.19.6 (from tensorboard)\n",
37
+ " Downloading protobuf-7.34.0-cp310-abi3-manylinux2014_x86_64.whl.metadata (595 bytes)\n",
38
+ "Requirement already satisfied: setuptools>=41.0.0 in /venv/main/lib/python3.12/site-packages (from tensorboard) (80.9.0)\n",
39
+ "Collecting tensorboard-data-server<0.8.0,>=0.7.0 (from tensorboard)\n",
40
+ " Downloading tensorboard_data_server-0.7.2-py3-none-manylinux_2_31_x86_64.whl.metadata (1.1 kB)\n",
41
+ "Collecting werkzeug>=1.0.1 (from tensorboard)\n",
42
+ " Downloading werkzeug-3.1.6-py3-none-any.whl.metadata (4.0 kB)\n",
43
+ "Requirement already satisfied: typing-extensions~=4.12 in /venv/main/lib/python3.12/site-packages (from grpcio>=1.48.2->tensorboard) (4.15.0)\n",
44
+ "Requirement already satisfied: markupsafe>=2.1.1 in /venv/main/lib/python3.12/site-packages (from werkzeug>=1.0.1->tensorboard) (3.0.3)\n",
45
+ "Requirement already satisfied: soupsieve>=1.6.1 in /venv/main/lib/python3.12/site-packages (from beautifulsoup4->gdown) (2.8.3)\n",
46
+ "Requirement already satisfied: fsspec>=2023.5.0 in /venv/main/lib/python3.12/site-packages (from huggingface_hub->timm) (2025.12.0)\n",
47
+ "Requirement already satisfied: hf-xet<2.0.0,>=1.2.0 in /venv/main/lib/python3.12/site-packages (from huggingface_hub->timm) (1.2.0)\n",
48
+ "Requirement already satisfied: httpx<1,>=0.23.0 in /venv/main/lib/python3.12/site-packages (from huggingface_hub->timm) (0.28.1)\n",
49
+ "Requirement already satisfied: shellingham in /venv/main/lib/python3.12/site-packages (from huggingface_hub->timm) (1.5.4)\n",
50
+ "Requirement already satisfied: typer-slim in /venv/main/lib/python3.12/site-packages (from huggingface_hub->timm) (0.21.0)\n",
51
+ "Requirement already satisfied: anyio in /venv/main/lib/python3.12/site-packages (from httpx<1,>=0.23.0->huggingface_hub->timm) (4.12.0)\n",
52
+ "Requirement already satisfied: certifi in /venv/main/lib/python3.12/site-packages (from httpx<1,>=0.23.0->huggingface_hub->timm) (2025.11.12)\n",
53
+ "Requirement already satisfied: httpcore==1.* in /venv/main/lib/python3.12/site-packages (from httpx<1,>=0.23.0->huggingface_hub->timm) (1.0.9)\n",
54
+ "Requirement already satisfied: idna in /venv/main/lib/python3.12/site-packages (from httpx<1,>=0.23.0->huggingface_hub->timm) (3.11)\n",
55
+ "Requirement already satisfied: h11>=0.16 in /venv/main/lib/python3.12/site-packages (from httpcore==1.*->httpx<1,>=0.23.0->huggingface_hub->timm) (0.16.0)\n",
56
+ "Requirement already satisfied: charset_normalizer<4,>=2 in /venv/main/lib/python3.12/site-packages (from requests[socks]->gdown) (3.4.4)\n",
57
+ "Requirement already satisfied: urllib3<3,>=1.21.1 in /venv/main/lib/python3.12/site-packages (from requests[socks]->gdown) (2.6.3)\n",
58
+ "Requirement already satisfied: PySocks!=1.5.7,>=1.5.6 in /venv/main/lib/python3.12/site-packages (from requests[socks]->gdown) (1.7.1)\n",
59
+ "Requirement already satisfied: sympy>=1.13.3 in /venv/main/lib/python3.12/site-packages (from torch->timm) (1.14.0)\n",
60
+ "Requirement already satisfied: networkx>=2.5.1 in /venv/main/lib/python3.12/site-packages (from torch->timm) (3.6.1)\n",
61
+ "Requirement already satisfied: jinja2 in /venv/main/lib/python3.12/site-packages (from torch->timm) (3.1.6)\n",
62
+ "Requirement already satisfied: cuda-bindings==13.0.3 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.0.3)\n",
63
+ "Requirement already satisfied: nvidia-cuda-nvrtc==13.0.88 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.0.88)\n",
64
+ "Requirement already satisfied: nvidia-cuda-runtime==13.0.96 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.0.96)\n",
65
+ "Requirement already satisfied: nvidia-cuda-cupti==13.0.85 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.0.85)\n",
66
+ "Requirement already satisfied: nvidia-cudnn-cu13==9.15.1.9 in /venv/main/lib/python3.12/site-packages (from torch->timm) (9.15.1.9)\n",
67
+ "Requirement already satisfied: nvidia-cublas==13.1.0.3 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.1.0.3)\n",
68
+ "Requirement already satisfied: nvidia-cufft==12.0.0.61 in /venv/main/lib/python3.12/site-packages (from torch->timm) (12.0.0.61)\n",
69
+ "Requirement already satisfied: nvidia-curand==10.4.0.35 in /venv/main/lib/python3.12/site-packages (from torch->timm) (10.4.0.35)\n",
70
+ "Requirement already satisfied: nvidia-cusolver==12.0.4.66 in /venv/main/lib/python3.12/site-packages (from torch->timm) (12.0.4.66)\n",
71
+ "Requirement already satisfied: nvidia-cusparse==12.6.3.3 in /venv/main/lib/python3.12/site-packages (from torch->timm) (12.6.3.3)\n",
72
+ "Requirement already satisfied: nvidia-cusparselt-cu13==0.8.0 in /venv/main/lib/python3.12/site-packages (from torch->timm) (0.8.0)\n",
73
+ "Requirement already satisfied: nvidia-nccl-cu13==2.28.9 in /venv/main/lib/python3.12/site-packages (from torch->timm) (2.28.9)\n",
74
+ "Requirement already satisfied: nvidia-nvshmem-cu13==3.4.5 in /venv/main/lib/python3.12/site-packages (from torch->timm) (3.4.5)\n",
75
+ "Requirement already satisfied: nvidia-nvtx==13.0.85 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.0.85)\n",
76
+ "Requirement already satisfied: nvidia-nvjitlink==13.0.88 in /venv/main/lib/python3.12/site-packages (from torch->timm) (13.0.88)\n",
77
+ "Requirement already satisfied: nvidia-cufile==1.15.1.6 in /venv/main/lib/python3.12/site-packages (from torch->timm) (1.15.1.6)\n",
78
+ "Requirement already satisfied: triton==3.6.0 in /venv/main/lib/python3.12/site-packages (from torch->timm) (3.6.0)\n",
79
+ "Requirement already satisfied: cuda-pathfinder~=1.1 in /venv/main/lib/python3.12/site-packages (from cuda-bindings==13.0.3->torch->timm) (1.3.3)\n",
80
+ "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /venv/main/lib/python3.12/site-packages (from sympy>=1.13.3->torch->timm) (1.3.0)\n",
81
+ "Requirement already satisfied: click>=8.0.0 in /venv/main/lib/python3.12/site-packages (from typer-slim->huggingface_hub->timm) (8.3.1)\n",
82
+ "Downloading tensorboard-2.20.0-py3-none-any.whl (5.5 MB)\n",
83
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.5/5.5 MB\u001b[0m \u001b[31m12.0 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0mm0:00:01\u001b[0m00:01\u001b[0m\n",
84
+ "\u001b[?25hDownloading tensorboard_data_server-0.7.2-py3-none-manylinux_2_31_x86_64.whl (6.6 MB)\n",
85
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m6.6/6.6 MB\u001b[0m \u001b[31m20.7 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0mm0:00:01\u001b[0m00:01\u001b[0m\n",
86
+ "\u001b[?25hDownloading absl_py-2.4.0-py3-none-any.whl (135 kB)\n",
87
+ "Downloading grpcio-1.78.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (6.7 MB)\n",
88
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m6.7/6.7 MB\u001b[0m \u001b[31m20.8 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0mm0:00:01\u001b[0m00:01\u001b[0m\n",
89
+ "\u001b[?25hDownloading markdown-3.10.2-py3-none-any.whl (108 kB)\n",
90
+ "Downloading protobuf-7.34.0-cp310-abi3-manylinux2014_x86_64.whl (324 kB)\n",
91
+ "Downloading werkzeug-3.1.6-py3-none-any.whl (225 kB)\n",
92
+ "Installing collected packages: werkzeug, tensorboard-data-server, protobuf, markdown, grpcio, absl-py, tensorboard\n",
93
+ "\u001b[2K \u001b[90m━━━━━━━━━━��━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7/7\u001b[0m [tensorboard]\u001b[0m [tensorboard]\n",
94
+ "\u001b[1A\u001b[2KSuccessfully installed absl-py-2.4.0 grpcio-1.78.0 markdown-3.10.2 protobuf-7.34.0 tensorboard-2.20.0 tensorboard-data-server-0.7.2 werkzeug-3.1.6\n",
95
+ "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.\u001b[0m\u001b[33m\n",
96
+ "\u001b[0mNote: you may need to restart the kernel to use updated packages.\n"
97
+ ]
98
+ }
99
+ ],
100
+ "source": [
101
+ "%pip install timm gdown tensorboard"
102
+ ]
103
+ },
104
+ {
105
+ "cell_type": "code",
106
+ "execution_count": 1,
107
+ "metadata": {},
108
+ "outputs": [
109
+ {
110
+ "name": "stdout",
111
+ "output_type": "stream",
112
+ "text": [
113
+ "Package Version\n",
114
+ "----------------------- ------------\n",
115
+ "anyio 4.12.0\n",
116
+ "asttokens 3.0.1\n",
117
+ "certifi 2025.11.12\n",
118
+ "charset-normalizer 3.4.4\n",
119
+ "click 8.3.1\n",
120
+ "comm 0.2.3\n",
121
+ "cuda-bindings 13.0.3\n",
122
+ "cuda-pathfinder 1.3.3\n",
123
+ "debugpy 1.8.19\n",
124
+ "decorator 5.2.1\n",
125
+ "executing 2.2.1\n",
126
+ "filelock 3.20.1\n",
127
+ "fsspec 2025.12.0\n",
128
+ "h11 0.16.0\n",
129
+ "hf-xet 1.2.0\n",
130
+ "httpcore 1.0.9\n",
131
+ "httpx 0.28.1\n",
132
+ "huggingface_hub 1.2.3\n",
133
+ "idna 3.11\n",
134
+ "ipykernel 7.1.0\n",
135
+ "ipython 9.8.0\n",
136
+ "ipython_pygments_lexers 1.1.1\n",
137
+ "ipywidgets 8.1.8\n",
138
+ "jedi 0.19.2\n",
139
+ "Jinja2 3.1.6\n",
140
+ "jupyter_client 8.7.0\n",
141
+ "jupyter_core 5.9.1\n",
142
+ "jupyterlab_widgets 3.0.16\n",
143
+ "MarkupSafe 3.0.3\n",
144
+ "matplotlib-inline 0.2.1\n",
145
+ "mpmath 1.3.0\n",
146
+ "nest-asyncio 1.6.0\n",
147
+ "networkx 3.6.1\n",
148
+ "numpy 2.4.1\n",
149
+ "nvidia-cublas 13.1.0.3\n",
150
+ "nvidia-cuda-cupti 13.0.85\n",
151
+ "nvidia-cuda-nvrtc 13.0.88\n",
152
+ "nvidia-cuda-runtime 13.0.96\n",
153
+ "nvidia-cudnn-cu13 9.15.1.9\n",
154
+ "nvidia-cufft 12.0.0.61\n",
155
+ "nvidia-cufile 1.15.1.6\n",
156
+ "nvidia-curand 10.4.0.35\n",
157
+ "nvidia-cusolver 12.0.4.66\n",
158
+ "nvidia-cusparse 12.6.3.3\n",
159
+ "nvidia-cusparselt-cu13 0.8.0\n",
160
+ "nvidia-nccl-cu13 2.28.9\n",
161
+ "nvidia-nvjitlink 13.0.88\n",
162
+ "nvidia-nvshmem-cu13 3.4.5\n",
163
+ "nvidia-nvtx 13.0.85\n",
164
+ "packaging 25.0\n",
165
+ "parso 0.8.5\n",
166
+ "pexpect 4.9.0\n",
167
+ "pillow 12.1.0\n",
168
+ "pip 25.3\n",
169
+ "platformdirs 4.5.1\n",
170
+ "prompt_toolkit 3.0.52\n",
171
+ "psutil 7.2.1\n",
172
+ "ptyprocess 0.7.0\n",
173
+ "pure_eval 0.2.3\n",
174
+ "Pygments 2.19.2\n",
175
+ "python-dateutil 2.9.0.post0\n",
176
+ "PyYAML 6.0.3\n",
177
+ "pyzmq 27.1.0\n",
178
+ "requests 2.32.5\n",
179
+ "sentencepiece 0.2.1\n",
180
+ "setuptools 80.9.0\n",
181
+ "shellingham 1.5.4\n",
182
+ "six 1.17.0\n",
183
+ "stack-data 0.6.3\n",
184
+ "sympy 1.14.0\n",
185
+ "torch 2.10.0+cu130\n",
186
+ "torchaudio 2.10.0+cu130\n",
187
+ "torchcodec 0.10.0\n",
188
+ "torchdata 0.10.0\n",
189
+ "torchtext 0.6.0\n",
190
+ "torchvision 0.25.0+cu130\n",
191
+ "tornado 6.5.4\n",
192
+ "tqdm 4.67.1\n",
193
+ "traitlets 5.14.3\n",
194
+ "triton 3.6.0\n",
195
+ "typer-slim 0.21.0\n",
196
+ "typing_extensions 4.15.0\n",
197
+ "urllib3 2.6.3\n",
198
+ "wcwidth 0.2.14\n",
199
+ "wheel 0.45.1\n",
200
+ "widgetsnbextension 4.0.15\n",
201
+ "Note: you may need to restart the kernel to use updated packages.\n"
202
+ ]
203
+ }
204
+ ],
205
+ "source": [
206
+ "%pip list"
207
+ ]
208
+ },
209
+ {
210
+ "cell_type": "code",
211
+ "execution_count": 6,
212
+ "metadata": {
213
+ "id": "xJYUsKdBCPVS"
214
+ },
215
+ "outputs": [],
216
+ "source": [
217
+ "import os\n",
218
+ "import shutil\n",
219
+ "import torch\n",
220
+ "import torch.nn as nn\n",
221
+ "import torch.optim as optim\n",
222
+ "from torchvision import datasets, transforms\n",
223
+ "from timm import create_model\n",
224
+ "from torch.optim.lr_scheduler import CosineAnnealingLR\n",
225
+ "from torch.utils.data import DataLoader\n",
226
+ "from torch.utils.tensorboard import SummaryWriter\n",
227
+ "from tqdm import tqdm # For progress bar\n",
228
+ "from torchvision.transforms import RandAugment\n",
229
+ "from timm.data import Mixup\n",
230
+ "from timm.loss import SoftTargetCrossEntropy\n",
231
+ "from timm.layers import DropPath # Updated import path\n",
232
+ "from timm.scheduler.cosine_lr import CosineLRScheduler\n",
233
+ "\n"
234
+ ]
235
+ },
236
+ {
237
+ "cell_type": "code",
238
+ "execution_count": null,
239
+ "metadata": {},
240
+ "outputs": [],
241
+ "source": [
242
+ "import gdown\n",
243
+ "\n",
244
+ "url = 'https://drive.google.com/'\n",
245
+ "output = 'dat2.zip'\n",
246
+ "gdown.download(url, output, quiet=False, fuzzy=True)"
247
+ ]
248
+ },
249
+ {
250
+ "cell_type": "code",
251
+ "execution_count": 10,
252
+ "metadata": {
253
+ "colab": {
254
+ "base_uri": "https://localhost:8080/"
255
+ },
256
+ "id": "cbW5kekcM0nQ",
257
+ "outputId": "0c0d198c-d492-4b04-8870-eea889d8dfea"
258
+ },
259
+ "outputs": [
260
+ {
261
+ "name": "stdout",
262
+ "output_type": "stream",
263
+ "text": [
264
+ "Files extracted to: /workspace/red\n"
265
+ ]
266
+ }
267
+ ],
268
+ "source": [
269
+ "import zipfile\n",
270
+ "import os\n",
271
+ "\n",
272
+ "zip_file_name = '/workspace/dat2.zip'\n",
273
+ "extract_dir = '/workspace/red' # Target directory for extraction\n",
274
+ "\n",
275
+ "# Create the target directory if it doesn't exist\n",
276
+ "if not os.path.exists(extract_dir):\n",
277
+ " os.makedirs(extract_dir)\n",
278
+ "\n",
279
+ "with zipfile.ZipFile(zip_file_name, 'r') as zip_ref:\n",
280
+ " # Extract all contents to the specified directory\n",
281
+ " zip_ref.extractall(extract_dir)\n",
282
+ "\n",
283
+ "print(f\"Files extracted to: {extract_dir}\")"
284
+ ]
285
+ },
286
+ {
287
+ "cell_type": "code",
288
+ "execution_count": 11,
289
+ "metadata": {
290
+ "id": "5D1NQo1LCStD"
291
+ },
292
+ "outputs": [],
293
+ "source": [
294
+ "# Paths and Constants\n",
295
+ "data_dir = \"/workspace/red/splits1\"\n",
296
+ "num_classes = 7\n",
297
+ "batch_size = 128 # Adjust based on GPU memory\n",
298
+ "num_epochs = 50 # Increased number of epochs for better convergence\n",
299
+ "learning_rate = 5e-4 # Lowered learning rate for fine-tuning\n",
300
+ "weight_decay = 0.01 # Adjusted weight decay\n",
301
+ "image_size = 224\n",
302
+ "log_interval = 100 # Log metrics every 100 batches\n"
303
+ ]
304
+ },
305
+ {
306
+ "cell_type": "code",
307
+ "execution_count": 12,
308
+ "metadata": {
309
+ "id": "JgcWhUooCfC7"
310
+ },
311
+ "outputs": [],
312
+ "source": [
313
+ "transform_train = transforms.Compose([\n",
314
+ " transforms.Resize((image_size, image_size), interpolation=transforms.InterpolationMode.BICUBIC),\n",
315
+ " RandAugment(), # Enhanced augmentation\n",
316
+ " transforms.ToTensor(),\n",
317
+ " transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),\n",
318
+ " transforms.RandomErasing(p=0.1),\n",
319
+ "])\n"
320
+ ]
321
+ },
322
+ {
323
+ "cell_type": "code",
324
+ "execution_count": 13,
325
+ "metadata": {
326
+ "id": "5YWSDRJrCgwF"
327
+ },
328
+ "outputs": [],
329
+ "source": [
330
+ "transform_test = transforms.Compose([\n",
331
+ " transforms.Resize((image_size, image_size), interpolation=transforms.InterpolationMode.BICUBIC),\n",
332
+ " transforms.ToTensor(),\n",
333
+ " transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),\n",
334
+ "])\n"
335
+ ]
336
+ },
337
+ {
338
+ "cell_type": "code",
339
+ "execution_count": 14,
340
+ "metadata": {
341
+ "id": "sqcG-hMtChPa"
342
+ },
343
+ "outputs": [],
344
+ "source": [
345
+ "train_dataset = datasets.ImageFolder(os.path.join(data_dir, 'train'), transform=transform_train)\n",
346
+ "train_loader = DataLoader(\n",
347
+ " train_dataset,\n",
348
+ " batch_size=batch_size,\n",
349
+ " shuffle=True,\n",
350
+ " num_workers=16, # Reduced from 8 to 2\n",
351
+ " pin_memory=True,\n",
352
+ " prefetch_factor=4,\n",
353
+ " persistent_workers=True\n",
354
+ ")\n"
355
+ ]
356
+ },
357
+ {
358
+ "cell_type": "code",
359
+ "execution_count": 15,
360
+ "metadata": {
361
+ "id": "h2QXKLTqCmT8"
362
+ },
363
+ "outputs": [],
364
+ "source": [
365
+ "val_images_dir = os.path.join(\"/workspace/red/splits1/val\", '')\n",
366
+ "val_dataset = datasets.ImageFolder(val_images_dir, transform=transform_test)\n",
367
+ "val_loader = DataLoader(\n",
368
+ " val_dataset,\n",
369
+ " batch_size=batch_size,\n",
370
+ " shuffle=False,\n",
371
+ " num_workers=16, # Reduced from 8 to 2\n",
372
+ " pin_memory=True,\n",
373
+ " prefetch_factor=4,\n",
374
+ " persistent_workers=True\n",
375
+ ")\n"
376
+ ]
377
+ },
378
+ {
379
+ "cell_type": "code",
380
+ "execution_count": 16,
381
+ "metadata": {
382
+ "colab": {
383
+ "base_uri": "https://localhost:8080/"
384
+ },
385
+ "id": "EQzvAErvCpJM",
386
+ "outputId": "704d863e-3bdf-487f-e62e-8c5f724ed99f"
387
+ },
388
+ "outputs": [
389
+ {
390
+ "data": {
391
+ "application/vnd.jupyter.widget-view+json": {
392
+ "model_id": "50ffe1bd5e6443719a4752151afd6b14",
393
+ "version_major": 2,
394
+ "version_minor": 0
395
+ },
396
+ "text/plain": [
397
+ "model.safetensors: 0%| | 0.00/88.2M [00:00<?, ?B/s]"
398
+ ]
399
+ },
400
+ "metadata": {},
401
+ "output_type": "display_data"
402
+ }
403
+ ],
404
+ "source": [
405
+ "model = create_model('vit_small_patch16_224.augreg_in1k', pretrained=True, num_classes=num_classes)\n"
406
+ ]
407
+ },
408
+ {
409
+ "cell_type": "code",
410
+ "execution_count": 17,
411
+ "metadata": {
412
+ "id": "KtTwgd7nC4VA"
413
+ },
414
+ "outputs": [],
415
+ "source": [
416
+ "def apply_stochastic_depth(model, drop_prob):\n",
417
+ " for module in model.modules():\n",
418
+ " if isinstance(module, DropPath):\n",
419
+ " module.drop_prob = drop_prob"
420
+ ]
421
+ },
422
+ {
423
+ "cell_type": "code",
424
+ "execution_count": 18,
425
+ "metadata": {
426
+ "id": "2lHD5QknC6FK"
427
+ },
428
+ "outputs": [],
429
+ "source": [
430
+ "apply_stochastic_depth(model, drop_prob=0.1)\n"
431
+ ]
432
+ },
433
+ {
434
+ "cell_type": "code",
435
+ "execution_count": 19,
436
+ "metadata": {
437
+ "id": "17jrn7IsC7WT"
438
+ },
439
+ "outputs": [],
440
+ "source": [
441
+ "for param in model.parameters():\n",
442
+ " param.requires_grad = True"
443
+ ]
444
+ },
445
+ {
446
+ "cell_type": "code",
447
+ "execution_count": 20,
448
+ "metadata": {
449
+ "colab": {
450
+ "base_uri": "https://localhost:8080/"
451
+ },
452
+ "id": "cLT8zi0uDH2_",
453
+ "outputId": "c81d6fba-6b58-462a-aa5f-163b5afd8a1d"
454
+ },
455
+ "outputs": [
456
+ {
457
+ "name": "stdout",
458
+ "output_type": "stream",
459
+ "text": [
460
+ "Using device: cuda\n"
461
+ ]
462
+ }
463
+ ],
464
+ "source": [
465
+ "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
466
+ "print(f\"Using device: {device}\")"
467
+ ]
468
+ },
469
+ {
470
+ "cell_type": "code",
471
+ "execution_count": 21,
472
+ "metadata": {
473
+ "id": "-gBcVTPXDKSK"
474
+ },
475
+ "outputs": [
476
+ {
477
+ "name": "stdout",
478
+ "output_type": "stream",
479
+ "text": [
480
+ "Using 2 GPUs\n"
481
+ ]
482
+ }
483
+ ],
484
+ "source": [
485
+ "if torch.cuda.device_count() > 1:\n",
486
+ " print(f\"Using {torch.cuda.device_count()} GPUs\")\n",
487
+ " model = nn.DataParallel(model) # This will use all available GPUs"
488
+ ]
489
+ },
490
+ {
491
+ "cell_type": "code",
492
+ "execution_count": 22,
493
+ "metadata": {
494
+ "id": "W-stw1OBDLqp"
495
+ },
496
+ "outputs": [],
497
+ "source": [
498
+ "model = model.to(device)\n"
499
+ ]
500
+ },
501
+ {
502
+ "cell_type": "code",
503
+ "execution_count": 23,
504
+ "metadata": {},
505
+ "outputs": [],
506
+ "source": [
507
+ "class_weights = torch.tensor([1.560, 3.737, 2.242, 0.541, 0.527, 0.970, 1.149], dtype=torch.float)\n",
508
+ "class_weights = class_weights.to(device)"
509
+ ]
510
+ },
511
+ {
512
+ "cell_type": "code",
513
+ "execution_count": 24,
514
+ "metadata": {
515
+ "id": "tI3m3KquDPXI"
516
+ },
517
+ "outputs": [],
518
+ "source": [
519
+ "criterion = nn.CrossEntropyLoss(weight=class_weights) # For Mixup and CutMix\n"
520
+ ]
521
+ },
522
+ {
523
+ "cell_type": "code",
524
+ "execution_count": 25,
525
+ "metadata": {
526
+ "id": "afX6EEL8DQvq"
527
+ },
528
+ "outputs": [],
529
+ "source": [
530
+ "optimizer = optim.AdamW(model.parameters(), lr=learning_rate, weight_decay=weight_decay)"
531
+ ]
532
+ },
533
+ {
534
+ "cell_type": "code",
535
+ "execution_count": 26,
536
+ "metadata": {
537
+ "id": "Yf0JflY9DR_t"
538
+ },
539
+ "outputs": [],
540
+ "source": [
541
+ "scheduler = CosineLRScheduler(\n",
542
+ " optimizer,\n",
543
+ " t_initial=num_epochs,\n",
544
+ " lr_min=1e-5,\n",
545
+ " warmup_t=5,\n",
546
+ " warmup_lr_init=1e-6,\n",
547
+ " warmup_prefix=True \n",
548
+ ")"
549
+ ]
550
+ },
551
+ {
552
+ "cell_type": "code",
553
+ "execution_count": 27,
554
+ "metadata": {
555
+ "id": "jmO4QHy3DT0j"
556
+ },
557
+ "outputs": [],
558
+ "source": [
559
+ "scaler = torch.amp.GradScaler(device='cuda') # Updated instantiation\n"
560
+ ]
561
+ },
562
+ {
563
+ "cell_type": "code",
564
+ "execution_count": 28,
565
+ "metadata": {
566
+ "id": "pW1uIqzeDVZv"
567
+ },
568
+ "outputs": [],
569
+ "source": [
570
+ "writer = SummaryWriter() # For TensorBoard logging\n"
571
+ ]
572
+ },
573
+ {
574
+ "cell_type": "code",
575
+ "execution_count": 29,
576
+ "metadata": {
577
+ "id": "eFQv3QNYDWyq"
578
+ },
579
+ "outputs": [],
580
+ "source": [
581
+ "def train_one_epoch(epoch):\n",
582
+ " model.train()\n",
583
+ " running_loss, correct, total = 0.0, 0, 0\n",
584
+ "\n",
585
+ " # Progress bar for training loop\n",
586
+ " train_loader_tqdm = tqdm(train_loader, desc=f\"Epoch {epoch+1}/{num_epochs} [Training]\", leave=False)\n",
587
+ " for batch_idx, (images, labels) in enumerate(train_loader_tqdm):\n",
588
+ " images, labels = images.to(device, non_blocking=True), labels.to(device, non_blocking=True)\n",
589
+ "\n",
590
+ " optimizer.zero_grad()\n",
591
+ "\n",
592
+ " with torch.cuda.amp.autocast():\n",
593
+ " outputs = model(images)\n",
594
+ " loss = criterion(outputs, labels)\n",
595
+ "\n",
596
+ " scaler.scale(loss).backward()\n",
597
+ " scaler.step(optimizer)\n",
598
+ " scaler.update()\n",
599
+ "\n",
600
+ " running_loss += loss.item() * images.size(0)\n",
601
+ " total += labels.size(0)\n",
602
+ "\n",
603
+ " # Since labels are soft, calculate accuracy based on predicted class vs hard labels\n",
604
+ " _, predicted = outputs.max(1)\n",
605
+ " correct += predicted.eq(labels).sum().item()\n",
606
+ "\n",
607
+ " # Update progress bar (accuracy in percentage)\n",
608
+ " if (batch_idx + 1) % log_interval == 0 or (batch_idx + 1) == len(train_loader):\n",
609
+ " current_loss = loss.item()\n",
610
+ " current_acc = 100. * correct / total\n",
611
+ " train_loader_tqdm.set_postfix(loss=f\"{current_loss:.4f}\", accuracy=f\"{current_acc:.2f}%\")\n",
612
+ "\n",
613
+ " epoch_loss = running_loss / total\n",
614
+ " epoch_acc = 100. * correct / total # Multiply by 100 to get percentage\n",
615
+ " writer.add_scalar('Loss/train', epoch_loss, epoch)\n",
616
+ " writer.add_scalar('Accuracy/train', epoch_acc, epoch)\n",
617
+ " print(f\"Epoch [{epoch+1}/{num_epochs}], Loss: {epoch_loss:.4f}, Acc: {epoch_acc:.2f}%\") # Acc in %\n"
618
+ ]
619
+ },
620
+ {
621
+ "cell_type": "code",
622
+ "execution_count": 30,
623
+ "metadata": {
624
+ "id": "Bi9RQ6CDDZuE"
625
+ },
626
+ "outputs": [],
627
+ "source": [
628
+ "def validate(epoch):\n",
629
+ " model.eval()\n",
630
+ " val_loss, correct, total = 0.0, 0, 0\n",
631
+ "\n",
632
+ " # Progress bar for validation loop\n",
633
+ " val_loader_tqdm = tqdm(val_loader, desc=f\"Epoch {epoch+1}/{num_epochs} [Validation]\", leave=False)\n",
634
+ "\n",
635
+ " criterion_val = nn.CrossEntropyLoss() # Standard loss for validation\n",
636
+ "\n",
637
+ " with torch.no_grad():\n",
638
+ " for batch_idx, (images, labels) in enumerate(val_loader_tqdm):\n",
639
+ " images, labels = images.to(device, non_blocking=True), labels.to(device, non_blocking=True)\n",
640
+ "\n",
641
+ " with torch.cuda.amp.autocast():\n",
642
+ " outputs = model(images)\n",
643
+ " loss = criterion_val(outputs, labels)\n",
644
+ "\n",
645
+ " val_loss += loss.item() * images.size(0)\n",
646
+ " total += labels.size(0)\n",
647
+ " _, predicted = outputs.max(1)\n",
648
+ " correct += predicted.eq(labels).sum().item()\n",
649
+ "\n",
650
+ " # Update progress bar (accuracy in percentage)\n",
651
+ " if (batch_idx + 1) % log_interval == 0 or (batch_idx + 1) == len(val_loader):\n",
652
+ " current_loss = loss.item()\n",
653
+ " current_acc = 100. * correct / total\n",
654
+ " val_loader_tqdm.set_postfix(loss=f\"{current_loss:.4f}\", accuracy=f\"{current_acc:.2f}%\")\n",
655
+ "\n",
656
+ " epoch_loss = val_loss / total\n",
657
+ " epoch_acc = 100. * correct / total # Multiply by 100 to get percentage\n",
658
+ " writer.add_scalar('Loss/val', epoch_loss, epoch)\n",
659
+ " writer.add_scalar('Accuracy/val', epoch_acc, epoch)\n",
660
+ " print(f\"Validation Loss: {epoch_loss:.4f}, Acc: {epoch_acc:.2f}%\") # Acc in %\n",
661
+ "\n",
662
+ " return epoch_acc"
663
+ ]
664
+ },
665
+ {
666
+ "cell_type": "code",
667
+ "execution_count": null,
668
+ "metadata": {
669
+ "colab": {
670
+ "base_uri": "https://localhost:8080/",
671
+ "height": 474
672
+ },
673
+ "id": "YJ53kRT3DcuJ",
674
+ "outputId": "ef6e52b1-8f2a-43c3-a5f4-49009ce167d8"
675
+ },
676
+ "outputs": [
677
+ {
678
+ "name": "stderr",
679
+ "output_type": "stream",
680
+ "text": [
681
+ "Epoch 1/50 [Training]: 0%| | 0/420 [00:00<?, ?it/s]/tmp/ipykernel_1207/2336286154.py:12: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\n",
682
+ " with torch.cuda.amp.autocast():\n",
683
+ " "
684
+ ]
685
+ },
686
+ {
687
+ "name": "stdout",
688
+ "output_type": "stream",
689
+ "text": [
690
+ "Epoch [1/50], Loss: 1.8604, Acc: 25.83%\n"
691
+ ]
692
+ },
693
+ {
694
+ "name": "stderr",
695
+ "output_type": "stream",
696
+ "text": [
697
+ "Epoch 1/50 [Validation]: 0%| | 0/53 [00:00<?, ?it/s]/tmp/ipykernel_1207/1125314546.py:14: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.\n",
698
+ " with torch.cuda.amp.autocast():\n",
699
+ " "
700
+ ]
701
+ },
702
+ {
703
+ "name": "stdout",
704
+ "output_type": "stream",
705
+ "text": [
706
+ "Validation Loss: 1.6683, Acc: 35.74%\n",
707
+ "New best model saved with accuracy: 35.74%\n"
708
+ ]
709
+ },
710
+ {
711
+ "name": "stderr",
712
+ "output_type": "stream",
713
+ "text": [
714
+ " "
715
+ ]
716
+ },
717
+ {
718
+ "name": "stdout",
719
+ "output_type": "stream",
720
+ "text": [
721
+ "Epoch [2/50], Loss: 1.1986, Acc: 58.37%\n"
722
+ ]
723
+ },
724
+ {
725
+ "name": "stderr",
726
+ "output_type": "stream",
727
+ "text": [
728
+ " "
729
+ ]
730
+ },
731
+ {
732
+ "name": "stdout",
733
+ "output_type": "stream",
734
+ "text": [
735
+ "Validation Loss: 0.9391, Acc: 64.96%\n",
736
+ "New best model saved with accuracy: 64.96%\n"
737
+ ]
738
+ },
739
+ {
740
+ "name": "stderr",
741
+ "output_type": "stream",
742
+ "text": [
743
+ " "
744
+ ]
745
+ },
746
+ {
747
+ "name": "stdout",
748
+ "output_type": "stream",
749
+ "text": [
750
+ "Epoch [3/50], Loss: 1.0441, Acc: 63.76%\n"
751
+ ]
752
+ },
753
+ {
754
+ "name": "stderr",
755
+ "output_type": "stream",
756
+ "text": [
757
+ " "
758
+ ]
759
+ },
760
+ {
761
+ "name": "stdout",
762
+ "output_type": "stream",
763
+ "text": [
764
+ "Validation Loss: 0.8123, Acc: 69.91%\n",
765
+ "New best model saved with accuracy: 69.91%\n"
766
+ ]
767
+ },
768
+ {
769
+ "name": "stderr",
770
+ "output_type": "stream",
771
+ "text": [
772
+ " "
773
+ ]
774
+ },
775
+ {
776
+ "name": "stdout",
777
+ "output_type": "stream",
778
+ "text": [
779
+ "Epoch [4/50], Loss: 0.9982, Acc: 65.22%\n"
780
+ ]
781
+ },
782
+ {
783
+ "name": "stderr",
784
+ "output_type": "stream",
785
+ "text": [
786
+ " "
787
+ ]
788
+ },
789
+ {
790
+ "name": "stdout",
791
+ "output_type": "stream",
792
+ "text": [
793
+ "Validation Loss: 0.8204, Acc: 68.54%\n"
794
+ ]
795
+ },
796
+ {
797
+ "name": "stderr",
798
+ "output_type": "stream",
799
+ "text": [
800
+ " "
801
+ ]
802
+ },
803
+ {
804
+ "name": "stdout",
805
+ "output_type": "stream",
806
+ "text": [
807
+ "Epoch [5/50], Loss: 1.0101, Acc: 64.34%\n"
808
+ ]
809
+ },
810
+ {
811
+ "name": "stderr",
812
+ "output_type": "stream",
813
+ "text": [
814
+ " "
815
+ ]
816
+ },
817
+ {
818
+ "name": "stdout",
819
+ "output_type": "stream",
820
+ "text": [
821
+ "Validation Loss: 0.9953, Acc: 63.25%\n"
822
+ ]
823
+ },
824
+ {
825
+ "name": "stderr",
826
+ "output_type": "stream",
827
+ "text": [
828
+ " "
829
+ ]
830
+ },
831
+ {
832
+ "name": "stdout",
833
+ "output_type": "stream",
834
+ "text": [
835
+ "Epoch [6/50], Loss: 1.0382, Acc: 63.27%\n"
836
+ ]
837
+ },
838
+ {
839
+ "name": "stderr",
840
+ "output_type": "stream",
841
+ "text": [
842
+ " "
843
+ ]
844
+ },
845
+ {
846
+ "name": "stdout",
847
+ "output_type": "stream",
848
+ "text": [
849
+ "Validation Loss: 0.9688, Acc: 63.53%\n"
850
+ ]
851
+ },
852
+ {
853
+ "name": "stderr",
854
+ "output_type": "stream",
855
+ "text": [
856
+ " "
857
+ ]
858
+ },
859
+ {
860
+ "name": "stdout",
861
+ "output_type": "stream",
862
+ "text": [
863
+ "Epoch [7/50], Loss: 0.9866, Acc: 64.88%\n"
864
+ ]
865
+ },
866
+ {
867
+ "name": "stderr",
868
+ "output_type": "stream",
869
+ "text": [
870
+ " "
871
+ ]
872
+ },
873
+ {
874
+ "name": "stdout",
875
+ "output_type": "stream",
876
+ "text": [
877
+ "Validation Loss: 0.9508, Acc: 64.74%\n"
878
+ ]
879
+ },
880
+ {
881
+ "name": "stderr",
882
+ "output_type": "stream",
883
+ "text": [
884
+ " "
885
+ ]
886
+ },
887
+ {
888
+ "name": "stdout",
889
+ "output_type": "stream",
890
+ "text": [
891
+ "Epoch [8/50], Loss: 0.9614, Acc: 65.99%\n"
892
+ ]
893
+ },
894
+ {
895
+ "name": "stderr",
896
+ "output_type": "stream",
897
+ "text": [
898
+ " "
899
+ ]
900
+ },
901
+ {
902
+ "name": "stdout",
903
+ "output_type": "stream",
904
+ "text": [
905
+ "Validation Loss: 0.8994, Acc: 67.26%\n"
906
+ ]
907
+ },
908
+ {
909
+ "name": "stderr",
910
+ "output_type": "stream",
911
+ "text": [
912
+ " "
913
+ ]
914
+ },
915
+ {
916
+ "name": "stdout",
917
+ "output_type": "stream",
918
+ "text": [
919
+ "Epoch [9/50], Loss: 0.9348, Acc: 67.00%\n"
920
+ ]
921
+ },
922
+ {
923
+ "name": "stderr",
924
+ "output_type": "stream",
925
+ "text": [
926
+ " "
927
+ ]
928
+ },
929
+ {
930
+ "name": "stdout",
931
+ "output_type": "stream",
932
+ "text": [
933
+ "Validation Loss: 0.8908, Acc: 67.36%\n"
934
+ ]
935
+ },
936
+ {
937
+ "name": "stderr",
938
+ "output_type": "stream",
939
+ "text": [
940
+ " "
941
+ ]
942
+ },
943
+ {
944
+ "name": "stdout",
945
+ "output_type": "stream",
946
+ "text": [
947
+ "Epoch [10/50], Loss: 0.9170, Acc: 67.31%\n"
948
+ ]
949
+ },
950
+ {
951
+ "name": "stderr",
952
+ "output_type": "stream",
953
+ "text": [
954
+ " "
955
+ ]
956
+ },
957
+ {
958
+ "name": "stdout",
959
+ "output_type": "stream",
960
+ "text": [
961
+ "Validation Loss: 0.8961, Acc: 66.42%\n"
962
+ ]
963
+ },
964
+ {
965
+ "name": "stderr",
966
+ "output_type": "stream",
967
+ "text": [
968
+ " "
969
+ ]
970
+ },
971
+ {
972
+ "name": "stdout",
973
+ "output_type": "stream",
974
+ "text": [
975
+ "Epoch [11/50], Loss: 0.9211, Acc: 67.15%\n"
976
+ ]
977
+ },
978
+ {
979
+ "name": "stderr",
980
+ "output_type": "stream",
981
+ "text": [
982
+ " "
983
+ ]
984
+ },
985
+ {
986
+ "name": "stdout",
987
+ "output_type": "stream",
988
+ "text": [
989
+ "Validation Loss: 0.8167, Acc: 70.44%\n",
990
+ "New best model saved with accuracy: 70.44%\n"
991
+ ]
992
+ },
993
+ {
994
+ "name": "stderr",
995
+ "output_type": "stream",
996
+ "text": [
997
+ " "
998
+ ]
999
+ },
1000
+ {
1001
+ "name": "stdout",
1002
+ "output_type": "stream",
1003
+ "text": [
1004
+ "Epoch [12/50], Loss: 0.8618, Acc: 69.12%\n"
1005
+ ]
1006
+ },
1007
+ {
1008
+ "name": "stderr",
1009
+ "output_type": "stream",
1010
+ "text": [
1011
+ " "
1012
+ ]
1013
+ },
1014
+ {
1015
+ "name": "stdout",
1016
+ "output_type": "stream",
1017
+ "text": [
1018
+ "Validation Loss: 0.7712, Acc: 70.79%\n",
1019
+ "New best model saved with accuracy: 70.79%\n"
1020
+ ]
1021
+ },
1022
+ {
1023
+ "name": "stderr",
1024
+ "output_type": "stream",
1025
+ "text": [
1026
+ " "
1027
+ ]
1028
+ },
1029
+ {
1030
+ "name": "stdout",
1031
+ "output_type": "stream",
1032
+ "text": [
1033
+ "Epoch [13/50], Loss: 0.8535, Acc: 69.34%\n"
1034
+ ]
1035
+ },
1036
+ {
1037
+ "name": "stderr",
1038
+ "output_type": "stream",
1039
+ "text": [
1040
+ " "
1041
+ ]
1042
+ },
1043
+ {
1044
+ "name": "stdout",
1045
+ "output_type": "stream",
1046
+ "text": [
1047
+ "Validation Loss: 0.8438, Acc: 68.26%\n"
1048
+ ]
1049
+ },
1050
+ {
1051
+ "name": "stderr",
1052
+ "output_type": "stream",
1053
+ "text": [
1054
+ " "
1055
+ ]
1056
+ },
1057
+ {
1058
+ "name": "stdout",
1059
+ "output_type": "stream",
1060
+ "text": [
1061
+ "Epoch [14/50], Loss: 0.8238, Acc: 70.29%\n"
1062
+ ]
1063
+ },
1064
+ {
1065
+ "name": "stderr",
1066
+ "output_type": "stream",
1067
+ "text": [
1068
+ " "
1069
+ ]
1070
+ },
1071
+ {
1072
+ "name": "stdout",
1073
+ "output_type": "stream",
1074
+ "text": [
1075
+ "Validation Loss: 0.7468, Acc: 72.20%\n",
1076
+ "New best model saved with accuracy: 72.20%\n"
1077
+ ]
1078
+ },
1079
+ {
1080
+ "name": "stderr",
1081
+ "output_type": "stream",
1082
+ "text": [
1083
+ " "
1084
+ ]
1085
+ },
1086
+ {
1087
+ "name": "stdout",
1088
+ "output_type": "stream",
1089
+ "text": [
1090
+ "Epoch [15/50], Loss: 0.7955, Acc: 71.32%\n"
1091
+ ]
1092
+ },
1093
+ {
1094
+ "name": "stderr",
1095
+ "output_type": "stream",
1096
+ "text": [
1097
+ " "
1098
+ ]
1099
+ },
1100
+ {
1101
+ "name": "stdout",
1102
+ "output_type": "stream",
1103
+ "text": [
1104
+ "Validation Loss: 0.7421, Acc: 72.94%\n",
1105
+ "New best model saved with accuracy: 72.94%\n"
1106
+ ]
1107
+ },
1108
+ {
1109
+ "name": "stderr",
1110
+ "output_type": "stream",
1111
+ "text": [
1112
+ " "
1113
+ ]
1114
+ },
1115
+ {
1116
+ "name": "stdout",
1117
+ "output_type": "stream",
1118
+ "text": [
1119
+ "Epoch [16/50], Loss: 0.7659, Acc: 72.32%\n"
1120
+ ]
1121
+ },
1122
+ {
1123
+ "name": "stderr",
1124
+ "output_type": "stream",
1125
+ "text": [
1126
+ " "
1127
+ ]
1128
+ },
1129
+ {
1130
+ "name": "stdout",
1131
+ "output_type": "stream",
1132
+ "text": [
1133
+ "Validation Loss: 0.8212, Acc: 69.08%\n"
1134
+ ]
1135
+ },
1136
+ {
1137
+ "name": "stderr",
1138
+ "output_type": "stream",
1139
+ "text": [
1140
+ " "
1141
+ ]
1142
+ },
1143
+ {
1144
+ "name": "stdout",
1145
+ "output_type": "stream",
1146
+ "text": [
1147
+ "Epoch [17/50], Loss: 0.7439, Acc: 72.93%\n"
1148
+ ]
1149
+ },
1150
+ {
1151
+ "name": "stderr",
1152
+ "output_type": "stream",
1153
+ "text": [
1154
+ " "
1155
+ ]
1156
+ },
1157
+ {
1158
+ "name": "stdout",
1159
+ "output_type": "stream",
1160
+ "text": [
1161
+ "Validation Loss: 0.7595, Acc: 72.08%\n"
1162
+ ]
1163
+ },
1164
+ {
1165
+ "name": "stderr",
1166
+ "output_type": "stream",
1167
+ "text": [
1168
+ " "
1169
+ ]
1170
+ },
1171
+ {
1172
+ "name": "stdout",
1173
+ "output_type": "stream",
1174
+ "text": [
1175
+ "Epoch [18/50], Loss: 0.7126, Acc: 73.81%\n"
1176
+ ]
1177
+ },
1178
+ {
1179
+ "name": "stderr",
1180
+ "output_type": "stream",
1181
+ "text": [
1182
+ " "
1183
+ ]
1184
+ },
1185
+ {
1186
+ "name": "stdout",
1187
+ "output_type": "stream",
1188
+ "text": [
1189
+ "Validation Loss: 0.7728, Acc: 72.69%\n"
1190
+ ]
1191
+ },
1192
+ {
1193
+ "name": "stderr",
1194
+ "output_type": "stream",
1195
+ "text": [
1196
+ " "
1197
+ ]
1198
+ },
1199
+ {
1200
+ "name": "stdout",
1201
+ "output_type": "stream",
1202
+ "text": [
1203
+ "Epoch [19/50], Loss: 0.6794, Acc: 75.10%\n"
1204
+ ]
1205
+ },
1206
+ {
1207
+ "name": "stderr",
1208
+ "output_type": "stream",
1209
+ "text": [
1210
+ " "
1211
+ ]
1212
+ },
1213
+ {
1214
+ "name": "stdout",
1215
+ "output_type": "stream",
1216
+ "text": [
1217
+ "Validation Loss: 0.7778, Acc: 71.33%\n"
1218
+ ]
1219
+ },
1220
+ {
1221
+ "name": "stderr",
1222
+ "output_type": "stream",
1223
+ "text": [
1224
+ " "
1225
+ ]
1226
+ },
1227
+ {
1228
+ "name": "stdout",
1229
+ "output_type": "stream",
1230
+ "text": [
1231
+ "Epoch [20/50], Loss: 0.6500, Acc: 76.01%\n"
1232
+ ]
1233
+ },
1234
+ {
1235
+ "name": "stderr",
1236
+ "output_type": "stream",
1237
+ "text": [
1238
+ " "
1239
+ ]
1240
+ },
1241
+ {
1242
+ "name": "stdout",
1243
+ "output_type": "stream",
1244
+ "text": [
1245
+ "Validation Loss: 0.7097, Acc: 74.52%\n",
1246
+ "New best model saved with accuracy: 74.52%\n"
1247
+ ]
1248
+ },
1249
+ {
1250
+ "name": "stderr",
1251
+ "output_type": "stream",
1252
+ "text": [
1253
+ " "
1254
+ ]
1255
+ },
1256
+ {
1257
+ "name": "stdout",
1258
+ "output_type": "stream",
1259
+ "text": [
1260
+ "Epoch [21/50], Loss: 0.6196, Acc: 76.85%\n"
1261
+ ]
1262
+ },
1263
+ {
1264
+ "name": "stderr",
1265
+ "output_type": "stream",
1266
+ "text": [
1267
+ " "
1268
+ ]
1269
+ },
1270
+ {
1271
+ "name": "stdout",
1272
+ "output_type": "stream",
1273
+ "text": [
1274
+ "Validation Loss: 0.6960, Acc: 75.07%\n",
1275
+ "New best model saved with accuracy: 75.07%\n"
1276
+ ]
1277
+ },
1278
+ {
1279
+ "name": "stderr",
1280
+ "output_type": "stream",
1281
+ "text": [
1282
+ " "
1283
+ ]
1284
+ },
1285
+ {
1286
+ "name": "stdout",
1287
+ "output_type": "stream",
1288
+ "text": [
1289
+ "Epoch [22/50], Loss: 0.6020, Acc: 77.35%\n"
1290
+ ]
1291
+ },
1292
+ {
1293
+ "name": "stderr",
1294
+ "output_type": "stream",
1295
+ "text": [
1296
+ " "
1297
+ ]
1298
+ },
1299
+ {
1300
+ "name": "stdout",
1301
+ "output_type": "stream",
1302
+ "text": [
1303
+ "Validation Loss: 0.7521, Acc: 73.93%\n"
1304
+ ]
1305
+ },
1306
+ {
1307
+ "name": "stderr",
1308
+ "output_type": "stream",
1309
+ "text": [
1310
+ " "
1311
+ ]
1312
+ },
1313
+ {
1314
+ "name": "stdout",
1315
+ "output_type": "stream",
1316
+ "text": [
1317
+ "Epoch [23/50], Loss: 0.5602, Acc: 78.87%\n"
1318
+ ]
1319
+ },
1320
+ {
1321
+ "name": "stderr",
1322
+ "output_type": "stream",
1323
+ "text": [
1324
+ " "
1325
+ ]
1326
+ },
1327
+ {
1328
+ "name": "stdout",
1329
+ "output_type": "stream",
1330
+ "text": [
1331
+ "Validation Loss: 0.7254, Acc: 74.60%\n"
1332
+ ]
1333
+ },
1334
+ {
1335
+ "name": "stderr",
1336
+ "output_type": "stream",
1337
+ "text": [
1338
+ " "
1339
+ ]
1340
+ },
1341
+ {
1342
+ "name": "stdout",
1343
+ "output_type": "stream",
1344
+ "text": [
1345
+ "Epoch [24/50], Loss: 0.5343, Acc: 79.65%\n"
1346
+ ]
1347
+ },
1348
+ {
1349
+ "name": "stderr",
1350
+ "output_type": "stream",
1351
+ "text": [
1352
+ " "
1353
+ ]
1354
+ },
1355
+ {
1356
+ "name": "stdout",
1357
+ "output_type": "stream",
1358
+ "text": [
1359
+ "Validation Loss: 0.6848, Acc: 77.01%\n",
1360
+ "New best model saved with accuracy: 77.01%\n"
1361
+ ]
1362
+ },
1363
+ {
1364
+ "name": "stderr",
1365
+ "output_type": "stream",
1366
+ "text": [
1367
+ " "
1368
+ ]
1369
+ },
1370
+ {
1371
+ "name": "stdout",
1372
+ "output_type": "stream",
1373
+ "text": [
1374
+ "Epoch [25/50], Loss: 0.5009, Acc: 80.83%\n"
1375
+ ]
1376
+ },
1377
+ {
1378
+ "name": "stderr",
1379
+ "output_type": "stream",
1380
+ "text": [
1381
+ " "
1382
+ ]
1383
+ },
1384
+ {
1385
+ "name": "stdout",
1386
+ "output_type": "stream",
1387
+ "text": [
1388
+ "Validation Loss: 0.6942, Acc: 75.85%\n"
1389
+ ]
1390
+ },
1391
+ {
1392
+ "name": "stderr",
1393
+ "output_type": "stream",
1394
+ "text": [
1395
+ " "
1396
+ ]
1397
+ },
1398
+ {
1399
+ "name": "stdout",
1400
+ "output_type": "stream",
1401
+ "text": [
1402
+ "Epoch [26/50], Loss: 0.4729, Acc: 81.98%\n"
1403
+ ]
1404
+ },
1405
+ {
1406
+ "name": "stderr",
1407
+ "output_type": "stream",
1408
+ "text": [
1409
+ " "
1410
+ ]
1411
+ },
1412
+ {
1413
+ "name": "stdout",
1414
+ "output_type": "stream",
1415
+ "text": [
1416
+ "Validation Loss: 0.7240, Acc: 76.18%\n"
1417
+ ]
1418
+ },
1419
+ {
1420
+ "name": "stderr",
1421
+ "output_type": "stream",
1422
+ "text": [
1423
+ " "
1424
+ ]
1425
+ },
1426
+ {
1427
+ "name": "stdout",
1428
+ "output_type": "stream",
1429
+ "text": [
1430
+ "Epoch [27/50], Loss: 0.4450, Acc: 82.59%\n"
1431
+ ]
1432
+ },
1433
+ {
1434
+ "name": "stderr",
1435
+ "output_type": "stream",
1436
+ "text": [
1437
+ " "
1438
+ ]
1439
+ },
1440
+ {
1441
+ "name": "stdout",
1442
+ "output_type": "stream",
1443
+ "text": [
1444
+ "Validation Loss: 0.7061, Acc: 76.64%\n"
1445
+ ]
1446
+ },
1447
+ {
1448
+ "name": "stderr",
1449
+ "output_type": "stream",
1450
+ "text": [
1451
+ " "
1452
+ ]
1453
+ },
1454
+ {
1455
+ "name": "stdout",
1456
+ "output_type": "stream",
1457
+ "text": [
1458
+ "Epoch [28/50], Loss: 0.4161, Acc: 83.64%\n"
1459
+ ]
1460
+ },
1461
+ {
1462
+ "name": "stderr",
1463
+ "output_type": "stream",
1464
+ "text": [
1465
+ " "
1466
+ ]
1467
+ },
1468
+ {
1469
+ "name": "stdout",
1470
+ "output_type": "stream",
1471
+ "text": [
1472
+ "Validation Loss: 0.7243, Acc: 76.22%\n"
1473
+ ]
1474
+ },
1475
+ {
1476
+ "name": "stderr",
1477
+ "output_type": "stream",
1478
+ "text": [
1479
+ " "
1480
+ ]
1481
+ },
1482
+ {
1483
+ "name": "stdout",
1484
+ "output_type": "stream",
1485
+ "text": [
1486
+ "Epoch [29/50], Loss: 0.3894, Acc: 84.72%\n"
1487
+ ]
1488
+ },
1489
+ {
1490
+ "name": "stderr",
1491
+ "output_type": "stream",
1492
+ "text": [
1493
+ " "
1494
+ ]
1495
+ },
1496
+ {
1497
+ "name": "stdout",
1498
+ "output_type": "stream",
1499
+ "text": [
1500
+ "Validation Loss: 0.7221, Acc: 77.03%\n",
1501
+ "New best model saved with accuracy: 77.03%\n"
1502
+ ]
1503
+ },
1504
+ {
1505
+ "name": "stderr",
1506
+ "output_type": "stream",
1507
+ "text": [
1508
+ " "
1509
+ ]
1510
+ },
1511
+ {
1512
+ "name": "stdout",
1513
+ "output_type": "stream",
1514
+ "text": [
1515
+ "Epoch [30/50], Loss: 0.3514, Acc: 86.06%\n"
1516
+ ]
1517
+ },
1518
+ {
1519
+ "name": "stderr",
1520
+ "output_type": "stream",
1521
+ "text": [
1522
+ " "
1523
+ ]
1524
+ },
1525
+ {
1526
+ "name": "stdout",
1527
+ "output_type": "stream",
1528
+ "text": [
1529
+ "Validation Loss: 0.7121, Acc: 78.37%\n",
1530
+ "New best model saved with accuracy: 78.37%\n"
1531
+ ]
1532
+ },
1533
+ {
1534
+ "name": "stderr",
1535
+ "output_type": "stream",
1536
+ "text": [
1537
+ " "
1538
+ ]
1539
+ },
1540
+ {
1541
+ "name": "stdout",
1542
+ "output_type": "stream",
1543
+ "text": [
1544
+ "Epoch [31/50], Loss: 0.3326, Acc: 86.66%\n"
1545
+ ]
1546
+ },
1547
+ {
1548
+ "name": "stderr",
1549
+ "output_type": "stream",
1550
+ "text": [
1551
+ " "
1552
+ ]
1553
+ },
1554
+ {
1555
+ "name": "stdout",
1556
+ "output_type": "stream",
1557
+ "text": [
1558
+ "Validation Loss: 0.7321, Acc: 77.68%\n"
1559
+ ]
1560
+ },
1561
+ {
1562
+ "name": "stderr",
1563
+ "output_type": "stream",
1564
+ "text": [
1565
+ " "
1566
+ ]
1567
+ },
1568
+ {
1569
+ "name": "stdout",
1570
+ "output_type": "stream",
1571
+ "text": [
1572
+ "Epoch [32/50], Loss: 0.3043, Acc: 87.87%\n"
1573
+ ]
1574
+ },
1575
+ {
1576
+ "name": "stderr",
1577
+ "output_type": "stream",
1578
+ "text": [
1579
+ " "
1580
+ ]
1581
+ },
1582
+ {
1583
+ "name": "stdout",
1584
+ "output_type": "stream",
1585
+ "text": [
1586
+ "Validation Loss: 0.7001, Acc: 80.12%\n",
1587
+ "New best model saved with accuracy: 80.12%\n"
1588
+ ]
1589
+ },
1590
+ {
1591
+ "name": "stderr",
1592
+ "output_type": "stream",
1593
+ "text": [
1594
+ " "
1595
+ ]
1596
+ },
1597
+ {
1598
+ "name": "stdout",
1599
+ "output_type": "stream",
1600
+ "text": [
1601
+ "Epoch [33/50], Loss: 0.2952, Acc: 88.21%\n"
1602
+ ]
1603
+ },
1604
+ {
1605
+ "name": "stderr",
1606
+ "output_type": "stream",
1607
+ "text": [
1608
+ " "
1609
+ ]
1610
+ },
1611
+ {
1612
+ "name": "stdout",
1613
+ "output_type": "stream",
1614
+ "text": [
1615
+ "Validation Loss: 0.7658, Acc: 78.04%\n"
1616
+ ]
1617
+ },
1618
+ {
1619
+ "name": "stderr",
1620
+ "output_type": "stream",
1621
+ "text": [
1622
+ " "
1623
+ ]
1624
+ },
1625
+ {
1626
+ "name": "stdout",
1627
+ "output_type": "stream",
1628
+ "text": [
1629
+ "Epoch [34/50], Loss: 0.2709, Acc: 89.20%\n"
1630
+ ]
1631
+ },
1632
+ {
1633
+ "name": "stderr",
1634
+ "output_type": "stream",
1635
+ "text": [
1636
+ " "
1637
+ ]
1638
+ },
1639
+ {
1640
+ "name": "stdout",
1641
+ "output_type": "stream",
1642
+ "text": [
1643
+ "Validation Loss: 0.7202, Acc: 79.50%\n"
1644
+ ]
1645
+ },
1646
+ {
1647
+ "name": "stderr",
1648
+ "output_type": "stream",
1649
+ "text": [
1650
+ " "
1651
+ ]
1652
+ },
1653
+ {
1654
+ "name": "stdout",
1655
+ "output_type": "stream",
1656
+ "text": [
1657
+ "Epoch [35/50], Loss: 0.2688, Acc: 89.49%\n"
1658
+ ]
1659
+ },
1660
+ {
1661
+ "name": "stderr",
1662
+ "output_type": "stream",
1663
+ "text": [
1664
+ " "
1665
+ ]
1666
+ },
1667
+ {
1668
+ "name": "stdout",
1669
+ "output_type": "stream",
1670
+ "text": [
1671
+ "Validation Loss: 0.7348, Acc: 79.95%\n"
1672
+ ]
1673
+ },
1674
+ {
1675
+ "name": "stderr",
1676
+ "output_type": "stream",
1677
+ "text": [
1678
+ " "
1679
+ ]
1680
+ },
1681
+ {
1682
+ "name": "stdout",
1683
+ "output_type": "stream",
1684
+ "text": [
1685
+ "Epoch [36/50], Loss: 0.2305, Acc: 90.76%\n"
1686
+ ]
1687
+ },
1688
+ {
1689
+ "name": "stderr",
1690
+ "output_type": "stream",
1691
+ "text": [
1692
+ " "
1693
+ ]
1694
+ },
1695
+ {
1696
+ "name": "stdout",
1697
+ "output_type": "stream",
1698
+ "text": [
1699
+ "Validation Loss: 0.7410, Acc: 80.30%\n",
1700
+ "New best model saved with accuracy: 80.30%\n"
1701
+ ]
1702
+ },
1703
+ {
1704
+ "name": "stderr",
1705
+ "output_type": "stream",
1706
+ "text": [
1707
+ " "
1708
+ ]
1709
+ },
1710
+ {
1711
+ "name": "stdout",
1712
+ "output_type": "stream",
1713
+ "text": [
1714
+ "Epoch [37/50], Loss: 0.2143, Acc: 91.46%\n"
1715
+ ]
1716
+ },
1717
+ {
1718
+ "name": "stderr",
1719
+ "output_type": "stream",
1720
+ "text": [
1721
+ " "
1722
+ ]
1723
+ },
1724
+ {
1725
+ "name": "stdout",
1726
+ "output_type": "stream",
1727
+ "text": [
1728
+ "Validation Loss: 0.7574, Acc: 80.36%\n",
1729
+ "New best model saved with accuracy: 80.36%\n"
1730
+ ]
1731
+ },
1732
+ {
1733
+ "name": "stderr",
1734
+ "output_type": "stream",
1735
+ "text": [
1736
+ " "
1737
+ ]
1738
+ },
1739
+ {
1740
+ "name": "stdout",
1741
+ "output_type": "stream",
1742
+ "text": [
1743
+ "Epoch [38/50], Loss: 0.1931, Acc: 92.13%\n"
1744
+ ]
1745
+ },
1746
+ {
1747
+ "name": "stderr",
1748
+ "output_type": "stream",
1749
+ "text": [
1750
+ " "
1751
+ ]
1752
+ },
1753
+ {
1754
+ "name": "stdout",
1755
+ "output_type": "stream",
1756
+ "text": [
1757
+ "Validation Loss: 0.7907, Acc: 79.33%\n"
1758
+ ]
1759
+ },
1760
+ {
1761
+ "name": "stderr",
1762
+ "output_type": "stream",
1763
+ "text": [
1764
+ " "
1765
+ ]
1766
+ },
1767
+ {
1768
+ "name": "stdout",
1769
+ "output_type": "stream",
1770
+ "text": [
1771
+ "Epoch [39/50], Loss: 0.1750, Acc: 92.91%\n"
1772
+ ]
1773
+ },
1774
+ {
1775
+ "name": "stderr",
1776
+ "output_type": "stream",
1777
+ "text": [
1778
+ " "
1779
+ ]
1780
+ },
1781
+ {
1782
+ "name": "stdout",
1783
+ "output_type": "stream",
1784
+ "text": [
1785
+ "Validation Loss: 0.8297, Acc: 80.44%\n",
1786
+ "New best model saved with accuracy: 80.44%\n"
1787
+ ]
1788
+ },
1789
+ {
1790
+ "name": "stderr",
1791
+ "output_type": "stream",
1792
+ "text": [
1793
+ " "
1794
+ ]
1795
+ },
1796
+ {
1797
+ "name": "stdout",
1798
+ "output_type": "stream",
1799
+ "text": [
1800
+ "Epoch [40/50], Loss: 0.1695, Acc: 93.31%\n"
1801
+ ]
1802
+ },
1803
+ {
1804
+ "name": "stderr",
1805
+ "output_type": "stream",
1806
+ "text": [
1807
+ " "
1808
+ ]
1809
+ },
1810
+ {
1811
+ "name": "stdout",
1812
+ "output_type": "stream",
1813
+ "text": [
1814
+ "Validation Loss: 0.8326, Acc: 80.50%\n",
1815
+ "New best model saved with accuracy: 80.50%\n"
1816
+ ]
1817
+ },
1818
+ {
1819
+ "name": "stderr",
1820
+ "output_type": "stream",
1821
+ "text": [
1822
+ " "
1823
+ ]
1824
+ },
1825
+ {
1826
+ "name": "stdout",
1827
+ "output_type": "stream",
1828
+ "text": [
1829
+ "Epoch [41/50], Loss: 0.1556, Acc: 93.82%\n"
1830
+ ]
1831
+ },
1832
+ {
1833
+ "name": "stderr",
1834
+ "output_type": "stream",
1835
+ "text": [
1836
+ " "
1837
+ ]
1838
+ },
1839
+ {
1840
+ "name": "stdout",
1841
+ "output_type": "stream",
1842
+ "text": [
1843
+ "Validation Loss: 0.8221, Acc: 80.55%\n",
1844
+ "New best model saved with accuracy: 80.55%\n"
1845
+ ]
1846
+ },
1847
+ {
1848
+ "name": "stderr",
1849
+ "output_type": "stream",
1850
+ "text": [
1851
+ " "
1852
+ ]
1853
+ },
1854
+ {
1855
+ "name": "stdout",
1856
+ "output_type": "stream",
1857
+ "text": [
1858
+ "Epoch [42/50], Loss: 0.1390, Acc: 94.33%\n"
1859
+ ]
1860
+ },
1861
+ {
1862
+ "name": "stderr",
1863
+ "output_type": "stream",
1864
+ "text": [
1865
+ " "
1866
+ ]
1867
+ },
1868
+ {
1869
+ "name": "stdout",
1870
+ "output_type": "stream",
1871
+ "text": [
1872
+ "Validation Loss: 0.8562, Acc: 80.65%\n",
1873
+ "New best model saved with accuracy: 80.65%\n"
1874
+ ]
1875
+ },
1876
+ {
1877
+ "name": "stderr",
1878
+ "output_type": "stream",
1879
+ "text": [
1880
+ " "
1881
+ ]
1882
+ },
1883
+ {
1884
+ "name": "stdout",
1885
+ "output_type": "stream",
1886
+ "text": [
1887
+ "Epoch [43/50], Loss: 0.1334, Acc: 94.76%\n"
1888
+ ]
1889
+ },
1890
+ {
1891
+ "name": "stderr",
1892
+ "output_type": "stream",
1893
+ "text": [
1894
+ " "
1895
+ ]
1896
+ },
1897
+ {
1898
+ "name": "stdout",
1899
+ "output_type": "stream",
1900
+ "text": [
1901
+ "Validation Loss: 0.8463, Acc: 80.92%\n",
1902
+ "New best model saved with accuracy: 80.92%\n"
1903
+ ]
1904
+ },
1905
+ {
1906
+ "name": "stderr",
1907
+ "output_type": "stream",
1908
+ "text": [
1909
+ " "
1910
+ ]
1911
+ },
1912
+ {
1913
+ "name": "stdout",
1914
+ "output_type": "stream",
1915
+ "text": [
1916
+ "Epoch [44/50], Loss: 0.1305, Acc: 94.73%\n"
1917
+ ]
1918
+ },
1919
+ {
1920
+ "name": "stderr",
1921
+ "output_type": "stream",
1922
+ "text": [
1923
+ " "
1924
+ ]
1925
+ },
1926
+ {
1927
+ "name": "stdout",
1928
+ "output_type": "stream",
1929
+ "text": [
1930
+ "Validation Loss: 0.8475, Acc: 80.89%\n"
1931
+ ]
1932
+ },
1933
+ {
1934
+ "name": "stderr",
1935
+ "output_type": "stream",
1936
+ "text": [
1937
+ " "
1938
+ ]
1939
+ },
1940
+ {
1941
+ "name": "stdout",
1942
+ "output_type": "stream",
1943
+ "text": [
1944
+ "Epoch [45/50], Loss: 0.1145, Acc: 95.42%\n"
1945
+ ]
1946
+ },
1947
+ {
1948
+ "name": "stderr",
1949
+ "output_type": "stream",
1950
+ "text": [
1951
+ " "
1952
+ ]
1953
+ },
1954
+ {
1955
+ "name": "stdout",
1956
+ "output_type": "stream",
1957
+ "text": [
1958
+ "Validation Loss: 0.8956, Acc: 80.97%\n",
1959
+ "New best model saved with accuracy: 80.97%\n"
1960
+ ]
1961
+ },
1962
+ {
1963
+ "name": "stderr",
1964
+ "output_type": "stream",
1965
+ "text": [
1966
+ " "
1967
+ ]
1968
+ },
1969
+ {
1970
+ "name": "stdout",
1971
+ "output_type": "stream",
1972
+ "text": [
1973
+ "Epoch [46/50], Loss: 0.1095, Acc: 95.56%\n"
1974
+ ]
1975
+ },
1976
+ {
1977
+ "name": "stderr",
1978
+ "output_type": "stream",
1979
+ "text": [
1980
+ " "
1981
+ ]
1982
+ },
1983
+ {
1984
+ "name": "stdout",
1985
+ "output_type": "stream",
1986
+ "text": [
1987
+ "Validation Loss: 0.8992, Acc: 81.41%\n",
1988
+ "New best model saved with accuracy: 81.41%\n"
1989
+ ]
1990
+ },
1991
+ {
1992
+ "name": "stderr",
1993
+ "output_type": "stream",
1994
+ "text": [
1995
+ " "
1996
+ ]
1997
+ },
1998
+ {
1999
+ "name": "stdout",
2000
+ "output_type": "stream",
2001
+ "text": [
2002
+ "Epoch [47/50], Loss: 0.1039, Acc: 95.84%\n"
2003
+ ]
2004
+ },
2005
+ {
2006
+ "name": "stderr",
2007
+ "output_type": "stream",
2008
+ "text": [
2009
+ " "
2010
+ ]
2011
+ },
2012
+ {
2013
+ "name": "stdout",
2014
+ "output_type": "stream",
2015
+ "text": [
2016
+ "Validation Loss: 0.8921, Acc: 81.18%\n"
2017
+ ]
2018
+ },
2019
+ {
2020
+ "name": "stderr",
2021
+ "output_type": "stream",
2022
+ "text": [
2023
+ " "
2024
+ ]
2025
+ },
2026
+ {
2027
+ "name": "stdout",
2028
+ "output_type": "stream",
2029
+ "text": [
2030
+ "Epoch [48/50], Loss: 0.0988, Acc: 96.01%\n"
2031
+ ]
2032
+ },
2033
+ {
2034
+ "name": "stderr",
2035
+ "output_type": "stream",
2036
+ "text": [
2037
+ " "
2038
+ ]
2039
+ },
2040
+ {
2041
+ "name": "stdout",
2042
+ "output_type": "stream",
2043
+ "text": [
2044
+ "Validation Loss: 0.8984, Acc: 81.24%\n"
2045
+ ]
2046
+ },
2047
+ {
2048
+ "name": "stderr",
2049
+ "output_type": "stream",
2050
+ "text": [
2051
+ " "
2052
+ ]
2053
+ },
2054
+ {
2055
+ "name": "stdout",
2056
+ "output_type": "stream",
2057
+ "text": [
2058
+ "Epoch [49/50], Loss: 0.0942, Acc: 96.17%\n"
2059
+ ]
2060
+ },
2061
+ {
2062
+ "name": "stderr",
2063
+ "output_type": "stream",
2064
+ "text": [
2065
+ " "
2066
+ ]
2067
+ },
2068
+ {
2069
+ "name": "stdout",
2070
+ "output_type": "stream",
2071
+ "text": [
2072
+ "Validation Loss: 0.9172, Acc: 81.13%\n"
2073
+ ]
2074
+ },
2075
+ {
2076
+ "name": "stderr",
2077
+ "output_type": "stream",
2078
+ "text": [
2079
+ " "
2080
+ ]
2081
+ },
2082
+ {
2083
+ "name": "stdout",
2084
+ "output_type": "stream",
2085
+ "text": [
2086
+ "Epoch [50/50], Loss: 0.0896, Acc: 96.47%\n"
2087
+ ]
2088
+ },
2089
+ {
2090
+ "name": "stderr",
2091
+ "output_type": "stream",
2092
+ "text": [
2093
+ "Epoch 50/50 [Validation]: 75%|███████▌ | 40/53 [00:03<00:00, 19.44it/s]"
2094
+ ]
2095
+ }
2096
+ ],
2097
+ "source": [
2098
+ "best_acc = 0\n",
2099
+ "for epoch in range(num_epochs):\n",
2100
+ " train_one_epoch(epoch)\n",
2101
+ " val_acc = validate(epoch)\n",
2102
+ "\n",
2103
+ " # Scheduler step\n",
2104
+ " scheduler.step(epoch+1)\n",
2105
+ "\n",
2106
+ " # Save best model\n",
2107
+ " if val_acc > best_acc:\n",
2108
+ " best_acc = val_acc\n",
2109
+ " os.makedirs('./models', exist_ok=True)\n",
2110
+ " # If using DataParallel, save the underlying model\n",
2111
+ " if isinstance(model, nn.DataParallel):\n",
2112
+ " torch.save(model.module.state_dict(), './models/best_vit_tiny_imagenet.pth')\n",
2113
+ " else:\n",
2114
+ " torch.save(model.state_dict(), './models/best_vit_tiny_imagenet.pth')\n",
2115
+ " print(f\"New best model saved with accuracy: {best_acc:.2f}%\")\n",
2116
+ "\n",
2117
+ "print(\"Training complete. Best validation accuracy:\", best_acc)"
2118
+ ]
2119
+ },
2120
+ {
2121
+ "cell_type": "code",
2122
+ "execution_count": null,
2123
+ "metadata": {},
2124
+ "outputs": [],
2125
+ "source": []
2126
+ }
2127
+ ],
2128
+ "metadata": {
2129
+ "accelerator": "GPU",
2130
+ "colab": {
2131
+ "gpuType": "T4",
2132
+ "provenance": []
2133
+ },
2134
+ "kernelspec": {
2135
+ "display_name": "Python3 (main venv)",
2136
+ "language": "python",
2137
+ "name": "main"
2138
+ },
2139
+ "language_info": {
2140
+ "codemirror_mode": {
2141
+ "name": "ipython",
2142
+ "version": 3
2143
+ },
2144
+ "file_extension": ".py",
2145
+ "mimetype": "text/x-python",
2146
+ "name": "python",
2147
+ "nbconvert_exporter": "python",
2148
+ "pygments_lexer": "ipython3",
2149
+ "version": "3.12.12"
2150
+ }
2151
+ },
2152
+ "nbformat": 4,
2153
+ "nbformat_minor": 4
2154
+ }
events.out.tfevents.1772175113.b6936449dab3.1207.0 ADDED
Binary file (9.88 kB). View file
 
model_training_results.csv ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Epoch,Model1_Train_Loss,Model1_Val_Acc,Model2_Train_Loss,Model2_Val_Acc
2
+ 1,,35.74,2.0360705567318855,22.568616724384142
3
+ 2,1.1986,64.96,1.9792786671214126,28.714201857764852
4
+ 3,1.0441,69.91,1.7821412383084092,52.3717207174597
5
+ 4,0.9982,68.54,1.6183974135435368,61.03818857129263
6
+ 5,1.0101,63.25,1.5248827010250547,61.7989283324995
7
+ 6,1.0382,63.53,1.4669698419753445,67.19869004684303
8
+ 7,0.9866,64.74,1.4290498651956256,70.68914373393275
9
+ 8,0.9614,67.26,1.3899061232092278,68.09367889549965
10
+ 9,0.9348,67.36,1.3519616517724033,72.85203231661302
11
+ 10,0.917,66.42,1.327325906479758,72.13604118488284
12
+ 11,0.9211,70.44,1.2994008415053335,72.52386902966192
13
+ 12,0.8618,70.79,1.274697577554073,75.3430823016565
14
+ 13,0.8535,68.26,1.2592898128135352,73.29952621459961
15
+ 14,0.8238,72.2,1.237077219064156,74.82100560727723
16
+ 15,0.7955,72.94,1.2224091088942934,75.82040918527754
17
+ 16,0.7659,69.08,1.2057472752612173,75.83532527295388
18
+ 17,0.7439,72.08,1.1897126553161292,76.04415603014051
19
+ 18,0.7126,72.69,1.176309052836952,76.1485714832752
20
+ 19,0.6794,71.33,1.1633472284346675,77.08830949059534
21
+ 20,0.65,74.52,1.145757243251116,78.43079130063478
22
+ 21,0.6196,75.07,1.1302614176387422,78.0429628549726
23
+ 22,0.602,73.93,1.1211656371942547,78.62470563328635
24
+ 23,0.5602,74.6,1.1059313349461442,78.53520640439235
25
+ 24,0.5343,77.01,1.0975406804438412,77.50597055658235
26
+ 25,0.5009,75.85,1.084783926392286,78.96778403830699
27
+ 26,0.4729,76.18,1.081091641239002,78.60978895838335
28
+ 27,0.445,76.64,1.066453042212856,78.5799557747306
29
+ 28,0.4161,76.22,1.0537544052566639,78.66945435380595
30
+ 29,0.3894,77.03,1.0435457479155235,79.05728320688507
31
+ 30,0.3514,78.37,1.0366827436326223,79.53460898205887
32
+ 31,0.3326,77.68,1.0267470826942955,79.54952623280819
33
+ 32,0.3043,80.12,1.0243510783574228,79.99702059738961
34
+ 33,0.2952,78.04,1.0170651106743152,79.59427570898379
35
+ 34,0.2709,79.5,1.0055033556011874,80.08651905469507
36
+ 35,0.2688,79.95,1.0005212784098667,80.38484843306439
37
+ 36,0.2305,80.3,0.9912698331632113,80.41468205144696
38
+ 37,0.2143,80.36,0.984084892501101,79.9671874558442
39
+ 38,0.1931,79.33,0.9812082930615074,80.16110167810626
40
+ 39,0.175,80.44,0.9815773544699381,80.54892974025026
41
+ 40,0.1695,80.5,0.9741421825292578,80.45943074351564
42
+ 41,0.1556,80.55,0.9691943751567859,80.65334544261486
43
+ 42,0.139,80.65,0.9659411455170389,80.66826100907063
44
+ 43,0.1334,80.92,0.9666519684084295,80.47434704172868
45
+ 44,0.1305,80.89,0.9551008048525267,80.72792770640662
46
+ 45,0.1145,80.97,0.9540892773838134,81.02625692203677
47
+ 46,0.1095,81.41,0.9528565328372153,80.90692550868579
48
+ 47,0.1039,81.18,0.9488438052709023,80.57876276685401
49
+ 48,0.0988,81.24,0.9472993114633423,80.7577597679557
50
+ 49,0.0942,81.13,0.9416460762754011,80.71301133194515
51
+ 50,0.0896,,0.9394862801549537,80.68317776591226
model_training_results2.xlsx ADDED
Binary file (33.1 kB). View file
 
test/New Text Document.txt ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Class Class_ID Precision Recall F1-Score Support
2
+ happy 3 0.918699 0.937759 0.928131 482
3
+ surprise 6 0.879310 0.796875 0.836066 128
4
+ angry 0 0.815789 0.849315 0.832215 73
5
+ sad 5 0.868750 0.727749 0.792023 191
6
+ neutral 4 0.724014 0.870690 0.790607 232
7
+ fear 2 0.647059 0.500000 0.564103 22
8
+ disgust 1 0.606557 0.506849 0.552239 73
9
+
10
+
11
+
12
+ ================================================================================
13
+ OVERALL METRICS SUMMARY
14
+ ================================================================================
15
+ Metric Score
16
+ Accuracy 0.836803
17
+ Precision (Macro) 0.780026
18
+ Precision (Weighted) 0.838746
19
+ Recall (Macro) 0.741320
20
+ Recall (Weighted) 0.836803
21
+ F1-Score (Macro) 0.756483
22
+ F1-Score (Weighted) 0.834761
test/download (1).png ADDED
test/download (3).png ADDED
test/download.png ADDED
test/perclassmetrics.csv ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ Class,Class_ID,Precision,Recall,F1-Score
2
+ happy,3,0.918699,0.937759,0.928131
3
+ surprise,6,0.87931,0.796875,0.836066
4
+ angry,0,0.815789,0.849315,0.832215
5
+ sad,5,0.86875,0.727749,0.792023
6
+ neutral,4,0.724014,0.87069,0.790607
7
+ fear,2,0.647059,0.5,0.564103
8
+ disgust,1,0.606557,0.506849,0.552239
val/New Text Document.txt ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Class Class_ID Precision Recall F1-Score Support
2
+ happy 3 0.924686 0.934461 0.929548 473
3
+ surprise 6 0.863636 0.766129 0.811966 124
4
+ neutral 4 0.794326 0.814545 0.804309 275
5
+ angry 0 0.741176 0.840000 0.787500 75
6
+ sad 5 0.775758 0.771084 0.773414 166
7
+ fear 2 0.800000 0.727273 0.761905 33
8
+ disgust 1 0.625000 0.583333 0.603448 60
9
+
10
+ Metric Score
11
+ Accuracy 0.838308
12
+ Precision (Macro) 0.789226
13
+ Precision (Weighted) 0.838450
14
+ Recall (Macro) 0.776689
15
+ Recall (Weighted) 0.838308
16
+ F1-Score (Macro) 0.781727
17
+ F1-Score (Weighted) 0.837764
val/download (1).png ADDED
val/download (2).png ADDED
val/download.png ADDED
val/perclass.csv ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ Class,Class_ID,Precision,Recall,F1-Score,Support
2
+ happy,3,0.924686,0.934461,0.929548,473
3
+ surprise,6,0.863636,0.766129,0.811966,124
4
+ neutral,4,0.794326,0.814545,0.804309,275
5
+ angry,0,0.741176,0.840000,0.787500,75
6
+ sad,5,0.775758,0.771084,0.773414,166
7
+ fear,2,0.800000,0.727273,0.761905,33
8
+ disgust,1,0.625000,0.583333,0.603448,60
vitsmall2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc3993eb6db745097c62b3a5ee00380edc8e0b998dbdb4a039083ad12628ff42
3
+ size 86734865