From d30836fb7c4018f2c07dd0f2584bed7c05bb96a1 Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 15:54:03 -0400 Subject: [PATCH 1/8] add PR template inspired by https://github.com/neuropsychology/NeuroKit/blob/97b5a97e660c867d01ffc683031e38c35ff7d034/.github/PULL_REQUEST_TEMPLATE.md --- .github/pull_request_template.md | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) create mode 100644 .github/pull_request_template.md diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md new file mode 100644 index 0000000..7631101 --- /dev/null +++ b/.github/pull_request_template.md @@ -0,0 +1,20 @@ +This is a template for making a pull-request. You can remove the text and sections and write your own thing if you wish, just make sure you give enough information about how and why. If you have any issues or difficulties, don't hesitate to open an issue. + + +# Description + +The aim is to add this feature ... + +# Proposed Changes + +I changed the `foo()` function so that ... + + +# Checklist + +Here are some things to check before creating the pull request. If you encounter any issues, don't hesitate to ask for help :) + +- [ ] I have read the [contributor's guide](https://github.com/arnab39/equiadapt/blob/main/CONTRIBUTING.md). +- [ ] The base branch of my pull request is the `dev` branch, not the `main` branch. +- [ ] I ran the [code checks](https://github.com/arnab39/equiadapt/blob/main/CONTRIBUTING.md#implement-your-changes) on the files I added or modified and fixed the errors. +- [ ] I updated the [changelog](https://github.com/arnab39/equiadapt/blob/main/CHANGELOG.md). \ No newline at end of file From 71deebc0fcdbec766d39e058f20ea814a3a9eaab Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 15:55:07 -0400 Subject: [PATCH 2/8] update format of changelog --- CHANGELOG.md | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index dd0325b..f02a1bb 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,3 +1,16 @@ # Changelog -## Version 0.1 (development) +All notable changes to this project will be documented in this file. + +The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), +and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). + +## [Unreleased] + +### Added + +### Fixed + +### Changed + +### Removed \ No newline at end of file From d52312f326b1f7dde394c43dcb19467f221de26b Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 15:57:31 -0400 Subject: [PATCH 3/8] update links in template --- .github/pull_request_template.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md index 7631101..bc5b355 100644 --- a/.github/pull_request_template.md +++ b/.github/pull_request_template.md @@ -14,7 +14,7 @@ I changed the `foo()` function so that ... Here are some things to check before creating the pull request. If you encounter any issues, don't hesitate to ask for help :) -- [ ] I have read the [contributor's guide](https://github.com/arnab39/equiadapt/blob/main/CONTRIBUTING.md). +- [ ] I have read the [contributor's guide](https://github.com/arnab39/reptrix/blob/main/CONTRIBUTING.md). - [ ] The base branch of my pull request is the `dev` branch, not the `main` branch. -- [ ] I ran the [code checks](https://github.com/arnab39/equiadapt/blob/main/CONTRIBUTING.md#implement-your-changes) on the files I added or modified and fixed the errors. -- [ ] I updated the [changelog](https://github.com/arnab39/equiadapt/blob/main/CHANGELOG.md). \ No newline at end of file +- [ ] I ran the [code checks](https://github.com/arnab39/reptrix/blob/main/CONTRIBUTING.md#implement-your-changes) on the files I added or modified and fixed the errors. +- [ ] I updated the [changelog](https://github.com/arnab39/reptrix/blob/main/CHANGELOG.md). \ No newline at end of file From e73ebcc8d85d0c16fcf80718afda962592fb1509 Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 15:57:42 -0400 Subject: [PATCH 4/8] add information about code checks to contributor's guide --- CONTRIBUTING.md | 47 ++++++++++++++++++++++++----------------------- 1 file changed, 24 insertions(+), 23 deletions(-) diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 884b817..7bd206f 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -183,12 +183,10 @@ conda activate reptrix and start making changes. Never work on the main branch! -2. Start your work on this branch. Don't forget to add [docstrings] to new - functions, modules and classes, especially if they are part of public APIs. +2. Start your work on this branch. Don't forget to add [docstrings] to the new + functions, modules and classes, especially if they are part of [equiadapt]. -3. Add yourself to the list of contributors in `AUTHORS.rst`. - -4. When you’re done editing, do: +3. When you’re done editing, do: ``` git add @@ -197,38 +195,41 @@ conda activate reptrix to record your changes in [git]. - ```{todo} if you are not using pre-commit, please remove the following item: - ``` - Please make sure to see the validation messages from [pre-commit] and fix any eventual issues. This should automatically use [flake8]/[black] to check/fix the code style in a way that is compatible with the project. - :::{important} - Don't forget to add unit tests and documentation in case your +> **Note**: + Please add unit tests and documentation in case your contribution adds an additional feature and is not just a bugfix. - Moreover, writing a [descriptive commit message] is highly recommended. In case of doubt, you can check the commit history with: + `git log --graph --decorate --pretty=oneline --abbrev-commit --all` + to look for recurring communication patterns. - ``` - git log --graph --decorate --pretty=oneline --abbrev-commit --all - ``` +#### Run code checks - to look for recurring communication patterns. - ::: +Please make sure to see the validation messages from pre-commit and fix any +eventual issues. This should automatically use [flake8]/[black] to check/fix +the code style in a way that is compatible with the project. -5. Please check that your changes don't break any unit tests with: +To run pre-commit manually, you can use: - ``` - tox - ``` +``` +pre-commit run --all-files +``` + +Please also check that your changes don't break any unit tests with: + +``` +tox +``` - (after having installed [tox] with `pip install tox` or `pipx`). +(after having installed [tox] with `pip install tox` or `pipx`). - You can also use [tox] to run several other pre-configured tasks in the - repository. Try `tox -av` to see a list of the available checks. +You can also use [tox] to run several other pre-configured tasks in the +repository. Try `tox -av` to see a list of the available checks. ### Submit your contribution From d9ae9bcca51118f7a589ea4c2d799be2a8b2ca08 Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 16:01:37 -0400 Subject: [PATCH 5/8] fix formatting in readme --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 7f8ba72..7c8ef62 100644 --- a/README.md +++ b/README.md @@ -8,8 +8,8 @@ Representation Metrics for assessing quality in pretrained deep models You can check out the [contributor's guide](CONTRIBUTING.md). -This project uses `pre-commit`_, you can install it before making any -changes:: +This project uses `pre-commit`, you can install it before making any +changes: pip install pre-commit cd reptrix From c45b2b0295a4291d61a7999020e6a95582cd75b1 Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 16:05:12 -0400 Subject: [PATCH 6/8] run pre-commit autoupdate --- .pre-commit-config.yaml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index 3099a08..128fe68 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -2,7 +2,7 @@ exclude: '^docs/conf.py' repos: - repo: https://github.com/pre-commit/pre-commit-hooks - rev: v4.5.0 + rev: v4.6.0 hooks: - id: trailing-whitespace - id: check-added-large-files @@ -41,7 +41,7 @@ repos: - id: isort - repo: https://github.com/psf/black - rev: 24.2.0 + rev: 24.4.0 hooks: - id: black language_version: python3 @@ -67,7 +67,7 @@ repos: # - id: codespell - repo: https://github.com/pre-commit/mirrors-mypy - rev: 'v1.8.0' + rev: 'v1.9.0' hooks: - id: mypy args: [--disallow-untyped-defs, --ignore-missing-imports] From 2f10cc5a410dcf57effe314913d879838f85333b Mon Sep 17 00:00:00 2001 From: danibene <34680344+danibene@users.noreply.github.com> Date: Sat, 13 Apr 2024 16:05:44 -0400 Subject: [PATCH 7/8] run pre-commit run --all-files --- .github/pull_request_template.md | 2 +- CHANGELOG.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md index bc5b355..5c609ac 100644 --- a/.github/pull_request_template.md +++ b/.github/pull_request_template.md @@ -17,4 +17,4 @@ Here are some things to check before creating the pull request. If you encounter - [ ] I have read the [contributor's guide](https://github.com/arnab39/reptrix/blob/main/CONTRIBUTING.md). - [ ] The base branch of my pull request is the `dev` branch, not the `main` branch. - [ ] I ran the [code checks](https://github.com/arnab39/reptrix/blob/main/CONTRIBUTING.md#implement-your-changes) on the files I added or modified and fixed the errors. -- [ ] I updated the [changelog](https://github.com/arnab39/reptrix/blob/main/CHANGELOG.md). \ No newline at end of file +- [ ] I updated the [changelog](https://github.com/arnab39/reptrix/blob/main/CHANGELOG.md). diff --git a/CHANGELOG.md b/CHANGELOG.md index f02a1bb..ab086d5 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -13,4 +13,4 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 ### Changed -### Removed \ No newline at end of file +### Removed From 54cc611d0abe2553b745ec0593abc2004cd888c2 Mon Sep 17 00:00:00 2001 From: Arna Ghosh Date: Thu, 30 May 2024 20:56:45 -0400 Subject: [PATCH 8/8] Updated tutorial notebook: fixed get_features, added time and LiDAR in description --- tutorial.ipynb | 171 ++++++++++++++++++++++++++++--------------------- 1 file changed, 98 insertions(+), 73 deletions(-) diff --git a/tutorial.ipynb b/tutorial.ipynb index 9ff639c..031de0d 100644 --- a/tutorial.ipynb +++ b/tutorial.ipynb @@ -14,9 +14,11 @@ "\n", "To assess the quality of the learned representations, we will use various metrics, including:\n", "\n", - "- **Alpha**: This metric computes the eigenvalues of the covariance matrix of the representations and fits a power-law distribution to them. The exponent of the power-law distribution is called the alpha exponent, which measures the heavy-tailedness of the distribution. A lower alpha exponent indicates that the representations are more discriminative.\n", + "- [**Alpha**](https://proceedings.neurips.cc/paper_files/paper/2022/hash/70596d70542c51c8d9b4e423f4bf2736-Abstract-Conference.html): This metric computes the eigenvalues of the covariance matrix of the representations and fits a power-law distribution to them. The exponent of the power-law distribution is called the alpha exponent, which measures the heavy-tailedness of the distribution. A lower alpha exponent indicates that the representations are more discriminative.\n", "\n", - "- **RankMe**: This metric computes the rank of the covariance matrix of the representations. A higher rank indicates representations of higher capacity.\n", + "- [**RankMe**](https://proceedings.mlr.press/v202/garrido23a): This metric computes the rank of the covariance matrix of the representations. A higher rank indicates representations of higher capacity.\n", + "\n", + "- [**LiDAR**](https://openreview.net/forum?id=f3g5XpL9Kb): This metric computes the rank of the Linear Discriminant Analysis (LDA) matrix constructed using representations of augmented versions of images. A higher rank indicates representations of higher discriminability.\n", "\n", "\n", "We will compute these metrics using the Reptrix library, which provides a convenient interface for representation analysis. Let's dive into the code and explore the evaluation process in detail.\n", @@ -40,7 +42,8 @@ "import torchvision\n", "from tqdm import tqdm\n", "from reptrix import alpha, rankme, lidar\n", - "import reptrix" + "import reptrix\n", + "import time" ] }, { @@ -79,14 +82,14 @@ " # Loop over the dataset and collect the representations\n", " for i, data in enumerate(tqdm(dataloader, 0)):\n", " inputs, _ = data\n", - " # apply 10 random augmentations for each image\n", + " # apply multiple augmentations for each image\n", " if transform:\n", " inputs = torch.cat([transform(inputs) for _ in range(num_augmentations)], dim=0)\n", " with torch.no_grad():\n", " features = encoder_function(inputs.to(device))\n", " if transform:\n", " # put the augmentations in an additonal dimension\n", - " features = features.reshape(-1, num_augmentations, features.shape[1])\n", + " features = features.reshape(num_augmentations, -1, features.shape[1]).transpose(1,0)\n", " all_features.append(features)\n", " \n", " \n", @@ -108,11 +111,11 @@ "metadata": {}, "outputs": [], "source": [ - "transform = torchvision.transforms.Compose([\n", - " torchvision.transforms.ToTensor(),\n", - " torchvision.transforms.Normalize((0.4467, 0.4398, 0.4066), \n", - " (0.2242, 0.2215, 0.2239))\n", - "])\n", + "transform_to_tensor = torchvision.transforms.ToTensor()\n", + "\n", + "STL_MEAN = (0.4467, 0.4398, 0.4066)\n", + "STL_STD = (0.2242, 0.2215, 0.2239)\n", + "transform_base = torchvision.transforms.Normalize(STL_MEAN, STL_STD)\n", "\n", "# Define additional SSL transformations for the Lidar metric evaluation\n", "transform_ssl = torchvision.transforms.Compose([\n", @@ -120,14 +123,16 @@ " torchvision.transforms.ColorJitter(brightness=0.4, contrast=0.4, saturation=0.4, hue=0.1),\n", " torchvision.transforms.RandomGrayscale(p=0.2),\n", " torchvision.transforms.RandomResizedCrop(\n", - " 96, scale=(0.2, 1.0), \n", + " 96, scale=(0.8, 1.0), \n", " ratio=(0.75, (4/3)), \n", " interpolation=torchvision.transforms.InterpolationMode.BICUBIC),\n", + " torchvision.transforms.Normalize(STL_MEAN, STL_STD)\n", "])\n", " \n", - "\n", + "dataset_folder = '/network/datasets/stl10.var/stl10_torchvision'\n", "# Get the STL10 test dataset to measure the quality of the representations learned by the model\n", - "testset = torchvision.datasets.STL10(root='./data', split='test', download=False, transform=transform)\n", + "# testset = torchvision.datasets.STL10(root='./data', split='test', download=False, transform=transform)\n", + "testset = torchvision.datasets.STL10(root=dataset_folder, split='test', download=False, transform=transform_to_tensor)\n", "\n", "# Define a dataloader to load the test dataset\n", "testloader = torch.utils.data.DataLoader(testset, batch_size=256, shuffle=False, num_workers=4)" @@ -158,10 +163,10 @@ "name": "stderr", "output_type": "stream", "text": [ - "Using cache found in /home/mila/a/arnab.mondal/.cache/torch/hub/facebookresearch_barlowtwins_main\n", - "/home/mila/a/arnab.mondal/.conda/envs/equiadapt/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n", + "Using cache found in /home/mila/g/ghosharn/.cache/torch/hub/facebookresearch_barlowtwins_main\n", + "/network/scratch/g/ghosharn/conda_envs/ffcv_ssl_fastssl/lib/python3.9/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n", " warnings.warn(\n", - "/home/mila/a/arnab.mondal/.conda/envs/equiadapt/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.\n", + "/network/scratch/g/ghosharn/conda_envs/ffcv_ssl_fastssl/lib/python3.9/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.\n", " warnings.warn(msg)\n" ] } @@ -183,7 +188,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "100%|██████████| 32/32 [00:02<00:00, 14.42it/s]\n" + "100%|██████████| 32/32 [00:03<00:00, 9.16it/s]\n" ] } ], @@ -194,27 +199,52 @@ "# Set the model to evaluation mode\n", "encoder.eval()\n", "\n", - "all_representations = get_features(encoder, testloader)" + "all_representations = get_features(encoder, testloader, transform=transform_base, num_augmentations=1)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "torch.Size([8000, 1, 2048])" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "all_representations.shape" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, "outputs": [], "source": [ + "start_time = time.time()\n", "metric_alpha = alpha.get_alpha(all_representations)\n", - "metric_rankme = rankme.get_rankme(all_representations)" + "alpha_time = time.time()\n", + "metric_rankme = rankme.get_rankme(all_representations)\n", + "rankme_time = time.time()\n", + "alpha_compute_time = alpha_time - start_time\n", + "rankme_compute_time = rankme_time - alpha_time" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 9, "metadata": {}, "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlEAAAHNCAYAAAA+MC57AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAABx2UlEQVR4nO3dd3gUVRfH8e+mU5JQAoFAAFFRakIvShWlCYKKgIWmotJFpFgAQUBEBYGogL4IKoqAIBaQIr23oFIElSYl9E5Cyrx/XJMQUtjU3SS/z/Pk0bk7O3N297I5mXvnXJtlWRYiIiIikioujg5AREREJDtSEiUiIiKSBkqiRERERNJASZSIiIhIGiiJEhEREUkDJVEiIiIiaaAkSkRERCQNlESJiIiIpIGSKBEREZE0UBIlIiIikgZKokRERETSQElUBvj888+x2WwcOnTI0aFINjdixAhsNpujw8hSzvTvZ+vWrdSrV498+fJhs9mYOHGi08Qmqffuu+9y7733EhMT4+hQxIl88sknlCpVioiIiHQfS0lUCmK/3JP72bRpk6NDzHY2bNjAiBEjuHDhgqNDEUkgMjKS9u3bc+7cOSZMmMAXX3xBUuuz59Q+HBERweDBgwkICCBPnjzUrl2bZcuW2fXc7du307x5c3x8fPD29uahhx4iNDT0ts8bPXo0NpuNSpUqJbvPjh07aNOmDYUKFSJv3rxUqlSJSZMm3fbYly5dYty4cQwePBgXF/t+1W3dupXevXtTsWJF8uXLR6lSpXjiiSfYv3+/Xc+/2ZUrVxg+fDjNmzenUKFC2Gw2Pv/88yT3Tc17n57PKaOlNxZ7+01q+teBAwfo2LEjJUuWJG/evNx7772MHDmSa9euxe3TtWtXbty4wdSpU1P7khOzJFkzZsywAGvkyJHWF198kejn9OnTlmVZVlRUlHX9+nUrJibGwRE7v/Hjx1uAdfDgQUeH4pSGDx9u5bZ/lrH/zhzdJ/bu3WsB1vTp0+Pakvq3nVP7cMeOHS03Nzdr4MCB1tSpU626detabm5u1tq1a1N83vbt2y0vLy/r7rvvtt577z3r3XfftcqUKWP5+PhY+/btS/Z5R48etfLmzWvly5fPqlixYpL7/PLLL5aHh4dVu3Zt64MPPrCmTZtmDR482Hr11Vdv+3omTJhg+fj4WNevX7/tvrEee+wxq1ixYlafPn2s6dOnW6NGjbL8/f2tfPnyWb///rvdx7Esyzp48KAFWKVKlbIaNWpkAdaMGTOS3Dc1731aP6fMkJ5Y7O03qelfR44csQoUKGCVLl3aGjt2rDV16lSra9euFmC1adMmwb6DBg2ySpcune7f27nr2zqVYr/ct27d6uhQcozU/AK6cuVK5gfkZHJTEhX7+TpLErV69WoLsObOnZvifjkxidq8ebMFWOPHj49ru379unXnnXdadevWTfG5LVu2tAoWLGidOXMmru348eNW/vz5rUcffTTZ53Xo0MFq0qSJ1bBhwySTqIsXL1r+/v5Wu3btrOjo6FS/pipVqlhPP/20XfvG9sX169dbERERCR7bv3+/5enpaT311FOpOn94eLh14sQJy7Isa+vWrckmUal579PzOWW09MZib79JTf8aPXq0BVh//PFHgvbOnTtbgHXu3Lm4tm3btlmAtWLFCvtfdBJyx7d1GtmbRCX3S2DlypVW9erVLU9PT6ts2bLWJ598kuQvyX///dfq1q2bVbRoUcvDw8OqUKGC9dlnnyXYJ/Z5Bw4csLp06WL5+vpaPj4+VteuXa2rV68m2PfSpUtWv379rNKlS1seHh5WkSJFrKZNm1rbt29PdLy9e/da7du3t7y9va1ChQpZffv2TfSXmz3xxe7XvXt3q3jx4paHh4dVpkwZ68UXX4z7Uoo9560/Bw8ejHts9+7dVqdOnawCBQpYwcHBlmVZVpcuXazSpUsnOt+t72Xs9p9//mk99dRTlo+Pj+Xn52e98cYbVkxMjHXkyBGrTZs2lre3t+Xv72+99957yXyiCR06dMh66aWXrHLlylleXl5WoUKFrMcffzzR552az8iyLGvt2rVWjRo1bts/knvd9nx2lmVZO3bssJo3b255e3tb+fLls5o0aWJt3Lgx7vFdu3ZZgPX999/HtcV+wVStWjXBsZo3b27VqlUrQVtq+m9Sn++t/37sfb/TG/fNunTpkqhfNmzYMFFsKfXhtPjmm2+sqlWrWl5eXta9995rLVu2zIqJibEqVKhgvf3222k6Zlq8+uqrlqurq3Xx4sUE7WPGjLEA68iRI8k+19vb22rfvn2i9latWlkeHh7W5cuXEz22evVqy9XV1frtt9+STaI+/vhjC7D27NljWZZJdOxNpv755x8LsD7//PNEj6XUF5NTrVo1q1q1anadOykpJVGpee/T8zlltPTGYm+/SU3/Gjx4sAXEjRLd3O7i4pLoD/PY7830cEv/gGDOd/HiRc6cOZOgzWazUbhw4WSfs3PnTpo3b07x4sV56623iI6OZuTIkRQpUiTBfmFhYdSpUwebzUbv3r0pUqQIixcv5tlnn+XSpUv0798/wf5PPPEEd9xxB2PHjmXHjh18+umnFC1alHHjxsXt8+KLLzJv3jx69+5NhQoVOHv2LOvWrWPv3r1Uq1Yt0fHKlCnD2LFj2bRpE5MmTeL8+fPMmjUrVfEdP36cWrVqceHCBXr06MG9997LsWPHmDdvHteuXcPDw4NHH32U/fv38/XXXzNhwgT8/PwAErwn7du35+6772bMmDFJzkexR4cOHShfvjzvvPMOP/30E2+//TaFChVi6tSpNGnShHHjxvHVV18xcOBAatasSYMGDVI83tatW9mwYUPcOPuhQ4f4+OOPadSoEXv27CFv3ryp/ox+//13HnroIYoUKcKIESOIiopi+PDh+Pv72/06b/fZAezevZv69evj4+PDoEGDcHd3Z+rUqTRq1IjVq1dTu3ZtKlWqRIECBVizZg1t2rQBYO3atbi4uLBr1y4uXbqEj48PMTExbNiwgR49esQdP7X9157P1973Oz1x3+qFF16gRIkSjBkzhr59+1KzZk38/f05duxYgv3s6cP2euONNxg9ejRdu3bl2Wef5cMPP6Rz585Mnz6df//9l969eyf73MjISC5evGjXeQoVKnTbOUE7d+6kXLly+Pj4JGivVasWAKGhoQQGBib53IiICPLkyZOoPW/evNy4cYM//viDOnXqxLVHR0fTp08fnnvuOSpXrpxsTMuXL8fHx4djx47Rtm1b9u/fT758+XjmmWeYMGECXl5eyT53w4YNAIm+725m73eNZVmEhYVRsWLFZPdJj9S89+n5nCBj+016Y7G336SmfzVq1Ihx48bx7LPP8tZbb1G4cGE2bNjAxx9/TN++fcmXL1+CY1SrVo3169cn/ybYI10pWA4X+1doUj+enp6J9rv5r9HWrVtbefPmtY4dOxbXduDAAcvNzS3BlYZnn33WKl68eIJLlZZlxpp9fX2ta9euWZYV/9dT9+7dE+zXrl07q3DhwgnafH19rV69eqX42mKPd+s4cc+ePS3A2rVrV6ri69y5s+Xi4pLkVTt75pPExtOpU6dEz0/tlagePXrEtUVFRVklS5a0bDab9c4778S1nz9/3sqTJ4/VpUuXRMe9VexrvNnGjRstwJo1a1ai89vzGbVt29by8vKyDh8+HNe2Z88ey9XV1e4rUbf77GLP4+HhYf39999xbcePH7e8vb2tBg0axLW1atUqwZWaRx991Hr00UctV1dXa/HixZZlmSta3HLlJ7X9N6nP99Z/P/a+3+mJOykrV65MNJyX1L/tjBjOW7NmjQVYgwcPjmubN2+eBViVKlVK0J5SrPb82BNnxYoVrSZNmiRq3717twVYn3zySbLPrVy5slWuXDkrKioqri0iIsIqVaqUBVjz5s1LsP+UKVMsX19f69SpU5ZlWcleiapSpYqVN29eK2/evFafPn2s+fPnW3369LEAq2PHjim+njfeeMMCkrwKllJfTMoXX3xhAUlefbdXSleiUvPep+dzsqyM7TfpjcXefpPa/jVq1CgrT548CV7L66+/nmQMPXr0sPLkyZNinLeju/PsEBISwrJlyxL8LF68ONn9o6OjWb58OW3btiUgICCu/a677qJFixZx25ZlMX/+fFq3bo1lWZw5cybup1mzZly8eJEdO3YkOPaLL76YYLt+/fqcPXuWS5cuxbUVKFCAzZs3c/z48du+tl69eiXY7tOnDwA///yz3fHFxMSwcOFCWrduTY0aNRKdIzW37N/6+tLiueeei/t/V1dXatSogWVZPPvss3HtBQoU4J577uGff/657fFu/isoMjKSs2fPctddd1GgQIFEnw/c/jOKjo7ml19+oW3btpQqVSpuv/Lly9OsWTO7X2dKn13seZYuXUrbtm0pW7Zs3H7FixfnySefZN26dXEx1a9fnx07dnD16lUA1q1bR8uWLQkODmbt2rWAucpjs9m4//77gYzpv0lJzfudlridwYcffkjBggV544034trKlCkDwF9//cXLL7+c4vODgoISfScl91OsWLHbxnP9+nU8PT0Ttcde7bl+/Xqyz+3Zsyf79+/n2WefZc+ePfzxxx907tyZEydOJHru2bNnGTZsGG+++eZtr95duXKFa9eu0blzZyZNmsSjjz7KpEmTeOGFF/jmm284cOBAss89e/Ysbm5u5M+fP9l97OmL+/bto1evXtStW5cuXbrcdv+0SM17n57PCTK236Q3Fnv7TWr6F5h/Rw0aNGDatGnMnz+f7t27M2bMGKZMmZIohoIFC3L9+vUEd+6llobz7FCrVq0kk4PknDp1iuvXr3PXXXcleuzmttOnT3PhwgWmTZvGtGnTkj3WzW7+pQumEwCcP38+7rLqu+++S5cuXQgMDKR69eq0bNmSzp07J/hFGuvuu+9OsH3nnXfi4uLCoUOH7I7v9OnTXLp0KcXblO11xx13pPsYt75Hvr6+eHl5xQ293Nx+9uzZ2x7v+vXrjB07lhkzZnDs2LEEl/6TujR+u8/o9OnTXL9+PdF7D3DPPffEJUG3k9JnB6Z/Xbt2jXvuuSfRc8uXL09MTAxHjx6lYsWK1K9fn6ioKDZu3EhgYCCnTp2ifv367N69O0EyUqFCBQoVKhR3/NT2X3s+39S832mJ29Fik9vWrVsn+Uu+W7dutx3WLViwIE2bNs2wmPLkyZNkzZzw8PC4x5Pz4osvcvToUcaPH8/MmTMBqFGjBoMGDWL06NEJXuMbb7xBoUKF4hL+28UE0KlTpwTtTz75JFOnTmXjxo1J/huy1+364smTJ2nVqhW+vr7MmzcPV1fXNJ8rJal579PzOUHG9pv0xmJvv0lN//rmm2/o0aMH+/fvp2TJkoAZgo+JiWHw4MF06tQpwTSc2O+W9NTmUxLlQLEF4J5++ulk/8qpUqVKgu3k/iHf/IvmiSeeoH79+ixYsIClS5cyfvx4xo0bx3fffZfgSlhSbu5M9sZnpXHuUlKS+oeXXAePjo5Osj2p98ie9y05ffr0YcaMGfTv35+6devi6+uLzWajY8eOSRbxS8+50iM9XwQ1atTAy8uLNWvWUKpUKYoWLUq5cuWoX78+H330EREREaxdu5Z27drFPSct/fd2X6yQuvc7LXE72j///MPly5cTzdc5ffo0kPgKY1Ju3LjBuXPn7DpfkSJFbpsAFC9ePNH8LyDur/2br6gnZfTo0QwcOJDdu3fj6+tL5cqVee211wAoV64cYOr3TJs2jYkTJya4Sh4eHk5kZCSHDh3Cx8cnLtkNCAhg9+7diRLKokWLAuaPkuQULlyYqKgoLl++jLe3d5L7pNQXL168SIsWLbhw4QJr16697etPj9S89+n9nDKy36Q3FrCv36Rmv48++oiqVavGJVCx2rRpw+eff87OnTsTJJHnz58nb968dn0vJUdJVCYoWrQoXl5e/PXXX4keu7mtSJEieHt7Ex0dnaF/VYLp4D179qRnz56cOnWKatWqMXr06ERJ1IEDBxL8RfbXX38RExNDmTJl7I4vJiYGHx8f/vjjj9vGlZZf9AULFkyysOHhw4dTfay0mDdvHl26dOH999+PawsPD09zscUiRYqQJ0+eJIcj/vzzT7uPk9JnF3uevHnzJnnMffv24eLiEjfx08PDg1q1arF27VpKlSpF/fr1AXOlJyIigq+++oqwsLAEk/Azq/+m5v1OS9zpld6K8rHJ0q1XRseOHZtke1I2bNhA48aN7TrfwYMH4/pEcoKDg1m5cmXcZPxYmzdvjnv8dgoWLJhgyHT58uWULFmSe++9F4Bjx44RExND37596du3b6Ln33HHHfTr14+JEycCUL16dZYtW8axY8cSXE2NTcBSGg6MPefBgwcTJfK3Ex4eTuvWrdm/fz/Lly+nQoUKqXp+aqXmvU/v55SR/SYj+gzcvt+kZr+wsLC4K/83i4yMBCAqKipB+8GDBylfvrxdcSZHc6IygaurK02bNmXhwoUJ/uL666+/EsylcnV15bHHHmP+/PlJJiCxX7apER0dnWjIo2jRogQEBCR56TUkJCTB9uTJkwFo0aKF3fG5uLjQtm1bfvjhB7Zt25Zov5uvwMTeHZGaBOTOO+/k4sWL/Pbbb3FtJ06cYMGCBXYfIz1cXV0TXUWaPHlyslfC7Dles2bNWLhwIUeOHIlr37t3L7/88ovdx0nps4s9z0MPPcT333+fYNmSsLAwZs+ezf3335/gy69+/fps3ryZlStXxiUjfn5+lC9fPu7Owtj22ONndP+NPW5q3u/Uxp1eaenDN/P19QVI8J7Nnj2bNWvWAPHDISnJ6DlRjz/+ONHR0QmGZSMiIpgxYwa1a9eOS7avXbvGvn37Et2tfKs5c+awdetW+vfvH3eHV6VKlViwYEGin4oVK1KqVCkWLFiQYN7iE088AcBnn32W4Niffvopbm5uNGrUKNnz161bFyDJ76OUREdH06FDBzZu3MjcuXPjjpOZ7H3vU7tvUjKy32R0n4Gk+01q9itXrhw7d+5MVGH+66+/xsXFJVFCvWPHDurVq3fbuFKiK1F2WLx4Mfv27UvUXq9evSTnGYFZA23p0qXcd999vPTSS0RHRzNlyhQqVaqUoFz9O++8w8qVK6lduzbPP/88FSpU4Ny5c+zYsYPly5fbfek11uXLlylZsiSPP/44QUFB5M+fn+XLl7N169YEf9nHOnjwIG3atKF58+Zs3LiRL7/8kieffJKgoKBUxTdmzBiWLl1Kw4YN6dGjB+XLl+fEiRPMnTuXdevWUaBAAcD8dQnw+uuv07FjR9zd3WndunWKr6ljx44MHjyYdu3a0bdvX65du8bHH39MuXLlkpzYndEefvhhvvjiC3x9falQoQIbN25k+fLlKZa4uJ233nqLJUuWUL9+fXr27ElUVBSTJ0+mYsWKCZLFlNzuswN4++23WbZsGffffz89e/bEzc2NqVOnEhERwbvvvpvgePXr12f06NEcPXo0QdLRoEEDpk6dSpkyZRJdJs/o/gupf7/TEnd6JNeHY5Mrm81Gw4YNWbVqVZLPL1++PHfccQeTJk0ib968uLi48M477/DEE0/w7bffMmLECAYMGJDi7f8ZPSeqdu3atG/fnqFDh3Lq1CnuuusuZs6cyaFDhxIkMVu2bKFx48YMHz6cESNGALBmzRpGjhzJQw89ROHChdm0aRMzZsygefPm9OvXL+65fn5+tG3bNtG5Y6883fpY1apV6d69O//73/+IioqKe0/nzp3L0KFDUxwuKlu2LJUqVWL58uV0797d7vfhlVdeYdGiRbRu3Zpz587x5ZdfJnj86aeftvtYAFOmTOHChQtxf0z/8MMP/Pvvv4AZtvb19bX7vQf7P6fkZGS/SU+fAfv7jb37Abz66qssXryY+vXr07t3bwoXLsyPP/7I4sWLee655xL0me3bt3Pu3DkeeeSR9L0R6bq3L4dLqcQBN92umlyxzRUrVlhVq1a1PDw8rDvvvNP69NNPrVdeecXy8vJKsF9YWJjVq1cvKzAw0HJ3d7eKFStmPfDAA9a0adPi9om9LffWImK3njsiIsJ69dVXraCgoLjiikFBQdZHH32U4Hmxx9uzZ4/1+OOPW97e3lbBggWt3r17JyrYaE98lmVZhw8ftjp37mwVKVIkroBkr169ElUAHjVqlFWiRAnLxcUlLvbkXl+spUuXWpUqVbI8PDyse+65x/ryyy+TLXFw6zG6dOli5cuXL9Exk7u1+lbnz5+3unXrZvn5+Vn58+e3mjVrZu3bt88qXbp0ghIJ9n5GsVavXm1Vr17d8vDwSFOxTXs+O8syt/g3a9bMyp8/v5U3b16rcePG1oYNGxLtd+nSJcvV1dXy9vZOcDvxl19+aQHWM888k2Q86em/Sb0/9r7f6Y37VvaWOLCspPuwZVnW5cuX7boFf+fOnVadOnUsT09Pq2DBgtbrr79uxcTEWN27d7fc3NySLBKZ2a5fv24NHDjQKlasmOXp6WnVrFnTWrJkSYJ9Yt+j4cOHx7X99ddf1kMPPWT5+flZnp6e1r333muNHTs20b/75KT07/DGjRvWiBEjrNKlS1vu7u7WXXfdZU2YMMGu437wwQdW/vz5E5XMSKkvNmzYMMXv/NQqXbq0XSUE7Hnv07JvZktrn7Es+/tNavvX5s2brRYtWljFihWz3N3drXLlylmjR4+2IiMjE+w3ePBgq1SpUule9sVmWZk821USaNu2Lbt3707x9tysMGLECN566y1Onz5t1xwMcR767JzXzz//zMMPP8yuXbtSvJIkme/ixYuULVuWd999N8EwoUhERARlypRhyJAhia5mpZbmRGWiW+tXHDhwgJ9//jnFsXwRyb5WrlxJx44dlUA5AV9fXwYNGsT48eOTvItWcq8ZM2bg7u6eIXUJdSUqExUvXpyuXbtStmxZDh8+zMcff0xERAQ7d+5MV32TjKCrGdmXPjsRx4iOjr7tDRP58+dPscin5CyaWJ6Jmjdvztdff83Jkyfx9PSkbt26jBkzxuEJlIiIpN7Ro0dvW6Tz1gnUkrPpSpSIiIgdwsPDWbduXYr7lC1bNtm7tiXnURIlIiIikgaaWC4iIiKSBpoTlQoxMTEcP34cb2/vdC/9ICIiIlnDsiwuX75MQEBAitXQU0tJVCocP378tmX1RURExDkdPXo0Q1cwUBKVCrGrgR89ejTBmmMZLTw8nKNHjxIYGIiXl1f8A5cuwSuvwIIFsGIF3LS8h0haJdvfRDKB+ptktfDwcPbu3UujRo3ifo9nlFyXRLVr145Vq1bxwAMPMG/evFQ9N3YIz8fHJ1OTKA8PD/Lnz4+Pj0/CLxkfH/jmG/jzT7hlhWsuXABfX9Awo6RSsv1NJBOov0lWi+1zQIZPxcl1E8v79evHrFmzHB1G2tlsiROoyEho3hzatoWzZx0SloiISG6T65KozLic53DDhsHmzbBokRniS2b1eBEREck4TjWct2bNGsaPH8/27ds5ceIECxYsoG3btgn2CQkJYfz48Zw8eZKgoCAmT55MrVq1HBOws6hfHz79FM6cgWPHoEkTeOMNk1y5OdVHLCK5XHR0NDExMURERDg6FMlB3N3dcXV1zfLzOtVv2KtXrxIUFET37t159NFHEz0+Z84cBgwYwCeffELt2rWZOHEizZo1488//6Ro0aIABAcHExUVlei5S5cuJSAgINNfg0O0bAm7dsEzz8Cvv4JlwahR5v9nz4ZSpRwdoYjkcpZlcfLkSc6fP09UVBTHjh1TqRjJUAUKFKBYsWJZ2q+cKolq0aIFLVq0SPbxDz74gOeff55u3boB8Mknn/DTTz/xv//9jyFDhgAQGhqaYfFEREQk+Gvp0qVLgJnp7+HhkWHnSeq8N27cSN1faoUKwaJFuL7/Pm4jR2KLjob167GCgoj8+GNibrmiJxIrTf1NJJVOnz7N5cuX8fPzw83NDQ8Pjwyt1yO5l2VZXLt2jdOnTxMVFUWRIkUSPB77HZcZnCqJSsmNGzfYvn07Q4cOjWtzcXGhadOmbNy4MVPOOXbsWN56661E7UePHs3UVbojIyM5ffo0NpsNd3f31D25Qwe8ypUj4JVXcD9+HNuFC3h06sT5Tp04NWyY7t6TRNLV30TsYFkWERER+Pv74+3tTWRkJK6urroSJRnG29ub6OhowsLCuHr1aoK+FRkZyZkzZzLlvNkmiTpz5gzR0dH4+/snaPf392ffvn12H6dp06bs2rWLq1evUrJkSebOnUvdunWT3Hfo0KEMGDAgbvvSpUsEBgYSGBiYqSUOIiIisCyLwMBAPD09U3+A0qWJbtQIl169cP3uOwC8ixQhT5kyGRuo5Ajp7m8itxEREcGxY8fw9vbG3d0dy7Jwd3fXlSjJUN7e3pw7d47ixYsn+C6LiIjg8uXLmXLObJNEZZTly5fbva+np2eSv1S8vLwyvb6Jh4cHnp6eaT9PsWIwb56ZcP7FF7iNH49bJg5BSvaW7v4mchs2mw1XV1dcXFwS/IhklNirm0l9l2XWFJxs04P9/PxwdXUlLCwsQXtYWBjFihVzUFROzmaD55+H1avh1g70yy9w/rxj4hIREckBsk0S5eHhQfXq1VmxYkVcW0xMDCtWrEh2OE7+c+u8g99/h0ceMTWl1q1zTEwiIjmAzWZj4cKFjg5DHMSpkqgrV64QGhoad4fdwYMHCQ0N5ciRIwAMGDCA6dOnM3PmTPbu3ctLL73E1atX4+7WEzv17AkREXD0KDRsaMohREc7OioREafTtWtXbDZbop/mzZsDcOLEiRTvKs+pypQpw8SJEx0dhsM51Zyobdu20bhx47jt2EndXbp04fPPP6dDhw6cPn2aYcOGcfLkSYKDg1myZEmiyeYZLSQkhJCQEKJzSqIxezY89RSsXQsxMaYo56+/wpdfQokSjo5ORMSpNG/enBkzZiRoi50vq+kkyYuOjsZms+XouW9O9coaNWqEZVmJfj7//PO4fXr37s3hw4eJiIhg8+bN1K5dO9Pj6tWrF3v27GHr1q2Zfq4sERhokqYRIyC2c69aBVWqmKVjREQkjqenJ8WKFUvwU7BgQSDxcN6GDRsIDg7Gy8uLGjVqsHDhQmw2W4Iahn/88QctWrQgf/78+Pv788wzzyS4Bb9Ro0b07duXQYMGUahQIYoVK8aIESPiHrcsixEjRlCqVCk8PT0JCAigb9++cY+XKVOGUaNG0alTJ/Lly0eJEiUICQlJ8JouXLjAc889R5EiRfDx8aFJkybs2rUrwT4//PADNWvWxMvLCz8/P9q1axcX3+HDh3n55ZfjrswBfP755xQoUIBFixZRoUIFPD09OXLkCI0aNaJ///4Jjt22bVu6du2aIOa3336bzp07kz9/fkqXLs2iRYs4ffo0jzzyCPnz56dKlSps27bN7s8tKzhVEiVZyM0Nhg83yVPJkqbt3DkzV6pvXwgPd2h4IiLZzaVLl2jdujWVK1dmx44djBo1isGDByfY58KFCzRp0oSqVauybds2lixZQlhYGE888USC/WbOnEm+fPnYvHkz7777LiNHjmTZsmUAzJ8/nwkTJjB16lQOHDjAwoULqVy5coLnjx8/nqCgIHbu3MmQIUPo169f3PMB2rdvz6lTp1i8eDHbt2+nWrVqPPDAA5w7dw6An376iXbt2tGyZUt27tzJihUr4pZY++677yhZsiQjR47kxIkTnDhxIu64165dY9y4cXz66afs3r07bjURe0yYMIH77ruPnTt30qpVK5555hk6d+7M008/zY4dO7jzzjvp3LkzlmXZfczM5lTDeeIA9eubJWOefRZi/5qaPBmOHInfFhHJYLVq2Th5MuvPW6wYpPZixo8//piowPJrr73Ga6+9lqBt9uzZ2Gw2pk+fjpeXFxUqVODYsWM8//zzcftMmTKFqlWrMmbMmLi2//3vfwQGBrJ//37KlSsHQJUqVRg+fDgAd999N1OmTGHFihU8+OCDHDlyhGLFitG0aVPc3d0pVapUojVk77vvvriVPMqVK8f69euZMGECDz74IOvWrWPLli2cOnUqbljyvffeY+HChcybN48ePXowevRoOnbsmKDgdFBQEACFChXC1dUVb2/vRMOZkZGRfPTRR3H7pkbLli154YUXABg2bBgff/wxNWvWpH379gAMHjyYunXrOtVd+UqixCwZ89138PHHMGAAREXBLX89iYhkpJMnzXrp2UHjxo35+OOPE7QVKlQo0X5//vknVapUSVCj6NbkZteuXaxcuTLJVS/+/vvvBEnUzYoXL86pU6cAcxVp4sSJlC1blubNm9OyZUtat26N200Lzt9613rdunXjJoLv2rWLK1euULhw4QT7XL9+nb///hswS6jdnPzZy8PDI1Hs9rr5ebFznW++whbbdurUKSVR4mRsNnPXXv36sGULqGyEiGQiR/0OTMt58+XLx1133ZUh579y5QqtW7dm3LhxiR4rXrx43P/fugSTzWYjJiYGgMDAQP7880+WL1/OsmXL6NmzJ+PHj2f16tV2Ld105coVihcvzqpVqxI9VqBAAQDy5MmTilcVL0+ePImW83FxcUk0BBcZGZnouTfHHnuMpNpi3wdnoCTKDjnu7ryUVK5sfm4WHQ3PPQcvvghZMJFfRHK+LVssXFxy1tp599xzD19++SURERFxw2S33pBUrVo15s+fT5kyZRJcOUqtPHny0Lp1a1q3bk2vXr249957+f3336lWrRoAmzZtSrD/pk2bKF++fFwMJ0+exM3NjTLJLAdWpUoVVqxYkWwJIQ8PD7t/JxYpUiTBvKno6Gj++OOPBHfjZ1eaWG6HHHd3XmqNHQuffw733w/vvGPKIoiI5BIRERGcPHkywU9SC9o++eSTxMTE0KNHD/bu3csvv/zCe++9B8RfRenVqxfnzp2jU6dObN26lb///ptffvmFbt262Z2UfP7553z22Wf88ccf/PPPP3z55ZfkyZOH0qVLx+2zfv163n33Xfbv309ISAhz586lX79+gFlDtm7durRt25alS5dy6NAhNmzYwOuvvx5399vw4cP5+uuvGT58OHv37uX3339PcPWsTJkyrFmzhmPHjt12cd8mTZrw008/8dNPP7Fv3z5eeuklLly4YNdrdXZKoiRl0dEQe0dHVBQMHQrNmsFNf1WIiORkS5YsoXjx4gl+7r///kT7+fj48MMPPxAaGkpwcDCvv/46w4YNA4ibJxUQEMD69euJjo7moYceonLlyvTv358CBQrYXU+pQIECTJ8+nfvuu48qVaqwfPlyfvjhhwRznF555RW2bdtG1apVefvtt/nggw9o1qwZYBK6n3/+mQYNGtCtWzfKlStHx44dOXz4cNy8o0aNGjF37lwWLVpEcHAwTZo0YcuWLXHHHzlyJIcOHeLOO++kSJEiKcbbvXt3unTpQufOnWnYsCFly5bNEVehAGyWM90r6OQuXbqEr68vFy9exMfHJ9POEx4ezuHDhyldurRzLAgbFQVvvQWjR0NsdylSBGbOhFxYqTencbr+JjlOeHg4Bw8e5I477sDDw4MbN27g4eGRo4swxvrqq6/o1q0bFy9eTPM8o9QqU6YM/fv3T1SbKae7uZ/d/F0WHh7O7t27qVGjRob//s75PVjSz83NLA2zYgUEBJi206ehZUt45RW4ccOx8YmIOIlZs2axbt06Dh48yMKFCxk8eDBPPPFEliVQkrWURIn9Gjc2NaUefji+7YMPoF49OHDAcXGJiDiJkydP8vTTT1O+fHlefvll2rdvz7Rp0xwdlmQS3Z0nqePnZ5aGmTwZXn3VXIXavt1sT5rk6OhERBxq0KBBDBo0yKExHDp0yKHnz010JcoOISEhVKhQgZo1a2bJ+caMcWXJksSF2JyGzWaWhtm0CcqVg4oVzV17IiIiuYiuRNmhV69e9OrVK25ieWb68UcYNcodKMHevVF8+CH8V27E+VStaq5CnT4NefMmfOzSJcjEyfciIiKOpitRTmbx4vj/nzrVjXr14L8q/M4pf364446Ebfv3Q5ky8P77qiklInF0M7hkJkf0LyVRTmbKFJgyJRIPD5N87NgB1arBvHkODsxeERHQsSOcPw8DB0KrVvDfek8ikjvFLt1x7do1B0ciOVls/7Jn6ZuMouE8J2OzwbPPRlOy5DEGDizFX3+5cOkStG8PvXvDe+858fAemBfQtCns3Gm2lyyBoCD44gvTLiK5jqurKwUKFODUqVPExMTg7u5OTExMrqgTJZnPsiyuXbvGqVOnKFCgAK6urll2biVRTureeyPYsOEGfft68c03pm3KFNiwAb79Fu6807HxJcvDA9591yRMnTtDWJhZrv2hh2DQIFNvKgv/ShAR51Dsv5V/T58+TVRUFG5ubokWqhVJjwIFCsT1s6yiJMqJeXvD7NmmPFPfvmakLHZ477PP4PHHHR1hCh56yNSU6tIFfvnFVDofNw5WrYKvv048j0pEcjSbzUbx4sXx8fHhyJEjBAQExC3SK5Je7u7uWXoFKpaSKCdns0GPHlC7thnSO3CA7DO85+8PP/9sCnIOHWqWj9m8GYKDzZIxbds6OkIRyWKurq64uLjg6empZYYk29OAtB2yuk5UUoKCTDWBjh3j26ZMgfvuc/K791xczATzDRvixyAvXzaX2URERLIxJVF26NWrF3v27GHr1q0OjSN2eG/q1PirT9u3m+G9+fMdGtrt1axpxiKfegqGDIEHHnB0RCIiIumiJCqbiR3e27QJ7r7btF26ZOZHxc6bclo+PuYuvbffTthuWbBwofmviIhINqEkKpsKDoZt26BDh/i2yZPN8N4//zgsrNuz2cwQ380mToR27cwcqbNnHRGViIhIqimJysZ8fMyNbp98knB4r2rVbDC8F+v4cTO8B2Zh46AgcwefiIiIk1MSlc3ZbPDCC9l0eA8gIAAWLAA/P7N97Bg0aQLDhpm7+URERJyUkqgcItsO7wG0bGlqSjVpYrYtyxTlbNQIjhxxaGgiIiLJURKVg2Tr4b2AAFi6FMaMgdiCaevXm+G9775zbGwiIiJJUBKVw9w8vHfXXaYtdnivXz8nH95zdTVFOdetgzJlTNuFC/DYY6Y4p4iIiBNREmUHZyi2mVrBweYq1M3De5MmZZPhvTp1zALGTzxhtu+8Ex591LExiYiI3EJJlB2cpdhmasUO7338cTYc3itQAL75Bj791LwIVTgXEREnoyQqh7PZ4MUXYePGbDi8Z7PBs8+aauc3O3wYunaFc+ccEpaIiAgoico1qlZNenjv/vuzwfDezaKizNIxM2eaMct16xwdkYiI5FJKonKRpIb3tm0za+9lmxvg9u+HvXvN/x89Cg0bwsiREB3t2LhERCTXURKVyyQ1vHfxorkBzumH9wAqVIDQUGjQwGzHxMDw4WZB43//dWhoIiKSuyiJyqVih/dib4ADM7xXvz4cPOi4uOwSGAi//gpvvRW/Dt/q1aam1KJFjo1NRERyDSVRuZiPj7kB7qOPwMPDtG3dahKsBQscG9ttubqapWFWrTJJFZiJ5o88An36QHi4Q8MTEZGcT0lULmezwUsvmeKcd95p2i5eNGWZssXwXv36ZnivXbv4tilTNOFcREQynZIoAczVpx07sunwXqFCpvDVRx+ZGfN9+0LTpo6OSkREcjglURInWw/vxV5S27EDxo1L+JhlwdWrjolLRERyLCVRkkBKw3v9+8ONGw4N7/YqVAAvr4Rtn34KlSqZFyUiIpJBlETZITuunZdescN77dvHt334oSnO6fTDezfbs8dM7jp0yAT/zjumLIKIiEg6KYmyQ3ZdOy+9fHxgzpzEw3vVqsHChQ4NzX758pmMEExBzqFD4aGH4MQJx8YlIiLZnpIoSVHs8N7GjfHDexcumJvhssXwXunSpobUG2+YFwOwYoWpKbV4sWNjExGRbE1JlNilWjVTnDNbDu+5ucGoUSZ5CggwbadPQ8uWMGBANqjjICIizkhJlNjN1zebD+81bgy7dsHDD8e3TZgA9eqZOVMiIiKpoCRKUiWl4b2XX84Gw3t+fmZpmA8/jM8ET582E8BERERSQUmUpElSw3sTJ5rhPae/qGOzmYKcmzaZkghffWUKdoqIiKSCkihJs9jhvZCQhMN75ctDs2am5uW2beamOKdUtSr89pspy36zkydN4CIiIilQEiXpYrNBz55meK9sWdMWHg5Ll8KQIVCzJhQuDG3bmmVk/vjDFBB3Gq6uCbdjYuCZZ8w8qfffV00pERFJlpIoyRDVqpninL17Q4kSCR+7eBG+/97UvKxcGYoVg06dYPp0+PtvJ0uqZs6E5cshMhIGDoRWrSAszNFRiYiIE1ISJRnG1xcmT4ajR+HPP+Hjj82cqcKFE+536pRZo69HD7jrLrjjDujeHb78Eo4fd0zscZ5+GgYPjt9essTUlFq2zHExiYiIU1ISJRnOZoNy5eDFF+Hbb03SFBoKH3xgqgt4eyfc//BhmDHDjKKVKGHmVPXqBfPnw9mzWRy8u7tZGmbpUnPJDMyVqIceMslVZGQWByQiIs5KSZRkOhcXczHn5Zfhhx/g3Dkzh2r0aHjggcTrBe/bZ2pRPf44FCli5n8PHAg//wyXL2dR0A8+aGpKNW8e3/buu+b2w3/+yaIgRETEmSmJkizn5gZ16sBrr5npR+fPw6+/mpVZ6tVLONfbssxVrPffN9OTChY0+7z5JqxcaSaxZ5qiReGnn8zJ3d1N25YtEBxsxixFRCRXUxIlDuflZYqJjxoF69ebpOqnn+CVV8xVqNgl78CUS9i4Ed5+G5o0MUlVmzYwe3YmXaVycTFLw2zYYCZwgZnoFRiYCScTEZHsxM3RAWQHISEhhISEEO20BY9yFm9vs6xdy5Zm++xZWLXKXK369Vcz3BcrPNwMEf7wg0nGWrWCDh3Mf/PmzcCgatQwtx+OHm0ug4mISK5nsyynusHcqV26dAlfX18uXryITyYuExIeHs7hw4cpXbo0XrdOGBKOHzdDeStWmJvnTpxIvE++fNC6tUmomjdPPO8qQ82ZYyaf9+mT8LJZNqH+JllJ/U2yWnh4OLt376ZGjRoZ/vtbw3mS7QQEwFNPwf/+Z6YmrVpl1vMrUiR+n6tXTRmFdu3A3x86dzZDhBm+tt/ff8Pzz5siWG3awJkzGXwCERFxVkqiJFtzdYWGDc3dfMePm3JOzz2XcCm8S5fgiy9MeYVixeDZZ00Fg6ioDAhgyZL4yVg//mhuQ1y5MgMOLCIizk5JlOQYbm7QtKmphH7ypCmJ0KUL3Hzl9vx5cwWrWTMoXtzUslq5Mh3r+/XqZS5xxV4GO37c1G14880MytJERMRZKYmSHMndHVq0gM8/N9OVFi40S83kyxe/z5kzMHWqucuvZEkzpWnz5jScrGVLU1PqgQfMtmWZ2wcbNjSVREVEJEdSEiU5npcXPPKIKYNw6hTMnWsKeebJE7/PyZMwZYqpX/Xww7B7dypPUry4GSMcOza+0NWGDaam1Lx5GfVSRETEiSiJklwlb16TQM2daxKq2bNNguXhEb/PTz9BlSpmvnhSd/4ly8UFhgyBdeugTBnTduGCGfLLslLrIiKSVZRESa6VP78Z4lu40CRU06aZYT2AmBj49FNTX3P4cLhyJRUHrlMHdu6EJ54w27NmJV4wUEREsj0lUSKAr6+58rR/vxmRi52Mfu0ajBxpkqmpU1MxV7xAAVNjYfNmM4v9Ztevm3lTIiKSrSmJErlJnjxmRO6vv8xEc7f/avqHhZk7+SpXhkWL7MyBbDaoVSthm2VBx45mTPHcuQyPX0REso6SKJEkFCkCkybBnj0m34m1b5+ZQ9WoEWzdmoYDh4SYLOy778yk83XrMihiERHJakqiRFJw991mEvqGDVCvXnz7mjXmIlOnTnDwYCoOWLp0fCXQo0dNGYSRI9NRqEpERBxFSZSIHerWNReN5s83iVWsb76Be++FV16xc3SudWtTU6phQ7MdE2Nmrj/wAPz7b6bELiIimUNJlIidbDZ49FFTQ2rKFPDzM+03bsAHH8Cdd8Inn5i8KEUlS5rVk0eONGURAFavNkvGLFqUqa9BREQyjpIokVRydzeln/7+G157zRTzBFMS6qWX4L774LffbnMQV1ezNMzq1RAYaNrOnTMTrvr00ZIxIiLZgJIokTTy8YHRo+HAAejcOb590yaoVg0GDYKrV29zkPvvh9BQaNcuvu306fiq5yIi4rSURImkU8mSMHOmWcj4nntMW3Q0jB8PFSuahZBTVKiQmWz18cdQoYIpSGWzZXrcIiKSPkqiRDJIo0ZmzvjIkeDpadoOH4ZWraB9ezh+PIUn22ymENWuXaby5802bzZjhSIi4lSURIlkIE9PM9Xpt9+gSZP49nnzzF18ISG3qWYQW90z1r//QsuWULUqbNyYKTGLiEjaKIkSyQTlysHy5fDFF/F38V2+DL17m3IJoaF2Hqh3bzPh/NAhqF/frElz29v/REQkKyiJskNISAgVKlSgZs2ajg5FshGbDZ5+Gv78E557Lr5961aoUcPUlrrtwsYffmhu9wNzCeu11+Chh24zNigiIllBSZQdevXqxZ49e9iapnU+JLcrVAimTzdVzsuXN23R0aa2VMWKsHhxCk8uXRpWrTJjhLGTzVesMDWlbjtjXUREMpOSKJEsUr++GcYbPTq+ttSRI2bK01NPmcoGSXJzM7PVf/0VAgJM25kzZsb6gAEQEZEV4YuIyC2URIlkIQ8PMyL3++9mpZdYs2ebq1SzZoFlJfPk2Nv/WreOb5swwSRTyT5JREQyi5IoEQe46y5YtgxmzICCBU3b2bPQpQs0a5bCosZ+fvD99zBpksnIwJRPV10pEZEspyRKxEFsNujaFfbuhY4d49uXLYNKleD995NZ/cVmM0vDbN5sxgZvrnYuIiJZRkmUiIP5+8PXX8MPP5jq5wDXrsHAgVCnDuzcmcwTg4PN2ODNLAuGDDG3AIqISKZSEiXiJB5+GPbsMReZYkfntm+HmjWhb19YuhQuXbrNQWbOhHHjoF49eO891ZQSEclESqJEnIi3t5nutH69KX8AphzC5MlmrlTBgmZx4z59YM4cOHbspidbFnz5pfn/qCh49VVz619YWJa/DhGR3EBJlIgTqlsXduwwlQ1i54+DubC0cydMmWLmUZUsCWXKmKKen/3PxsVvFsPgwfFP+OUXU1Nq6dIsfw0iIjmdkigRJ+XhYWpsHjli5kz16mXyoVtvxDt8GL76ylRFLxbozjPH3mHX+KVYxYqZHcLCzGWswYPhxo2sfyEiIjmUkigRJ+fvb646TZliinWeP2+qnL/xhikdlSdP/L7h4WZEL/jVB6ntuYu/724e/+C778L998Pff2f1SxARyZGURIlkM76+0Lw5jBoFK1fChQtmDlXv3vE1pwC2Hi7K3Qd+4hXb+0Ta3P9r3GqyLxERSTclUSLZnIeHuRlv8mSzLvHXX8ODD5phPwsXPrAGUNfawAHu4opvgNlRRETSTUmUSA7i5WWG/pYuNVXPR4wwaxhvpwbV2EGDK4vZeMAv4ZO09p6ISJooiRLJoUqXhuHD4Z9/4JVX4Are7IyuwhNP3LTY8alTeFapQoEUF+0TEZGkKIkSyeFcXOCdd6BBA7P977+mJEJ0ZAx06YLtyBH8x4zB/fHH4cwZxwYrIpKNKIkSyQXc3OCbb8ydfmCG+0aPiomv6Am4/vyzqaGwcqWDohQRyV6URInkEsWLm0TK5b9/9SPedmPpQ+9xY+FCogoVMo3Hj8MDD5g7+JJc/VhERGIpiRLJRRo1gtGjzf9bFjz5JByu0IxD339PdJMm8Q+MHm3G/w4dclSoIiJOT0mUSC4zaJBZ7Bjg7Fl4+mkPrvsWIfKHH8zkKTc38+DGjRAcDN9957BYRUScmZIokVzGxQVmzjRr7gFs2eLCe+8VMQ8MHgzr1sEdd5gHL16Eq1cdFquIiDNzc3QAIpL1ChWCuXPhvvvMcnqzZhViyRKLsmWhbNnalH9sJ0+ueZG8BdzxffwZ8tz+kCIiuY6uRInkUjVqwIcfxm+fOmVj0yaYPRvefM+XO7fMJnDpZ5QqBe+9B9eu/bfjqlWqKSUigpIokVzthRfgo48iqV79GsWL35oY2YjCnTNn4NVXoWxZ+LHLXGjcGB57DM6dc0jMIiLOQkmUSC5ms0G3btF89dVR/vkngmvXYM8e+PFHmDQJ2rc3+wCEh13gvlk9zMaCBVhVgmDtWscFLyLiYEqiRCROnjxQvjy0agV9+sC338Lvv8MTT8BFCtCZWZyhMAC2Y/9iNWoEb70F0dGODVxExAGURIlIiipWhDlz4LffwPOx1gSxi1U0BMAWE2NWOW7SBI4edWygIiJZTEmUiNilcmWYNw9+2F6C7qVW8CYjiY79ClmzxtSUWrjQkSGKiGQpJVEikirVqsHaDa4srPQmDVjDYUqZB86dg3bt4KOPHBugiEgWyVVJ1NGjR2nUqBEVKlSgSpUqzJ0719EhiWRLJUqYi0+2++4jmFDm8ygAET5+0LatY4MTEckiuSqJcnNzY+LEiezZs4elS5fSv39/rqoas0iaFCwIS5fC/Q8X5HHm8QKf8OilmYz7IkBlpEQkV8hVSVTx4sUJDg4GoFixYvj5+XFOtW5E0ixvXrO0XpcuNqbxAj/TkiFDoEMHuHIFOH/e3OZ34YKjQxURyXBOlUStWbOG1q1bExAQgM1mY2ESk1RDQkIoU6YMXl5e1K5dmy1btqTpXNu3byc6OprAwMB0Ri2Su7m7w4wZ5ia9WHPnQr26Flc6PQ9TpkDVqmZBYxGRHMSpkqirV68SFBRESEhIko/PmTOHAQMGMHz4cHbs2EFQUBDNmjXj1KlTcfsEBwdTqVKlRD/Hjx+P2+fcuXN07tyZadOmZfprEskNbDYYPhy+/x58fEzb5T8OEbV0hdk4dAjq14exYyEmxmFxiohkJJtlOefsBZvNxoIFC2h70yTV2rVrU7NmTaZMmQJATEwMgYGB9OnThyFDhth13IiICB588EGef/55nnnmmdvuGxEREbd96dIlAgMDCQsLwyf2N0UmiIiI4MiRI5QqVQpPT89MO48IZHx/27/fxhNPuPPnny4EcoTZPMn9rI97PLpxYyI//RQCAtJ9Lsl+9P0mWS0iIoI9e/ZQr149Ll68mKG/v90y7EiZ7MaNG2zfvp2hQ4fGtbm4uNC0aVM22jlMYFkWXbt2pUmTJrdNoADGjh3LW2+9laj96NGj5M+f3/7gUykyMpLTp09js9lwd3fPtPOIQMb3N09P+OorFwYNKsavv5aiEasYxkje4G1csHBduRKrRg1OvvMOVxs2zIBXINmJvt8kq0VGRnLmzJlMOXa2SaLOnDlDdHQ0/v7+Cdr9/f3Zt2+fXcdYv349c+bMoUqVKnHzrb744gsqV66c5P5Dhw5lwIABcduxV6ICAwMz/UqUZVkEBgbqLzXJdJnV3374AaZNi2TsWDeGnxrJrzThS56mJMdwO3+eki+8QFSfPkSNGmUyL8kV9P0mWS0iIoLLly9nyrGzTRKVEe6//35iUjEfw9PTM8l/5F5eXnh5eWVkaIl4eHjg6emZ6ecRgczrb/37w/PPw+TJMG5cI4IvhPI/utOGHwBwmzyZhcfrUX18R+64I0NPLU5M32+S1Tw8PDLluE41sTwlfn5+uLq6EhYWlqA9LCyMYsWKOSgqEbmdfPlgyBA4eBBeesOPZ7y/pw+TiMCDeTxG+/kduOsu6N5dy++JSPaSbZIoDw8PqlevzooVK+LaYmJiWLFiBXXr1nVgZCJijwIFYNQoOHHSRt2v+tCvzhZ6MB2wERNjyiTcfTcMfvkGZ886OloRkdtzqiTqypUrhIaGEhoaCsDBgwcJDQ3lyJEjAAwYMIDp06czc+ZM9u7dy0svvcTVq1fp1q2bA6MWkdTImxeefBI+2RjEb0cL8tZb4OtrHmsS8TPPT6zAY6W20rcv7Nnj2FhFRFLiVEnUtm3bqFq1KlWrVgVM0lS1alWGDRsGQIcOHXjvvfcYNmwYwcHBhIaGsmTJkkSTzTNaSEgIFSpUoGbNmpl6HpHcpmRJGDYM/vkHRvU8wed05S7+Ztm1enhOHk+lijE0agRJ1N0VEXE4p60T5YwuXbqEr69vhteZuFV4eDiHDx+mdOnSmngpmc5p+tuRI9xo+wQeOzfHNS2hGV2YySn8GT8eBg50XHiSMZymv0muER4ezu7du6lRo0aG//52qitRIpKLlSqFx+a1MGQIls0GQHN+4Teq8CBLGTwYVq1ybIgiIjdTEiUizsPdHcaOxbZ0Kfx3160/p1hKM8bGDOKZDje4aQUnERGHUhIlIs6naVPYtQtatIhrGsR45p+6n35tDhIZ6cDYRET+oyRKRJxT0aLw44/w/vtY/y0Pcg9/snW7jZdf1jrGIuJ4SqLsoLvzRBzExQUGDMC2cSPhJe/iJdfpHKYMISHwyCOonpSIOJSSKDv06tWLPXv2sHXrVkeHIpI7Va+O119/8MDUJ+KafvwR7g+6TOjnoY6LS0RyNSVRIpI9eHry7LOweDH4+Zmmwcf6UL5bbebeN5GjR1StRUSylpIoEclWmjeH0FAYUmERXZmJJzdov+Flfi/TmsHdT/Pvv46OUERyCyVRIpLtlCgBo7Y2Z+N9r8S1tbR+ot+MIJ6941f69kWlEEQk0ymJEpFsyS2vB3XXvcelb37mSt4iAARwgsVRTfGf/Dr3lI2kUyf4+GPYvRu0NoOIZDQlUSKSrfl0aEH+v3/jRsOmALhg8TpjWBrRgI3fHKJnT6hUCR58EK5edXCwIpKjKImyg0ociDi5YsXw+PUXeOcdLDc3AOqyiVCCKcefAKxYAZ06QXS0IwMVkZxESZQdVOJAJBtwcYHBg7GtWwd33GHa6tbl+XF34+trNn/4AQYMcFyIIpKzKIkSkZyldm3YuRNeeokCCz5n4CAX5s+H/y5QMWkSvPOO5kiJSPopiRKRnMfXFz76CPz9AXjgAZg2DRqyipf4iKFDLR5/HM6dc3CcIpKtKYkSkVyhW+szLPJ+io/oxXweY+V35wgOhm3bHB2ZiGRXSqJEJHdYtAify6Z41KMsYBdBlD66lmbNYN8+B8cmItmSkigRyR26d4dFi6BwYQAC+ZdVNKL3ubdo2SxaxTlFJNWURIlI7tG6NezaBY0aAeBKDG8xghlHmtCt6VGOHXNseCKSvSiJsoPqRInkICVKwPLlMGoUlqsrAA1Zw+y9wQy+ZyFz5zo4PhHJNpRE2UF1okRyGFdXeOMNbKtXExlQCoDCnOOTq0/T64lTPP00XLjg2BBFxPkpiRKR3Ou++3D/I5QbrR8DoA+TOU1RvvoKKleGX391cHwi4tSURIlI7lawIB7fz8VauowHv+waV93833+h2QNRjB1jqTCniCRJSZSIiM2G7cGmPPmUjd9/hyZNTPNE+nHn6x14seMFwsMdG6KIOB8lUSIiNwkMhGXL4NsnF9KLj3iCuQz9Npi+NTZw8qSjoxMRZ6IkSkTkFi4u0L6DCzfyFQCgDIf5aHcDZt07mvVrojW8JyKAkigRkaS1aYPHnl1cCb4fADeiGXTxDSIaPkj9ssd44w04dMixIYqIYymJEhFJTqlS5N+6kisDhhH939dlE1ay8FAQoaN/pGxZaNMGTpxwcJwi4hBKokREUuLmRv733yJ66a9cKVACAD/O8iOtmWD1Y8kPN2jYEFU7F8mFlETZQRXLRcTjwYbk/2uXufT0n1ruoUTjyoEDKJESyYWURNlBFctFBDCLFy9cCJMnQ/HilFrzJWXKmqVj/v4b2rVDpRBEchElUSIiqWGzQe/e8PfflKgTyOrVUKaMeejy1r0M7HFJd++J5BJKokRE0iJPHgBKljQXp/y9LvIzLXn5i6p0vHMrgwfDzz/DpUuODVNEMo+SKBGRdAoKgnW1X+EODnEn//DlwXrEvDueh1vFULYsbNjg6AhFJDMoiRIRyQB3zXidM3fVAcCdKMYziJ9pievZMJ58Ei5fdnCAIpLhlESJiGSEO+7Ab88aGDoUy2YDoDm/8BtVKHd4KQMGODg+EclwSqJERDKKuzuMGYNt2TIoVgwAf06xlGbc/ekgOj1+Q4U5RXIQJVEiIhntgQfgt9+gZcu4pkGMp+f8B6gaFMNvvzkwNhHJMEqiREQyQ5Ei8MMP8MEHRLu6A/AtTxB22oXGjWH7dgfHJyLppiRKRCSzuLjAyy/jumUT157vy9ZavQE4dw4aNYLFix0bnoikj5IoO2jZFxFJl2rVyDvtQ5Yus1G/vmm6cgWWtZrImMd38Ndfjg1PRNJGSZQdtOyLiGQEHx/45Rd47DF4kKV8YL3MwPl1mHL3RKoGW0yfDlFRjo5SROylJEpEJAvlyQPffgtTyk0GwINIJvIyb+96mNd6nKZqVdi1y8FBiohdlESJiGQxFxco9/t8wnsPjGtrxc/sIogif/xK3brw1VcODFBE7JKuJGrChAkA7N69m+jo6AwJSEQkV/DwwGvyeDO7vGhRAAI4wXKa8sb11+j6dCT9+kFkpIPjFJFkuaXnycHBwQC89tpr7Nu3jzx58lCxYkUqV65MpUqVePjhhzMiRhGRnKt5czN+17kzLFuGCxavMZbGrOTJSbM5duwOvvkG3NL1bS0imSFdV6IaN24MwPfff8+ff/7JunXr6Nu3L35+fixfvjxDAhQRyfGKFYMlS2DcOKz/sqW6bGIaPZg/35RDmDEDLMuxYYpIQhn6t03+/PmpXbs2tWvXzsjDiojkfC4uMGgQtoYNoVMnosLO0DfyE4iE9evNz4EDMGaMowMVkVjpnlgeHR3N7t27mTNnDm+++Sbt2rXLiLhERHKn2rVh507clvzEhEV3EhgY/9AH46PZv99xoYlIQqm6EvXPP//w+++/88cff8T9HDhwgMjISDw8PChfvjyVK1fOrFhFRHIHX1+oX5/mwOHDMHQoTB53lbVR9Znb6FnundSTdo/acNH91SIOZXcS9fTTT/P1119js9nImzcvV69epVWrVgwbNozKlStz99134+rqmpmxiojkOjYbvPkmBE3uR7VrO6l2ojcL2y/jgcqfMWFWYf67v0dEHMDuv2PmzZvHpEmTuHLlCsePH6d3794sXbqUrVu3Urp0aSVQIiKZJF+eGJq0yR+33ZbvmfV7MK/fv5rNmx0YmEguZ3cS9fLLL9O5c2e8vLzInz8/H374IevXr2flypVUrFiRJUuWZGacIiK5l4sL/l9PxFr0Azd8CgMQyL8sutqElQ2G89sOrRUj4gh2J1Fjx47F29s7QVv16tXZsmUL/fr1o0OHDjz55JOcPn06w4MUERGwtX4Yjz27iK7fCABXYhhyYyTX6zRm//Ijjg1OJBdK97REm81Gv3792LNnDxEREdx7770ZEZeIiCSlRAlcVy7nxrC3icJMo6gduQ6/B4NZ1nMBWjxCJOtk2L0dJUqUYP78+cyaNSujDuk0QkJCqFChAjVr1nR0KCIi4OqKx1uvc+WnNZzwKAVAIc6z4uM/efBBuHTJwfGJ5BIZfoNsq1atMvqQDterVy/27NnD1q1bHR2KiEicAi3r4ftPKFtLP84ymvIug1i5Eho0gB07HB2dSM6nKiMiItlY3hIFqXnwWzx+WkjBQuYrfdcuqF4dulfdyY8/WFouRiSTKIkSEcnubDYatszH2rVQvrxpup+1TA+twbU2HWha4wKbNjk2RJGcSEmUiEgOUaEChIbC5Hev863bU7gSwxPM5bMdwbzWaIOG+EQymJIoEZEcxMMDer+ah2JzPuRG/oIAlOEwSyMa8EuD0axbrdv3RDKKkigRkRzI9mg7PHaHElPvfgDciGbo1Te40ehB3upxTKUQRDKAkigRkZyqVClcVq/k+qDhRP/3dd+ElfSaHsRPL/3o4OBEsj8lUSIiOZmbG3nGjcBa8SuXfEoA4MdZ2kxvzclBHzg4OJHsTUmUiEgu4NakIT7/7GJbyUcAuEYeHhjfnKeegj//dHBwItmUkigRkdyicGGKb1zAm4VC6EUIe6jA7Nnmrr4PPkD1pERSSUmUiEguUqKkjSFHelJ+XDf8/ExbTAwMfSWCLQ++DhcvOjZAkWxESZSISC6TLx8MGgQHD8LgwaZtHIOpvWIM1+6tClu2ODZAkWxCSZSISC6VPz+88w681esUXZgJQN6TB4muex833n7XXKISkWQpiRIRyeWGfFCUvvfvZCN1AHCNicLjzcFYzZrDyZMOjk7EeSmJEhHJ5Tw8YObqMuycuIbx7q8Rgw0A2/JlxFSuAkuWODhCEeekJEpERHBxgZ793Km+ZDTNXZZxnOKm/cxpaNECBg6EGzccHKWIc1ESJSIicZo0gcFLH6BRgV38RMv4B95/n/BJ0xwXmIgTUhIlIiIJPPAA/LKjCG8G/0h/JnADdzZQl6CPX9SNeyI3URIlIiKJ3HEHbNpso9jY/jyQdxNPMpv9/7hx333w8cf/7aS79ySXUxIlIiJJ8vCAIUNg5u/V8K9VBoCoKOjZE5a9sx2CgmDHDscGKeJASqJERCRFZcvCunXQv7/Zzs9l7nitI/zxB1adOjBxotaMkVxJSZQdQkJCqFChAjVr1nR0KCIiDuHubtbX69YNfLnIeasAALbISHj5ZayHH4bTpx0bpEgWUxJlh169erFnzx62bt3q6FBERBzGZoOpU6HhkyW5j/W8y6vxj/38sxne+/VXB0YokrWURImIiN3c3eGrr2BrqAehnd6lGUsIo6h58MQJaNoUXnsNIiMdG6hIFlASJSIiqRYUBLNnw+PTmhHELpbyoHnAsmDsWGjQwKxwLJKDKYkSEZE0e+45qNu2GM1Zwqu8SyRuAFjbt8O5cw6OTiRzKYkSEZE0i50nVbmKC+/xKvXYwN+UZdqd4/gzf3VHhyeSqZREiYhIuhQtCtu3w4svwjZqEsQuXtzXn8qVYc6c/3aKjIS9ex0ap0hGUxIlIiLp5uYGH30ECxZAfv/8gI3ISOjYETp1gsjXR0DVqrh+8olqSkmOoSRKREQyhM0GbdvCn39Cu3bx7Ue+WY/r+LEQEYH7yy8T0Ls3nD3rsDhFMoqSKBERyVC+vjB3LnzyCeTPD9uowST6xj3uvWIFnrVrw+rVDoxSJP2URImISIZzdYUXXoCNG8G3iCcvM5GH+YEzFAbAduwYNGkCw4ebBflEsiElUSIikmkqVYKlS+Gee+AnHiaIXfxKY/NgTAyMHAmNG8ORI44NVCQNlESJiEimCg6GrVuhd2847R7AgyzjNUYThavZYd06s1NoqAOjFEk9JVEiIpLpvL1h8mT49dcbFC5iMZbXaMAajriUNjuUKQPlyzs0RpHUUhIlIiJZpkYNizlzDlOxYgwbqUeVmFBmu3Xmy1ZfY3l4Ojo8kVRREiUiIlkqICCKX3+9Qd26cJECPBU1k2fevoc6deD33//bafduUwpdNaXEiSmJEhGRLOfjA0uWwPPPx7dt2QL16sH8L69Dhw6mBHr79nD+vOMCFUmBkigREXEIHx+YNg2+/hoKFDBtV67At88sMleiAObPN5PO1693VJgiyVISJSIiDtWxIxw7Zi4+AXxLB9qygEtuBU3DkSPQoAGMGgXR0Y4LVOQWSqJERMTh8uY1V6RCQszyMd/TlopRu1hDfbNDTAwMGwZNm5qMS8QJKIkSERGnYLNBz54wZ44pifAvgTThV4Yzghjbf7+uVq2CKlVg0SKHxioCSqJERMTJtG8P//4Ld98N0bgxkuE0slZywbuk2eHcOXj0UTh40LGBSq6nJEpERJyOjw98/71ZEQZgLQ2483IomwPamobhw+GOOxwWnwgoiRIRESdVvjz8+qupdO7iAucoTJ3j3zGl0TxihryWcGfLUk0pyXJKokRExKn17g0//giengA2+qx6jDr3ufLXXzft9M478NRTcPGig6KU3EhJlIiIOL0WLcyde7G2boWqVWHmTLA2bIQ33zS391Wtaqp2imQBJVEiIpItPPusuSnvrrvM9pUr0LUrjB9yFitfftN48CDcdx+8+64piyCSiZREiYhIttG6NezcCd27x7cNXvsw9b1DuVyprmmIioLBg6F5czh50jGBSq6gJEpERLKV/Pnhs89MPSlfX9O2/lgZiu5dzY7mr5mCUwDLlpmaUkuWOC5YydGURImISLb0xBOwa5cZvQMIj3an+pLRfNFlOVbx4qbx9GkzoWrgQLhxw3HBSo6Uq5KoCxcuUKNGDYKDg6lUqRLTp093dEgiIpIOpUvDypXQp098W+fPm/B05d+IbPZwfOPPP5thPpEMlKuSKG9vb9asWUNoaCibN29mzJgxnD171tFhiYhIOri7w8SJ0K9ffNvspX6UP7CIY69ONGN+c+aYBfpEMlCuSqJcXV3J+98/ooiICCzLwlJxNhGRbM/FxSRS8+dDvnym7e9/bNw9pR9hmw9B5coJn/Dvv3D5claHKTmMUyVRa9asoXXr1gQEBGCz2Vi4cGGifUJCQihTpgxeXl7Url2bLamsB3LhwgWCgoIoWbIkr776Kn5+fhkUvYiIONqjj8K2bVCunNm+fh3urF6ArVtv2unGDWjXDqpVg+3bHRKn5AxOlURdvXqVoKAgQm6uqHaTOXPmMGDAAIYPH86OHTsICgqiWbNmnDp1Km6f2PlOt/4cP34cgAIFCrBr1y4OHjzI7NmzCQsLy5LXJiIiWePee83dey7//Ya7ehXq1zfFOi0LeOstk2n99RfUrQsffKCaUpImNstJx7NsNhsLFiygbdu2cW21a9emZs2aTJkyBYCYmBgCAwPp06cPQ4YMSfU5evbsSZMmTXj88ceTfDwiIoKIiIi47UuXLhEYGEhYWBg+Pj6pPp+9IiIiOHLkCKVKlcLTrHMgkmnU3yQrZWV/27PHxnPPubNzZ/z1gs6do5g6aD8eXTrjctNVqOiHHiJy+nQoWjRTY5KsFxERwZ49e6hXrx4XL17M0N/fbhl2pEx248YNtm/fztChQ+PaXFxcaNq0KRs3brTrGGFhYeTNmxdvb28uXrzImjVreOmll5Ldf+zYsbz11luJ2o8ePUr+/PlT/yLsFBkZyenTp7HZbLi7u2faeURA/U2yVlb2t3z5YOZMG++958esWYUAmDXLjUOHSvHuhFncPfMDCn32GQCuS5di1ajBiXHjuFavXqbGJVkrMjKSM2fOZMqxs00SdebMGaKjo/H390/Q7u/vz759++w6xuHDh+nRo0fchPI+ffpQ+dbJhjcZOnQoAwYMiNuOvRIVGBiY6VeiLMsiMDBQVwYk06m/SVZyRH+bOhXq17/B8897ALBmTX46P3cvP/44hfxt2+L+7LPYTp3C7fRpSj77LNEDBhA1fLi57U+yvYiICC5n0k0E2SaJygi1atUiNDTU7v09PT2T/Efu5eWFl5dXBkaWmIeHB56enpl+HhFQf5Os5Yj+9txzEBAA3brBqVOwf78LDRp48e23D1P/t9+gc2dYuhSbZeH2/vu4rV8Pq1eDh0eWxSiZxyOTPkenmlieEj8/P1xdXRNNBA8LC6NYsWIOikpERLKLli1NXhQQYLZPnoRWrWDHMX9YvBjGjwe3/64tNGigBEpuK9skUR4eHlSvXp0VK1bEtcXExLBixQrq1q3rwMhERCS7uPdeWL/e3K0HplRUzZrQopULBx8bCBs2QMeO8Pbbjg1UsgWnSqKuXLlCaGho3JDbwYMHCQ0N5ciRIwAMGDCA6dOnM3PmTPbu3ctLL73E1atX6datW6bGFRISQoUKFahZs2amnkdERDJfmTJmTeLY+eMxMWa7YUPYGFUTvv468Xyo2bPNQn0iN3GqJGrbtm1UrVqVqlWrAiZpqlq1KsOGDQOgQ4cOvPfeewwbNozg4GBCQ0NZsmRJosnmGa1Xr17s2bOHrQmqtYmISHaVNy8sXQrDhkGRIqbt6FGTWM2de8vOoaFmMlXt2jBlyn/FpkScuE6UM7p06RK+vr4ZXmfiVuHh4Rw+fJjSpUtroq9kOvU3yUrO2N/CwqBxY9i712wXKADvvQfdu4PNhhnemzMn/glt2sD//geFCzsiXEml8PBwdu/eTY0aNTL897dTXYkSERHJav7+sHVr/ITzCxfM3Xz9+/+3w8yZN20AixZBUJCZpS65mpIoERHJ9fLlg507oWnT+LZJk8wNe5aHJ0yYAD/+CLHrrR47Zi5fDRsGUVGOCVocTkmUiIgIZsWXZctg+vT4tkGDoGtXs2YxrVqZyeVNmpgHLQtGjYJGjeC/G6Akd1ESZQfdnScikns89xwMHx6/PWsWVKsGf/+NGfNbuhTGjAFXV7PD+vXw6KOacJ4LKYmyg+7OExHJXUaMgC+/jK+9uXu3uXNv2zZM8jR0KKxdC6VLm50++ui/WeiSmyiJEhERScJTT8GaNVC2rNk+dQruv9/cuQdA3bqm/MF330GtWgmfrKtSuYKSKBERkWTUrWuuPsVWOI+IgFdfhRdfhOvXMfUQWrdO+KSoKGje3Kx8rGQqR1MSJSIikoKCBc00qCefjG+bOtXMLz92LIknvP22ecKLL0L79nD+fJbFKllLSZSIiMhteHnBV1/BuHHg6WnaNm2CqlXhiy9uuuBkWXDuXPwT58+H4GAz+VxyHCVRIiIidho0yORDsUvFnD4NnTvDCy+YxYyx2UyBqQULzCUsMOUPGjQw5RCiox0Wu2Q8JVF2UIkDERGJVb26WSKmVav4tunToVKl/8ogALRta2pKxU6miokxhTmbNk1mDFCyIyVRdlCJAxERuVnhwqaA+QcfmKE+MBecmjSBQ4f+2ykwEH791dRLcPnv1+2qVVClilk6RrI9JVEiIiJp9PLLpoZUxYpm+8gRc0df3BQoNzdTuXPlSihZ0rSdO2cmnku2pyRKREQkHcqWhRUr4J57zPbJk2YK1OTJN+3UoIGpKdW2rbkSFVdsSrIzJVEiIiLp5O9vrj41aGC2Y2Kgb18zET1ufeLChU1hzpUr48cAYx04oJpS2ZCSKBERkQxQuLBZwLhXr/i28eOhcWO4ePG/BpsNChVK+MS9e00ZhE6dbtpRsgMlUSIiIhnEw8MM440bFz+XfN06s4DxP/8k8YSoKJM8XbsGc+aYZGrTpqwMWdJBSZSIiEgGstnMMN4vv4CPj2n75x+oUQO2b79lZzc3eP118PU124cOmQX63nnHjAmKU1MSZQfViRIRkdRq2tRchYotzHn+vKkttWbNLTu2b29qStWrZ7ajo2HoUHjoIThxIktjltRREmUH1YkSEZG0qFwZduyAOnXMdlgYNGoEL70EV6/etGPp0rB6tbkqZbOZthUrICgIFi/O6rDFTkqiREREMlHJkrBwYfyde5YFn3xiLjQlWJvYzc0sXrxiBQQEmLbTp6FlSxgzJqvDFjsoiRIREclk/v6mePmQIfFtGzZAw4Zw/PgtOzdubIb3Hn7YbNtscN99WRar2E9JlIiISBZwdYWxY83wXtGipu33382de3EVzmP5+ZmlYT78EEaONNmWOB0lUSIiIlmoalUz4bxMGbMdO09q0qRbbsiz2UzFzjfeSHiAmBhT8fzy5SyKWJKjJEpERCSL3X03bN4MsTd9R0VBv37Qrt0tE86T8v778Oqr5hJWopoJkpWURImIiDhA0aLmitTzz8e3LVpkEqy9e5N50qVLppInwF9/mdWOP/hANaUcREmUiIiIg3h4wLRp8MMPkCePaTtxwkyB2rgxiSf4+CS8hBUZCa+8YgpQnTqVZXGLoSRKRETEwR5+2FyVKlvWbJ8+bYp1TpmSxEWmO+80Ow8eHN+2ZImpKbV8eZbFLEqi7KKK5SIiktmqVYO1a+MLl1+7Bn36QI8epoh5Ah4eZmmYpUtN/QSAkydN8akhQ8wVKsl0SqLsoIrlIiKSFQICzJp7zz0X3/bZZ9CxI0REJPGEBx80NaWaNTPblmXmTMXOm5JMpSRKRETEieTPD9OnwzffmCLmAPPmQZMmphxCIv7+8PPPpuyBuzuUKwf9+2dlyLmWkigREREn1KFDwgnnGzZA9erJ3Lnn4mImmG/YAN9+azKxm1lWpsebGymJEhERcVLNm5vlYgoVMtvHjplinR9/nMwTatQwE8xvdvCgKYUQGpqZoeZKSqJEREScWJ06sHu3SZ7AzI3q2RNGj7bjAlNkJDz5pCmLULs2TJ6sq1IZSEmUiIiIkytWDFavhvbt49veeAO6dYOLF1N44tmzcOOG+f8bN8wyMo88AmfOZGq8uYWSKBERkWzA29tMd7p5Kb2ZM+Hee1NY/aVYMTNP6uWX49t++MEM+a1alZnh5gpKokRERLKRkSNN8pQ3r9k+eRIaNDAjdkny9DRLw/z0E/j5mbbjx83tfm++aRbukzRREiUiIpKN2GzQubOZJx4YaNquXTNzpz77LIUntmxpako1aWK2LQvefhsaNYLDhzM56pxJSZSIiEg2dPfdJie69974tueeM4U5r11L5kkBAabK+dix4Opq2tavT2HFY0mJkigREZFsqmBB2LkzYYXzOXPMWnxXriTzJFdXszTMunVQpgwMGGBqKUiqKYmyg9bOExERZ+XlZSqcf/FFfIXzlSvhvvtMiahk1aljxgTHjEnYbllw6FAmRZuzKImyg9bOExERZ/f002ZkrkABs/3bb1CzpinWmSxfXzPx/GaffALly5v/qqZUipREiYiI5BC1apmKBnffbbbPnjVrFE+YYOcB9uwx5RDCw+Gll0xhqvPnMy3e7E5JlIiISA5Svjxs2QItWpjtmBgz7en1100B8xSVLQvPPx+/PX++qSm1bl2mxZudKYkSERHJYQoUMDU1X389vm3MGLOAcYoXlry8zNIwCxfGL9h39Cg0bGgKVEVHZ2LU2Y+SKBERkRzI1RVGjTJ1Nm020/b773DXXea/KXrkEVM/oWFDsx0TA8OHwwMPwL//Zmrc2YmSKBERkRzKZjNTnH74Ib7t3Dm4/36zFl+KSpaEFSvMFSiX/9KF1avN8N6mTZkWc3aiJEpERCSHa9UKtm2DO+4w25cuwUMPmSlPKXJ1NUvDrF4dXx7d2zthhc9cTEmUiIhILlC9uhnGi51wfuMGdOgAn35qx5Pvv9/UlGrfHmbPjq+jkMspiRIREckl8uWD7783yROYeeLPP2/mTt22JFShQvDtt1CvXsL2Y8fg889zZU0pJVEiIiK5iLs7fPWVmSsVa9gweOwxuHo1lQeLjoZnnoFu3aBTJ7h4MUNjdXZKokRERHIZV1dz19748fFtCxaYob5U5UG//GLWmAGzaF9wcK6adK4kSkREJJcaONAkT/nyme21a83Nd//8Y+cBWrY0Q3y+vmb70CEzf+qdd0xZhBxOSZSIiEgu1rYt/PhjfCJ1+LBZvHjZMjsP0L69qSkVO1cqOhqGDjW3/504kRkhOw0lUSIiIrlco0awe7dZMgbg5EmTA40YYed88dKlTRmE11+Pr+y5YoW5rLV4cSZF7XhKouwQEhJChQoVqFmzpqNDERERyRSlS8OaNWbB4lhvvQVdu9q5BrGbG7z9tkmeAgJM2+nTZsjvt98yI2SHUxJlh169erFnzx62bt3q6FBEREQyjZ+fmSs+dmx826xZUKcOnDlj50EaNzbDew8/bLZfeAGqVMnwWJ2BkigRERGJY7PBkCEwdaq5iw9g/36oWhU2b7bzIH5+sGgRzJhhbgPMoZREiYiISCI9epilYgoXNtv//mvWI54zx84D2GxmLDBv3oTtX3wBTz9t1p7J5pREiYiISJKCg83Vpzp1zHZEBHTsaNYkTlMFgwMH4KWXTLXPatUgm0+TURIlIiIiybrzTjPh/Nln49uGD4cuXSA8PJUH+/tvcHGJ//969eC997JtTSklUSIiIpIid3eYPj3hhPMvvzRXqMLCUnGg5s3NQsa1apntqCh49VVzB1+qDuQclESJiIjIbcVOOP/qK8iTx7Tt2mXmSf37byoOVLYsrFsHgwfHt/3yi6kptXRphsac2ZREiYiIiN2efBK2bIHAQLP955+mwvn+/ak4iLu7WRpm6VLw9zdtYWHQrJlJrm7cyPC4M4OSKBEREUmVSpXMPKmyZc32kSPmitTOnak80IMPmkKczZvHt82alcpVkB1HSZSIiIikWpkysGEDVK5stk+eNGsPz5+fygMVLQo//QTvvw+enmayVZEiGR1uplASJSIiImni7w+//hq/9vC1a6YEwqxZqTyQiwsMGAAHD8IDDyR87Nw5uHo1Q+LNaEqiREREJM38/MxyeR07mu2oKFP+4N137Vy8+GbFiyfctizo3BmqVzd39TkZJVEiIiKSLl5eZhSuV6/4tsGDzcWldJWAmjzZDPX9+SfUrg2TJqUhM8s8SqJEREQk3VxdTc4zenR828SJ8NRTptJ5mrRoYSqbg7ljr18/aNMmFashZy4lUSIiIpIhbDZ47TX47LP4xYu/+QZatUrjUnl3321mrw8YEN/244+mptTKlRkSc3ooiRIREZEM1b07LFwYX5RzxQpo1CiNRck9Pc2dez//HH/X3vHjZgL6m2+aSVgOoiRKREREMtzDD5vkqVAhs71zp1km5q+/0njAFi1MifSmTc22ZcHbb5sCVVeuZEjMqaUkSkRERDJF3bpmhZdSpcz2oUMm59mzJ40HLF7cLBHzzjvg5mba7rgD8uXLiHBTTUmUiIiIZJry5c20pnvuMdvHj0P9+rB2bRoP6OJibv1bu9ZUPP/oIzMZywGURImIiEimKlHCXJGqXt1snztnRud27EjHQevUMWvv+fgkbP/lF/j993Qc2H5KouwQEhJChQoVqFmzpqNDERERyZZii3Led5/ZvnoVmjSBTZsy8CSHD5uqn7VqwccfZ3pNKSVRdujVqxd79uxh69atjg5FREQk2/L1hWXL4hOpixdNIrV4cQad4O234cIFCA+Hnj3hscfMZa9MoiRKREREskyePGbErUkTs339Ojz6qCn/lG6TJ0Pv3vHbCxbgWbs2XukaN0yekigRERHJUvnymbJPjz1mtsPD4ZFHTG2pdPHyMonUwoVxtRVs//5LyRdeSOeBk6YkSkRERLKcpyfMnh2fSMXEmP+fOTMDDv7II6amVMOGANgyaW6UkigRERFxCA8PmDMHnnjCbMfEQNeupkB5upUsCStWEDlsGFYmlUBQEiUiIiIO4+oKX3+dcCrTwIGmFFS6LyC5uhI9dCj/TpuWzgMlTUmUiIiIOJSLC0yaBEOHxre9+y4MG5YxVQrCq1ZN/0GSoCRKREREHM5mgzFj4JNP4tveftusMeyslESJiIiI03jhBXNVKtbo0WapPGekJEpEREScSp8+JnmKNXQojBjhsHCSpSRKREREnM7QoTBoUPz2W2/Be+85Lp6kKIkSERERp2OzwbhxMGFCfNurr8L48Y6L6VZKokRERMRp9e9vkqdYgwaZouTOQEmUiIiIOLVx48xwXqx+/WDGDMfFE0tJlIiIiDg1m83UjHr9dbNtWfD88xmw1l46KYkSERGRbGHUKFMCASA6Gjp1gl9+cVw8SqJEREQkW7DZ4KOPoGNHsx0eDu3awbp1jolHSZSIiIhkGy4uMHMmNGtmtq9fh1atYMcOB8SS9acUERERSTsPD/j+e3joIbN96ZJJqvbvz9o4lESJiIhItuPpCd99B/fdZ7bPnDGJ1MmTWReDkigRERHJlvLlg59+gooVzfahQ/Dww3D5ctacX0mUiIiIZFu+vrB0KZQqZba3b4dHH4WIiMw/t5IoERERydYCAmDxYihY0GwvXw4dOkBUVOaeV0mUiIiIZHsVKsCPP0LevGb7++/h5Zcz95xKokRERCRHqFcPFi0CNzezPWUKfPaZa6adT0mUiIiI5BgPPABTp8Zv9+/vxo4dXplyLiVRIiIikqN07w79+5v/j4qyMWhQ8Uw5j5IoERERyXHGj4cHHzT/f+GCW6acQ0mUiIiI5DhubjBnDtx5Z0ymnSNXJlHXrl2jdOnSDBw40NGhiIiISCYpWBDmz4/Exyc6U46fK5Oo0aNHU6dOHUeHISIiIpnsnnss3n//eKYcO9clUQcOHGDfvn20aNHC0aGIiIhIFqhaNTxTjutUSdSaNWto3bo1AQEB2Gw2Fi5cmGifkJAQypQpg5eXF7Vr12bLli2pOsfAgQMZO3ZsBkUsIiIiuZVTJVFXr14lKCiIkJCQJB+fM2cOAwYMYPjw4ezYsYOgoCCaNWvGqVOn4vYJDg6mUqVKiX6OHz/O999/T7ly5ShXrlxWvSQRERHJoTLnnr80atGiRYrDbB988AHPP/883bp1A+CTTz7hp59+4n//+x9DhgwBIDQ0NNnnb9q0iW+++Ya5c+dy5coVIiMj8fHxYdiwYUnuHxERQcRNKxhevHgRgNOnTxMenjmXBmPPe+7cOfLmzYunp2emnUcE1N8ka6m/SVaLiIjg/PnzAFiWlbEHt5wUYC1YsCBuOyIiwnJ1dU3QZlmW1blzZ6tNmzapPv6MGTOsV155JcV9hg8fbgH60Y9+9KMf/egnB/wcPXo01flCSpzqSlRKzpw5Q3R0NP7+/gna/f392bdvX6acc+jQoQwYMCBuOyYmhnPnzlG4cGFq1arF1q1bU3x+zZo1U9wnuccvXbpEYGAgR48excfHJ+0vwEFu97qd+XzpOVZanmvvc+zZT/0t+51P/U39LSvPl95jpfb5ztDfIL7P7dmzh4CAgNvGkxrZJonKaF27dr3tPp6enokuNxcoUAAAV1fX234B3G6f2z3u4+OTLb9k7HlvnPV86TlWWp5r73PU35Kn/pbxz1F/S15u7W9peb4z9TeAEiVK4OKSsVPBnWpieUr8/PxwdXUlLCwsQXtYWBjFihXL8nh69eqV7n3sOUZ2lNWvKyPPl55jpeW59j5H/S156m8Z/xz1t+Tl1v6Wlufnhv5ms6yMnmWVMWw2GwsWLKBt27ZxbbVr16ZWrVpMnjwZMMNrpUqVonfv3nETy3OCS5cu4evry8WLF7PlX2qSvai/SVZSf5Oslpl9zqmG865cucJff/0Vt33w4EFCQ0MpVKgQpUqVYsCAAXTp0oUaNWpQq1YtJk6cyNWrV+Pu1sspPD09GT58uO5ckSyh/iZZSf1Nslpm9jmnuhK1atUqGjdunKi9S5cufP755wBMmTKF8ePHc/LkSYKDg5k0aRK1a9fO4khFREQkt3OqJEpEREQku8g2E8tFREREnImSKBEREZE0UBIlIiIikgZKokRERETSQElUNvPjjz9yzz33cPfdd/Ppp586OhzJBdq1a0fBggV5/PHHHR2K5HBHjx6lUaNGVKhQgSpVqjB37lxHhyQ52IULF6hRowbBwcFUqlSJ6dOnp/oYujsvG4mKiqJChQqsXLkSX19fqlevzoYNGyhcuLCjQ5McbNWqVVy+fJmZM2cyb948R4cjOdiJEycICwsjODiYkydPUr16dfbv30++fPkcHZrkQNHR0URERJA3b16uXr1KpUqV2LZtW6p+p+pKVDayZcsWKlasSIkSJcifPz8tWrRg6dKljg5LcrhGjRrh7e3t6DAkFyhevDjBwcEAFCtWDD8/P86dO+fYoCTHcnV1JW/evABERERgWRapva6kJCoLrVmzhtatWxMQEIDNZmPhwoWJ9gkJCaFMmTJ4eXlRu3ZttmzZEvfY8ePHKVGiRNx2iRIlOHbsWFaELtlUevucSGpkZH/bvn070dHRBAYGZnLUkl1lRH+7cOECQUFBlCxZkldffRU/P79UxaAkKgtdvXqVoKAgQkJCknx8zpw5DBgwgOHDh7Njxw6CgoJo1qwZp06dyuJIJadQn5OslFH97dy5c3Tu3Jlp06ZlRdiSTWVEfytQoAC7du3i4MGDzJ49m7CwsNQFYYlDANaCBQsStNWqVcvq1atX3HZ0dLQVEBBgjR071rIsy1q/fr3Vtm3buMf79etnffXVV1kSr2R/aelzsVauXGk99thjWRGm5BBp7W/h4eFW/fr1rVmzZmVVqJIDpOf7LdZLL71kzZ07N1Xn1ZUoJ3Hjxg22b99O06ZN49pcXFxo2rQpGzduBKBWrVr88ccfHDt2jCtXrrB48WKaNWvmqJAlm7Onz4lkFHv6m2VZdO3alSZNmvDMM884KlTJAezpb2FhYVy+fBmAixcvsmbNGu65555Uncct40KW9Dhz5gzR0dH4+/snaPf392ffvn0AuLm58f7779O4cWNiYmIYNGiQ7syTNLOnzwE0bdqUXbt2cfXqVUqWLMncuXOpW7duVocr2Zw9/W39+vXMmTOHKlWqxM1v+eKLL6hcuXJWhyvZnD397fDhw/To0SNuQnmfPn1S3deURGUzbdq0oU2bNo4OQ3KR5cuXOzoEySXuv/9+YmJiHB2G5BK1atUiNDQ0XcfQcJ6T8PPzw9XVNdGktrCwMIoVK+agqCQnU5+TrKT+Jlkpq/qbkign4eHhQfXq1VmxYkVcW0xMDCtWrNDQiWQK9TnJSupvkpWyqr9pOC8LXblyhb/++itu++DBg4SGhlKoUCFKlSrFgAED6NKlCzVq1KBWrVpMnDiRq1ev0q1bNwdGLdmZ+pxkJfU3yUpO0d9SdxOhpMfKlSstINFPly5d4vaZPHmyVapUKcvDw8OqVauWtWnTJscFLNme+pxkJfU3yUrO0N+0dp6IiIhIGmhOlIiIiEgaKIkSERERSQMlUSIiIiJpoCRKREREJA2URImIiIikgZIoERERkTRQEiUiIiKSBkqiRERERNJASZSICNCoUSP69+/v6DBEJBtRxXIREeDcuXO4u7vj7e3t6FBEJJtQEiUiIiKSBhrOExFBw3kiknpKokRERETSQEmUiIiISBooiRIRERFJAyVRIiIiImmgJEpEREQkDZREiYiIiKSBkigRERGRNFCxTREREZE00JUoERERkTRQEiUiIiKSBkqiRERERNJASZSIiIhIGiiJEhEREUkDJVEiIiIiaaAkSkRERCQNlESJiIiIpIGSKBEREZE0UBIlIiIikgZKokRERETSQEmUiIiISBr8Hy8IDQQOX18wAAAAAElFTkSuQmCC", + "image/png": "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", "text/plain": [ "
" ] @@ -233,21 +263,24 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "100%|██████████| 32/32 [01:22<00:00, 2.58s/it]\n" + " 0%| | 0/32 [00:00" ] @@ -379,14 +414,16 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "100%|██████████| 32/32 [01:21<00:00, 2.55s/it]\n" + " 0%| | 0/32 [00:00