Colab Sample Demo (MAE).ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Colab Sample Demo (MAE).ipynb","provenance":[],"collapsed_sections":[],"machine_shape":"hm"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","metadata":{"id":"cn9yRTb9-tDz"},"source":["<p><img alt=\"Colaboratory logo\" height=\"45px\" src=\"/img/colab_favicon.ico\" align=\"left\" hspace=\"10px\" vspace=\"0px\"></p>\n","\n","<h1>Colab 实验平台使用模板，以MAE为例</h1>\n","\n","# 根据如下步骤熟悉如何使用CoLab实验平台\n","\n","12/19/2021 by：张天翊\n","\n","* 简介：colab是一个虚拟的服务器分发平台，每次会分给你一个服务器（cpu 56G内存+ 某个GPU）使用可以在后台执行最多24h\n","\n","* 优点：可以记录结果 + 交互式计算 + 给大家类似本地linux服务器的环境 + 提供论文复现环境\n","* 为了最大化大家的内卷效率：\n","实验室决定把多个人学习程序的时候在这个上面执行，或者是短时间的/有复现需求的任务放在colab上，正式长时间模型还是在线下linux服务器跑（路径配置与colab相同）"]},{"cell_type":"markdown","metadata":{"id":"yF-fjSKYApiA"},"source":["# 本例子中的文件系统\n","\n","## 服务器路径结构：\n","* /home 工作目录\n","* /home/Insight 项目文件路径\n","* /home/Insight/runs 存放tensorboard输出\n","* /home/Insight/saved_models 存模型\n","* /home/Insight/code 代码总目录（与github一样）\n","* /home/Insight/code/colab_sample 某某 colab_sample的 个人代码路径（与pycharm同步路径一致）\n","\n","\n","* /data 数据目录\n","* /data/Insight 项目数据路径\n","* /data/Insight/samole_data.zip 数据集压缩包（或文件夹）\n","\n","\n","## github路径结构\n","* /Insight     项目的总路径\n","* /Insight/colab_sample 个人(colab_sample)代码文件夹\n","\n","## pycharm同步文件夹\n","* 将本地代码文件夹 与服务器上的/home/Insight/code/colab_sample 个人代码文件夹 同步即可\n","\n","## Google Drive路径结构\n","* /mydrive/Insight/code\n","* /mydrive/Insight/runs\n","* /mydrive/Insight/saved_models\n","* /mydrive/dataset/samole_data.zip 数据集压缩包"]},{"cell_type":"markdown","metadata":{"id":"6I_U0BkV2YNT"},"source":["# 配置运行时资源\n","* 在开始之前根据需要先把实例切换成gpu+高RAM （右上角'链接'-'查看资源'-'更改运行时类型'的位置设置RAM磁盘啥的，默认的不一定合适） \n","* 硬件加速器选None表示这个是CPU环境，选GPU调用有显卡的服务器来进行实验，TPU则只支持基于tensorflow、keras写的代码，我们统一使用PPytorch\n","\n","* 管理会话查看正在运行的运行时（右上角'链接'-'查看资源'-'管理会话'，注意别把别人的终止掉了） \n","\n","\n","# 检查显卡\n","\n","* 首先检查用的什么卡，最好是v100、P100（这两个区别不是很大）不行的话就T4，再不行就躺平吧\n","* 目前发现可以开1-3个notebook跑，也就是最多可以用3个单卡环境，无法多卡交互\n","\n","* 如果显示：NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running. 表示没有显卡"]},{"cell_type":"code","metadata":{"id":"g23_xZCs2IT4","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1650181552788,"user_tz":-480,"elapsed":18,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"16b7ec4f-e183-4b78-a609-8844398b6c38"},"source":["# 查看显卡类型，感叹号在ipynb调用linux指令\n","!nvidia-smi"],"execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Sun Apr 17 07:45:51 2022       \n","+-----------------------------------------------------------------------------+\n","| NVIDIA-SMI 460.32.03    Driver Version: 460.32.03    CUDA Version: 11.2     |\n","|-------------------------------+----------------------+----------------------+\n","| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n","| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n","|                               |                      |               MIG M. |\n","|===============================+======================+======================|\n","|   0  Tesla P100-PCIE...  Off  | 00000000:00:04.0 Off |                    0 |\n","| N/A   35C    P0    28W / 250W |      0MiB / 16280MiB |      0%      Default |\n","|                               |                      |                  N/A |\n","+-------------------------------+----------------------+----------------------+\n","                                                                               \n","+-----------------------------------------------------------------------------+\n","| Processes:                                                                  |\n","|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n","|        ID   ID                                                   Usage      |\n","|=============================================================================|\n","|  No running processes found                                                 |\n","+-----------------------------------------------------------------------------+\n"]}]},{"cell_type":"markdown","metadata":{"id":"8CMIHKMS2ssK"},"source":["## 显示时间"]},{"cell_type":"code","metadata":{"id":"i28S2PXC22CK","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1650181552789,"user_tz":-480,"elapsed":12,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"f2b01b25-55b6-44d9-8625-1be6c3aefb89"},"source":["# 如果一段时间不需要用显卡的话，可以整个僵尸参数占用点显存，避免出现警告\n","# import torch\n","# a = torch.Tensor([1000, 1000, 1000]).cuda()  # 大约占用 1.1G 显存\n","!date --date='+8 hour'  # 查看系统时间，转换为北京时间"],"execution_count":2,"outputs":[{"output_type":"stream","name":"stdout","text":["Sun Apr 17 15:45:52 UTC 2022\n"]}]},{"cell_type":"markdown","metadata":{"id":"viaUMP4R21ea"},"source":["# 挂载Google Drive\n","\n","* 直接执行会需要点链接验证，点确定就行\n","* 或 点击 左边菜单栏 文件-回收文件夹的符号实现挂载"]},{"cell_type":"code","metadata":{"id":"cGciEUpy3Qp-","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1650181576705,"user_tz":-480,"elapsed":23920,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"0c24ae8c-f5d7-49c5-c0f7-09402a7edd06"},"source":["from google.colab import drive\n","drive.mount('/content/drive')"],"execution_count":3,"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}]},{"cell_type":"markdown","metadata":{"id":"kocu7d1UHCpH"},"source":["## 挂载Github（调取公用账号，给予colab代码库权限）\n","用来把代码库从github上拿下来，并装载到colab虚拟机上\n","\n","ref: https://blog.csdn.net/u011119817/article/details/108722832\n","ref: https://docs.github.com/en/github/authenticating-to-github/keeping-your-account-and-data-secure/creating-a-personal-access-token"]},{"cell_type":"code","metadata":{"id":"jxdf2UZKJSIu","executionInfo":{"status":"ok","timestamp":1650181576706,"user_tz":-480,"elapsed":9,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}}},"source":["import os\n","# GitHub user\n","user = 'Idonttellu'\n","# GitHub personal-access-token\n","password = 'Idonttellu'\n","os.environ['GITHUB_AUTH'] = user + ':' + password"],"execution_count":4,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"T01ZzrGL3Jfy"},"source":["## 创建文件 与 建立路径环境\n","* 在此之前，先挂载google云盘。云盘上已经有数据了（[这是云盘的地址](https://drive.google.com/drive/u/1/my-drive)）："]},{"cell_type":"code","metadata":{"id":"0-vKhDTB3f2K","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1650181581724,"user_tz":-480,"elapsed":5025,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"48fbcf7e-b56c-477b-a79f-513cb5ed5012"},"source":["# 创建文件\n","# 在此之前，先挂载google云盘。云盘上已经有代码和数据了，访问下面的网址可以增删内容：\n","# https://drive.google.com/drive/u/1/my-drive\n","\n","# 先清空CoLab（用于避免重复执行的bug）\n","!rm -rf /data\n","!rm -rf /home/Insight\n","\n","# 建立实验路径文件夹环境（与本地实验室linux服务器一致，便于代码直接在本地跑）\n","!mkdir /home/Insight  # 实验路径\n","!mkdir /home/Insight/runs  # 存tensorboard输出+json输出的路径\n","!mkdir /home/Insight/code  # 存代码的路径\n","!mkdir /home/Insight/saved_models  # 存模型的路径\n","\n","!mkdir /data  # 数据路径\n","!mkdir /data/Insight\n","\n","print('Folder Tree Creation completed!')\n","\n","# 复制代码（来自colab）\n","# !cp -r /content/drive/MyDrive/zhanglab_headneck/Insight/* /home/zhanglab_headneck/Insight/code\n","# print('code transfer from colab completed!')\n","\n","# 复制代码（来自github）大家的代码都加载到虚拟机/home/Insight/code，之后进入自己的代码文件夹\n","!git clone https://github.com/sagizty/Insight.git /home/Insight/code\n","print('code transfer from github completed!')\n","\n","# 如果是私有仓库需要登陆密码，使用方法如下，加一个$GITHUB_AUTH@,注意前面需要赋值，\n","# 该账号需要有权限访问私有仓库，因此在自己的仓库把公用账号Idonttellu拉进来就行\n","# !git clone https://$GITHUB_AUTH@github.com/sagizty/Insight.git /home/Insight/code\n","\n","# 复制runs(除非对比需要调用，否则不应使用，节约时间)\n","# !cp -r /content/drive/MyDrive/zhanglab_headneck/runs/* /home/zhanglab_headneck/runs\n","# print('tensorboard log transfer completed!')\n","\n","# 复制已有模型(除非调用已有模型，否则不应使用，节约时间，*代表全部，可以根据需要只要部分)\n","# !cp -r /content/drive/MyDrive/zhanglab_headneck/saved_models/* /home/zhanglab_headneck/saved_models\n","# print('model state dicts transfered')\n","\n","# 复制数据集\n","# 复制压缩包\n","!cp /content/drive/MyDrive/dataset/sample_data.zip /data/Insight/\n","# 解压缩(生成coco文件夹 /data/zhanglab_headneck/coco)\n","!unzip -q /data/Insight/sample_data.zip -d /data/Insight/\n","# 调整删除zip\n","!rm -f /data/Insight/sample_data.zip\n","print('data transfer completed!')"],"execution_count":5,"outputs":[{"output_type":"stream","name":"stdout","text":["Folder Tree Creation completed!\n","Cloning into '/home/Insight/code'...\n","remote: Enumerating objects: 395, done.\u001b[K\n","remote: Counting objects: 100% (264/264), done.\u001b[K\n","remote: Compressing objects: 100% (252/252), done.\u001b[K\n","remote: Total 395 (delta 106), reused 0 (delta 0), pack-reused 131\u001b[K\n","Receiving objects: 100% (395/395), 28.00 MiB | 16.76 MiB/s, done.\n","Resolving deltas: 100% (164/164), done.\n","code transfer from github completed!\n","data transfer completed!\n"]}]},{"cell_type":"markdown","metadata":{"id":"XGIRGJyM48Vb"},"source":["## 安装库与程序环境\n","\n","## 之后把工作路径切换到自己的文件夹！！！！！！！！！！！！"]},{"cell_type":"code","metadata":{"id":"-QRojs2c4_bi","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1650181590191,"user_tz":-480,"elapsed":8477,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"9421dd27-a5dd-4fc6-d1b4-188cbd7a0975"},"source":["# 切换工作目录(请切换到自己的文件夹！！！！！！！！！！！！)\n","import os\n","my_file = 'colab_sample' # 切换工作目录(请切换到自己的文件夹！！！！！！！！！！！！)\n","\n","my_path = os.path.join(\"/home/Insight/code\", my_file)\n","os.chdir(my_path)\n","\n","# 输出工作路径\n","!pwd\n","\n","# 安装依赖\n","!pip install tensorboardX\n","!pip install timm\n","!pip install notifyemail\n","# 关于：使用tensorboard\n","# https://blog.csdn.net/wry3482/article/details/105930449\n","# https://zhuanlan.zhihu.com/p/64479055\n","# https://www.jianshu.com/p/86f7373160bf\n","\n","# 保存和加载训练到一半的模型\n","# https://www.cnblogs.com/xiximayou/p/12450761.html"],"execution_count":6,"outputs":[{"output_type":"stream","name":"stdout","text":["/home/Insight/code/colab_sample\n","Collecting tensorboardX\n","  Downloading tensorboardX-2.5-py2.py3-none-any.whl (125 kB)\n","\u001b[K     |████████████████████████████████| 125 kB 14.9 MB/s \n","\u001b[?25hRequirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from tensorboardX) (1.15.0)\n","Requirement already satisfied: protobuf>=3.8.0 in /usr/local/lib/python3.7/dist-packages (from tensorboardX) (3.17.3)\n","Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from tensorboardX) (1.21.5)\n","Installing collected packages: tensorboardX\n","Successfully installed tensorboardX-2.5\n","Collecting timm\n","  Downloading timm-0.5.4-py3-none-any.whl (431 kB)\n","\u001b[K     |████████████████████████████████| 431 kB 14.7 MB/s \n","\u001b[?25hRequirement already satisfied: torch>=1.4 in /usr/local/lib/python3.7/dist-packages (from timm) (1.10.0+cu111)\n","Requirement already satisfied: torchvision in /usr/local/lib/python3.7/dist-packages (from timm) (0.11.1+cu111)\n","Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torch>=1.4->timm) (4.1.1)\n","Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from torchvision->timm) (1.21.5)\n","Requirement already satisfied: pillow!=8.3.0,>=5.3.0 in /usr/local/lib/python3.7/dist-packages (from torchvision->timm) (7.1.2)\n","Installing collected packages: timm\n","Successfully installed timm-0.5.4\n","Collecting notifyemail\n","  Downloading notifyemail-1.0.2-py3-none-any.whl (31 kB)\n","Installing collected packages: notifyemail\n","Successfully installed notifyemail-1.0.2\n"]}]},{"cell_type":"markdown","metadata":{"id":"sCnrlyYb5DTK"},"source":["## 开始正式内容\n","* 命令行直接调用代码即可(前面用！代表是传给系统的linux指令，不然就传给python 解释器了）\n","* 这种情况下使用Parser调节参数设置即可\n","\n","* 或者直接写脚本，如果比较简单的话"]},{"cell_type":"markdown","source":["# 第一种用法是直接把虚拟机作为linux服务器，使用命令行调用执行代码，推荐\n","\n","---"],"metadata":{"id":"T_QK0uj4uowY"}},{"cell_type":"markdown","source":["前面:\n","1. 我们把github的代码装载到了服务器上\n","2. 我们已经切换到了虚拟机上自己的代码文件夹 /home/Insight/code/colab_sample"],"metadata":{"id":"Z7qhkRUUvBBD"}},{"cell_type":"code","source":["!python sample.py  # 直接用!python叫系统执行"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Q57RyaBvvZ1a","executionInfo":{"status":"ok","timestamp":1650181590192,"user_tz":-480,"elapsed":17,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"7f22f371-97bb-432a-fb0f-a366033408a4"},"execution_count":7,"outputs":[{"output_type":"stream","name":"stdout","text":["hello world\n","working path: /home/Insight/code/colab_sample\n","7\n","2\n","3\n","1\n","5\n","6\n","4\n","hello colab-github！！！\n"]}]},{"cell_type":"markdown","source":["# 第二种用法是直接把python写在这里，这种方法主要是为了小调试，不推荐\n","\n","---\n","\n"],"metadata":{"id":"gNKugDrVuYsQ"}},{"cell_type":"code","metadata":{"id":"OhwuAE3e5uKS","executionInfo":{"status":"ok","timestamp":1650181596358,"user_tz":-480,"elapsed":6172,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}}},"source":["'''\n","MAE 简单复现  ver: 2021.12.11\n","本实验是单个图片的MAE复现，意在说明MAE的过程。\n","本身没有进行代码结构性与逻辑细节的优化。\n","这个里面用的vit和我们习惯的timm版本略有不同，后续需要做匹配\n","总的来说，这个脚本作为一个流程例子用来学习MAE流程和部分细节还是不错的，\n","但是需要匹配性的重新做才可以使用到论文中。\n","这个脚本离真正论文能用的还有很远距离。\n","来自：https://zhuanlan.zhihu.com/p/439554945\n","\n","加了训练脚本（但是是单张图片的，后续进一步优化）\n","我加了一些内容和注释,并且修改成了适配timm的样子\n","'''\n","import matplotlib.pyplot as plt\n","from PIL import Image\n","import timm\n","import torch, os\n","import torch.optim as optim\n","from Transformers import *\n","from tensorboardX import SummaryWriter\n","\n","\n","class MAE(nn.Module):\n","    def __init__(self, encoder, decoder_dim, mask_ratio=0.75, decoder_depth=1,\n","                 num_decoder_heads=8, decoder_dim_per_head=64):\n","        \"\"\"\n","        MAE 框架，自带decoder模块\n","        :param encoder: 输入Transformer模型\n","        :param decoder_dim: Encoder 输出的维度可能和 Decoder 要求的输入维度不一致\n","        :param mask_ratio: paper提倡这个比例最好是 75%\n","        :param decoder_depth: 实质就是多层堆叠的 Transformer（这个也有一个gap问题，前后结构是否一样值得研究） TODO\n","        :param num_decoder_heads:\n","        :param decoder_dim_per_head:\n","        \"\"\"\n","        super().__init__()\n","        assert 0. < mask_ratio < 1., f'mask ratio must be kept between 0 and 1, got: {mask_ratio}'\n","\n","        # Encoder(这里 CW 用 ViT 实现)\n","        self.encoder = encoder\n","        self.patch_h = encoder.patch_embed.patch_size[0]\n","        self.patch_w = encoder.patch_embed.patch_size[1]\n","\n","        encoder_dim = encoder.embed_dim\n","\n","        channels = 3\n","        patch_dim = channels * self.patch_h * self.patch_w\n","\n","        # 由于原生的 ViT 有 cls_token，因此其 position embedding 的倒数第2个维度是：\n","        # 实际划分的 patch 数量加上 1个 cls_token\n","        # num_patches_plus_cls_token, encoder_dim = encoder.pos_embed.shape[-2:]  fixme 之前不匹配\n","        num_patches_plus_cls_token = encoder.patch_embed.num_patches + 1\n","\n","        # Patch embedding 后续考虑做成使用encoder的，但是原文使用的是MLP映射而不是CNN映射\n","        self.patch_embed = nn.Linear(patch_dim, encoder_dim)\n","\n","        # Input channels of encoder patch embedding: patch size**2 x 3\n","        # 这个用作预测头部的输出通道，从而能够对 patch 中的所有像素值进行预测\n","        num_pixels_per_patch = encoder_dim  # encoder.patch_embed.weight.size(1)  fixme 之前不匹配\n","        # print(num_pixels_per_patch)\n","\n","        # Encoder-Decoder：Encoder 输出的维度可能和 Decoder 要求的输入维度不一致，因此需要转换\n","        self.enc_to_dec = nn.Linear(encoder_dim, decoder_dim) if encoder_dim != decoder_dim else nn.Identity()\n","\n","        # Mask token\n","        # paper提倡这个比例最好是 75%\n","        self.mask_ratio = mask_ratio\n","        # mask token 的实质：1个可学习的共享向量\n","        self.mask_embed = nn.Parameter(torch.randn(decoder_dim))\n","\n","        # Decoder：实质就是多层堆叠的 Transformer\n","        self.decoder = Transformer(\n","            decoder_dim,\n","            decoder_dim * 4,\n","            depth=decoder_depth,\n","            num_heads=num_decoder_heads,\n","            dim_per_head=decoder_dim_per_head,\n","        )\n","        # 在 Decoder 中用作对 mask tokens 的 position embedding\n","        # Filter out cls_token 注意第1个维度去掉 cls_token\n","        self.decoder_pos_embed = nn.Embedding(num_patches_plus_cls_token - 1, decoder_dim)\n","\n","        # Prediction head 输出的维度数等于1个 patch 的像素值数量\n","        self.head = nn.Linear(decoder_dim, num_pixels_per_patch)\n","\n","    def forward(self, x):\n","        device = x.device\n","        b, c, h, w = x.shape\n","\n","        '''i. Patch partition'''\n","        # 首先需要将图像划分成 patch，划分方式实质就是维度的变换\n","        num_patches = (h // self.patch_h) * (w // self.patch_w)\n","        # (b, c=3, h, w)->(b, n_patches, patch_size**2*c)\n","        patches = x.view(\n","            b, c,\n","            h // self.patch_h, self.patch_h,\n","            w // self.patch_w, self.patch_w\n","        ).permute(0, 2, 4, 3, 5, 1).reshape(b, num_patches, -1)\n","\n","        '''ii. Divide into masked & un-masked groups'''\n","\n","        # 根据 mask 比例计算需要 mask 掉的 patch 数量\n","        # num_patches = (h // self.patch_h) * (w // self.patch_w)\n","        num_masked = int(self.mask_ratio * num_patches)\n","\n","        # Shuffle:生成对应 patch 的随机索引\n","        # torch.rand() 服从均匀分布(normal distribution)\n","        # torch.rand() 只是生成随机数，argsort() 是为了获得成索引\n","        # (b, n_patches)\n","        shuffle_indices = torch.rand(b, num_patches, device=device).argsort()\n","        # mask 和 unmasked patches 对应的索引\n","        mask_ind, unmask_ind = shuffle_indices[:, :num_masked], shuffle_indices[:, num_masked:]\n","\n","        # 对应 batch 维度的索引：(b,1)\n","        batch_ind = torch.arange(b, device=device).unsqueeze(-1)\n","        # 利用先前生成的索引对 patches 进行采样，分为 mask 和 unmasked 两组 TODO\n","        mask_patches, unmask_patches = patches[batch_ind, mask_ind], patches[batch_ind, unmask_ind]\n","\n","        '''iii. Encode'''\n","\n","        # 将 patches 通过 emebdding 转换成 tokens\n","        unmask_tokens = self.patch_embed(unmask_patches)\n","        # unmask_tokens = self.encoder.patch_embed(unmask_patches)  fixme 之前不匹配\n","\n","        # 为 tokens 加入 position embeddings\n","        # 注意这里索引加1是因为索引0对应 ViT 的 cls_token\n","        unmask_tokens += self.encoder.pos_embed.repeat(b, 1, 1)[batch_ind, unmask_ind + 1]\n","        # 真正的编码过程\n","        encoded_tokens = self.encoder.blocks(unmask_tokens)\n","        # encoded_tokens = self.encoder.transformer(unmask_tokens)  fixme 之前不匹配\n","\n","        '''iv. Decode'''\n","\n","        # 对编码后的 tokens 维度进行转换，从而符合 Decoder 要求的输入维度\n","        enc_to_dec_tokens = self.enc_to_dec(encoded_tokens)\n","\n","        # 由于 mask token 实质上只有1个，因此要对其进行扩展，从而和 masked patches 一一对应\n","        # (decoder_dim)->(b, n_masked, decoder_dim)\n","        mask_tokens = self.mask_embed[None, None, :].repeat(b, num_masked, 1)\n","        # 为 mask tokens 加入位置信息\n","        mask_tokens += self.decoder_pos_embed(mask_ind)\n","\n","        # 将 mask tokens 与 编码后的 tokens 拼接起来\n","        # (b, n_patches, decoder_dim)\n","        concat_tokens = torch.cat([mask_tokens, enc_to_dec_tokens], dim=1)\n","        # Un-shuffle：恢复原先 patches 的次序\n","        dec_input_tokens = torch.empty_like(concat_tokens, device=device)\n","        dec_input_tokens[batch_ind, shuffle_indices] = concat_tokens\n","        # 将全量 tokens 喂给 Decoder 解码\n","        decoded_tokens = self.decoder(dec_input_tokens)\n","\n","        '''v. Mask pixel Prediction'''\n","\n","        dec_mask_tokens = decoded_tokens[batch_ind, mask_ind, :]\n","        # (b, n_masked, n_pixels_per_patch=patch_size**2 x c)\n","        pred_mask_pixel_values = self.head(dec_mask_tokens)\n","\n","        return pred_mask_pixel_values, mask_patches\n","\n","    # @torch.no_grad\n","    def predict(self, x):\n","        self.eval()\n","\n","        device = x.device\n","        b, c, h, w = x.shape\n","\n","        '''i. Patch partition'''\n","\n","        num_patches = (h // self.patch_h) * (w // self.patch_w)\n","        # (b, c=3, h, w)->(b, n_patches, patch_size**2*c)\n","        patches = x.view(\n","            b, c,\n","            h // self.patch_h, self.patch_h,\n","            w // self.patch_w, self.patch_w\n","        ).permute(0, 2, 4, 3, 5, 1).reshape(b, num_patches, -1)\n","\n","        '''ii. Divide into masked & un-masked groups'''\n","\n","        num_masked = int(self.mask_ratio * num_patches)\n","\n","        # Shuffle\n","        # (b, n_patches)\n","        shuffle_indices = torch.rand(b, num_patches, device=device).argsort()\n","        mask_ind, unmask_ind = shuffle_indices[:, :num_masked], shuffle_indices[:, num_masked:]\n","\n","        # (b, 1)\n","        batch_ind = torch.arange(b, device=device).unsqueeze(-1)\n","        mask_patches, unmask_patches = patches[batch_ind, mask_ind], patches[batch_ind, unmask_ind]\n","\n","        '''iii. Encode'''\n","\n","        # unmask_tokens = self.encoder.patch_embed(unmask_patches)  fixme 之前不匹配\n","        unmask_tokens = self.patch_embed(unmask_patches)\n","\n","        # Add position embeddings\n","        unmask_tokens += self.encoder.pos_embed.repeat(b, 1, 1)[batch_ind, unmask_ind + 1]\n","        # encoded_tokens = self.encoder.transformer(unmask_tokens)  fixme 之前不匹配\n","        encoded_tokens = self.encoder.blocks(unmask_tokens)\n","        '''iv. Decode'''\n","\n","        enc_to_dec_tokens = self.enc_to_dec(encoded_tokens)\n","\n","        # (decoder_dim)->(b, n_masked, decoder_dim)\n","        mask_tokens = self.mask_embed[None, None, :].repeat(b, num_masked, 1)\n","        # Add position embeddings\n","        mask_tokens += self.decoder_pos_embed(mask_ind)\n","\n","        # (b, n_patches, decoder_dim)\n","        concat_tokens = torch.cat([mask_tokens, enc_to_dec_tokens], dim=1)\n","        # dec_input_tokens = concat_tokens\n","        dec_input_tokens = torch.empty_like(concat_tokens, device=device)\n","        # Un-shuffle\n","        dec_input_tokens[batch_ind, shuffle_indices] = concat_tokens\n","        decoded_tokens = self.decoder(dec_input_tokens)\n","\n","        '''v. Mask pixel Prediction'''\n","\n","        dec_mask_tokens = decoded_tokens[batch_ind, mask_ind, :]\n","        # (b, n_masked, n_pixels_per_patch=patch_size**2 x c)\n","        pred_mask_pixel_values = self.head(dec_mask_tokens)\n","\n","        # 比较下预测值和真实值\n","        mse_per_patch = (pred_mask_pixel_values - mask_patches).abs().mean(dim=-1)\n","        mse_all_patches = mse_per_patch.mean()\n","\n","        print(f'mse per (masked)patch: {mse_per_patch} mse all (masked)patches: {mse_all_patches} '\n","              f'total {num_masked} masked patches')\n","        print(f'all close: {torch.allclose(pred_mask_pixel_values, mask_patches, rtol=1e-1, atol=1e-1)}')\n","\n","        '''vi. Reconstruction'''\n","\n","        recons_patches = patches.detach()\n","        # Un-shuffle (b, n_patches, patch_size**2 * c)\n","        recons_patches[batch_ind, mask_ind] = pred_mask_pixel_values\n","        # 模型重建的效果图\n","        # Reshape back to image\n","        # (b, n_patches, patch_size**2 * c)->(b, c, h, w)\n","        recons_img = recons_patches.view(\n","            b, h // self.patch_h, w // self.patch_w,\n","            self.patch_h, self.patch_w, c\n","        ).permute(0, 5, 1, 3, 2, 4).reshape(b, c, h, w)\n","\n","        mask_patches = torch.randn_like(mask_patches, device=mask_patches.device)\n","        # mask 效果图\n","        patches[batch_ind, mask_ind] = mask_patches\n","        patches_to_img = patches.view(\n","            b, h // self.patch_h, w // self.patch_w,\n","            self.patch_h, self.patch_w, c\n","        ).permute(0, 5, 1, 3, 2, 4).reshape(b, c, h, w)\n","\n","        return recons_img, patches_to_img\n","\n","\n","def train(mae, dataloader, optimizer, criterion, epoch=1000, writer=None):\n","\n","    for i in range(epoch):\n","        img_ts=dataloader  # TODO 这个只是假的dataloader\n","        pred_mask_pixel_values, mask_patches = mae(img_ts)\n","        optimizer.zero_grad()\n","        loss = criterion(pred_mask_pixel_values, mask_patches)\n","\n","        loss.backward()\n","        optimizer.step()\n","        # 比较下预测值和真实值\n","        mse_per_patch = (pred_mask_pixel_values - mask_patches).abs().mean(dim=-1)\n","        mse_all_patches = mse_per_patch.mean()\n","        print(f'Epoch: {i+1} ')\n","        print(f'mse all (masked)patches: {mse_all_patches} ')\n","        print(f'all close: {torch.allclose(pred_mask_pixel_values, mask_patches, rtol=1e-1, atol=1e-1)}')\n","\n","        if writer is not None:\n","            # ...log the running loss\n","            writer.add_scalar('train' + ' minibatch loss',\n","                              float(mse_all_patches),\n","                              i)\n","    # 结束记录内容给tensorboard\n","    if writer is not None:\n","        writer.close()\n"],"execution_count":8,"outputs":[]},{"cell_type":"markdown","source":["分段执行，和jyputer一样"],"metadata":{"id":"YL7vwmfHwbfV"}},{"cell_type":"code","source":["BASE_DIR = r'./'\n","device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n","\n","draw_path = '/home/Insight/runs/MAE'\n","save_model_path = '/home/Insight/saved_models'\n","if not os.path.exists(draw_path):\n","    os.makedirs(draw_path)\n","if not os.path.exists(save_model_path):\n","    os.makedirs(save_model_path)\n","\n","writer = SummaryWriter(draw_path)\n","\n","# 读入图像并缩放到适合模型输入的尺寸\n","img_raw = Image.open(os.path.join(BASE_DIR, 'photo.jpg'))\n","h, w = img_raw.height, img_raw.width\n","ratio = h / w\n","print(f\"image hxw: {h} x {w} mode: {img_raw.mode}\")\n","\n","img_size, patch_size = (224, 224), (16, 16)\n","img = img_raw.resize(img_size)\n","rh, rw = img.height, img.width\n","print(f'resized image hxw: {rh} x {rw} mode: {img.mode}')\n","img.save(os.path.join(BASE_DIR, 'resized_photo.jpg'))\n","\n","# 将图像转换成张量\n","from torchvision.transforms import ToTensor, ToPILImage\n","\n","img_ts = ToTensor()(img).unsqueeze(0).to(device)\n","print(f\"input tensor shape: {img_ts.shape} dtype: {img_ts.dtype} device: {img_ts.device}\")\n","\n","# 实例化模型\n","# encoder = ViT(img_size, patch_size, dim=512, mlp_dim=1024, dim_per_head=64)\n","encoder = timm.create_model('vit_base_patch16_224', pretrained=True, num_classes=1000)\n","decoder_dim = 512\n","mae = MAE(encoder, decoder_dim, decoder_depth=6)\n","# weight = torch.load(os.path.join(BASE_DIR, 'mae.pth'), map_location='cpu')  # 加载训练好的权重\n","mae.to(device)\n","\n","# 训练\n","criterion = nn.MSELoss()\n","optimizer = optim.Adam(mae.parameters(), lr=0.0001, weight_decay=0.01)\n","\n","train(mae, img_ts, optimizer, criterion, epoch=1000, writer=writer)  # TODO 这个用的假dataloader\n","\n","torch.save(mae.state_dict(), os.path.join(save_model_path,'MAE.pth'))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"NDSCkkQxguY5","executionInfo":{"status":"ok","timestamp":1650181660015,"user_tz":-480,"elapsed":63662,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"02ed1c65-089f-4dcb-d6b7-859e49bd992f"},"execution_count":9,"outputs":[{"output_type":"stream","name":"stdout","text":["image hxw: 951 x 886 mode: RGB\n","resized image hxw: 224 x 224 mode: RGB\n","input tensor shape: torch.Size([1, 3, 224, 224]) dtype: torch.float32 device: cuda:0\n","Epoch: 1 \n","mse all (masked)patches: 0.8639051914215088 \n","all close: False\n","Epoch: 2 \n","mse all (masked)patches: 0.701176106929779 \n","all close: False\n","Epoch: 3 \n","mse all (masked)patches: 0.6338961720466614 \n","all close: False\n","Epoch: 4 \n","mse all (masked)patches: 0.6046916842460632 \n","all close: False\n","Epoch: 5 \n","mse all (masked)patches: 0.5863736867904663 \n","all close: False\n","Epoch: 6 \n","mse all (masked)patches: 0.5727626085281372 \n","all close: False\n","Epoch: 7 \n","mse all (masked)patches: 0.559879720211029 \n","all close: False\n","Epoch: 8 \n","mse all (masked)patches: 0.5512166619300842 \n","all close: False\n","Epoch: 9 \n","mse all (masked)patches: 0.5394402146339417 \n","all close: False\n","Epoch: 10 \n","mse all (masked)patches: 0.5304932594299316 \n","all close: False\n","Epoch: 11 \n","mse all (masked)patches: 0.5240459442138672 \n","all close: False\n","Epoch: 12 \n","mse all (masked)patches: 0.5204212069511414 \n","all close: False\n","Epoch: 13 \n","mse all (masked)patches: 0.5107011795043945 \n","all close: False\n","Epoch: 14 \n","mse all (masked)patches: 0.4989766776561737 \n","all close: False\n","Epoch: 15 \n","mse all (masked)patches: 0.49673372507095337 \n","all close: False\n","Epoch: 16 \n","mse all (masked)patches: 0.48766782879829407 \n","all close: False\n","Epoch: 17 \n","mse all (masked)patches: 0.4794529378414154 \n","all close: False\n","Epoch: 18 \n","mse all (masked)patches: 0.4743230938911438 \n","all close: False\n","Epoch: 19 \n","mse all (masked)patches: 0.47495678067207336 \n","all close: False\n","Epoch: 20 \n","mse all (masked)patches: 0.46635332703590393 \n","all close: False\n","Epoch: 21 \n","mse all (masked)patches: 0.46156954765319824 \n","all close: False\n","Epoch: 22 \n","mse all (masked)patches: 0.4608357548713684 \n","all close: False\n","Epoch: 23 \n","mse all (masked)patches: 0.46226105093955994 \n","all close: False\n","Epoch: 24 \n","mse all (masked)patches: 0.4539899528026581 \n","all close: False\n","Epoch: 25 \n","mse all (masked)patches: 0.4525796175003052 \n","all close: False\n","Epoch: 26 \n","mse all (masked)patches: 0.450038343667984 \n","all close: False\n","Epoch: 27 \n","mse all (masked)patches: 0.44615572690963745 \n","all close: False\n","Epoch: 28 \n","mse all (masked)patches: 0.44530168175697327 \n","all close: False\n","Epoch: 29 \n","mse all (masked)patches: 0.44169729948043823 \n","all close: False\n","Epoch: 30 \n","mse all (masked)patches: 0.43994301557540894 \n","all close: False\n","Epoch: 31 \n","mse all (masked)patches: 0.4322627782821655 \n","all close: False\n","Epoch: 32 \n","mse all (masked)patches: 0.43558257818222046 \n","all close: False\n","Epoch: 33 \n","mse all (masked)patches: 0.4333517849445343 \n","all close: False\n","Epoch: 34 \n","mse all (masked)patches: 0.42885100841522217 \n","all close: False\n","Epoch: 35 \n","mse all (masked)patches: 0.42615580558776855 \n","all close: False\n","Epoch: 36 \n","mse all (masked)patches: 0.4188821315765381 \n","all close: False\n","Epoch: 37 \n","mse all (masked)patches: 0.424525648355484 \n","all close: False\n","Epoch: 38 \n","mse all (masked)patches: 0.41895994544029236 \n","all close: False\n","Epoch: 39 \n","mse all (masked)patches: 0.4180571734905243 \n","all close: False\n","Epoch: 40 \n","mse all (masked)patches: 0.4181917607784271 \n","all close: False\n","Epoch: 41 \n","mse all (masked)patches: 0.4144126772880554 \n","all close: False\n","Epoch: 42 \n","mse all (masked)patches: 0.40981757640838623 \n","all close: False\n","Epoch: 43 \n","mse all (masked)patches: 0.41027647256851196 \n","all close: False\n","Epoch: 44 \n","mse all (masked)patches: 0.4066788852214813 \n","all close: False\n","Epoch: 45 \n","mse all (masked)patches: 0.4022556245326996 \n","all close: False\n","Epoch: 46 \n","mse all (masked)patches: 0.39970189332962036 \n","all close: False\n","Epoch: 47 \n","mse all (masked)patches: 0.40197864174842834 \n","all close: False\n","Epoch: 48 \n","mse all (masked)patches: 0.4010923504829407 \n","all close: False\n","Epoch: 49 \n","mse all (masked)patches: 0.3999311029911041 \n","all close: False\n","Epoch: 50 \n","mse all (masked)patches: 0.401620477437973 \n","all close: False\n","Epoch: 51 \n","mse all (masked)patches: 0.3944433331489563 \n","all close: False\n","Epoch: 52 \n","mse all (masked)patches: 0.39514705538749695 \n","all close: False\n","Epoch: 53 \n","mse all (masked)patches: 0.3930519223213196 \n","all close: False\n","Epoch: 54 \n","mse all (masked)patches: 0.39015907049179077 \n","all close: False\n","Epoch: 55 \n","mse all (masked)patches: 0.39220109581947327 \n","all close: False\n","Epoch: 56 \n","mse all (masked)patches: 0.3870716392993927 \n","all close: False\n","Epoch: 57 \n","mse all (masked)patches: 0.38801342248916626 \n","all close: False\n","Epoch: 58 \n","mse all (masked)patches: 0.38642778992652893 \n","all close: False\n","Epoch: 59 \n","mse all (masked)patches: 0.3838232159614563 \n","all close: False\n","Epoch: 60 \n","mse all (masked)patches: 0.38166049122810364 \n","all close: False\n","Epoch: 61 \n","mse all (masked)patches: 0.38275545835494995 \n","all close: False\n","Epoch: 62 \n","mse all (masked)patches: 0.378233939409256 \n","all close: False\n","Epoch: 63 \n","mse all (masked)patches: 0.38038814067840576 \n","all close: False\n","Epoch: 64 \n","mse all (masked)patches: 0.377549409866333 \n","all close: False\n","Epoch: 65 \n","mse all (masked)patches: 0.37515121698379517 \n","all close: False\n","Epoch: 66 \n","mse all (masked)patches: 0.37098413705825806 \n","all close: False\n","Epoch: 67 \n","mse all (masked)patches: 0.37123140692710876 \n","all close: False\n","Epoch: 68 \n","mse all (masked)patches: 0.365955114364624 \n","all close: False\n","Epoch: 69 \n","mse all (masked)patches: 0.37131574749946594 \n","all close: False\n","Epoch: 70 \n","mse all (masked)patches: 0.3692561089992523 \n","all close: False\n","Epoch: 71 \n","mse all (masked)patches: 0.37231284379959106 \n","all close: False\n","Epoch: 72 \n","mse all (masked)patches: 0.3692416846752167 \n","all close: False\n","Epoch: 73 \n","mse all (masked)patches: 0.3658464550971985 \n","all close: False\n","Epoch: 74 \n","mse all (masked)patches: 0.3616153597831726 \n","all close: False\n","Epoch: 75 \n","mse all (masked)patches: 0.36326998472213745 \n","all close: False\n","Epoch: 76 \n","mse all (masked)patches: 0.36402761936187744 \n","all close: False\n","Epoch: 77 \n","mse all (masked)patches: 0.360393226146698 \n","all close: False\n","Epoch: 78 \n","mse all (masked)patches: 0.35911789536476135 \n","all close: False\n","Epoch: 79 \n","mse all (masked)patches: 0.35737064480781555 \n","all close: False\n","Epoch: 80 \n","mse all (masked)patches: 0.35747310519218445 \n","all close: False\n","Epoch: 81 \n","mse all (masked)patches: 0.35953468084335327 \n","all close: False\n","Epoch: 82 \n","mse all (masked)patches: 0.3535137176513672 \n","all close: False\n","Epoch: 83 \n","mse all (masked)patches: 0.3557579219341278 \n","all close: False\n","Epoch: 84 \n","mse all (masked)patches: 0.35036763548851013 \n","all close: False\n","Epoch: 85 \n","mse all (masked)patches: 0.3562769591808319 \n","all close: False\n","Epoch: 86 \n","mse all (masked)patches: 0.3531622886657715 \n","all close: False\n","Epoch: 87 \n","mse all (masked)patches: 0.350694864988327 \n","all close: False\n","Epoch: 88 \n","mse all (masked)patches: 0.3495137691497803 \n","all close: False\n","Epoch: 89 \n","mse all (masked)patches: 0.34999799728393555 \n","all close: False\n","Epoch: 90 \n","mse all (masked)patches: 0.3471413254737854 \n","all close: False\n","Epoch: 91 \n","mse all (masked)patches: 0.3501530885696411 \n","all close: False\n","Epoch: 92 \n","mse all (masked)patches: 0.3466203510761261 \n","all close: False\n","Epoch: 93 \n","mse all (masked)patches: 0.34539794921875 \n","all close: False\n","Epoch: 94 \n","mse all (masked)patches: 0.3447358012199402 \n","all close: False\n","Epoch: 95 \n","mse all (masked)patches: 0.3444928526878357 \n","all close: False\n","Epoch: 96 \n","mse all (masked)patches: 0.3378399908542633 \n","all close: False\n","Epoch: 97 \n","mse all (masked)patches: 0.33988457918167114 \n","all close: False\n","Epoch: 98 \n","mse all (masked)patches: 0.33967578411102295 \n","all close: False\n","Epoch: 99 \n","mse all (masked)patches: 0.3340795040130615 \n","all close: False\n","Epoch: 100 \n","mse all (masked)patches: 0.33676573634147644 \n","all close: False\n","Epoch: 101 \n","mse all (masked)patches: 0.341340571641922 \n","all close: False\n","Epoch: 102 \n","mse all (masked)patches: 0.3359587490558624 \n","all close: False\n","Epoch: 103 \n","mse all (masked)patches: 0.33544522523880005 \n","all close: False\n","Epoch: 104 \n","mse all (masked)patches: 0.33631110191345215 \n","all close: False\n","Epoch: 105 \n","mse all (masked)patches: 0.3310392498970032 \n","all close: False\n","Epoch: 106 \n","mse all (masked)patches: 0.32916343212127686 \n","all close: False\n","Epoch: 107 \n","mse all (masked)patches: 0.3313950300216675 \n","all close: False\n","Epoch: 108 \n","mse all (masked)patches: 0.3275112211704254 \n","all close: False\n","Epoch: 109 \n","mse all (masked)patches: 0.328208863735199 \n","all close: False\n","Epoch: 110 \n","mse all (masked)patches: 0.3283732831478119 \n","all close: False\n","Epoch: 111 \n","mse all (masked)patches: 0.32583725452423096 \n","all close: False\n","Epoch: 112 \n","mse all (masked)patches: 0.3259781301021576 \n","all close: False\n","Epoch: 113 \n","mse all (masked)patches: 0.3275197148323059 \n","all close: False\n","Epoch: 114 \n","mse all (masked)patches: 0.3233051300048828 \n","all close: False\n","Epoch: 115 \n","mse all (masked)patches: 0.32540857791900635 \n","all close: False\n","Epoch: 116 \n","mse all (masked)patches: 0.32420429587364197 \n","all close: False\n","Epoch: 117 \n","mse all (masked)patches: 0.3209933638572693 \n","all close: False\n","Epoch: 118 \n","mse all (masked)patches: 0.32052767276763916 \n","all close: False\n","Epoch: 119 \n","mse all (masked)patches: 0.32158297300338745 \n","all close: False\n","Epoch: 120 \n","mse all (masked)patches: 0.3205036520957947 \n","all close: False\n","Epoch: 121 \n","mse all (masked)patches: 0.31658250093460083 \n","all close: False\n","Epoch: 122 \n","mse all (masked)patches: 0.3199161887168884 \n","all close: False\n","Epoch: 123 \n","mse all (masked)patches: 0.31490007042884827 \n","all close: False\n","Epoch: 124 \n","mse all (masked)patches: 0.3171260952949524 \n","all close: False\n","Epoch: 125 \n","mse all (masked)patches: 0.31704413890838623 \n","all close: False\n","Epoch: 126 \n","mse all (masked)patches: 0.31392404437065125 \n","all close: False\n","Epoch: 127 \n","mse all (masked)patches: 0.31409379839897156 \n","all close: False\n","Epoch: 128 \n","mse all (masked)patches: 0.31212782859802246 \n","all close: False\n","Epoch: 129 \n","mse all (masked)patches: 0.3109298050403595 \n","all close: False\n","Epoch: 130 \n","mse all (masked)patches: 0.3101118505001068 \n","all close: False\n","Epoch: 131 \n","mse all (masked)patches: 0.310153990983963 \n","all close: False\n","Epoch: 132 \n","mse all (masked)patches: 0.30888432264328003 \n","all close: False\n","Epoch: 133 \n","mse all (masked)patches: 0.30864042043685913 \n","all close: False\n","Epoch: 134 \n","mse all (masked)patches: 0.3091789186000824 \n","all close: False\n","Epoch: 135 \n","mse all (masked)patches: 0.30894166231155396 \n","all close: False\n","Epoch: 136 \n","mse all (masked)patches: 0.30588555335998535 \n","all close: False\n","Epoch: 137 \n","mse all (masked)patches: 0.3040779232978821 \n","all close: False\n","Epoch: 138 \n","mse all (masked)patches: 0.30471694469451904 \n","all close: False\n","Epoch: 139 \n","mse all (masked)patches: 0.30400457978248596 \n","all close: False\n","Epoch: 140 \n","mse all (masked)patches: 0.3024696111679077 \n","all close: False\n","Epoch: 141 \n","mse all (masked)patches: 0.30489659309387207 \n","all close: False\n","Epoch: 142 \n","mse all (masked)patches: 0.2998306453227997 \n","all close: False\n","Epoch: 143 \n","mse all (masked)patches: 0.3021726906299591 \n","all close: False\n","Epoch: 144 \n","mse all (masked)patches: 0.30003443360328674 \n","all close: False\n","Epoch: 145 \n","mse all (masked)patches: 0.29761502146720886 \n","all close: False\n","Epoch: 146 \n","mse all (masked)patches: 0.29993805289268494 \n","all close: False\n","Epoch: 147 \n","mse all (masked)patches: 0.2963225245475769 \n","all close: False\n","Epoch: 148 \n","mse all (masked)patches: 0.2955719232559204 \n","all close: False\n","Epoch: 149 \n","mse all (masked)patches: 0.29567739367485046 \n","all close: False\n","Epoch: 150 \n","mse all (masked)patches: 0.2941613793373108 \n","all close: False\n","Epoch: 151 \n","mse all (masked)patches: 0.29269787669181824 \n","all close: False\n","Epoch: 152 \n","mse all (masked)patches: 0.2942391037940979 \n","all close: False\n","Epoch: 153 \n","mse all (masked)patches: 0.2933633625507355 \n","all close: False\n","Epoch: 154 \n","mse all (masked)patches: 0.2924988567829132 \n","all close: False\n","Epoch: 155 \n","mse all (masked)patches: 0.28839007019996643 \n","all close: False\n","Epoch: 156 \n","mse all (masked)patches: 0.2884889543056488 \n","all close: False\n","Epoch: 157 \n","mse all (masked)patches: 0.28987547755241394 \n","all close: False\n","Epoch: 158 \n","mse all (masked)patches: 0.28845348954200745 \n","all close: False\n","Epoch: 159 \n","mse all (masked)patches: 0.2901272475719452 \n","all close: False\n","Epoch: 160 \n","mse all (masked)patches: 0.28637996315956116 \n","all close: False\n","Epoch: 161 \n","mse all (masked)patches: 0.2866813838481903 \n","all close: False\n","Epoch: 162 \n","mse all (masked)patches: 0.28260594606399536 \n","all close: False\n","Epoch: 163 \n","mse all (masked)patches: 0.28427502512931824 \n","all close: False\n","Epoch: 164 \n","mse all (masked)patches: 0.28239667415618896 \n","all close: False\n","Epoch: 165 \n","mse all (masked)patches: 0.28183290362358093 \n","all close: False\n","Epoch: 166 \n","mse all (masked)patches: 0.28453654050827026 \n","all close: False\n","Epoch: 167 \n","mse all (masked)patches: 0.280490905046463 \n","all close: False\n","Epoch: 168 \n","mse all (masked)patches: 0.28043222427368164 \n","all close: False\n","Epoch: 169 \n","mse all (masked)patches: 0.2798701226711273 \n","all close: False\n","Epoch: 170 \n","mse all (masked)patches: 0.2791080176830292 \n","all close: False\n","Epoch: 171 \n","mse all (masked)patches: 0.2776795029640198 \n","all close: False\n","Epoch: 172 \n","mse all (masked)patches: 0.276224285364151 \n","all close: False\n","Epoch: 173 \n","mse all (masked)patches: 0.27820903062820435 \n","all close: False\n","Epoch: 174 \n","mse all (masked)patches: 0.2764035761356354 \n","all close: False\n","Epoch: 175 \n","mse all (masked)patches: 0.27560532093048096 \n","all close: False\n","Epoch: 176 \n","mse all (masked)patches: 0.27360156178474426 \n","all close: False\n","Epoch: 177 \n","mse all (masked)patches: 0.2745380103588104 \n","all close: False\n","Epoch: 178 \n","mse all (masked)patches: 0.2720503807067871 \n","all close: False\n","Epoch: 179 \n","mse all (masked)patches: 0.2715320587158203 \n","all close: False\n","Epoch: 180 \n","mse all (masked)patches: 0.2698350250720978 \n","all close: False\n","Epoch: 181 \n","mse all (masked)patches: 0.27210381627082825 \n","all close: False\n","Epoch: 182 \n","mse all (masked)patches: 0.26673632860183716 \n","all close: False\n","Epoch: 183 \n","mse all (masked)patches: 0.27131569385528564 \n","all close: False\n","Epoch: 184 \n","mse all (masked)patches: 0.26808488368988037 \n","all close: False\n","Epoch: 185 \n","mse all (masked)patches: 0.265107661485672 \n","all close: False\n","Epoch: 186 \n","mse all (masked)patches: 0.2662237286567688 \n","all close: False\n","Epoch: 187 \n","mse all (masked)patches: 0.26633864641189575 \n","all close: False\n","Epoch: 188 \n","mse all (masked)patches: 0.26401394605636597 \n","all close: False\n","Epoch: 189 \n","mse all (masked)patches: 0.26543134450912476 \n","all close: False\n","Epoch: 190 \n","mse all (masked)patches: 0.2626297175884247 \n","all close: False\n","Epoch: 191 \n","mse all (masked)patches: 0.2644844055175781 \n","all close: False\n","Epoch: 192 \n","mse all (masked)patches: 0.2616291046142578 \n","all close: False\n","Epoch: 193 \n","mse all (masked)patches: 0.2635394036769867 \n","all close: False\n","Epoch: 194 \n","mse all (masked)patches: 0.25985923409461975 \n","all close: False\n","Epoch: 195 \n","mse all (masked)patches: 0.26254934072494507 \n","all close: False\n","Epoch: 196 \n","mse all (masked)patches: 0.25890931487083435 \n","all close: False\n","Epoch: 197 \n","mse all (masked)patches: 0.2587358355522156 \n","all close: False\n","Epoch: 198 \n","mse all (masked)patches: 0.2572062313556671 \n","all close: False\n","Epoch: 199 \n","mse all (masked)patches: 0.257618248462677 \n","all close: False\n","Epoch: 200 \n","mse all (masked)patches: 0.25825899839401245 \n","all close: False\n","Epoch: 201 \n","mse all (masked)patches: 0.25632673501968384 \n","all close: False\n","Epoch: 202 \n","mse all (masked)patches: 0.2554510533809662 \n","all close: False\n","Epoch: 203 \n","mse all (masked)patches: 0.2550792098045349 \n","all close: False\n","Epoch: 204 \n","mse all (masked)patches: 0.25449344515800476 \n","all close: False\n","Epoch: 205 \n","mse all (masked)patches: 0.2525787055492401 \n","all close: False\n","Epoch: 206 \n","mse all (masked)patches: 0.2530257999897003 \n","all close: False\n","Epoch: 207 \n","mse all (masked)patches: 0.25140026211738586 \n","all close: False\n","Epoch: 208 \n","mse all (masked)patches: 0.2522294819355011 \n","all close: False\n","Epoch: 209 \n","mse all (masked)patches: 0.2518361210823059 \n","all close: False\n","Epoch: 210 \n","mse all (masked)patches: 0.2481805980205536 \n","all close: False\n","Epoch: 211 \n","mse all (masked)patches: 0.249953031539917 \n","all close: False\n","Epoch: 212 \n","mse all (masked)patches: 0.2491435557603836 \n","all close: False\n","Epoch: 213 \n","mse all (masked)patches: 0.24714577198028564 \n","all close: False\n","Epoch: 214 \n","mse all (masked)patches: 0.24819499254226685 \n","all close: False\n","Epoch: 215 \n","mse all (masked)patches: 0.24645639955997467 \n","all close: False\n","Epoch: 216 \n","mse all (masked)patches: 0.24574927985668182 \n","all close: False\n","Epoch: 217 \n","mse all (masked)patches: 0.24573157727718353 \n","all close: False\n","Epoch: 218 \n","mse all (masked)patches: 0.24382439255714417 \n","all close: False\n","Epoch: 219 \n","mse all (masked)patches: 0.2434867024421692 \n","all close: False\n","Epoch: 220 \n","mse all (masked)patches: 0.23965874314308167 \n","all close: False\n","Epoch: 221 \n","mse all (masked)patches: 0.24277111887931824 \n","all close: False\n","Epoch: 222 \n","mse all (masked)patches: 0.23618780076503754 \n","all close: False\n","Epoch: 223 \n","mse all (masked)patches: 0.24042053520679474 \n","all close: False\n","Epoch: 224 \n","mse all (masked)patches: 0.24045011401176453 \n","all close: False\n","Epoch: 225 \n","mse all (masked)patches: 0.24043208360671997 \n","all close: False\n","Epoch: 226 \n","mse all (masked)patches: 0.2375408560037613 \n","all close: False\n","Epoch: 227 \n","mse all (masked)patches: 0.23611901700496674 \n","all close: False\n","Epoch: 228 \n","mse all (masked)patches: 0.2345651537179947 \n","all close: False\n","Epoch: 229 \n","mse all (masked)patches: 0.23547591269016266 \n","all close: False\n","Epoch: 230 \n","mse all (masked)patches: 0.23614293336868286 \n","all close: False\n","Epoch: 231 \n","mse all (masked)patches: 0.2355174869298935 \n","all close: False\n","Epoch: 232 \n","mse all (masked)patches: 0.2358502745628357 \n","all close: False\n","Epoch: 233 \n","mse all (masked)patches: 0.2334643006324768 \n","all close: False\n","Epoch: 234 \n","mse all (masked)patches: 0.23673050105571747 \n","all close: False\n","Epoch: 235 \n","mse all (masked)patches: 0.2330416440963745 \n","all close: False\n","Epoch: 236 \n","mse all (masked)patches: 0.23415015637874603 \n","all close: False\n","Epoch: 237 \n","mse all (masked)patches: 0.23280443251132965 \n","all close: False\n","Epoch: 238 \n","mse all (masked)patches: 0.22827023267745972 \n","all close: False\n","Epoch: 239 \n","mse all (masked)patches: 0.23061853647232056 \n","all close: False\n","Epoch: 240 \n","mse all (masked)patches: 0.22957371175289154 \n","all close: False\n","Epoch: 241 \n","mse all (masked)patches: 0.2321012020111084 \n","all close: False\n","Epoch: 242 \n","mse all (masked)patches: 0.22992928326129913 \n","all close: False\n","Epoch: 243 \n","mse all (masked)patches: 0.22791095077991486 \n","all close: False\n","Epoch: 244 \n","mse all (masked)patches: 0.22477862238883972 \n","all close: False\n","Epoch: 245 \n","mse all (masked)patches: 0.22605465352535248 \n","all close: False\n","Epoch: 246 \n","mse all (masked)patches: 0.22593644261360168 \n","all close: False\n","Epoch: 247 \n","mse all (masked)patches: 0.22475072741508484 \n","all close: False\n","Epoch: 248 \n","mse all (masked)patches: 0.22602438926696777 \n","all close: False\n","Epoch: 249 \n","mse all (masked)patches: 0.22308385372161865 \n","all close: False\n","Epoch: 250 \n","mse all (masked)patches: 0.2229708731174469 \n","all close: False\n","Epoch: 251 \n","mse all (masked)patches: 0.2240828424692154 \n","all close: False\n","Epoch: 252 \n","mse all (masked)patches: 0.22352415323257446 \n","all close: False\n","Epoch: 253 \n","mse all (masked)patches: 0.22007079422473907 \n","all close: False\n","Epoch: 254 \n","mse all (masked)patches: 0.22122406959533691 \n","all close: False\n","Epoch: 255 \n","mse all (masked)patches: 0.2184547781944275 \n","all close: False\n","Epoch: 256 \n","mse all (masked)patches: 0.22276759147644043 \n","all close: False\n","Epoch: 257 \n","mse all (masked)patches: 0.21762700378894806 \n","all close: False\n","Epoch: 258 \n","mse all (masked)patches: 0.2160341590642929 \n","all close: False\n","Epoch: 259 \n","mse all (masked)patches: 0.2197761982679367 \n","all close: False\n","Epoch: 260 \n","mse all (masked)patches: 0.21719565987586975 \n","all close: False\n","Epoch: 261 \n","mse all (masked)patches: 0.21798045933246613 \n","all close: False\n","Epoch: 262 \n","mse all (masked)patches: 0.21806283295154572 \n","all close: False\n","Epoch: 263 \n","mse all (masked)patches: 0.2172820270061493 \n","all close: False\n","Epoch: 264 \n","mse all (masked)patches: 0.21539224684238434 \n","all close: False\n","Epoch: 265 \n","mse all (masked)patches: 0.21446649730205536 \n","all close: False\n","Epoch: 266 \n","mse all (masked)patches: 0.21293872594833374 \n","all close: False\n","Epoch: 267 \n","mse all (masked)patches: 0.2133188396692276 \n","all close: False\n","Epoch: 268 \n","mse all (masked)patches: 0.2141149640083313 \n","all close: False\n","Epoch: 269 \n","mse all (masked)patches: 0.21090613305568695 \n","all close: False\n","Epoch: 270 \n","mse all (masked)patches: 0.2142641395330429 \n","all close: False\n","Epoch: 271 \n","mse all (masked)patches: 0.21021856367588043 \n","all close: False\n","Epoch: 272 \n","mse all (masked)patches: 0.20962314307689667 \n","all close: False\n","Epoch: 273 \n","mse all (masked)patches: 0.21067577600479126 \n","all close: False\n","Epoch: 274 \n","mse all (masked)patches: 0.20955975353717804 \n","all close: False\n","Epoch: 275 \n","mse all (masked)patches: 0.20779964327812195 \n","all close: False\n","Epoch: 276 \n","mse all (masked)patches: 0.2103259265422821 \n","all close: False\n","Epoch: 277 \n","mse all (masked)patches: 0.20751048624515533 \n","all close: False\n","Epoch: 278 \n","mse all (masked)patches: 0.20659635961055756 \n","all close: False\n","Epoch: 279 \n","mse all (masked)patches: 0.2041628509759903 \n","all close: False\n","Epoch: 280 \n","mse all (masked)patches: 0.20603257417678833 \n","all close: False\n","Epoch: 281 \n","mse all (masked)patches: 0.20411363244056702 \n","all close: False\n","Epoch: 282 \n","mse all (masked)patches: 0.20439468324184418 \n","all close: False\n","Epoch: 283 \n","mse all (masked)patches: 0.20158764719963074 \n","all close: False\n","Epoch: 284 \n","mse all (masked)patches: 0.2051527500152588 \n","all close: False\n","Epoch: 285 \n","mse all (masked)patches: 0.20052042603492737 \n","all close: False\n","Epoch: 286 \n","mse all (masked)patches: 0.20052936673164368 \n","all close: False\n","Epoch: 287 \n","mse all (masked)patches: 0.20292454957962036 \n","all close: False\n","Epoch: 288 \n","mse all (masked)patches: 0.2035515159368515 \n","all close: False\n","Epoch: 289 \n","mse all (masked)patches: 0.1998082399368286 \n","all close: False\n","Epoch: 290 \n","mse all (masked)patches: 0.20015770196914673 \n","all close: False\n","Epoch: 291 \n","mse all (masked)patches: 0.2000778317451477 \n","all close: False\n","Epoch: 292 \n","mse all (masked)patches: 0.20026005804538727 \n","all close: False\n","Epoch: 293 \n","mse all (masked)patches: 0.19917018711566925 \n","all close: False\n","Epoch: 294 \n","mse all (masked)patches: 0.1974126249551773 \n","all close: False\n","Epoch: 295 \n","mse all (masked)patches: 0.1967884749174118 \n","all close: False\n","Epoch: 296 \n","mse all (masked)patches: 0.19741801917552948 \n","all close: False\n","Epoch: 297 \n","mse all (masked)patches: 0.19583643972873688 \n","all close: False\n","Epoch: 298 \n","mse all (masked)patches: 0.19258305430412292 \n","all close: False\n","Epoch: 299 \n","mse all (masked)patches: 0.1959764063358307 \n","all close: False\n","Epoch: 300 \n","mse all (masked)patches: 0.19474796950817108 \n","all close: False\n","Epoch: 301 \n","mse all (masked)patches: 0.19437311589717865 \n","all close: False\n","Epoch: 302 \n","mse all (masked)patches: 0.19449591636657715 \n","all close: False\n","Epoch: 303 \n","mse all (masked)patches: 0.1924305260181427 \n","all close: False\n","Epoch: 304 \n","mse all (masked)patches: 0.193630650639534 \n","all close: False\n","Epoch: 305 \n","mse all (masked)patches: 0.19275258481502533 \n","all close: False\n","Epoch: 306 \n","mse all (masked)patches: 0.19136542081832886 \n","all close: False\n","Epoch: 307 \n","mse all (masked)patches: 0.1921001672744751 \n","all close: False\n","Epoch: 308 \n","mse all (masked)patches: 0.18972621858119965 \n","all close: False\n","Epoch: 309 \n","mse all (masked)patches: 0.1899198591709137 \n","all close: False\n","Epoch: 310 \n","mse all (masked)patches: 0.1904064267873764 \n","all close: False\n","Epoch: 311 \n","mse all (masked)patches: 0.1878795027732849 \n","all close: False\n","Epoch: 312 \n","mse all (masked)patches: 0.18684259057044983 \n","all close: False\n","Epoch: 313 \n","mse all (masked)patches: 0.1870085597038269 \n","all close: False\n","Epoch: 314 \n","mse all (masked)patches: 0.18670868873596191 \n","all close: False\n","Epoch: 315 \n","mse all (masked)patches: 0.18593160808086395 \n","all close: False\n","Epoch: 316 \n","mse all (masked)patches: 0.18810544908046722 \n","all close: False\n","Epoch: 317 \n","mse all (masked)patches: 0.18472230434417725 \n","all close: False\n","Epoch: 318 \n","mse all (masked)patches: 0.18637049198150635 \n","all close: False\n","Epoch: 319 \n","mse all (masked)patches: 0.18277740478515625 \n","all close: False\n","Epoch: 320 \n","mse all (masked)patches: 0.18512406945228577 \n","all close: False\n","Epoch: 321 \n","mse all (masked)patches: 0.1844872087240219 \n","all close: False\n","Epoch: 322 \n","mse all (masked)patches: 0.18368856608867645 \n","all close: False\n","Epoch: 323 \n","mse all (masked)patches: 0.18229883909225464 \n","all close: False\n","Epoch: 324 \n","mse all (masked)patches: 0.18146046996116638 \n","all close: False\n","Epoch: 325 \n","mse all (masked)patches: 0.18303067982196808 \n","all close: False\n","Epoch: 326 \n","mse all (masked)patches: 0.17871560156345367 \n","all close: False\n","Epoch: 327 \n","mse all (masked)patches: 0.18108922243118286 \n","all close: False\n","Epoch: 328 \n","mse all (masked)patches: 0.17871402204036713 \n","all close: False\n","Epoch: 329 \n","mse all (masked)patches: 0.17899896204471588 \n","all close: False\n","Epoch: 330 \n","mse all (masked)patches: 0.17889565229415894 \n","all close: False\n","Epoch: 331 \n","mse all (masked)patches: 0.18069641292095184 \n","all close: False\n","Epoch: 332 \n","mse all (masked)patches: 0.17872728407382965 \n","all close: False\n","Epoch: 333 \n","mse all (masked)patches: 0.177641823887825 \n","all close: False\n","Epoch: 334 \n","mse all (masked)patches: 0.1798698455095291 \n","all close: False\n","Epoch: 335 \n","mse all (masked)patches: 0.1803935468196869 \n","all close: False\n","Epoch: 336 \n","mse all (masked)patches: 0.17593444883823395 \n","all close: False\n","Epoch: 337 \n","mse all (masked)patches: 0.1772352159023285 \n","all close: False\n","Epoch: 338 \n","mse all (masked)patches: 0.17667409777641296 \n","all close: False\n","Epoch: 339 \n","mse all (masked)patches: 0.17527003586292267 \n","all close: False\n","Epoch: 340 \n","mse all (masked)patches: 0.17527888715267181 \n","all close: False\n","Epoch: 341 \n","mse all (masked)patches: 0.17721988260746002 \n","all close: False\n","Epoch: 342 \n","mse all (masked)patches: 0.1766217052936554 \n","all close: False\n","Epoch: 343 \n","mse all (masked)patches: 0.1735030561685562 \n","all close: False\n","Epoch: 344 \n","mse all (masked)patches: 0.17463333904743195 \n","all close: False\n","Epoch: 345 \n","mse all (masked)patches: 0.17537741363048553 \n","all close: False\n","Epoch: 346 \n","mse all (masked)patches: 0.17133766412734985 \n","all close: False\n","Epoch: 347 \n","mse all (masked)patches: 0.1734136939048767 \n","all close: False\n","Epoch: 348 \n","mse all (masked)patches: 0.17063257098197937 \n","all close: False\n","Epoch: 349 \n","mse all (masked)patches: 0.17304441332817078 \n","all close: False\n","Epoch: 350 \n","mse all (masked)patches: 0.17235341668128967 \n","all close: False\n","Epoch: 351 \n","mse all (masked)patches: 0.16857676208019257 \n","all close: False\n","Epoch: 352 \n","mse all (masked)patches: 0.17048117518424988 \n","all close: False\n","Epoch: 353 \n","mse all (masked)patches: 0.16963639855384827 \n","all close: False\n","Epoch: 354 \n","mse all (masked)patches: 0.1676536351442337 \n","all close: False\n","Epoch: 355 \n","mse all (masked)patches: 0.16569232940673828 \n","all close: False\n","Epoch: 356 \n","mse all (masked)patches: 0.16925331950187683 \n","all close: False\n","Epoch: 357 \n","mse all (masked)patches: 0.1656796932220459 \n","all close: False\n","Epoch: 358 \n","mse all (masked)patches: 0.1671544462442398 \n","all close: False\n","Epoch: 359 \n","mse all (masked)patches: 0.16817986965179443 \n","all close: False\n","Epoch: 360 \n","mse all (masked)patches: 0.16662055253982544 \n","all close: False\n","Epoch: 361 \n","mse all (masked)patches: 0.16821788251399994 \n","all close: False\n","Epoch: 362 \n","mse all (masked)patches: 0.16330096125602722 \n","all close: False\n","Epoch: 363 \n","mse all (masked)patches: 0.16533219814300537 \n","all close: False\n","Epoch: 364 \n","mse all (masked)patches: 0.16301314532756805 \n","all close: False\n","Epoch: 365 \n","mse all (masked)patches: 0.1645638644695282 \n","all close: False\n","Epoch: 366 \n","mse all (masked)patches: 0.15985430777072906 \n","all close: False\n","Epoch: 367 \n","mse all (masked)patches: 0.16403338313102722 \n","all close: False\n","Epoch: 368 \n","mse all (masked)patches: 0.16281050443649292 \n","all close: False\n","Epoch: 369 \n","mse all (masked)patches: 0.1644110381603241 \n","all close: False\n","Epoch: 370 \n","mse all (masked)patches: 0.16308966279029846 \n","all close: False\n","Epoch: 371 \n","mse all (masked)patches: 0.16312448680400848 \n","all close: False\n","Epoch: 372 \n","mse all (masked)patches: 0.1601535975933075 \n","all close: False\n","Epoch: 373 \n","mse all (masked)patches: 0.16082604229450226 \n","all close: False\n","Epoch: 374 \n","mse all (masked)patches: 0.15884967148303986 \n","all close: False\n","Epoch: 375 \n","mse all (masked)patches: 0.16014744341373444 \n","all close: False\n","Epoch: 376 \n","mse all (masked)patches: 0.1594305783510208 \n","all close: False\n","Epoch: 377 \n","mse all (masked)patches: 0.16018524765968323 \n","all close: False\n","Epoch: 378 \n","mse all (masked)patches: 0.15820810198783875 \n","all close: False\n","Epoch: 379 \n","mse all (masked)patches: 0.1586911380290985 \n","all close: False\n","Epoch: 380 \n","mse all (masked)patches: 0.15951621532440186 \n","all close: False\n","Epoch: 381 \n","mse all (masked)patches: 0.1617138832807541 \n","all close: False\n","Epoch: 382 \n","mse all (masked)patches: 0.16141188144683838 \n","all close: False\n","Epoch: 383 \n","mse all (masked)patches: 0.15915390849113464 \n","all close: False\n","Epoch: 384 \n","mse all (masked)patches: 0.1574249565601349 \n","all close: False\n","Epoch: 385 \n","mse all (masked)patches: 0.15592968463897705 \n","all close: False\n","Epoch: 386 \n","mse all (masked)patches: 0.15595334768295288 \n","all close: False\n","Epoch: 387 \n","mse all (masked)patches: 0.15613655745983124 \n","all close: False\n","Epoch: 388 \n","mse all (masked)patches: 0.1565369814634323 \n","all close: False\n","Epoch: 389 \n","mse all (masked)patches: 0.15858714282512665 \n","all close: False\n","Epoch: 390 \n","mse all (masked)patches: 0.15830577909946442 \n","all close: False\n","Epoch: 391 \n","mse all (masked)patches: 0.15374234318733215 \n","all close: False\n","Epoch: 392 \n","mse all (masked)patches: 0.15258701145648956 \n","all close: False\n","Epoch: 393 \n","mse all (masked)patches: 0.153436079621315 \n","all close: False\n","Epoch: 394 \n","mse all (masked)patches: 0.15208719670772552 \n","all close: False\n","Epoch: 395 \n","mse all (masked)patches: 0.1536085307598114 \n","all close: False\n","Epoch: 396 \n","mse all (masked)patches: 0.15363788604736328 \n","all close: False\n","Epoch: 397 \n","mse all (masked)patches: 0.15155717730522156 \n","all close: False\n","Epoch: 398 \n","mse all (masked)patches: 0.1519624888896942 \n","all close: False\n","Epoch: 399 \n","mse all (masked)patches: 0.15399347245693207 \n","all close: False\n","Epoch: 400 \n","mse all (masked)patches: 0.1544983983039856 \n","all close: False\n","Epoch: 401 \n","mse all (masked)patches: 0.15068629384040833 \n","all close: False\n","Epoch: 402 \n","mse all (masked)patches: 0.1531050205230713 \n","all close: False\n","Epoch: 403 \n","mse all (masked)patches: 0.15287351608276367 \n","all close: False\n","Epoch: 404 \n","mse all (masked)patches: 0.1504346877336502 \n","all close: False\n","Epoch: 405 \n","mse all (masked)patches: 0.15082795917987823 \n","all close: False\n","Epoch: 406 \n","mse all (masked)patches: 0.1495073437690735 \n","all close: False\n","Epoch: 407 \n","mse all (masked)patches: 0.14964789152145386 \n","all close: False\n","Epoch: 408 \n","mse all (masked)patches: 0.1495531052350998 \n","all close: False\n","Epoch: 409 \n","mse all (masked)patches: 0.1492619663476944 \n","all close: False\n","Epoch: 410 \n","mse all (masked)patches: 0.14510716497898102 \n","all close: False\n","Epoch: 411 \n","mse all (masked)patches: 0.1494055539369583 \n","all close: False\n","Epoch: 412 \n","mse all (masked)patches: 0.1504059135913849 \n","all close: False\n","Epoch: 413 \n","mse all (masked)patches: 0.15041787922382355 \n","all close: False\n","Epoch: 414 \n","mse all (masked)patches: 0.14716137945652008 \n","all close: False\n","Epoch: 415 \n","mse all (masked)patches: 0.14797700941562653 \n","all close: False\n","Epoch: 416 \n","mse all (masked)patches: 0.14439210295677185 \n","all close: False\n","Epoch: 417 \n","mse all (masked)patches: 0.14931249618530273 \n","all close: False\n","Epoch: 418 \n","mse all (masked)patches: 0.14582696557044983 \n","all close: False\n","Epoch: 419 \n","mse all (masked)patches: 0.146165668964386 \n","all close: False\n","Epoch: 420 \n","mse all (masked)patches: 0.15011052787303925 \n","all close: False\n","Epoch: 421 \n","mse all (masked)patches: 0.14858442544937134 \n","all close: False\n","Epoch: 422 \n","mse all (masked)patches: 0.14777231216430664 \n","all close: False\n","Epoch: 423 \n","mse all (masked)patches: 0.14366279542446136 \n","all close: False\n","Epoch: 424 \n","mse all (masked)patches: 0.14552049338817596 \n","all close: False\n","Epoch: 425 \n","mse all (masked)patches: 0.14437632262706757 \n","all close: False\n","Epoch: 426 \n","mse all (masked)patches: 0.14777760207653046 \n","all close: False\n","Epoch: 427 \n","mse all (masked)patches: 0.1407472938299179 \n","all close: False\n","Epoch: 428 \n","mse all (masked)patches: 0.14378581941127777 \n","all close: False\n","Epoch: 429 \n","mse all (masked)patches: 0.1426994800567627 \n","all close: False\n","Epoch: 430 \n","mse all (masked)patches: 0.14482271671295166 \n","all close: False\n","Epoch: 431 \n","mse all (masked)patches: 0.141408309340477 \n","all close: False\n","Epoch: 432 \n","mse all (masked)patches: 0.14365342259407043 \n","all close: False\n","Epoch: 433 \n","mse all (masked)patches: 0.14335845410823822 \n","all close: False\n","Epoch: 434 \n","mse all (masked)patches: 0.14182068407535553 \n","all close: False\n","Epoch: 435 \n","mse all (masked)patches: 0.14148657023906708 \n","all close: False\n","Epoch: 436 \n","mse all (masked)patches: 0.1403057873249054 \n","all close: False\n","Epoch: 437 \n","mse all (masked)patches: 0.14221414923667908 \n","all close: False\n","Epoch: 438 \n","mse all (masked)patches: 0.14123941957950592 \n","all close: False\n","Epoch: 439 \n","mse all (masked)patches: 0.1404017210006714 \n","all close: False\n","Epoch: 440 \n","mse all (masked)patches: 0.14133258163928986 \n","all close: False\n","Epoch: 441 \n","mse all (masked)patches: 0.13713732361793518 \n","all close: False\n","Epoch: 442 \n","mse all (masked)patches: 0.13982391357421875 \n","all close: False\n","Epoch: 443 \n","mse all (masked)patches: 0.13898701965808868 \n","all close: False\n","Epoch: 444 \n","mse all (masked)patches: 0.13719654083251953 \n","all close: False\n","Epoch: 445 \n","mse all (masked)patches: 0.13682669401168823 \n","all close: False\n","Epoch: 446 \n","mse all (masked)patches: 0.13872483372688293 \n","all close: False\n","Epoch: 447 \n","mse all (masked)patches: 0.13625766336917877 \n","all close: False\n","Epoch: 448 \n","mse all (masked)patches: 0.13672131299972534 \n","all close: False\n","Epoch: 449 \n","mse all (masked)patches: 0.13810251653194427 \n","all close: False\n","Epoch: 450 \n","mse all (masked)patches: 0.13754905760288239 \n","all close: False\n","Epoch: 451 \n","mse all (masked)patches: 0.13507643342018127 \n","all close: False\n","Epoch: 452 \n","mse all (masked)patches: 0.13871072232723236 \n","all close: False\n","Epoch: 453 \n","mse all (masked)patches: 0.13639764487743378 \n","all close: False\n","Epoch: 454 \n","mse all (masked)patches: 0.13928526639938354 \n","all close: False\n","Epoch: 455 \n","mse all (masked)patches: 0.1358065903186798 \n","all close: False\n","Epoch: 456 \n","mse all (masked)patches: 0.13758599758148193 \n","all close: False\n","Epoch: 457 \n","mse all (masked)patches: 0.1365167796611786 \n","all close: False\n","Epoch: 458 \n","mse all (masked)patches: 0.13548552989959717 \n","all close: False\n","Epoch: 459 \n","mse all (masked)patches: 0.13743886351585388 \n","all close: False\n","Epoch: 460 \n","mse all (masked)patches: 0.13499228656291962 \n","all close: False\n","Epoch: 461 \n","mse all (masked)patches: 0.13448642194271088 \n","all close: False\n","Epoch: 462 \n","mse all (masked)patches: 0.13534100353717804 \n","all close: False\n","Epoch: 463 \n","mse all (masked)patches: 0.13420172035694122 \n","all close: False\n","Epoch: 464 \n","mse all (masked)patches: 0.13411398231983185 \n","all close: False\n","Epoch: 465 \n","mse all (masked)patches: 0.13194377720355988 \n","all close: False\n","Epoch: 466 \n","mse all (masked)patches: 0.13043241202831268 \n","all close: False\n","Epoch: 467 \n","mse all (masked)patches: 0.13096275925636292 \n","all close: False\n","Epoch: 468 \n","mse all (masked)patches: 0.1328447312116623 \n","all close: False\n","Epoch: 469 \n","mse all (masked)patches: 0.13172969222068787 \n","all close: False\n","Epoch: 470 \n","mse all (masked)patches: 0.13138432800769806 \n","all close: False\n","Epoch: 471 \n","mse all (masked)patches: 0.13309703767299652 \n","all close: False\n","Epoch: 472 \n","mse all (masked)patches: 0.13356514275074005 \n","all close: False\n","Epoch: 473 \n","mse all (masked)patches: 0.1317010372877121 \n","all close: False\n","Epoch: 474 \n","mse all (masked)patches: 0.12989407777786255 \n","all close: False\n","Epoch: 475 \n","mse all (masked)patches: 0.13633783161640167 \n","all close: False\n","Epoch: 476 \n","mse all (masked)patches: 0.13457632064819336 \n","all close: False\n","Epoch: 477 \n","mse all (masked)patches: 0.1314258873462677 \n","all close: False\n","Epoch: 478 \n","mse all (masked)patches: 0.13205355405807495 \n","all close: False\n","Epoch: 479 \n","mse all (masked)patches: 0.12983731925487518 \n","all close: False\n","Epoch: 480 \n","mse all (masked)patches: 0.13015390932559967 \n","all close: False\n","Epoch: 481 \n","mse all (masked)patches: 0.1316300630569458 \n","all close: False\n","Epoch: 482 \n","mse all (masked)patches: 0.1301528811454773 \n","all close: False\n","Epoch: 483 \n","mse all (masked)patches: 0.1307813674211502 \n","all close: False\n","Epoch: 484 \n","mse all (masked)patches: 0.12998270988464355 \n","all close: False\n","Epoch: 485 \n","mse all (masked)patches: 0.1304738074541092 \n","all close: False\n","Epoch: 486 \n","mse all (masked)patches: 0.13031038641929626 \n","all close: False\n","Epoch: 487 \n","mse all (masked)patches: 0.1296563744544983 \n","all close: False\n","Epoch: 488 \n","mse all (masked)patches: 0.1276608407497406 \n","all close: False\n","Epoch: 489 \n","mse all (masked)patches: 0.12839269638061523 \n","all close: False\n","Epoch: 490 \n","mse all (masked)patches: 0.1276729553937912 \n","all close: False\n","Epoch: 491 \n","mse all (masked)patches: 0.1270432472229004 \n","all close: False\n","Epoch: 492 \n","mse all (masked)patches: 0.13085469603538513 \n","all close: False\n","Epoch: 493 \n","mse all (masked)patches: 0.12788933515548706 \n","all close: False\n","Epoch: 494 \n","mse all (masked)patches: 0.12797091901302338 \n","all close: False\n","Epoch: 495 \n","mse all (masked)patches: 0.1272878348827362 \n","all close: False\n","Epoch: 496 \n","mse all (masked)patches: 0.12742212414741516 \n","all close: False\n","Epoch: 497 \n","mse all (masked)patches: 0.12788815796375275 \n","all close: False\n","Epoch: 498 \n","mse all (masked)patches: 0.12777945399284363 \n","all close: False\n","Epoch: 499 \n","mse all (masked)patches: 0.12607300281524658 \n","all close: False\n","Epoch: 500 \n","mse all (masked)patches: 0.12784993648529053 \n","all close: False\n","Epoch: 501 \n","mse all (masked)patches: 0.12572559714317322 \n","all close: False\n","Epoch: 502 \n","mse all (masked)patches: 0.1242060735821724 \n","all close: False\n","Epoch: 503 \n","mse all (masked)patches: 0.12741778790950775 \n","all close: False\n","Epoch: 504 \n","mse all (masked)patches: 0.12752658128738403 \n","all close: False\n","Epoch: 505 \n","mse all (masked)patches: 0.12679916620254517 \n","all close: False\n","Epoch: 506 \n","mse all (masked)patches: 0.12895876169204712 \n","all close: False\n","Epoch: 507 \n","mse all (masked)patches: 0.1272534728050232 \n","all close: False\n","Epoch: 508 \n","mse all (masked)patches: 0.12480983138084412 \n","all close: False\n","Epoch: 509 \n","mse all (masked)patches: 0.12211764603853226 \n","all close: False\n","Epoch: 510 \n","mse all (masked)patches: 0.1268579661846161 \n","all close: False\n","Epoch: 511 \n","mse all (masked)patches: 0.12412561476230621 \n","all close: False\n","Epoch: 512 \n","mse all (masked)patches: 0.12889128923416138 \n","all close: False\n","Epoch: 513 \n","mse all (masked)patches: 0.12398357689380646 \n","all close: False\n","Epoch: 514 \n","mse all (masked)patches: 0.1231202632188797 \n","all close: False\n","Epoch: 515 \n","mse all (masked)patches: 0.12421806156635284 \n","all close: False\n","Epoch: 516 \n","mse all (masked)patches: 0.12002882361412048 \n","all close: False\n","Epoch: 517 \n","mse all (masked)patches: 0.12353330850601196 \n","all close: False\n","Epoch: 518 \n","mse all (masked)patches: 0.1185324415564537 \n","all close: False\n","Epoch: 519 \n","mse all (masked)patches: 0.12392662465572357 \n","all close: False\n","Epoch: 520 \n","mse all (masked)patches: 0.12476374208927155 \n","all close: False\n","Epoch: 521 \n","mse all (masked)patches: 0.1257869452238083 \n","all close: False\n","Epoch: 522 \n","mse all (masked)patches: 0.12083616107702255 \n","all close: False\n","Epoch: 523 \n","mse all (masked)patches: 0.11950559914112091 \n","all close: False\n","Epoch: 524 \n","mse all (masked)patches: 0.12219163775444031 \n","all close: False\n","Epoch: 525 \n","mse all (masked)patches: 0.11953600496053696 \n","all close: False\n","Epoch: 526 \n","mse all (masked)patches: 0.12391060590744019 \n","all close: False\n","Epoch: 527 \n","mse all (masked)patches: 0.12174195796251297 \n","all close: False\n","Epoch: 528 \n","mse all (masked)patches: 0.11999081820249557 \n","all close: False\n","Epoch: 529 \n","mse all (masked)patches: 0.12189746648073196 \n","all close: False\n","Epoch: 530 \n","mse all (masked)patches: 0.11728045344352722 \n","all close: False\n","Epoch: 531 \n","mse all (masked)patches: 0.12067574262619019 \n","all close: False\n","Epoch: 532 \n","mse all (masked)patches: 0.12220288068056107 \n","all close: False\n","Epoch: 533 \n","mse all (masked)patches: 0.1206214502453804 \n","all close: False\n","Epoch: 534 \n","mse all (masked)patches: 0.11973509192466736 \n","all close: False\n","Epoch: 535 \n","mse all (masked)patches: 0.11813294887542725 \n","all close: False\n","Epoch: 536 \n","mse all (masked)patches: 0.11756937205791473 \n","all close: False\n","Epoch: 537 \n","mse all (masked)patches: 0.11843779683113098 \n","all close: False\n","Epoch: 538 \n","mse all (masked)patches: 0.12112278491258621 \n","all close: False\n","Epoch: 539 \n","mse all (masked)patches: 0.11598221212625504 \n","all close: False\n","Epoch: 540 \n","mse all (masked)patches: 0.1210426539182663 \n","all close: False\n","Epoch: 541 \n","mse all (masked)patches: 0.11932966858148575 \n","all close: False\n","Epoch: 542 \n","mse all (masked)patches: 0.12015273422002792 \n","all close: False\n","Epoch: 543 \n","mse all (masked)patches: 0.11606563627719879 \n","all close: False\n","Epoch: 544 \n","mse all (masked)patches: 0.11996698379516602 \n","all close: False\n","Epoch: 545 \n","mse all (masked)patches: 0.1179991364479065 \n","all close: False\n","Epoch: 546 \n","mse all (masked)patches: 0.1178082525730133 \n","all close: False\n","Epoch: 547 \n","mse all (masked)patches: 0.12009771913290024 \n","all close: False\n","Epoch: 548 \n","mse all (masked)patches: 0.11814350634813309 \n","all close: False\n","Epoch: 549 \n","mse all (masked)patches: 0.12229685485363007 \n","all close: False\n","Epoch: 550 \n","mse all (masked)patches: 0.11880980432033539 \n","all close: False\n","Epoch: 551 \n","mse all (masked)patches: 0.1182316318154335 \n","all close: False\n","Epoch: 552 \n","mse all (masked)patches: 0.11685602366924286 \n","all close: False\n","Epoch: 553 \n","mse all (masked)patches: 0.11760351061820984 \n","all close: False\n","Epoch: 554 \n","mse all (masked)patches: 0.11844826489686966 \n","all close: False\n","Epoch: 555 \n","mse all (masked)patches: 0.11370519548654556 \n","all close: False\n","Epoch: 556 \n","mse all (masked)patches: 0.11589039862155914 \n","all close: False\n","Epoch: 557 \n","mse all (masked)patches: 0.12059143930673599 \n","all close: False\n","Epoch: 558 \n","mse all (masked)patches: 0.11984391510486603 \n","all close: False\n","Epoch: 559 \n","mse all (masked)patches: 0.11792819947004318 \n","all close: False\n","Epoch: 560 \n","mse all (masked)patches: 0.11648073047399521 \n","all close: False\n","Epoch: 561 \n","mse all (masked)patches: 0.11758162081241608 \n","all close: False\n","Epoch: 562 \n","mse all (masked)patches: 0.11550749838352203 \n","all close: False\n","Epoch: 563 \n","mse all (masked)patches: 0.11833324283361435 \n","all close: False\n","Epoch: 564 \n","mse all (masked)patches: 0.11494117230176926 \n","all close: False\n","Epoch: 565 \n","mse all (masked)patches: 0.11831259727478027 \n","all close: False\n","Epoch: 566 \n","mse all (masked)patches: 0.11464731395244598 \n","all close: False\n","Epoch: 567 \n","mse all (masked)patches: 0.11608855426311493 \n","all close: False\n","Epoch: 568 \n","mse all (masked)patches: 0.1174580305814743 \n","all close: False\n","Epoch: 569 \n","mse all (masked)patches: 0.1161249503493309 \n","all close: False\n","Epoch: 570 \n","mse all (masked)patches: 0.11594723165035248 \n","all close: False\n","Epoch: 571 \n","mse all (masked)patches: 0.11275391280651093 \n","all close: False\n","Epoch: 572 \n","mse all (masked)patches: 0.11697123944759369 \n","all close: False\n","Epoch: 573 \n","mse all (masked)patches: 0.11647047847509384 \n","all close: False\n","Epoch: 574 \n","mse all (masked)patches: 0.11426140367984772 \n","all close: False\n","Epoch: 575 \n","mse all (masked)patches: 0.11621502041816711 \n","all close: False\n","Epoch: 576 \n","mse all (masked)patches: 0.1123330220580101 \n","all close: False\n","Epoch: 577 \n","mse all (masked)patches: 0.11602414399385452 \n","all close: False\n","Epoch: 578 \n","mse all (masked)patches: 0.11363827437162399 \n","all close: False\n","Epoch: 579 \n","mse all (masked)patches: 0.11372682452201843 \n","all close: False\n","Epoch: 580 \n","mse all (masked)patches: 0.11309412121772766 \n","all close: False\n","Epoch: 581 \n","mse all (masked)patches: 0.11471862345933914 \n","all close: False\n","Epoch: 582 \n","mse all (masked)patches: 0.11301899701356888 \n","all close: False\n","Epoch: 583 \n","mse all (masked)patches: 0.11291934549808502 \n","all close: False\n","Epoch: 584 \n","mse all (masked)patches: 0.11469724029302597 \n","all close: False\n","Epoch: 585 \n","mse all (masked)patches: 0.11001851409673691 \n","all close: False\n","Epoch: 586 \n","mse all (masked)patches: 0.11143893003463745 \n","all close: False\n","Epoch: 587 \n","mse all (masked)patches: 0.11624261736869812 \n","all close: False\n","Epoch: 588 \n","mse all (masked)patches: 0.11659181118011475 \n","all close: False\n","Epoch: 589 \n","mse all (masked)patches: 0.11340466886758804 \n","all close: False\n","Epoch: 590 \n","mse all (masked)patches: 0.11221271753311157 \n","all close: False\n","Epoch: 591 \n","mse all (masked)patches: 0.11272148787975311 \n","all close: False\n","Epoch: 592 \n","mse all (masked)patches: 0.10986429452896118 \n","all close: False\n","Epoch: 593 \n","mse all (masked)patches: 0.11457828432321548 \n","all close: False\n","Epoch: 594 \n","mse all (masked)patches: 0.10986905544996262 \n","all close: False\n","Epoch: 595 \n","mse all (masked)patches: 0.11203795671463013 \n","all close: False\n","Epoch: 596 \n","mse all (masked)patches: 0.11133573949337006 \n","all close: False\n","Epoch: 597 \n","mse all (masked)patches: 0.11210069060325623 \n","all close: False\n","Epoch: 598 \n","mse all (masked)patches: 0.11036426573991776 \n","all close: False\n","Epoch: 599 \n","mse all (masked)patches: 0.10923624038696289 \n","all close: False\n","Epoch: 600 \n","mse all (masked)patches: 0.11131644248962402 \n","all close: False\n","Epoch: 601 \n","mse all (masked)patches: 0.10985395312309265 \n","all close: False\n","Epoch: 602 \n","mse all (masked)patches: 0.11233547329902649 \n","all close: False\n","Epoch: 603 \n","mse all (masked)patches: 0.10869190096855164 \n","all close: False\n","Epoch: 604 \n","mse all (masked)patches: 0.11187172681093216 \n","all close: False\n","Epoch: 605 \n","mse all (masked)patches: 0.1101110428571701 \n","all close: False\n","Epoch: 606 \n","mse all (masked)patches: 0.1109526976943016 \n","all close: False\n","Epoch: 607 \n","mse all (masked)patches: 0.11006004363298416 \n","all close: False\n","Epoch: 608 \n","mse all (masked)patches: 0.10813377797603607 \n","all close: False\n","Epoch: 609 \n","mse all (masked)patches: 0.10832244902849197 \n","all close: False\n","Epoch: 610 \n","mse all (masked)patches: 0.1127583235502243 \n","all close: False\n","Epoch: 611 \n","mse all (masked)patches: 0.10993383824825287 \n","all close: False\n","Epoch: 612 \n","mse all (masked)patches: 0.11015652120113373 \n","all close: False\n","Epoch: 613 \n","mse all (masked)patches: 0.10964273661375046 \n","all close: False\n","Epoch: 614 \n","mse all (masked)patches: 0.10888689756393433 \n","all close: False\n","Epoch: 615 \n","mse all (masked)patches: 0.10786136984825134 \n","all close: False\n","Epoch: 616 \n","mse all (masked)patches: 0.10922995954751968 \n","all close: False\n","Epoch: 617 \n","mse all (masked)patches: 0.10940086841583252 \n","all close: False\n","Epoch: 618 \n","mse all (masked)patches: 0.10614576935768127 \n","all close: False\n","Epoch: 619 \n","mse all (masked)patches: 0.10953469574451447 \n","all close: False\n","Epoch: 620 \n","mse all (masked)patches: 0.1106535941362381 \n","all close: False\n","Epoch: 621 \n","mse all (masked)patches: 0.10728655010461807 \n","all close: False\n","Epoch: 622 \n","mse all (masked)patches: 0.10688699781894684 \n","all close: False\n","Epoch: 623 \n","mse all (masked)patches: 0.11062709987163544 \n","all close: False\n","Epoch: 624 \n","mse all (masked)patches: 0.10696343332529068 \n","all close: False\n","Epoch: 625 \n","mse all (masked)patches: 0.10603994131088257 \n","all close: False\n","Epoch: 626 \n","mse all (masked)patches: 0.10683369636535645 \n","all close: False\n","Epoch: 627 \n","mse all (masked)patches: 0.10717528313398361 \n","all close: False\n","Epoch: 628 \n","mse all (masked)patches: 0.10533041507005692 \n","all close: False\n","Epoch: 629 \n","mse all (masked)patches: 0.10643548518419266 \n","all close: False\n","Epoch: 630 \n","mse all (masked)patches: 0.10533233731985092 \n","all close: False\n","Epoch: 631 \n","mse all (masked)patches: 0.10604355484247208 \n","all close: False\n","Epoch: 632 \n","mse all (masked)patches: 0.10931476950645447 \n","all close: False\n","Epoch: 633 \n","mse all (masked)patches: 0.11097854375839233 \n","all close: False\n","Epoch: 634 \n","mse all (masked)patches: 0.10868487507104874 \n","all close: False\n","Epoch: 635 \n","mse all (masked)patches: 0.10941388458013535 \n","all close: False\n","Epoch: 636 \n","mse all (masked)patches: 0.10730218887329102 \n","all close: False\n","Epoch: 637 \n","mse all (masked)patches: 0.10774712264537811 \n","all close: False\n","Epoch: 638 \n","mse all (masked)patches: 0.10215045511722565 \n","all close: False\n","Epoch: 639 \n","mse all (masked)patches: 0.10208047926425934 \n","all close: False\n","Epoch: 640 \n","mse all (masked)patches: 0.10951494425535202 \n","all close: False\n","Epoch: 641 \n","mse all (masked)patches: 0.10837696492671967 \n","all close: False\n","Epoch: 642 \n","mse all (masked)patches: 0.1053084135055542 \n","all close: False\n","Epoch: 643 \n","mse all (masked)patches: 0.10660189390182495 \n","all close: False\n","Epoch: 644 \n","mse all (masked)patches: 0.11298426985740662 \n","all close: False\n","Epoch: 645 \n","mse all (masked)patches: 0.10524032264947891 \n","all close: False\n","Epoch: 646 \n","mse all (masked)patches: 0.10738911479711533 \n","all close: False\n","Epoch: 647 \n","mse all (masked)patches: 0.10817107558250427 \n","all close: False\n","Epoch: 648 \n","mse all (masked)patches: 0.10791799426078796 \n","all close: False\n","Epoch: 649 \n","mse all (masked)patches: 0.10653063654899597 \n","all close: False\n","Epoch: 650 \n","mse all (masked)patches: 0.10533203184604645 \n","all close: False\n","Epoch: 651 \n","mse all (masked)patches: 0.10551375895738602 \n","all close: False\n","Epoch: 652 \n","mse all (masked)patches: 0.10616695880889893 \n","all close: False\n","Epoch: 653 \n","mse all (masked)patches: 0.10849002003669739 \n","all close: False\n","Epoch: 654 \n","mse all (masked)patches: 0.10208119451999664 \n","all close: False\n","Epoch: 655 \n","mse all (masked)patches: 0.10222684592008591 \n","all close: False\n","Epoch: 656 \n","mse all (masked)patches: 0.10747864842414856 \n","all close: False\n","Epoch: 657 \n","mse all (masked)patches: 0.10965492576360703 \n","all close: False\n","Epoch: 658 \n","mse all (masked)patches: 0.10451465845108032 \n","all close: False\n","Epoch: 659 \n","mse all (masked)patches: 0.10640893876552582 \n","all close: False\n","Epoch: 660 \n","mse all (masked)patches: 0.1080460175871849 \n","all close: False\n","Epoch: 661 \n","mse all (masked)patches: 0.10658882558345795 \n","all close: False\n","Epoch: 662 \n","mse all (masked)patches: 0.10458826273679733 \n","all close: False\n","Epoch: 663 \n","mse all (masked)patches: 0.10548733174800873 \n","all close: False\n","Epoch: 664 \n","mse all (masked)patches: 0.10302518308162689 \n","all close: False\n","Epoch: 665 \n","mse all (masked)patches: 0.1052686870098114 \n","all close: False\n","Epoch: 666 \n","mse all (masked)patches: 0.10100430995225906 \n","all close: False\n","Epoch: 667 \n","mse all (masked)patches: 0.10584162920713425 \n","all close: False\n","Epoch: 668 \n","mse all (masked)patches: 0.10224250704050064 \n","all close: False\n","Epoch: 669 \n","mse all (masked)patches: 0.10588712990283966 \n","all close: False\n","Epoch: 670 \n","mse all (masked)patches: 0.10567421466112137 \n","all close: False\n","Epoch: 671 \n","mse all (masked)patches: 0.1048206239938736 \n","all close: False\n","Epoch: 672 \n","mse all (masked)patches: 0.10445907711982727 \n","all close: False\n","Epoch: 673 \n","mse all (masked)patches: 0.10775908082723618 \n","all close: False\n","Epoch: 674 \n","mse all (masked)patches: 0.1044207513332367 \n","all close: False\n","Epoch: 675 \n","mse all (masked)patches: 0.10437584668397903 \n","all close: False\n","Epoch: 676 \n","mse all (masked)patches: 0.10336881130933762 \n","all close: False\n","Epoch: 677 \n","mse all (masked)patches: 0.10487633943557739 \n","all close: False\n","Epoch: 678 \n","mse all (masked)patches: 0.10370863974094391 \n","all close: False\n","Epoch: 679 \n","mse all (masked)patches: 0.10446088761091232 \n","all close: False\n","Epoch: 680 \n","mse all (masked)patches: 0.10344446450471878 \n","all close: False\n","Epoch: 681 \n","mse all (masked)patches: 0.10068944096565247 \n","all close: False\n","Epoch: 682 \n","mse all (masked)patches: 0.10271959751844406 \n","all close: False\n","Epoch: 683 \n","mse all (masked)patches: 0.10422033071517944 \n","all close: False\n","Epoch: 684 \n","mse all (masked)patches: 0.10057689994573593 \n","all close: False\n","Epoch: 685 \n","mse all (masked)patches: 0.10692913830280304 \n","all close: False\n","Epoch: 686 \n","mse all (masked)patches: 0.09908602386713028 \n","all close: False\n","Epoch: 687 \n","mse all (masked)patches: 0.09838651865720749 \n","all close: False\n","Epoch: 688 \n","mse all (masked)patches: 0.10060928016901016 \n","all close: False\n","Epoch: 689 \n","mse all (masked)patches: 0.10625424236059189 \n","all close: False\n","Epoch: 690 \n","mse all (masked)patches: 0.10252690315246582 \n","all close: False\n","Epoch: 691 \n","mse all (masked)patches: 0.10360696911811829 \n","all close: False\n","Epoch: 692 \n","mse all (masked)patches: 0.10450040549039841 \n","all close: False\n","Epoch: 693 \n","mse all (masked)patches: 0.10087842494249344 \n","all close: False\n","Epoch: 694 \n","mse all (masked)patches: 0.1030954122543335 \n","all close: False\n","Epoch: 695 \n","mse all (masked)patches: 0.10670056939125061 \n","all close: False\n","Epoch: 696 \n","mse all (masked)patches: 0.10505811870098114 \n","all close: False\n","Epoch: 697 \n","mse all (masked)patches: 0.09936470538377762 \n","all close: False\n","Epoch: 698 \n","mse all (masked)patches: 0.10348770767450333 \n","all close: False\n","Epoch: 699 \n","mse all (masked)patches: 0.10150734335184097 \n","all close: False\n","Epoch: 700 \n","mse all (masked)patches: 0.0994783267378807 \n","all close: False\n","Epoch: 701 \n","mse all (masked)patches: 0.10038217157125473 \n","all close: False\n","Epoch: 702 \n","mse all (masked)patches: 0.10367193818092346 \n","all close: False\n","Epoch: 703 \n","mse all (masked)patches: 0.10665174573659897 \n","all close: False\n","Epoch: 704 \n","mse all (masked)patches: 0.1028587594628334 \n","all close: False\n","Epoch: 705 \n","mse all (masked)patches: 0.10148938000202179 \n","all close: False\n","Epoch: 706 \n","mse all (masked)patches: 0.10224700719118118 \n","all close: False\n","Epoch: 707 \n","mse all (masked)patches: 0.10468859225511551 \n","all close: False\n","Epoch: 708 \n","mse all (masked)patches: 0.09885641187429428 \n","all close: False\n","Epoch: 709 \n","mse all (masked)patches: 0.10201475024223328 \n","all close: False\n","Epoch: 710 \n","mse all (masked)patches: 0.10069109499454498 \n","all close: False\n","Epoch: 711 \n","mse all (masked)patches: 0.10413110256195068 \n","all close: False\n","Epoch: 712 \n","mse all (masked)patches: 0.100614033639431 \n","all close: False\n","Epoch: 713 \n","mse all (masked)patches: 0.10104898363351822 \n","all close: False\n","Epoch: 714 \n","mse all (masked)patches: 0.10299283266067505 \n","all close: False\n","Epoch: 715 \n","mse all (masked)patches: 0.10008794814348221 \n","all close: False\n","Epoch: 716 \n","mse all (masked)patches: 0.0990152508020401 \n","all close: False\n","Epoch: 717 \n","mse all (masked)patches: 0.10202490538358688 \n","all close: False\n","Epoch: 718 \n","mse all (masked)patches: 0.10045606642961502 \n","all close: False\n","Epoch: 719 \n","mse all (masked)patches: 0.0998888686299324 \n","all close: False\n","Epoch: 720 \n","mse all (masked)patches: 0.10202990472316742 \n","all close: False\n","Epoch: 721 \n","mse all (masked)patches: 0.10186800360679626 \n","all close: False\n","Epoch: 722 \n","mse all (masked)patches: 0.1012657880783081 \n","all close: False\n","Epoch: 723 \n","mse all (masked)patches: 0.09854507446289062 \n","all close: False\n","Epoch: 724 \n","mse all (masked)patches: 0.1022694855928421 \n","all close: False\n","Epoch: 725 \n","mse all (masked)patches: 0.10082423686981201 \n","all close: False\n","Epoch: 726 \n","mse all (masked)patches: 0.10097071528434753 \n","all close: False\n","Epoch: 727 \n","mse all (masked)patches: 0.0991867259144783 \n","all close: False\n","Epoch: 728 \n","mse all (masked)patches: 0.10094661265611649 \n","all close: False\n","Epoch: 729 \n","mse all (masked)patches: 0.10279379040002823 \n","all close: False\n","Epoch: 730 \n","mse all (masked)patches: 0.09932860732078552 \n","all close: False\n","Epoch: 731 \n","mse all (masked)patches: 0.09849333763122559 \n","all close: False\n","Epoch: 732 \n","mse all (masked)patches: 0.10449486970901489 \n","all close: False\n","Epoch: 733 \n","mse all (masked)patches: 0.10281850397586823 \n","all close: False\n","Epoch: 734 \n","mse all (masked)patches: 0.1017860695719719 \n","all close: False\n","Epoch: 735 \n","mse all (masked)patches: 0.10330472141504288 \n","all close: False\n","Epoch: 736 \n","mse all (masked)patches: 0.09952743351459503 \n","all close: False\n","Epoch: 737 \n","mse all (masked)patches: 0.09650658071041107 \n","all close: False\n","Epoch: 738 \n","mse all (masked)patches: 0.09697709232568741 \n","all close: False\n","Epoch: 739 \n","mse all (masked)patches: 0.10402815043926239 \n","all close: False\n","Epoch: 740 \n","mse all (masked)patches: 0.10276374220848083 \n","all close: False\n","Epoch: 741 \n","mse all (masked)patches: 0.10204195231199265 \n","all close: False\n","Epoch: 742 \n","mse all (masked)patches: 0.10137522220611572 \n","all close: False\n","Epoch: 743 \n","mse all (masked)patches: 0.10175249725580215 \n","all close: False\n","Epoch: 744 \n","mse all (masked)patches: 0.10087969154119492 \n","all close: False\n","Epoch: 745 \n","mse all (masked)patches: 0.10016373544931412 \n","all close: False\n","Epoch: 746 \n","mse all (masked)patches: 0.10057050734758377 \n","all close: False\n","Epoch: 747 \n","mse all (masked)patches: 0.10309357941150665 \n","all close: False\n","Epoch: 748 \n","mse all (masked)patches: 0.09880697727203369 \n","all close: False\n","Epoch: 749 \n","mse all (masked)patches: 0.09825470298528671 \n","all close: False\n","Epoch: 750 \n","mse all (masked)patches: 0.10004019737243652 \n","all close: False\n","Epoch: 751 \n","mse all (masked)patches: 0.10044882446527481 \n","all close: False\n","Epoch: 752 \n","mse all (masked)patches: 0.09642572700977325 \n","all close: False\n","Epoch: 753 \n","mse all (masked)patches: 0.1004738062620163 \n","all close: False\n","Epoch: 754 \n","mse all (masked)patches: 0.09920179843902588 \n","all close: False\n","Epoch: 755 \n","mse all (masked)patches: 0.09753815084695816 \n","all close: False\n","Epoch: 756 \n","mse all (masked)patches: 0.10023866593837738 \n","all close: False\n","Epoch: 757 \n","mse all (masked)patches: 0.10002516210079193 \n","all close: False\n","Epoch: 758 \n","mse all (masked)patches: 0.09748261421918869 \n","all close: False\n","Epoch: 759 \n","mse all (masked)patches: 0.09732435643672943 \n","all close: False\n","Epoch: 760 \n","mse all (masked)patches: 0.10068649798631668 \n","all close: False\n","Epoch: 761 \n","mse all (masked)patches: 0.09969185292720795 \n","all close: False\n","Epoch: 762 \n","mse all (masked)patches: 0.09991496056318283 \n","all close: False\n","Epoch: 763 \n","mse all (masked)patches: 0.10140199214220047 \n","all close: False\n","Epoch: 764 \n","mse all (masked)patches: 0.099762924015522 \n","all close: False\n","Epoch: 765 \n","mse all (masked)patches: 0.09841041266918182 \n","all close: False\n","Epoch: 766 \n","mse all (masked)patches: 0.09658703207969666 \n","all close: False\n","Epoch: 767 \n","mse all (masked)patches: 0.09622414410114288 \n","all close: False\n","Epoch: 768 \n","mse all (masked)patches: 0.10102752596139908 \n","all close: False\n","Epoch: 769 \n","mse all (masked)patches: 0.09921173751354218 \n","all close: False\n","Epoch: 770 \n","mse all (masked)patches: 0.09853695333003998 \n","all close: False\n","Epoch: 771 \n","mse all (masked)patches: 0.0984090119600296 \n","all close: False\n","Epoch: 772 \n","mse all (masked)patches: 0.09828758239746094 \n","all close: False\n","Epoch: 773 \n","mse all (masked)patches: 0.09907824546098709 \n","all close: False\n","Epoch: 774 \n","mse all (masked)patches: 0.0950997844338417 \n","all close: False\n","Epoch: 775 \n","mse all (masked)patches: 0.10061722993850708 \n","all close: False\n","Epoch: 776 \n","mse all (masked)patches: 0.09951544553041458 \n","all close: False\n","Epoch: 777 \n","mse all (masked)patches: 0.09830929338932037 \n","all close: False\n","Epoch: 778 \n","mse all (masked)patches: 0.10028592497110367 \n","all close: False\n","Epoch: 779 \n","mse all (masked)patches: 0.09564721584320068 \n","all close: False\n","Epoch: 780 \n","mse all (masked)patches: 0.09641443192958832 \n","all close: False\n","Epoch: 781 \n","mse all (masked)patches: 0.09899481385946274 \n","all close: False\n","Epoch: 782 \n","mse all (masked)patches: 0.10113897919654846 \n","all close: False\n","Epoch: 783 \n","mse all (masked)patches: 0.09890575706958771 \n","all close: False\n","Epoch: 784 \n","mse all (masked)patches: 0.09891126304864883 \n","all close: False\n","Epoch: 785 \n","mse all (masked)patches: 0.09749431908130646 \n","all close: False\n","Epoch: 786 \n","mse all (masked)patches: 0.09680669009685516 \n","all close: False\n","Epoch: 787 \n","mse all (masked)patches: 0.09554757922887802 \n","all close: False\n","Epoch: 788 \n","mse all (masked)patches: 0.10044638812541962 \n","all close: False\n","Epoch: 789 \n","mse all (masked)patches: 0.1003890261054039 \n","all close: False\n","Epoch: 790 \n","mse all (masked)patches: 0.09983132779598236 \n","all close: False\n","Epoch: 791 \n","mse all (masked)patches: 0.09772460162639618 \n","all close: False\n","Epoch: 792 \n","mse all (masked)patches: 0.09736625850200653 \n","all close: False\n","Epoch: 793 \n","mse all (masked)patches: 0.09691941738128662 \n","all close: False\n","Epoch: 794 \n","mse all (masked)patches: 0.09899075329303741 \n","all close: False\n","Epoch: 795 \n","mse all (masked)patches: 0.09610658138990402 \n","all close: False\n","Epoch: 796 \n","mse all (masked)patches: 0.10015702247619629 \n","all close: False\n","Epoch: 797 \n","mse all (masked)patches: 0.09837634116411209 \n","all close: False\n","Epoch: 798 \n","mse all (masked)patches: 0.09807346016168594 \n","all close: False\n","Epoch: 799 \n","mse all (masked)patches: 0.09596104174852371 \n","all close: False\n","Epoch: 800 \n","mse all (masked)patches: 0.09914889186620712 \n","all close: False\n","Epoch: 801 \n","mse all (masked)patches: 0.09762076288461685 \n","all close: False\n","Epoch: 802 \n","mse all (masked)patches: 0.09721603989601135 \n","all close: False\n","Epoch: 803 \n","mse all (masked)patches: 0.09889385104179382 \n","all close: False\n","Epoch: 804 \n","mse all (masked)patches: 0.09893172979354858 \n","all close: False\n","Epoch: 805 \n","mse all (masked)patches: 0.10057180374860764 \n","all close: False\n","Epoch: 806 \n","mse all (masked)patches: 0.09703858941793442 \n","all close: False\n","Epoch: 807 \n","mse all (masked)patches: 0.09787409752607346 \n","all close: False\n","Epoch: 808 \n","mse all (masked)patches: 0.09860978275537491 \n","all close: False\n","Epoch: 809 \n","mse all (masked)patches: 0.09855221211910248 \n","all close: False\n","Epoch: 810 \n","mse all (masked)patches: 0.09809053689241409 \n","all close: False\n","Epoch: 811 \n","mse all (masked)patches: 0.10119132697582245 \n","all close: False\n","Epoch: 812 \n","mse all (masked)patches: 0.09819986671209335 \n","all close: False\n","Epoch: 813 \n","mse all (masked)patches: 0.09348303079605103 \n","all close: False\n","Epoch: 814 \n","mse all (masked)patches: 0.09411467611789703 \n","all close: False\n","Epoch: 815 \n","mse all (masked)patches: 0.10017647594213486 \n","all close: False\n","Epoch: 816 \n","mse all (masked)patches: 0.09753809124231339 \n","all close: False\n","Epoch: 817 \n","mse all (masked)patches: 0.09287548810243607 \n","all close: False\n","Epoch: 818 \n","mse all (masked)patches: 0.09657000005245209 \n","all close: False\n","Epoch: 819 \n","mse all (masked)patches: 0.09407813847064972 \n","all close: False\n","Epoch: 820 \n","mse all (masked)patches: 0.09821564704179764 \n","all close: False\n","Epoch: 821 \n","mse all (masked)patches: 0.09616279602050781 \n","all close: False\n","Epoch: 822 \n","mse all (masked)patches: 0.09965988248586655 \n","all close: False\n","Epoch: 823 \n","mse all (masked)patches: 0.09720739722251892 \n","all close: False\n","Epoch: 824 \n","mse all (masked)patches: 0.09753093123435974 \n","all close: False\n","Epoch: 825 \n","mse all (masked)patches: 0.09688473492860794 \n","all close: False\n","Epoch: 826 \n","mse all (masked)patches: 0.09628358483314514 \n","all close: False\n","Epoch: 827 \n","mse all (masked)patches: 0.09937013685703278 \n","all close: False\n","Epoch: 828 \n","mse all (masked)patches: 0.10002122819423676 \n","all close: False\n","Epoch: 829 \n","mse all (masked)patches: 0.10067015886306763 \n","all close: False\n","Epoch: 830 \n","mse all (masked)patches: 0.09682142734527588 \n","all close: False\n","Epoch: 831 \n","mse all (masked)patches: 0.1016412079334259 \n","all close: False\n","Epoch: 832 \n","mse all (masked)patches: 0.09232959151268005 \n","all close: False\n","Epoch: 833 \n","mse all (masked)patches: 0.09402312338352203 \n","all close: False\n","Epoch: 834 \n","mse all (masked)patches: 0.09439955651760101 \n","all close: False\n","Epoch: 835 \n","mse all (masked)patches: 0.0970379039645195 \n","all close: False\n","Epoch: 836 \n","mse all (masked)patches: 0.09702468663454056 \n","all close: False\n","Epoch: 837 \n","mse all (masked)patches: 0.09816596657037735 \n","all close: False\n","Epoch: 838 \n","mse all (masked)patches: 0.09586530178785324 \n","all close: False\n","Epoch: 839 \n","mse all (masked)patches: 0.09439650923013687 \n","all close: False\n","Epoch: 840 \n","mse all (masked)patches: 0.10130441188812256 \n","all close: False\n","Epoch: 841 \n","mse all (masked)patches: 0.09765375405550003 \n","all close: False\n","Epoch: 842 \n","mse all (masked)patches: 0.09588155150413513 \n","all close: False\n","Epoch: 843 \n","mse all (masked)patches: 0.09367211908102036 \n","all close: False\n","Epoch: 844 \n","mse all (masked)patches: 0.09756498038768768 \n","all close: False\n","Epoch: 845 \n","mse all (masked)patches: 0.09591668099164963 \n","all close: False\n","Epoch: 846 \n","mse all (masked)patches: 0.09799464792013168 \n","all close: False\n","Epoch: 847 \n","mse all (masked)patches: 0.09724241495132446 \n","all close: False\n","Epoch: 848 \n","mse all (masked)patches: 0.0989973396062851 \n","all close: False\n","Epoch: 849 \n","mse all (masked)patches: 0.09526809304952621 \n","all close: False\n","Epoch: 850 \n","mse all (masked)patches: 0.09801219403743744 \n","all close: False\n","Epoch: 851 \n","mse all (masked)patches: 0.09456084668636322 \n","all close: False\n","Epoch: 852 \n","mse all (masked)patches: 0.09519342333078384 \n","all close: False\n","Epoch: 853 \n","mse all (masked)patches: 0.09794931858778 \n","all close: False\n","Epoch: 854 \n","mse all (masked)patches: 0.09551528096199036 \n","all close: False\n","Epoch: 855 \n","mse all (masked)patches: 0.09876559674739838 \n","all close: False\n","Epoch: 856 \n","mse all (masked)patches: 0.09302815049886703 \n","all close: False\n","Epoch: 857 \n","mse all (masked)patches: 0.09846431761980057 \n","all close: False\n","Epoch: 858 \n","mse all (masked)patches: 0.0961185023188591 \n","all close: False\n","Epoch: 859 \n","mse all (masked)patches: 0.0945085659623146 \n","all close: False\n","Epoch: 860 \n","mse all (masked)patches: 0.09513869136571884 \n","all close: False\n","Epoch: 861 \n","mse all (masked)patches: 0.09575973451137543 \n","all close: False\n","Epoch: 862 \n","mse all (masked)patches: 0.10016004741191864 \n","all close: False\n","Epoch: 863 \n","mse all (masked)patches: 0.09397797286510468 \n","all close: False\n","Epoch: 864 \n","mse all (masked)patches: 0.09005110710859299 \n","all close: False\n","Epoch: 865 \n","mse all (masked)patches: 0.0950833410024643 \n","all close: False\n","Epoch: 866 \n","mse all (masked)patches: 0.09479659050703049 \n","all close: False\n","Epoch: 867 \n","mse all (masked)patches: 0.0944681316614151 \n","all close: False\n","Epoch: 868 \n","mse all (masked)patches: 0.09741954505443573 \n","all close: False\n","Epoch: 869 \n","mse all (masked)patches: 0.09592881053686142 \n","all close: False\n","Epoch: 870 \n","mse all (masked)patches: 0.09653542190790176 \n","all close: False\n","Epoch: 871 \n","mse all (masked)patches: 0.09530586749315262 \n","all close: False\n","Epoch: 872 \n","mse all (masked)patches: 0.09608128666877747 \n","all close: False\n","Epoch: 873 \n","mse all (masked)patches: 0.09537798911333084 \n","all close: False\n","Epoch: 874 \n","mse all (masked)patches: 0.09470368176698685 \n","all close: False\n","Epoch: 875 \n","mse all (masked)patches: 0.09498431533575058 \n","all close: False\n","Epoch: 876 \n","mse all (masked)patches: 0.09340853989124298 \n","all close: False\n","Epoch: 877 \n","mse all (masked)patches: 0.09814812988042831 \n","all close: False\n","Epoch: 878 \n","mse all (masked)patches: 0.09476280957460403 \n","all close: False\n","Epoch: 879 \n","mse all (masked)patches: 0.09516038745641708 \n","all close: False\n","Epoch: 880 \n","mse all (masked)patches: 0.09642858803272247 \n","all close: False\n","Epoch: 881 \n","mse all (masked)patches: 0.0955403596162796 \n","all close: False\n","Epoch: 882 \n","mse all (masked)patches: 0.09563327580690384 \n","all close: False\n","Epoch: 883 \n","mse all (masked)patches: 0.09560510516166687 \n","all close: False\n","Epoch: 884 \n","mse all (masked)patches: 0.09436355531215668 \n","all close: False\n","Epoch: 885 \n","mse all (masked)patches: 0.09584546089172363 \n","all close: False\n","Epoch: 886 \n","mse all (masked)patches: 0.09664859622716904 \n","all close: False\n","Epoch: 887 \n","mse all (masked)patches: 0.09555423259735107 \n","all close: False\n","Epoch: 888 \n","mse all (masked)patches: 0.09463825821876526 \n","all close: False\n","Epoch: 889 \n","mse all (masked)patches: 0.09711672365665436 \n","all close: False\n","Epoch: 890 \n","mse all (masked)patches: 0.09285622090101242 \n","all close: False\n","Epoch: 891 \n","mse all (masked)patches: 0.09494578838348389 \n","all close: False\n","Epoch: 892 \n","mse all (masked)patches: 0.09362786263227463 \n","all close: False\n","Epoch: 893 \n","mse all (masked)patches: 0.09249386191368103 \n","all close: False\n","Epoch: 894 \n","mse all (masked)patches: 0.09750834107398987 \n","all close: False\n","Epoch: 895 \n","mse all (masked)patches: 0.09369806945323944 \n","all close: False\n","Epoch: 896 \n","mse all (masked)patches: 0.09788879752159119 \n","all close: False\n","Epoch: 897 \n","mse all (masked)patches: 0.09514486789703369 \n","all close: False\n","Epoch: 898 \n","mse all (masked)patches: 0.09791480749845505 \n","all close: False\n","Epoch: 899 \n","mse all (masked)patches: 0.09387695044279099 \n","all close: False\n","Epoch: 900 \n","mse all (masked)patches: 0.0948096290230751 \n","all close: False\n","Epoch: 901 \n","mse all (masked)patches: 0.0958024114370346 \n","all close: False\n","Epoch: 902 \n","mse all (masked)patches: 0.09443160891532898 \n","all close: False\n","Epoch: 903 \n","mse all (masked)patches: 0.09562230110168457 \n","all close: False\n","Epoch: 904 \n","mse all (masked)patches: 0.09563787281513214 \n","all close: False\n","Epoch: 905 \n","mse all (masked)patches: 0.09303497523069382 \n","all close: False\n","Epoch: 906 \n","mse all (masked)patches: 0.09453827887773514 \n","all close: False\n","Epoch: 907 \n","mse all (masked)patches: 0.09561777114868164 \n","all close: False\n","Epoch: 908 \n","mse all (masked)patches: 0.09455171972513199 \n","all close: False\n","Epoch: 909 \n","mse all (masked)patches: 0.09694818407297134 \n","all close: False\n","Epoch: 910 \n","mse all (masked)patches: 0.09713298827409744 \n","all close: False\n","Epoch: 911 \n","mse all (masked)patches: 0.09723382443189621 \n","all close: False\n","Epoch: 912 \n","mse all (masked)patches: 0.09717274457216263 \n","all close: False\n","Epoch: 913 \n","mse all (masked)patches: 0.09518064558506012 \n","all close: False\n","Epoch: 914 \n","mse all (masked)patches: 0.09575144946575165 \n","all close: False\n","Epoch: 915 \n","mse all (masked)patches: 0.09670798480510712 \n","all close: False\n","Epoch: 916 \n","mse all (masked)patches: 0.09525343030691147 \n","all close: False\n","Epoch: 917 \n","mse all (masked)patches: 0.09585599601268768 \n","all close: False\n","Epoch: 918 \n","mse all (masked)patches: 0.09637647867202759 \n","all close: False\n","Epoch: 919 \n","mse all (masked)patches: 0.0934678465127945 \n","all close: False\n","Epoch: 920 \n","mse all (masked)patches: 0.09377682209014893 \n","all close: False\n","Epoch: 921 \n","mse all (masked)patches: 0.09233825653791428 \n","all close: False\n","Epoch: 922 \n","mse all (masked)patches: 0.09589643776416779 \n","all close: False\n","Epoch: 923 \n","mse all (masked)patches: 0.09652549028396606 \n","all close: False\n","Epoch: 924 \n","mse all (masked)patches: 0.09012114256620407 \n","all close: False\n","Epoch: 925 \n","mse all (masked)patches: 0.09363330900669098 \n","all close: False\n","Epoch: 926 \n","mse all (masked)patches: 0.09383684396743774 \n","all close: False\n","Epoch: 927 \n","mse all (masked)patches: 0.09378132969141006 \n","all close: False\n","Epoch: 928 \n","mse all (masked)patches: 0.09463471919298172 \n","all close: False\n","Epoch: 929 \n","mse all (masked)patches: 0.09506525844335556 \n","all close: False\n","Epoch: 930 \n","mse all (masked)patches: 0.09645497798919678 \n","all close: False\n","Epoch: 931 \n","mse all (masked)patches: 0.09170262515544891 \n","all close: False\n","Epoch: 932 \n","mse all (masked)patches: 0.09761334210634232 \n","all close: False\n","Epoch: 933 \n","mse all (masked)patches: 0.09849527478218079 \n","all close: False\n","Epoch: 934 \n","mse all (masked)patches: 0.09480041265487671 \n","all close: False\n","Epoch: 935 \n","mse all (masked)patches: 0.09426068514585495 \n","all close: False\n","Epoch: 936 \n","mse all (masked)patches: 0.09379865229129791 \n","all close: False\n","Epoch: 937 \n","mse all (masked)patches: 0.0938999354839325 \n","all close: False\n","Epoch: 938 \n","mse all (masked)patches: 0.09662433713674545 \n","all close: False\n","Epoch: 939 \n","mse all (masked)patches: 0.095900759100914 \n","all close: False\n","Epoch: 940 \n","mse all (masked)patches: 0.09698738157749176 \n","all close: False\n","Epoch: 941 \n","mse all (masked)patches: 0.09321283549070358 \n","all close: False\n","Epoch: 942 \n","mse all (masked)patches: 0.09450668096542358 \n","all close: False\n","Epoch: 943 \n","mse all (masked)patches: 0.094060517847538 \n","all close: False\n","Epoch: 944 \n","mse all (masked)patches: 0.0955059751868248 \n","all close: False\n","Epoch: 945 \n","mse all (masked)patches: 0.09439146518707275 \n","all close: False\n","Epoch: 946 \n","mse all (masked)patches: 0.09763700515031815 \n","all close: False\n","Epoch: 947 \n","mse all (masked)patches: 0.09668084979057312 \n","all close: False\n","Epoch: 948 \n","mse all (masked)patches: 0.09434664994478226 \n","all close: False\n","Epoch: 949 \n","mse all (masked)patches: 0.09554171562194824 \n","all close: False\n","Epoch: 950 \n","mse all (masked)patches: 0.09381704032421112 \n","all close: False\n","Epoch: 951 \n","mse all (masked)patches: 0.09200722724199295 \n","all close: False\n","Epoch: 952 \n","mse all (masked)patches: 0.09530128538608551 \n","all close: False\n","Epoch: 953 \n","mse all (masked)patches: 0.09356793016195297 \n","all close: False\n","Epoch: 954 \n","mse all (masked)patches: 0.09509509056806564 \n","all close: False\n","Epoch: 955 \n","mse all (masked)patches: 0.09581015259027481 \n","all close: False\n","Epoch: 956 \n","mse all (masked)patches: 0.0958411917090416 \n","all close: False\n","Epoch: 957 \n","mse all (masked)patches: 0.09518695622682571 \n","all close: False\n","Epoch: 958 \n","mse all (masked)patches: 0.09746260195970535 \n","all close: False\n","Epoch: 959 \n","mse all (masked)patches: 0.09448167681694031 \n","all close: False\n","Epoch: 960 \n","mse all (masked)patches: 0.09542274475097656 \n","all close: False\n","Epoch: 961 \n","mse all (masked)patches: 0.09563227742910385 \n","all close: False\n","Epoch: 962 \n","mse all (masked)patches: 0.09378333389759064 \n","all close: False\n","Epoch: 963 \n","mse all (masked)patches: 0.09312561899423599 \n","all close: False\n","Epoch: 964 \n","mse all (masked)patches: 0.09344135969877243 \n","all close: False\n","Epoch: 965 \n","mse all (masked)patches: 0.09631715714931488 \n","all close: False\n","Epoch: 966 \n","mse all (masked)patches: 0.09885092079639435 \n","all close: False\n","Epoch: 967 \n","mse all (masked)patches: 0.10063052922487259 \n","all close: False\n","Epoch: 968 \n","mse all (masked)patches: 0.09593400359153748 \n","all close: False\n","Epoch: 969 \n","mse all (masked)patches: 0.09321855008602142 \n","all close: False\n","Epoch: 970 \n","mse all (masked)patches: 0.09832177311182022 \n","all close: False\n","Epoch: 971 \n","mse all (masked)patches: 0.09424596279859543 \n","all close: False\n","Epoch: 972 \n","mse all (masked)patches: 0.09383273869752884 \n","all close: False\n","Epoch: 973 \n","mse all (masked)patches: 0.09496928751468658 \n","all close: False\n","Epoch: 974 \n","mse all (masked)patches: 0.09613549709320068 \n","all close: False\n","Epoch: 975 \n","mse all (masked)patches: 0.09875994920730591 \n","all close: False\n","Epoch: 976 \n","mse all (masked)patches: 0.09463000297546387 \n","all close: False\n","Epoch: 977 \n","mse all (masked)patches: 0.09652470052242279 \n","all close: False\n","Epoch: 978 \n","mse all (masked)patches: 0.0960085466504097 \n","all close: False\n","Epoch: 979 \n","mse all (masked)patches: 0.09635419398546219 \n","all close: False\n","Epoch: 980 \n","mse all (masked)patches: 0.09356957674026489 \n","all close: False\n","Epoch: 981 \n","mse all (masked)patches: 0.09612584114074707 \n","all close: False\n","Epoch: 982 \n","mse all (masked)patches: 0.09462849050760269 \n","all close: False\n","Epoch: 983 \n","mse all (masked)patches: 0.09084930270910263 \n","all close: False\n","Epoch: 984 \n","mse all (masked)patches: 0.0921381488442421 \n","all close: False\n","Epoch: 985 \n","mse all (masked)patches: 0.09779561311006546 \n","all close: False\n","Epoch: 986 \n","mse all (masked)patches: 0.09413997828960419 \n","all close: False\n","Epoch: 987 \n","mse all (masked)patches: 0.09532200545072556 \n","all close: False\n","Epoch: 988 \n","mse all (masked)patches: 0.09282392263412476 \n","all close: False\n","Epoch: 989 \n","mse all (masked)patches: 0.09269218891859055 \n","all close: False\n","Epoch: 990 \n","mse all (masked)patches: 0.09311262518167496 \n","all close: False\n","Epoch: 991 \n","mse all (masked)patches: 0.0936117097735405 \n","all close: False\n","Epoch: 992 \n","mse all (masked)patches: 0.09649363160133362 \n","all close: False\n","Epoch: 993 \n","mse all (masked)patches: 0.0938173457980156 \n","all close: False\n","Epoch: 994 \n","mse all (masked)patches: 0.0921391025185585 \n","all close: False\n","Epoch: 995 \n","mse all (masked)patches: 0.09324530512094498 \n","all close: False\n","Epoch: 996 \n","mse all (masked)patches: 0.09508305042982101 \n","all close: False\n","Epoch: 997 \n","mse all (masked)patches: 0.09136485308408737 \n","all close: False\n","Epoch: 998 \n","mse all (masked)patches: 0.09325723350048065 \n","all close: False\n","Epoch: 999 \n","mse all (masked)patches: 0.09677261114120483 \n","all close: False\n","Epoch: 1000 \n","mse all (masked)patches: 0.09493381530046463 \n","all close: False\n"]}]},{"cell_type":"code","source":["# 推理\n","# 模型重建的效果图，mask 效果图\n","recons_img_ts, masked_img_ts = mae.predict(img_ts)\n","recons_img_ts, masked_img_ts = recons_img_ts.cpu().squeeze(0), masked_img_ts.cpu().squeeze(0)\n","\n","# 将结果保存下来以便和原图比较\n","recons_img = ToPILImage()(recons_img_ts)\n","recons_img.save(os.path.join(BASE_DIR, 'recons_photo.jpg'))\n","\n","masked_img = ToPILImage()(masked_img_ts)\n","masked_img.save(os.path.join(BASE_DIR, 'masked_photo.jpg'))\n","\n","# 画图\n","img = Image.open(os.path.join(BASE_DIR, 'photo.jpg'))\n","plt.figure(\"original photo\")  # 图像窗口名称\n","plt.imshow(img)\n","plt.axis('off')  # 关掉坐标轴为 off\n","plt.title('original photo')  # 图像题目\n","plt.show()\n","\n","img = Image.open(os.path.join(BASE_DIR, 'masked_photo.jpg'))\n","plt.figure(\"masked photo\")  # 图像窗口名称\n","plt.imshow(img)\n","plt.axis('off')  # 关掉坐标轴为 off\n","plt.title('masked photo')  # 图像题目\n","plt.show()\n","\n","img = Image.open(os.path.join(BASE_DIR, 'recons_photo.jpg'))\n","plt.figure(\"reconstructed photo\")  # 图像窗口名称\n","plt.imshow(img)\n","plt.axis('off')  # 关掉坐标轴为 off\n","plt.title('reconstructed photo')  # 图像题目\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"4Dks1qKTfptK","executionInfo":{"status":"ok","timestamp":1650181661667,"user_tz":-480,"elapsed":1660,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"700c4c1f-c20e-4617-9333-7771482f602e"},"execution_count":10,"outputs":[{"output_type":"stream","name":"stdout","text":["mse per (masked)patch: tensor([[0.1299, 0.0321, 0.0783, 0.2180, 0.1312, 0.0802, 0.1562, 0.0800, 0.0876,\n","         0.1425, 0.0380, 0.0813, 0.0786, 0.1710, 0.0866, 0.1254, 0.1642, 0.0266,\n","         0.1014, 0.1074, 0.0532, 0.0323, 0.1165, 0.1435, 0.1048, 0.0954, 0.0441,\n","         0.1398, 0.0759, 0.1277, 0.1184, 0.0465, 0.1268, 0.1692, 0.0805, 0.0801,\n","         0.0615, 0.1143, 0.1363, 0.0737, 0.0963, 0.0805, 0.1187, 0.1506, 0.0985,\n","         0.0936, 0.1362, 0.0253, 0.1061, 0.1410, 0.0793, 0.1610, 0.0939, 0.1753,\n","         0.0303, 0.1295, 0.0454, 0.0700, 0.0435, 0.0920, 0.0361, 0.1130, 0.1794,\n","         0.0492, 0.0776, 0.1087, 0.0930, 0.0226, 0.0227, 0.1179, 0.1398, 0.0889,\n","         0.1239, 0.0232, 0.1373, 0.0332, 0.1269, 0.0219, 0.1280, 0.0822, 0.0741,\n","         0.0284, 0.0755, 0.0894, 0.1242, 0.1333, 0.1426, 0.0920, 0.1097, 0.1493,\n","         0.1249, 0.1605, 0.1836, 0.0836, 0.0837, 0.1316, 0.0678, 0.0281, 0.0868,\n","         0.0201, 0.0568, 0.0191, 0.1465, 0.0840, 0.1344, 0.0725, 0.0648, 0.0936,\n","         0.1108, 0.1485, 0.0739, 0.0930, 0.0524, 0.2150, 0.1158, 0.0925, 0.0372,\n","         0.0913, 0.0316, 0.1559, 0.1547, 0.0600, 0.0832, 0.1187, 0.1782, 0.2049,\n","         0.1086, 0.1358, 0.1404, 0.2410, 0.0138, 0.0746, 0.0492, 0.0849, 0.0832,\n","         0.0964, 0.0644, 0.1360, 0.0728, 0.0286, 0.0426, 0.0700, 0.0947, 0.0962,\n","         0.1918, 0.0853, 0.1062]], device='cuda:0', grad_fn=<MeanBackward1>) mse all (masked)patches: 0.09826221317052841 total 147 masked patches\n","all close: False\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":["# %reload_ext tensorboard\n","%tensorboard --logdir '/home/Insight/runs'"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ZqGVaU_MfHok","executionInfo":{"status":"error","timestamp":1650181727131,"user_tz":-480,"elapsed":727,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"outputId":"ff2d4ef9-9755-4159-e37a-a0006af9d198"},"execution_count":16,"outputs":[{"output_type":"stream","name":"stderr","text":["UsageError: Line magic function `%tensorboard` not found.\n"]}]},{"cell_type":"markdown","metadata":{"id":"sDNK_Ece5Nbq"},"source":["# 完成任务之后，把输出的材料保存到google drive\n","* 因为虚拟机上的东西生命周期到了就消失了"]},{"cell_type":"code","metadata":{"id":"8MAHKp_U5Pl6","executionInfo":{"status":"ok","timestamp":1650181708718,"user_tz":-480,"elapsed":4014,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"38ff42d0-32a7-441d-e267-3fd546b3c60b"},"source":["# 复制tensorboard记录\n","!/bin/cp -rf /home/Insight/runs/*  /content/drive/MyDrive/Insight/runs\n","print('runs copy completed!')\n","# 复制saved models模型参数\n","!/bin/cp -rf /home/Insight/saved_models/* /content/drive/MyDrive/Insight/saved_models\n","print('models copy completed!')"],"execution_count":13,"outputs":[{"output_type":"stream","name":"stdout","text":["runs copy completed!\n","models copy completed!\n"]}]},{"cell_type":"markdown","metadata":{"id":"MN-TzUmsDZT5"},"source":["## 显示时间"]},{"cell_type":"code","metadata":{"id":"ysfaUWASDaHo","executionInfo":{"status":"ok","timestamp":1650181708719,"user_tz":-480,"elapsed":10,"user":{"displayName":"Rush MSHT","userId":"17678018738220178064"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"ec40f284-c8d3-45ef-89b0-d871f0622653"},"source":["# 如果一段时间不需要用显卡的话，可以整个僵尸参数占用点显存，避免出现警告\n","# import torch\n","# a = torch.Tensor([1000, 1000, 1000]).cuda()  # 大约占用 1.1G 显存\n","!date --date='+8 hour'  # 查看系统时间，转换为北京时间"],"execution_count":14,"outputs":[{"output_type":"stream","name":"stdout","text":["Sun Apr 17 15:48:27 UTC 2022\n"]}]}]}