From d845831b04e527ed4c085f2bcb536a64672a8d48 Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Tue, 19 Mar 2024 13:20:45 -0500
Subject: [PATCH 1/9] Initial Text Changes
---
docs/guides/artifacts/intro.md | 49 ++++++++++++++++++++--------------
1 file changed, 29 insertions(+), 20 deletions(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 1d9ef39d57..0ffd01ddc8 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -12,19 +12,19 @@ import { CTAButtons } from '@site/src/components/CTAButtons/CTAButtons.tsx';
-Use W&B Artifacts to track and version any serialized data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and trained model as output. In addition to logging hyper-parameters and metadata to a run, you can use an artifact to log the dataset used to train the model as input and the resulting model checkpoints as outputs. You will always be able answer the question “what version of my dataset was this model trained on”.
+Use W&B Artifacts to track and version any serialized data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and trained model as output. In addition to logging hyper-parameters and metadata to a run, you can use an artifact to log the dataset used to train the model as input and the resulting model checkpoints as outputs.
-In summary, with W&B Artifacts, you can:
-* [View where a model came from, including data it was trained on](./explore-and-traverse-an-artifact-graph.md).
-* [Version every dataset change or model checkpoint](./create-a-new-artifact-version.md).
-* [Easily reuse models and datasets across your team](./download-and-use-an-artifact.md).
+## Use cases
+You can use artifacts throughout your entire ML workflow as inputs and outputs of [runs](../runs/intro.md):
+This means that you can:
+* [View where a model came from, including trained data](./explore-and-traverse-an-artifact-graph.md).
+* [Version every dataset change or model checkpoint](./create-a-new-artifact-version.md).
+* [Easily reuse models and datasets across your team](./download-and-use-an-artifact.md).
-The diagram above demonstrates how you can use artifacts throughout your entire ML workflow; as inputs and outputs of [runs](../runs/intro.md).
-
-## How it works
+## Create an artifact
Create an artifact with four lines of code:
1. Create a [W&B run](../runs/intro.md).
@@ -33,7 +33,7 @@ Create an artifact with four lines of code:
4. Log your artifact to W&B.
-```python showLineNumbers
+```python
run = wandb.init(project="artifacts-example", job_type="add-dataset")
artifact = wandb.Artifact(name="my_data", type="dataset")
artifact.add_dir(local_path="./dataset.h5") # Add dataset directory to artifact
@@ -41,18 +41,27 @@ run.log_artifact(artifact) # Logs the artifact version "my_data:v0"
```
:::tip
-The preceding code snippet, and the [colab linked on this page](https://colab.research.google.com/github/wandb/examples/blob/master/colabs/wandb-artifacts/Artifacts_Quickstart_with_W&B.ipynb), show how to track files by uploading them to W&B. See the [track external files](./track-external-files.md) page for information on how to add references to files or directories that are stored in external object storage (for example, in an Amazon S3 bucket).
+See the [track external files](./track-external-files.md) page for information on how to add references to files or directories stored in external object storage, like an Amazon S3 bucket.
:::
-## How to get started
+## Download an artifact
+Indicate the artifact you want to use with the [`use_artifact`](../../ref/python/run.md#use_artifact) method, which returns a run object:
+
+```python
+artifact = run.use_artifact("my_data") #returns a run object using the "my_data" artifact
+```
+
+Then, use the returned object to download all contents of the artifact:
+
+```python
+datadir = artifact.download() #downloads the "my_data" artifact
+```
-Depending on your use case, explore the following resources to get started with W&B Artifacts:
+:::tip
+For alternate ways to download artifacts and to see additional parameters, see the guide on [downloading and using artifacts](./download-and-use-an-artifact.md)
+:::
-* If this is your first time using W&B Artifacts, we recommend you go through the [Artifacts Colab notebook](https://colab.research.google.com/github/wandb/examples/blob/master/colabs/wandb-artifacts/Artifacts_Quickstart_with_W&B.ipynb#scrollTo=fti9TCdjOfHT).
-* Read the [artifacts walkthrough](./artifacts-walkthrough.md) for a step-by-step outline of the W&B Python SDK commands you could use to create, track, and use a dataset artifact.
-* Explore this chapter to learn how to:
- * [Construct an artifact](./construct-an-artifact.md) or a [new artifact version](./create-a-new-artifact-version.md)
- * [Update an artifact](./update-an-artifact.md)
- * [Download and use an artifact](./download-and-use-an-artifact.md).
- * [Delete artifacts](./delete-artifacts.md).
-* Explore the [Python SDK Artifact APIs](../../ref/python/artifact.md) and [Artifact CLI Reference Guide](../../ref/cli/wandb-artifact/README.md).
+## Next steps
+* Read the [artifacts walkthrough](./artifacts-walkthrough.md) for a detailed step-by-step outline of how to create, track, and use an artifact.
+* Learn how to [version](./create-a-new-artifact-version.md), [update](./update-an-artifact.md), or [delete](./delete-artifacts.md) artifacts.
+* Explore the [Python SDK](../../ref/python/artifact.md) and [CLI](../../ref/cli/wandb-artifact/README.md) reference guides.
From 6c6e92bb7d6b3123a1f5f0149dea64e83ae6fb53 Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Tue, 26 Mar 2024 12:20:21 -0500
Subject: [PATCH 2/9] Update docs/guides/artifacts/intro.md
Co-authored-by: Noa <40642416+noaleetz@users.noreply.github.com>
---
docs/guides/artifacts/intro.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 0ffd01ddc8..157e75facc 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -20,7 +20,7 @@ You can use artifacts throughout your entire ML workflow as inputs and outputs o
This means that you can:
-* [View where a model came from, including trained data](./explore-and-traverse-an-artifact-graph.md).
+* [Visualize a model's or dataset's lineage and dependencies](./explore-and-traverse-an-artifact-graph.md).
* [Version every dataset change or model checkpoint](./create-a-new-artifact-version.md).
* [Easily reuse models and datasets across your team](./download-and-use-an-artifact.md).
From 80b0a4b4aa7353a52e9ecfe13d27e1c7ad4a368d Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Fri, 29 Mar 2024 15:23:02 -0500
Subject: [PATCH 3/9] Changes to use cases section, addresses DOCS-533,
DOCS-539
---
docs/guides/artifacts/intro.md | 21 +++++++++++++++------
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 157e75facc..70b9d14a16 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -15,14 +15,16 @@ import { CTAButtons } from '@site/src/components/CTAButtons/CTAButtons.tsx';
Use W&B Artifacts to track and version any serialized data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and trained model as output. In addition to logging hyper-parameters and metadata to a run, you can use an artifact to log the dataset used to train the model as input and the resulting model checkpoints as outputs.
## Use cases
-You can use artifacts throughout your entire ML workflow as inputs and outputs of [runs](../runs/intro.md):
+You can use artifacts throughout your entire ML workflow as inputs and outputs of [runs](../runs/intro.md). You can use datasets, models, or even other artifacts as inputs for processing.
-This means that you can:
-* [Visualize a model's or dataset's lineage and dependencies](./explore-and-traverse-an-artifact-graph.md).
-* [Version every dataset change or model checkpoint](./create-a-new-artifact-version.md).
-* [Easily reuse models and datasets across your team](./download-and-use-an-artifact.md).
+Some of the use cases of artifacts including:
+* **Training models**: Input a dataset to track the training of your model and monitor progress with the W&B dashboard.
+* **Pre-processing datasets**: Clean and transform a dataset and track its changes over time.
+* **Evaluating models**: Track a model's performance with your data using W&B [Tables](../tables/tables-walkthrough.md)
+* **Optimizing models**: Optimize models and track performance over time.
+
## Create an artifact
@@ -57,11 +59,18 @@ Then, use the returned object to download all contents of the artifact:
datadir = artifact.download() #downloads the "my_data" artifact
```
+Optionally, you can retrieve the artifact's URL for reference:
+
+```python
+print("reference file:", artifact.manifest.entries[ref].ref) #prints a URL pointing to the "my_data" artifact for future reference
+```
+
:::tip
For alternate ways to download artifacts and to see additional parameters, see the guide on [downloading and using artifacts](./download-and-use-an-artifact.md)
:::
## Next steps
-* Read the [artifacts walkthrough](./artifacts-walkthrough.md) for a detailed step-by-step outline of how to create, track, and use an artifact.
* Learn how to [version](./create-a-new-artifact-version.md), [update](./update-an-artifact.md), or [delete](./delete-artifacts.md) artifacts.
+* Learn about [artifact automation](./project-scoped-automations.md).
+* Learn about the [model registry](../model_registry/intro.md), a space that houses trained models.
* Explore the [Python SDK](../../ref/python/artifact.md) and [CLI](../../ref/cli/wandb-artifact/README.md) reference guides.
From 22b451b066db071a57dd963028e4d24b7d6543ee Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Wed, 3 Apr 2024 14:11:39 -0500
Subject: [PATCH 4/9] Update docs/guides/artifacts/intro.md
Co-authored-by: Noa <40642416+noaleetz@users.noreply.github.com>
---
docs/guides/artifacts/intro.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 70b9d14a16..45d4f7fb13 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -47,7 +47,7 @@ See the [track external files](./track-external-files.md) page for information o
:::
## Download an artifact
-Indicate the artifact you want to use with the [`use_artifact`](../../ref/python/run.md#use_artifact) method, which returns a run object:
+Indicate the artifact you want to mark as input to your run with the [`use_artifact`](../../ref/python/run.md#use_artifact) method, which returns an artifact object:
```python
artifact = run.use_artifact("my_data") #returns a run object using the "my_data" artifact
From 23e4be7a530555ea42fd3bae21647d65903c95ed Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Tue, 9 Apr 2024 11:29:23 -0500
Subject: [PATCH 5/9] redos intro section again.
---
docs/guides/artifacts/intro.md | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 45d4f7fb13..97a86fbfcc 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -19,11 +19,12 @@ You can use artifacts throughout your entire ML workflow as inputs and outputs o
-Some of the use cases of artifacts including:
-* **Training models**: Input a dataset to track the training of your model and monitor progress with the W&B dashboard.
-* **Pre-processing datasets**: Clean and transform a dataset and track its changes over time.
-* **Evaluating models**: Track a model's performance with your data using W&B [Tables](../tables/tables-walkthrough.md)
-* **Optimizing models**: Optimize models and track performance over time.
+| Use Case | Input | Output |
+|------------------------|-----------------------------|------------------------------|
+| Model Training | Dataset (training data) | Trained [Model](../models.md) |
+| Dataset Pre-Processing | Dataset (raw data) | Dataset (pre-processed data) |
+| Model Evaluation | Model + Dataset (test data) | [Table](../tables/intro.md) |
+| Model Optimization | Model | Optimized Model |
## Create an artifact
From d1af48b73e3a44693f57b6a5fa54058964a19b53 Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Wed, 10 Apr 2024 13:40:53 -0500
Subject: [PATCH 6/9] Addressing slack feedback.
---
docs/guides/artifacts/intro.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 97a86fbfcc..31943cf9c7 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -12,7 +12,7 @@ import { CTAButtons } from '@site/src/components/CTAButtons/CTAButtons.tsx';
-Use W&B Artifacts to track and version any serialized data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and trained model as output. In addition to logging hyper-parameters and metadata to a run, you can use an artifact to log the dataset used to train the model as input and the resulting model checkpoints as outputs.
+Use W&B Artifacts to track and version data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and trained model as output. In addition to logging hyperparameters and metadata to a run, you can use an artifact to log the dataset used to train the model as input and the resulting model checkpoints as outputs.
## Use cases
You can use artifacts throughout your entire ML workflow as inputs and outputs of [runs](../runs/intro.md). You can use datasets, models, or even other artifacts as inputs for processing.
From 996de42302dca54b4bfd8aab4333c5f3e89a541b Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Wed, 10 Apr 2024 16:50:49 -0500
Subject: [PATCH 7/9] Additional edits
---
docs/guides/artifacts/intro.md | 14 ++++----------
1 file changed, 4 insertions(+), 10 deletions(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 31943cf9c7..5842e81001 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -51,27 +51,21 @@ See the [track external files](./track-external-files.md) page for information o
Indicate the artifact you want to mark as input to your run with the [`use_artifact`](../../ref/python/run.md#use_artifact) method, which returns an artifact object:
```python
-artifact = run.use_artifact("my_data") #returns a run object using the "my_data" artifact
+artifact = run.use_artifact("my_data:latest") #returns a run object using the "my_data" artifact
```
Then, use the returned object to download all contents of the artifact:
```python
-datadir = artifact.download() #downloads the "my_data" artifact
-```
-
-Optionally, you can retrieve the artifact's URL for reference:
-
-```python
-print("reference file:", artifact.manifest.entries[ref].ref) #prints a URL pointing to the "my_data" artifact for future reference
+datadir = artifact.download() #downloads the full "my_data" artifact to the default directory.
```
:::tip
-For alternate ways to download artifacts and to see additional parameters, see the guide on [downloading and using artifacts](./download-and-use-an-artifact.md)
+You can pass a custom path into the `root` [parameter](../../ref/python/artifact.md) to download an artifact to a specific directory. For alternate ways to download artifacts and to see additional parameters, see the guide on [downloading and using artifacts](./download-and-use-an-artifact.md)
:::
## Next steps
* Learn how to [version](./create-a-new-artifact-version.md), [update](./update-an-artifact.md), or [delete](./delete-artifacts.md) artifacts.
-* Learn about [artifact automation](./project-scoped-automations.md).
+* Learn how to trigger downstream workflows in response to changes to your artifacts with [artifact automation](./project-scoped-automations.md).
* Learn about the [model registry](../model_registry/intro.md), a space that houses trained models.
* Explore the [Python SDK](../../ref/python/artifact.md) and [CLI](../../ref/cli/wandb-artifact/README.md) reference guides.
From 525202eb6cd27e296f5c2f5a296d7baa966cde62 Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Wed, 24 Apr 2024 17:52:08 -0500
Subject: [PATCH 8/9] Addresses language changes
---
docs/guides/artifacts/intro.md | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/docs/guides/artifacts/intro.md b/docs/guides/artifacts/intro.md
index 5842e81001..d7f516c205 100644
--- a/docs/guides/artifacts/intro.md
+++ b/docs/guides/artifacts/intro.md
@@ -12,7 +12,7 @@ import { CTAButtons } from '@site/src/components/CTAButtons/CTAButtons.tsx';
-Use W&B Artifacts to track and version data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and trained model as output. In addition to logging hyperparameters and metadata to a run, you can use an artifact to log the dataset used to train the model as input and the resulting model checkpoints as outputs.
+Use W&B Artifacts to track and version data as the inputs and outputs of your [W&B Runs](../runs/intro.md). For example, a model training run might take in a dataset as input and produce a trained model as output. In addition to logging hyperparameters, metadata and metrics to a run, you can use an artifact to log, track and version the dataset used to train the model as input and another artifact for the resulting model checkpoints as outputs.
## Use cases
You can use artifacts throughout your entire ML workflow as inputs and outputs of [runs](../runs/intro.md). You can use datasets, models, or even other artifacts as inputs for processing.
@@ -21,25 +21,25 @@ You can use artifacts throughout your entire ML workflow as inputs and outputs o
| Use Case | Input | Output |
|------------------------|-----------------------------|------------------------------|
-| Model Training | Dataset (training data) | Trained [Model](../models.md) |
+| Model Training | Dataset (training and validation data) | Trained [Model](../models.md) |
| Dataset Pre-Processing | Dataset (raw data) | Dataset (pre-processed data) |
-| Model Evaluation | Model + Dataset (test data) | [Table](../tables/intro.md) |
+| Model Evaluation | Model + Dataset (test data) | [W&B Table](../tables/intro.md) |
| Model Optimization | Model | Optimized Model |
## Create an artifact
Create an artifact with four lines of code:
-1. Create a [W&B run](../runs/intro.md).
+1. Create a [W&B Run](../runs/intro.md).
2. Create an artifact object with the [`wandb.Artifact`](../../ref/python/artifact.md) API.
-3. Add one or more files, such as a model file or dataset, to your artifact object.
+3. Add one or more files, such as a model file or dataset, to your artifact object. In this example, you'll add a single file.
4. Log your artifact to W&B.
```python
run = wandb.init(project="artifacts-example", job_type="add-dataset")
artifact = wandb.Artifact(name="my_data", type="dataset")
-artifact.add_dir(local_path="./dataset.h5") # Add dataset directory to artifact
+artifact.add_file(local_path="./dataset.h5") # Add dataset file to artifact
run.log_artifact(artifact) # Logs the artifact version "my_data:v0"
```
From bc7a5d92c8dfabdb766b7dd3ea796a26d51bdff2 Mon Sep 17 00:00:00 2001
From: katjacksonWB <144173614+katjacksonWB@users.noreply.github.com>
Date: Mon, 13 May 2024 16:54:46 -0500
Subject: [PATCH 9/9] Adds new image to the docs.
---
.../artifacts/artifacts_landing_page2.png | Bin 66853 -> 49079 bytes
1 file changed, 0 insertions(+), 0 deletions(-)
diff --git a/static/images/artifacts/artifacts_landing_page2.png b/static/images/artifacts/artifacts_landing_page2.png
index ea0f1bcc6cb4d161bfc156b853d552679a5b10bc..848a11ac0cc306aefa5840dbcbd6dd92291a4c66 100644
GIT binary patch
literal 49079
zcmeFY_ghn46E?a5c~DRhq(r)af`XJskQzXxDM%4%(u*QpBtWPEm0lu(^d=xpdJ7#?
z0#--7lhvRi2VQ25P_L`Y>&#aky!gaLN&e5{bLJ)NBp1Sfw
z2%_nPAd217|9~r3Zb1(cEo0~aaWAFADfO1n6gz#pgVZa=sU
zK^17aLn~?sYSp}_eESiSV)e*3{)2}<&*AYRsV3#)=}Eyq>CB6_WE_XWDZ98;FTDQw
zFWUvvGF3jx>sEd@aaGZ)`z*(xpdHs&@CuCtRJ!cw(ceTOJ%azz(}t4_YP<%gx{v)#$Ja{7YDBxI296b`2?Wv+=k97%_wjgq
zAMp%$P4d5G`(3x_**$u8l*~R;z7Hrx_f^7=gc}+^5;?tAr-ckpFZ@-R?B!His9ZgU
z3!ak47Q9wi_+@E*hHptd=k)Q6>A(k;3Ngd~|D>`bWbu2c~|s
zhND5r&xqx}9fryi8#$6(B4d!l1fAsj`6XQ*iU-xq!v{g4^Fz8}`<{`>d))rZx!x;*HY`(rhZ
zY}gHb*`H*i3>%`F^pVOq~?a(3JZf(JLPDe|N#%d*qe!NBq_%jHY^MqaWXHN|rtkmZmYK8_;*g
z#vV&M!~b_Y4qLFDo+>BqmF5eC_|OxvI;(TT|=gCIg#=NenWm5C@Y3{&H2fG<&&$|DWJ1F$@d{ke!&RQ7Qr{#Ym=EC
zZ!^p7U(N9V`?%3{F)v0wE-&wVzFme)N{j^UZ(_}t=rT@w(>smX$an9A3~VdbE1z^q
z9=FJw$g`c8H~T!^~V{>l*qS&A$7`ks6$MEq|1)WZ7N_nP&BzOJxE)q&9`
z4k(biUVlLv(vgc@9NQmQlc?}QS(k%{wL3D^ZGhO~v|v@-G7#j?K=))9FZYJ|bErp!
zQ+t=&CCnaU`&cjZs}|6|hh5x1)YE#?Lj`3y
z_FiuB?Y6C5e-Y36^*XR8#GC=v4YFKZV&M)T@6P-*WRu7X2papYme(o4y8yXkXZtJ8
ztzd^B$bHnE1A>{yo+gcUnb}VtzY3pWDVql-ai8WqS2ZhGle=;g4A&=cbvK!1P?aOI
zuAOqW#?c_Wi~^!nR)N}dB3xoK34XHd`C9&SzOk8}%lGWJi{71q0()6hh)GN}zSd48be2Ggmqp?i0a3MV6=lmtulPJ=i-w
zOjuvX>0_$VC-V`^#%a!TvXEF2*iV7)8SYi2Yvb=-^xeI(EZU*OWzY0oTV`p`NgFJ4
zS(N4envW0fif7*L>|TebDubiYTPi3=9T*jiHn#l|>hPY;fSRVlQ~w|K0^WU_ed9V?&b4nK*%D6CEm
zIS}gXGW~Z#qZPX~fz3a&eqk|I?}%2ul&q(%a5>u9G!vjv0BMI?Q!aDNnPC;WG+r)}
zj`*C!D+B2mVj{qxLE8iU7e|v6zru-g{n2wvH;q4^hXQBfJ%oTvIi-<;;zk?J(NT|S
zpym7I_gAOINyN2^GJU0hLbTiWfhp9^2F#gg$r%G(=?L+%Lzce+9H|0y?sNWVL6!>b*ya7konnou!Gi2sLT)jPu=H;iZ8b
ztc)ok*acb~LWt3W*g^qWt}8DPV>-RAM-3}YsT^-{EjWVls|M&H*y;VLS3M=jXK(Wt1@u6!H$3OSvs?Y_^z|
zD3)~Z<$1`G$eEIIABR>IQ)o){z*8~Vhz%w#JKt260<5<
z3aAHF(Ykk0TOl3xK>Qj{M{#Dra(G8E6-1>C_ILTB^{MG@ulpIjifB0Tk!&9_6Fgy2
zQ4;-GT3p<+6Q>>Vi6StF9GM6g2}X6+VZXe+y4Bm8qgFL%dUys>LbXdfkMqYyy311o
zW9lJ$?a1ql&uDSNA;?S>ARVZgneIa7$Ufqh7JCHmKPSY!ZnAee16f}9d-GLQk+pG@
z*e7e5=j%EBd*b8yb3)e_Kds*R2Wp;_*ji;i7k=9njpG3)(#t#
z{j)2nBI|Sgv~y5f8yVvdL{wL>(jG?F#my(DLLq1~U#!cItngQ=z?v2tN6Q-&$0_;*}V?%ii1XhTp+B=Q@RpwEpi1{+Nj83+?z
z0u$(Q?CI8?_a#(_F80T3_&b$
z!*Lq-xHI#JSA*7X~9F~gnBge@T3vw5EI9ngeawW^oyI)dkQkiir
zj9(nm`A;PZH3AgS)+;iqyEb$b=H&0LK7b&PCuy=@e)kN-*8y|+i#_IuXgxz>i-OPA
zGfJp{4>xd$rO{J6Y55eil}3J8lnk0WYPzuD50ubAEOqJGxUe56mnK=L#|Fr+t!dRj
zKDWbD8;PQVwl~{>OwPvnljw7ST&_TY=48`vV41wCD*n)}_XF=48ff5N`}!<s@0aWnJwk;lhTDA9>ioD9_eL%pg
z8Rd8es9cQvQk{sNesO`-t
zh7efl=CE`vQUL@oIzS|{FMj`i^io%{UYl$RrWv;aQc@zi+$Oz;yc#1=G$b?<7ffsHz9<>>})wxtVZaNwMsfOCGS%SdPRK$Z$>cuyV{@jI&`
z>P>1&UodSjV$*eRfYeogt4E=FaC^`a$2ASj(I>J(Yq*AhR)pQKPiVu6U5o8PbQE^z@3
zki{tBF;~vZ6aRqo-v<;BOqfK^_3vazY^`gnKX~abGI%}Ulz7PbN7~7j4-&K7!Rx_n
ztFfMzEMZ`9qga57uxj4llU|d_R|Nzfo7-(B+^~1T8I3jIv1blG1%!atm9fh3Lgw4nNVfBznXo`tS5h*(q)&V9e85^o~cKxn%z`
z-*|QPm}FG@Q&tbMTmtcx7)8JMK`ftIUj*~RDM-nkmm13I6GL=%7;XG8_fnbPwN5+8
z^!uadF8fyTMSt{CntQi~38q>4MuOFHg!y&!xlu|=AOTAq%?HU2RP{$Rd3__Q*Ic12
znOI6lXDK50ZQBfMP
z(!3fG(ZVr=XswC@x6HBo*WIE#h~X|2DQNqKtmxqX&Ksy<;ib|qa!?7|3F%HIY>_GxuTT>
zS0NF!d=p%#9K(oV&6!UPUr19dGr*o)2*!Ig;W6ykoy^+i68eefxU}Xe+`go*3$Y5H
zvHMW=;Abxg2L=FIaD--Gqf>>YUNWcqHhlSX!BTKH?0-%FSd}CY~{ZWywZkEk-b=ZBkV1zipjRhCqEb=C1&#G
z>YMqpeXm-@+I}C#k?+pi+wyEU&yKQr*ePQX6pO9
z=$D9k>6Q;fjxL$qCL1)LEcOSfu#-1XJx?Ehmy3G0vTjXg;el6Ar+7FKzwg|!TSSX<
zYhHwjvUcF4&5*qx*>;<+!I3AseXr;z6jzA;TVJH|BF#OBFL%_rnYuX!!A2EszT03M
zI`t(wsio=7ZrWy-N&D>HDk;W2&`KmGw2-emSMnoSj1FW^rXF@aY-5iQ9x=Q3&d@CG
zcw@M}j@w1Ln#?U=lt!bOAf
zxb+V_qN{bcbk`P@zqUzompdW~wYf9ShH9_Lndd8H>Ez?GFz3lkx@4Tmoj7a#a?|3N@O1L2{E4*ctK|Z6)B%O~3+0Bf
zaw9EWetPw)Oo|p1kM%uU8Jg1avX<85l${WaDFX*Ree$wm#cJL9)5`F$6Z7tIzOm~-U
zylQOY=-#H$DV?=6G_UXAHcr~Vn7Eg%lo6>y=?&asV6cq*oXwyO!#oLvT6_7EsLC<%
z;3rJnVGyq@Um~sG6LQo_2W7SXOr(GW!%sbK)OxoVCBn`<9qV%~V)_>?t{zz|{%^}O
z(w7rcZ<8nU&FAD8IOkPNp0dzNTtv4br+Yb>SV4%J*G1h%!@cP8Qb$JhNi`?gnLq9U
zO73|-k|W<>Vr|j(EgbINqF7<4{a9FzjrxxoSBBhmxTXOjT#W)AF@Qb4vFOLmePZvk
z&{&E`YJUmExMpq>=qx<$jc0~a1|T-&z1UBhp7{!baBgtsGI!zB{q`7fm!k=Hi=mlq
zg7zEVKZ{i$KXZu;ru>?kU1iTX!|Q!p+zof^+*}FQ^;MS4lYj0op7;#mIng%RSbu$g
zQCeJu>%N-e;d^5X^br=v$T^kRvKCQGl1o5;+UaQhsA|N57xakG*4!aTgRnTlE>k9D
zdG(X(r^dM1y-BZW<9f^ZUXR=kIyE3@@-apAj{c0#k3Tn4%ujY)ClXq=u-%v&VXWIZ
zi|i`g50fqUDm$83m3jPV{-}Mv%A_{;hxzUSV&0@ILyy0VEtHFtN{-Qt
zob)6`p@5}G45FH_2aleWX5xfDCIs_AZJ)_*{$0WilZ55mqOWF>V{#FTz*|Z_i0n0~
zO^YS(TPIELs4IkgL2Ic-^x`SSg^9!`==@ZdWUm&|7yz70#-IGvu-(cO@@qF623j`;
z#}EEJ_5*PMo2LhfKn1}RJcJ<@3`K$FrAYzXDu_sovlUluy032~|JKmedu7mA%|pC*pnDcc
zzT!`m&K5NHJEWHy^=(T2A=1j*IN?8>zLYv$cf&tFd;Ds|oU0Fx*_m#RDBZR6T=c8y
z>TQM91R=0)jp7(A5!eQ?*m_sWhLp)t@l+Wt20#Oy|i#i5_4@4QM4Lt_Wt
ze5XOE!CX3J+6D26wOj2LYo6o{WrZh2P56fs0qsi!zu_8x<5h6RBWGFz$4Gzh?iX!i
ze_xMOeq1lcr7}lvio#`kTFgz3tLSE(yg
zNFrW1<;wCnw?qiq8oKGh!Fja0HTy&UD%#qEDJc~@J1yK^#vs~>$ahf@KhEK9sj%<#
zX3D@Z#=R@`Tag^K9@J`g-3KI|?H{xx^?%jIN+TLQT%e9KV|6+#uG;yxra7V3FNnv}l1H})&?=Nbh=uZJT+y*!m
zqRBr{OPI_I@OxfF-2JlDNpt?gQFQxAi=9s@hn(wV$`&ECNh!n2F9t;!$$AT%r-C`?Luqz^;aNlf2bj19%up$i2Au1>EN
z+#eb|A`Ke^xN6`Ic@^pHhqfp0SA9kbTSjjNFP5VT@V&O9DlRYiKSl_#hkQdP+f<;~
zrB|5GYmSD&NmUFvu6y%Kip*UUmIYv^w*`(*t7rS8tX+MOC!e--aNXttgDEm1jLvt*
zqTdqmhVND*T-F!gthL8norJlOxcAqi3-v&tp|QkX)ynd%&ZQ%^Q?RqBxgkev(%u@+
zC*?1wF;~+t{p}mP>KJaoW_9Ua+eSvXTUs{&js)NE1?<{WK5;%>!-;8e7e&U5$O=JA
z1%pzf44<`R|7)CkpSGTKA9ee~Y3JAj>}B_Bm*l@mK`^ljq~z>Y=+zxb+SyQ+8|Gyn
z!xSU?BWAmcdY9*8mx(viMat8-92GEcxZGasn_jk_-0(1APE{l&r!GW2M0f61Xiua?
zM~Nzgz&@I9OQ(d7i?DWlNd-a&pko`x-Q$%+$_T@7Yh^**{2s!sG99xUfXX%rF{I9^ZKK
ziurP_H#EZnQI-bL^)nD^JA
zi*)h7(waTS)5=tW>ZLoHjeeK<_nv6cm_bgY~FDFmPex9B)dv~Ilab9vqm?lt#k$Rn=08i#hnfQ2NN^Kq~&GL?gZUC#2x4+#;Y&+WRKJc0tep2S^
z-@l@!$&tQ{9odm1pN-2hCwt9BW<#Hv$j-}hdYA`v;kn$?UbWxS#RabspY|6lzW1|o
zH04!A9kV}_C4!O#j6=4Q!V*%}J!Ds$emAu(s(2qYx#zCK==`MzJ{7Khc9Mwksuhk}
z7#8wE%w<0})eOOIH(b)#^CdW-CjH8Y>Ilik0QrzvTo^;N?w0GqgnReLLm&RH*OvBK_uZ?@&i;jZX!(UzF@P3m+fugeJE5`prjB7+bJ~W74O6O69H`
zpM$nkn6Nn-@*=2K-#b=KXzL)`_K-vt+E#`#76B&a$ew3a=azXJqKoiRBksIoZ7#C^)OerCU(u&O`kfi~<>@LpRjDce-E_|BwZHKX$ULlF
z+q2};O%&*stv?3W(x>)I9s8?@=v~R?#|aXoJ?_ag54)3Zb_#=@=#5!-Cy#VRQlv>`
zZ;V^&uN}K?tm6gh6N54p#GGc=YUqfjFptr)|JTGDav~q>`Ygk{;^Fj~Ua@Lkv{27Q
zx%?R0NTVE{uGxlihE7(spatKjYOzgqlU*MY&0-tG6%@R-4rDmfM4cvYd40@%;q+K?
zsM#=mZ2m@wXOy<~D~I_Ynx|f0wY8AkqTx;ITydF*w^V;q(}7nQCj~OuQM?*;=Zb=s
z-lP+WeJ$Sl+0WR6iTg2AT^(iJOmA%v)*15SP2L&_7b1Iu_np;vSASR>9TaCeUx{cX
zZh}MVLjRXwk+{QJ3-8`5&SkA`I3v$VA@<;vS$AQDn5P?Ed+LG#3u~@+IQ|?cbJ~Ri
zQ^k()?^R*-@;el1Jd9Xn`E48&Q{=|yT+5v;I$H{#&
z63C2Esr2WOrSYsyav#pMra^)Gk<`W0u&hF@oV+HPVHqxwr~%5gsm8Awywp`_aIT
z72h>&dxD}?=Q2YTsIHKUVl2A+V=f21G1Oj>Y7d>qLp+v#N`CCSCS;h$T52}ASGCK%
zLOOVJ;k6a!>Gt>dTV^*W@x2xFtjs)9Md-ewnPQafM@0o4RJ#z?p}zmZ6T?
z-;1MLW0`IUaON|xNH-q|8}4>$UtUQNC3rU-SEV=h%}pH@c3D^Zm~+r)9{v1PReQSO
zIhS6rk1@|w(|lC*BWqLAP|0}1L9%-nTr?V-Nn?!$Q+%$^ZVJ9ti)z`;%sYW^E2kvZ
zr4vo^Tvzs8naZfzctM&pP^Gd**@S<@T((x>XqjjeJvTOs2rrJ{8BJ`Jf-?Qu3Y
zk~Mb4gO&Psl3+Y*yX(oyLPLkmcm~~qczuDx>jOS|S67m|{_r){u^c0FjuNi)wTkZD
z4lbs-ADHGRn+Z-DyhS%wI&Fh%L{pA$&Tcp1BYC~urh+eUP1+>Qs%FGv4HPI%uCKD8Hp~V2Xspt3+Z*a=>q;MZ-Xya=bpglf;XH`
zoEt7Sjl;W9J2gvm&~r_ujuSL@w3itSoxOKXbS&*y1GnL*QT@haG>oZJ)PItB(4zG)
zLV~+O%^Q)+szPio+Dnf-zfPF&SST)2x!h#BReF7dbyvNNk$O!lp5T~GbSJ0I!s`=O
zTcd$}q_yA^mQ)HS+gB>GwtG)T{H-{l#T1aE*wV!W6Bn6KW;w9qWDEJ5D+-rRMc~%c
zg7yiQMc;O|EKKrGg6#Y6q$aUr995|7ak5c+3Mu`Mqd(B?Wc0;Yx8$*tLcbYaQ3HD4Q-AUUu
zg8#Pu8D#CgZmUK-_~q(%lvc3@vSyXD;>rxIreEn>Npiyyw4<4d247>wN$NQ;ZVbHF
zYDSh8H;=oZNE$P==o2hQU0-n#jd+P<>bB_L3y|XcGhI*WlEFwi85A33{Bwut+oslr
z3rqF&H&jX%|8J_uVzLV@&KrY_vf)q;MRVL!JXq>xeA9O+Rr`VCKOQM}R$XB#1lYHc
zGrn)A#6BzR<2w8A#-;l|RmJT2c1z!AevV(3P(T0R!H3MK6ZvyhO?#$@X|LAAXa&&-
z-sh&Gguvqs!&Zd_*_p${lSX6Gi`LD_;itjt*L9%)=Ny=VIJ$IM?s#Y1UA1Z<8{(o~
z3EtZnZkejS*-~Vb8QICCvW0Y!qwT!h!qQpefZBM98D1Va*RZ`Zl-I8fzjiLog{4r$
z!j=97w1_tGh@nzKtGq4g|BU1x%Qo@Tp~G1LMg?D0=H+U}lN)E9tmKth_l+~xnuM+}
zx&weE*9k8><_8m#)|e(|Um3vYbC`0}+@xsxb^~^~D>d}3noQi-b!|nx-e7z=P&@S|
zDJ&g#{Rj;!ib=8_^ZJHTTpA1L8Ey7m3rQ9QWh({RN$h827&YXTxTD#f-k*C~(Yztr
zs`7=7Z08SaD+k0|wVB3rqxFL?=sXF@2|7zq5C4XsQliJF$-Y&<;KE}{+0WFfSE~g1
z^^};|RLd5#&3#6Ss7iVETQTynUzvblmqCD(jSDUopImAdmA_G9qL8$7Gt!7vgmzSy
zr?uGnWW5rW7|=OAl`P^nh;=UWo!1|>RarF%-W9Ay{SKkOMY|+*iwn5g!eXTpR!lA$
zHTx%qmPF*hH6)suA8`QN{3zs_KdP|w#S$Aqq0kwFeE-r!*=S>xB0qZxq|fN8
zE3Tj~LIY;)ODFp$Cvnw;f?!}>(K(?IE
zi;a5mB@>K`F@Vq7@`(=8vq90k&Z#YRUp1Wmg(C6Dx%$1$wPwZ7|{LTLi}OvRfTmtbP~
zi#MHWUl=c?)WN423*+ElZLY@?mqEl>IRW)hNgX{?8((RknkB$(p6?1f**Se_IF;i7
zNP6~yD%p6IVUODUnY{%decgOIW1G+YV$2<1mIv>$9~4W$?>67b*nid2R3|KDcWh;i^t^Wn3k|DXU7@sj-oG4A4A=Q
za-V;9d^mp@HBMb`tCDfL!C2;m6Bf+X8!Sc_Df9L
zDN?#Qj<~EyBAkJe6hWYwl&`hJ#{EYm{yO_miTLC6*eXCye8gUHf<(N*Fa{INcKIQg
z=d{01CEjDo;8k$(N5g9|cf+pH3WfN;{TyEAq=>&J+ev>YMgeW|HJ*bu2V9hGn6Rkr
zy^2QysvmDUINW>)=z}Cs=d=Bz63L01H8+X*r@#`wQ<{>6_xxV7uw$Q=d
zdKG$N0JC8yY)0X`!zs<#gC?Lud_2{Ozlgk6FfdR%gS(&L=CGJo%dF;TvEG
z7m!s1LvSffY>-5YHzT&MYK>Ii-j%MoV1}z%leNz?lZwz{fFG<#L
z6=qy4qf*iiPoT)6BW22rZIJSdO_Nol*8l~mqj0!kB{x~W+wsW%+bt+WxqX>5*1qDA
zxy4kmKN7+)&KnYYi!st<9J?a^7(}B<01*&Y3j@!*vL$kS)2vv{aC|Zk;nfdn@ed%e
zvt!I1+QyP>k-u*Z_#x!y0O-iEMVf#k`2k4}P;j!`D@UN7;3G8?fr$*Mm+836v^aru
zvFulx=xv}L12hqEGUyT&*bTdONL4uc{#Y=n~mu>-jjP@J`v|kNf0OB;u4!8nD^#5kx
zdW1l)JhU=OL%b{nHQEeeD7Vylrjyph2S5_PJILEST0LSRp}%BBey+=91ODoxVadOP
z;g2iZ00-;c{z;62c-Jf<<-Y%ra
zKjCi6P+NhnPOgaX76kc_v-OLqu(E)$LV>F=rj9}~Ggdd&N~+c>U4;uU)5wJ>{~Djc
zBE{LO^KQhn!);gCPSVN;^vU}I
zw<4k#62Fv7TlN1_ArJNFWm`d(y(6eJEDHgHoTh-X=n`ibFJ?JK1ELuDI8ed31W
zFg_c*2cPmi2U#YBp@&M1(=B_9jeriwM6~inEf;nj3aG&I`@K}gvz34V
z-FM*O2&(L5WbTv3?3SMP_l~*SWmwYnDGLaC6zOay%x|lsBJX;r2=#~p{A+s~c644U
z)3cIN2F^^#zY)Ew-_#00&(>Z)U~iv2n`=~8@fom>7B3p#g)F}*|Lan9=PLXyYsZy7
zK)#NI9aigUYl^7@M#)Hr2dH&g*k)4%Uir|UX#O>>krwnUq(>5$^V1N1FK=6>$wK8Y
z+xOsm)_XG4Fa8KkNfD33cNqYaN+!VR?&HR%9xp$5E?*Nc0;R95^s^F>*afDq$@IZ)
zz&0YYbEVThFmd8o>;akn_5dAp_AK>_hWBEipvx}abpTd+znm#sFkqHQR$t`ZKbIhw
zAnhFDjuq%az+@}gx7x9o2>(PLUlRaqIV|tWS?Fw(_$!DgX&(lX+hMJY-V(PfS(qq|34$S9&1mfUo0Y+vIqtttV{lC7`kA3wSD|lM|l>n8imCa8aBj4Y)tr*(fnaVsoFdXx07|PC>3r@|<2O=$Rw+<>gqa_@#djyzQ7os8Pt^lKu
zT?WzbUPFqkRey*Ripb2O1VLv{QNLi}6^)7{Pz2uQ;ROut1o4$^#(Px&j&5V50d+cJ
zL=IM=CB@VMZpa}|=&+x(JwhKlQxhpm1=J996`C7Y$va{!XxG2FQ)ubc&9)f*elS&9(FDd
z#I;88_pe^Po9;IYI>jvCkau290{ZcEm7S}PkF~2cQAJmt@rf?e@NGb!60H#Q*t8eO
zCw68&;^DTcIDLy71NhHh{}y^9byvu$9vcws8AJQIVP{{f$B(*RfU;V6r(!4pP1Z9X
zFhZ}YstE`913S(GM`K`5Ry-PKpr|N6#f^*X*vBdx#}AokAlMBCM?eJsW7{6#>QxTCd*hMz)BrNLG?z_@GdLE$2LRRxO;#K*EUfFqg?tUxJPkWy_!K%`eiVe)~3
zVepoLFZP6mlJ;Bx$(u_X64_dsYk0m5MYzr>>#jcN)
zyF7T@==oqL{}c>@j&xReF~Pl-5u51?nJ->8V}Y$O&Op#lJ^md77o*r{4tl_fB~vbQ
z0g0~hdrqU@1(g}!`G|SrK1Yy8vY$-RX9Tb;T+|flH{2i1yFYal{=x1w=o_H5y-fl6
z*lri;M{C>%4JPoNnkLJD^`03Ul;T7knXI~P!t^OoO2m;JOW5;e`fOPCI2I20>qZX!
z*pZ9<;Q~<~01m**BQ;jKPJ2L>@q#g;v3|CX_&X^;K_tC&z@FAwpZVuqrp`_H4?Zqw
zzdiMxnmYNoqkX6G@tBkIclpaJUDn?Tv0>M7eG$ng3Rh|^k9o~edBo`%Yhhc8#hNe*
z1Iby63g`>3_3(}S^tT>w)CB5vS0eA~rdhqql!_bi6828GGwNPC2qS9fnPaXQ4&TQj#
zA{EkU%-hC(I^%L@3aSKo+n?;VPZjsDhD&kJ0V1$^rAm`al#q^EMUMb__%0Ktwyc$A
zY+6tv>$jY5bD%ZAUu!okTl$M&0L|(c+=_S|L`-+1G}`5fjV?BP$D;#ZOCpC!KYH;w
z`E2)UkvFNLBIJplzklDpCk3!Q&2Ma+2_{+bTv3GF`GZ{%TeNn}t^>5j`;F|>|Nau#
zyl4WS|K_RU7NR0-edNt&K78eTR7CM;t*%~LTKKSJ);79Gg}a@7k0{{zBuK7JFyb{9
zv5q`_^(fmCOkiOGxp}UtO8uK$=Sj!S$S2*RqVEXPKgJJET
zw7#s_l70~|pIK0#bLEs+L?!l6TWc6b?k!R#$WSd;Ap2t!k8>eTm^aaPT>z(@`~P(M
zSYFW7T+@*A!iy&`z`TL5pUr_bol93MMD3zZp*1v7;oTmSb5$yr{U*l-Rr7CFYuZpJ
zv5Y6FuJN**?lb8M@KLwouq+i+DmHJv4=CpvA_iuPpeg7~puKm{AvN$dVI4i;Aa$6X
z;{dwE002C~XoRa!iTg+@UKJ|_KI7>RR*LIyb`xfi^mso9%x7t(d&MrkhvE7B6Ohrv
z{UJ5zCrc|gw`2PX=oh;3Ni1UXCkTQnHPTS3@BIiZOC>GK^4G2sE*M_O7qLdW1-k{-
zXDhWh00{b=cP@NNhl3D{bnnOj!!V!0Wa9l~(*u_8oq-ZQgH}XgTk2=L4bCsflI){awIjO=yA^%V?)F4rs%#OGx^
zJ3cDRlq%h9l8CZqSSd&AQX2_33WPLDw^Zp4IL*g{PDdtYx<6QF>TDxcab@5HxV^7(
zD){k}#h{?}QTUP#xaS{haqVo)}a3dE*ETpO(}crJM9Q6i@A}
zDHagVX5IDOKp~OzD$6-xK}KpC;5(^hTlbLz1EwlyE01k;?!NYRnmd@0qYf
z1L_UIn9kh%%TK;vR9dhWU-?ux6Fp25EPPCf4N!&+B3l>dEyYpT6LZlOQbFu
zGdvymXTyIiM7L2Z-g)KR1Xakun8mZBqvn(NkX|v=A4vG!UA-^$9ur+1xMSVMF5L%7wt6CN|=xeo!)zucv~B`Agh|c$&ijB`Wf?(
zZow!jgtHMVv-5Oo>-?@wZbQX#ThC+i(S6z-#irUV{}HIfM2)GFmE6KC+jFn-myw6%
zGi*|a0SOlJH@)?U{z0{Q`;%N*v4OE0Pov?Tp-3Tq$Vd}J2t7f7@G!bTG8rz>fbKf?
zdzZAfE(D!|UI~Kpzfer8ZSgN&vB29bHiP3go^tkq%$t!D`7u@galk3**^nK-ReTRT
z_(c?`atf8!0<{I&OU++%s!C#pc?z^tZtr>pI0|C!PzGYjJwVQWxP`_s^Mw!WQv8AM
zYnMzx!UuLAKf7JBGCbzW0SjO#16et8pGZ#vh$yN-j)9zQp9h?ZL!-x-3AO8DbJ=23
zrR-0JoBK(ZDBA+VjNsNg1_9fzqt93_VsrC7BaK1n>=aaeHI5#-f#{PI8U1HKwb3Ph
zJx*Hyf}PT3a%C-3>)a~?4ghkPi65>$4iL802bHA+5bR}zFW#{qmgM4N!$^uiH5VoY
z(^ugxrhE_;4=u>?by!%F__DOJrKd`L<#)KG@MoYDavha69r6B@8BPf=e}ZyFr7v=k
z^U!DD67^;HJ45S^0-TwU10Q*SP8)(-e+HCw4IIl0*6WD^W1jC}y{88r3QXb{T_EB2
zVf`l?#B}Gf-G+tqBXYQ0r95tY`5R`KSvmq8Di}j(08LU=q*c%xovHlXj#u*)-@#0K#gD&nD^B&k&Nup6c@G4||%Ps+2tYi?`aq8h_yC!eZ4#v?C6{o%FU(TcsgH21@ZQR5(g
z5AnV4e{B-Bvg*zAfz>RruV~>)g87v0q=v=u@^Do!=9R6Qiq|K;@e>(}?F()`?JMa~
z59tYToCKY5P%lHp!4%5D^tj^4!=~Mb>HCdmocCFq4yQi^!`h{}EDNPg3W8vl_G0O$
zlRVU`FH&y-=H^=jxhaIqP5tWmmNo9R4SQ!_<}vn0_{`sK4b(#z3fWaJ+JEKI#s|@<
zw=M3wW}U}J8R^K
zHM<U*!IAM~AqxoYnZg)Ew#=M18-
z;zBWI)z&U`pkv^u%A)(1d3PlZSCW5}2q=Wx52)ufbo4(CyL30|*Il&nE3eRBT4U;ZugS}|@*B?1UXpNk2
z*BgE8YlETgWJFX78_d~e*ajim1ypX
zP0Q6mM3Mvm!`I(JR_#5*Ct}@fE*!oRahk6$6%y3+hqvyT{3LCr{+bXgrl|Hlz@|r#@2L&7yY46{qa;%6G`Ty
z)}(j06;RT$)QV0Qcs`Q9H5>f7p_h4bYVWkB
zM%dBmT~24y4DYLO8IN4u3yNug;(OhBx=~RmrhE7Ynp?wF25$>yHD&tReSqr-{c+2|V@JvA;}TR7`za|88{rNTxMsIGByOUGjY
zn@4d%FseJBCnM4sO)gkngl_&Z;h`7*KkS|9TT)#c_U*3jx|_^WF~`!=3hQQXISx8k^K
zK>>UraMewDhAg)b!*51b
zjWDrQq&>vaNoCD7sb=9T<1Cv_8(&MC%oKx2s1|;(cjyL6C^a
zBPf)|Pq^1o`$5xKizXt{f;RTTi-<`d#QS&C#>ma-Q5gqgxdSm?mZMHhtYIoMt9Udn
z!<`+@nu~v_Qev%1W+BXGz3cCA4yCp}T=f*M&{kK6e1{J|zq|GB5$<91+c9G2%eiB_
zW)b6|sKtsq8fVHojE=?V5iUnatFH$DhO)|!MQws=GX?#DLN}5WV&q-=ShZ%vf@PXJ
zQX7ia$PP4u<{X7~!c4e>-d~(C_sVWeMRq;q;p2&lb>EgfDx%WNT98^*gh^dLb8)%%
zc(QSkB>27IyvFXqK58xp$<{RW3AtJ*vRzxN
zb=Oy^wul2RFKKi;KWQBmcU<%(y%mwLkD2I0-*2?Qyn1Jx6c8-Sn!I#T&MzTSKoHZ7T|9N!>IZ|H52Z5cnD8gc(hyEhhQ8berdyoU7t*ts~N5Gy7xfx?Ye=$^9YJ_#t1
zf!xpEZOxmaEMgo!W{e8*(`_mQ{VWbJMppS=e$|ZSVObql+cG!;Ei@0E3H~#&xRZm@
zc&r(I*ZGWc$?q$%^@ZhPFf}~QyX`3t>Q*PVN})#Mm<=``BR0SlrdDi&;GlNhsrgkC
zZeQ6xMfox6`iTnXw@ray=$pN-4rZPdOygTwSKN*!D&W{ECo+tn;V>zEM~#GEA#pznoF2UlvHxrLg32Qf3M-8c#u*+5D-=
zAR_r=I7~c#64|z}R@1(i7X~;2Z3}y3VW_#efZf$TR1+_5>u98mR<{XB<#Yx;QhxiU
zPtze{h2JzUo+5-z(l+09^KWf>R5Z6uA>OCtg3Q#Fo!W}y4X4`0*W{ht6%ZSf3hbCJ
zrbNBea1s0*!$9!;aNb>Md@=AebIC+yLU+fcZI?YTpWdmXzPAfz03z?(&IKx|$Hl99
zLAl|DK_OTsm61?T^TLu84Y)=P86sBx_``B&fLj(SFkE=t&nr@xu5?EKiCunEKt0N{QFhTa0b0|FV
z+R@BD-Hy2F!aakxExWq)S|c|JXR@%r`!AO@i~QwPUNE1@AUfpO%GhoB$t+x=bK7Lh
z@`jP`F?MFj;TzcSpBXX_905Tbsur>sPM8+H>+S?$)Ug@81{zbK^J
z_5!L9hnaVZ>*)@B`TSv35QH3x7>GR~_}#g1!4q3^qr<+ehV!9RoGyJQd9>SKo<|8d
z8mCxcNAX1zvtx!7hE9msV&--rMqHR8l{%DlAU4_H`x~_wQ|0*hJAcgqR=s|aN1t5N
zIMZ5tM3>vAVy)M-+RJM5BzdM;0(OUI&#J0X3{cQLg|;W=a{-ZrXP}R2Najp<4{=An
zOgD$o75i$^6_SBctT9FhfFkjTB=0*4r5i~>fL>_T3s&!TOYxPXR9W%b!VAN~{KFU7
zjpnxus9$>bh2+f|i?b3Qvg1OqQmKYoWUjmNsFluZXh1IAi-SKDJ?lJ*uJKI`Qn`!T
zqw2$&iJHZPYUm@A)Qehw#`*rZLS>?I&wS@l8H{PwG`fTu+zY=!#zdnEP8V
zwT26u4>#u%gDW@5jo)|NE81VN@vikw|Do~3V7H?;@??1MBcCp2_ol7mNQ{SmJ*KDg
zq#^khO~yeS;RCdk?iA8^R!MKS6kZqn%S;OYW*G0fMrK}&OX8k-Ez{A*8=wiLSi?T&
zuGUrU>#nshtDq=g{Hn*_6L-Q|qla<
zdKy-aUH-v-4m0Ue)LS3mBpxWh3
z)aBaA?yNUe3E&#HvP;!_znU*sw+f93d$52=5)c3mH(E<@q`ZH6t^WH0_ItUTVt@5_
zEGy6AMA`FpWo*!Ut`ckr_AZmEi&l58tTLs_9KxKf;pJll@4`z756v_AXwZ!8k-BDEbeD1p&oq
zgYTu?Snpu^@yhJfX?6qk{qwPWH5I%R%{2v$*obgP?WAMZ`{6UNv?oNB$hh@QITkf?Y6#(G
z8^Q!*bTjUgH2lv&)yGkr?7SI>fso~L29|GAOFtO<1!8Pqq&PR(tF
zm9_PmE8kVwptFIE)B$Jjl;9kGJuWEsTe97DhK>whMpRvShTJ@8M3gUn4*EtDg3$V?
zeU^MdrQ(bF*y1b|_ibu?CO-5y)=sLq$aAAZD)w{Qo8u{7qV|K~DO*?1ADt2nT3??K
z*bkQ|@I>D5uooIDGZ=DL*KOHYI%42@S98@1>z&=+;CRlV1q{2`kUF3@kkWKM?HOxi
zjJXnzM~pQiwB386V{!E#;iSpsjf3MNR1-Al)`<38+GLus37ne5_hx+x2`JKYS|27t
zHMUm}c3dc6;ArWS87zG4t3{K>1FYnbxFX2hx&bP#$uPqa!4405suxyc_0lRVMcmiD
z*R8$UQSsFdySUlaW85R+ag@qfRpz{{wkv7qvr};rrr(6ubJfbOUQW_#4|2L9&;B1B
z`?~Z^e|ChLa;gpNa7_wFtuxK1mMXN`{E0Tk9^`xL{4`hXtnP~i?H126tOq#sNLG*sr8QNo!R
z-g(7DN??J%H1WtAzK+T*$wyztaZi&G33q&h8Nmp1@y8U&!J5-4?a-;}BgovBE;lTx
z?>)AqNk0B#K=5HYeAkYC4st4KcGKPmL`2aHBl#xiW`euucQX8D=QAe24=`e6N0EHm
zwz)u`X;9QYZbrS?udH@?h4ypt(Z{UzB*EZhaR08V*F(*QonbGI
zkCDFf!MHQreR9+w<)_rI`WP=9!ZU}Srp&s!zLYN{jnKSJ|;P-nbw@~{4!nDUhWDF|J|nhrxtM4eSPAUr~Axa7*~2p
zWPc!7?J)YXYQ(0i8uA1_rX``PnMaCup*-^e?1_&;Wd~qUL&xf)W&9{!5T9Cn~g=?!kYGbw5o%%O77!P#@H;&KLbu+Cz
z6cOT#eXRo(>SqSy2Z+pnOiV(NxeR)2BH1i@Hf6N}NV=d<-`S<6TUIZ?j#4a3(=!O{e5}$?{aDYITMy87^H0`m7tn_=!gcb_Ov)*
zCP*Y+Slmvu(b-YG-@|nzh9_X+B
z}T!7qKIL7>0cBJ0Biu7V6+ieqq{RA~-3E8NPk5$K{ahnl~$#Kz0e$*cEn4`MsZ%
zjc&%dI+`0;#0G1Dx>K7Rk;N9Z@v(1i936vi^G>rvBhvK2z~ksE;+yB@Rz^XB;9+k?
zwJpAB-b_pCJ7^_nPJxyou5Uvu6{g^?%kZE$_t($iK}I}J85XpV5*90DO1j{`LG@A6
zFDe3x+?K96pHV+0E1hj=J{fSug~T;cQt#<|p4T`a{W<@H3VKr9D!2^OuFjcX{ZIUh
zQEb5&_)j+a6%nqo0`fHKu(;mE%qQfNPbFvGUts70*6&IQ^3h@#AN4(Q=zw>&MT^%f>uNs>YhhzdZWTZ0l>ix3t4QG`-vLq
z8<)1!`yCCd;eN=FaS6qHG$Dg<`sCLtxG9ggv5RFr7pK8_$G+T1#=DAclAZP6#OZhS
zu&9E2F~OYWHWp~p_)3j8*Hv>DCEo}ydP1}yM=j}dOWm&5U4h>Gu}tr0)%i@FD>>Zy
zc~m#wuBfoq?@okC=pZ1N>C1AH@__HXnQTeb(LG|?C%O3%0X$+96PL6+xOU>+Ybm{F
zI#NbsWN3|b(WR*^xjHgHqrX>K!iJ0xn-q<}_8MN7Q
zqkZ4U7jOrE@>OM3DfK
zQ@)&@tgMtBVidG(FJ?HVg6<5vbP`O?mt2op)<31u9fjv0qvqsRsV&RzqtfUFqhbdQ
zHaYpWizu#?*S}!za!sqSv!?uEgX=vv#S79vd?bAwnV7lXyy@zSuKD_>B`Ig`apc+j
zmoxE=a$ruXo~@Si`c5GDlokzh+7w-eon>vAivIF>sLuYNGsoU%
ztLB~|TY4z6X;Jx%7v6+#mqY5zBSPWoK6HOp&tMm8
zW<(yB%|SA=36~_&-UB|FPHn69&Pdc1v&dtuAPm8EhbPusA>XJS_u?(}WmY6*jrr$s
zsvxvuhnd_5^1!QZeuv@DTRS2~9)6E_-c&Ao>IG;YF#GAnjU4*`Di1OvyHruxEB@`i
zaQlbn-kfsS`smvJikLmD^+NyaO?g#w#bkqJ#)4TSWn{>e0n-Fspn+mF1TE4!^0-ag-SNOj8lPx884Hp{W&w0_I)hqRZ`>Q8E^i-Ky;)$Ln^2XOPmxDCg;ht-u?miyrB?YB>{*xg7lX^PS&;vTl(VQqt2dJrk-Bg4wxs(GiL7I
zO<_I@nbrSAug^KD&_aRS?c(Dy7of6gJhS(6g7%6ov9-eG0`-cUwwb&!=*YP6zk)5KQUGR7Cr8CNc0*a|)^K!Nit0J_{dxPrDsxamT
zBfaUgbKZ|*(hdSkytB1eq{;Vy`v<3pb*C3*~sFb%8iDAAH(3^-8Oe;YA!Pa1vH<
z^C=Ky&0S9bZC{}#+E(sL`763D9uOHQp#)5~*l*!lDeq_h@<|Kik7)!I)k)&v8M8^F
zzg~-3kTPV&F7@j6#l=+sH#C_I^@6IDcGy0L;h8h}Pp@t5RaIhX~s^)uSMK3^PRay4Je;496qSeQP7dlleB`-MIqO4_`D
zlU0=MiL`9xwTYkVNF0w0xBj<%9XxT}bYCj_KMl&f>%JtHJ>KZlMK0{xM4OvO4*6gX
zwrQ49^w9~WP-dX9oJq>Mm&F_S$kE?Rt
zq;SZt4@~E4g`UuLEb3N*I6`^6GoCOCza0sIR&BHiYhn$^rHfxzbX;__OlYH7aKCK$OZfcp+U&B&
zS>maFT+!@g2EP4833-de;#*Q0@7JSYM68KjfZ_G{Slsz5j>vM$^pYj6p~oT%BW6{Sp&Lh%Vvb)dv-{$r!
z_giUge8JmtLD?0kQQu6_9DZ{;(afw-xD+8dWed5>Qw|u-RmNlE0LN18x5w}i{g|0)
zAnAA$qhp!Y9{|<&?OLsZ6XY`8oJY<$L8F4tOjnv;{*bj2h>z~G2FG~=YPjh>92+2a
zL;!S~e{WZ0E~24bOsG5UanIMtJV8}Sd8?hRN1$_YQjN=qYhRN=F$ozRfHZJGmzyTe
zm%1JN8=y@*P1%>KBZFNcv&}|t<5(fn5g|U6YyPnKUX0fbfKJt2lo8lj%ii5WvH&nB
z#Ky}>o6D2cA2wn9?C84X2WZmj>!p@3KhB)NTwIm;myRVJ7lQCxj5Q6>q?y#cqbm+{
z*eU#!%ajYrdZaU*13gm@n@6ZnU)%w_+^N3({DvqV)G!2G%^lPjKXG7LJ_t%
zbUfa-BKh$$;tU=%I(Ws0YEdL^8r5EK58p%7O&{X_@%j8{7$O>
zm1biIC{Y7Gz&(8YWka0{6|4%}>|!k8_2o1Ow?f-rwjx3x?=@j;f1+62P1~4Hg#0kf
z9MfK|B4s148v!aXMRniO)v%a*S-Z9MUJZ1=&0fhIv?@<5pDQRQPW65K~%jbCD<#AyHY>78*NrZ2hJ3Nmc7ZS6QnGU{&|98ymdPo
z*au`iHxk@yCr#QME~P!bv?&DGQlxOWHrFT^8~D?&W&LBy!@Jw|A@UV`Gs4ZQ*B=mT
z6t;g0B{G8`48Um|Ocb#h1L>z);3LC?^7xjE0gg^~zi=j&VTL#w;b0>h9P>*g
z7=Dy~^nPQax7uee+kU`v^#&|Wsh0DgI~2`%Q~p>DQ0R`4d)6#Be|~T!xe~A-OKoV{
z5cr0lmA~P23#ZwmYIgZ|yY57TO{jH2__*M4bECQThD`7DsFlKW)2@%pzREQ?i*!yA
zd{8~4iH_a8lOD&!xuvVFG(g!=bqVE1;A1^USjR`4Wb-B^rGo88_xXUnmfX{aJvVPs
z?xTm-1k8x@XRG?D7%RfYAHU?QVqj_GT2P7l0|dkh8Y$Jl>mHNgK%?;I{JezGtGrP0SM&dpuJzJbg
ziibLwh>^gW`r}g)a(QdIE5r{nJKEqDWC7aF5_FpJ6iXH@vap?*#2~KuYfu_gB_#Ua
zcE8&IasUWvw$g%GOZ|E|5f$IcosPLuw>CVVOlIIhqwjzcI_6td{Of8f+n?!Ft9Wv!
z6dCKe<@<8&E=GQRnG*D5h55%=S_)xIt0o#3K?`LDmjyi2xGFPdXA*DDMV_@V7+4fr
z{hW8X^6h~2a+sjA_4gZ_8!$bQAa&B8ht%}TavFG4QZDq(yT=2k3jCVr(Vf}*CC);0
zpujwXNMqeCm<*E-ovXQAr<+AxQ++x0{hYSSZUdTr*grLi4L~jiqPlaI;B3ZVR$)Gp1;gb+mg~
zy{R~La=Day5MAPC@y?up?NPnAag`pmw@WW`-Z1^5W#zXIH$JG8qNF;IF=>{m=iT_R
z?oAO`I9S<+elNsS*%3Vp%eAF(jI560xIWzJx@q@|<&{0bMOrO>LrxYYR
z+<%Q^mI-pqXI_4W>go1Q8}Yyig1E@ugZB~n`p9>2B&y+?3Nvg~D6qbanvNe@+}bfB
zIErO9V>ewJT9F<`&u-&*BT=_%o|k*)V<~mfhv@osg5F1ck0N-n`{gBopS
zaPNBq4cw14q;!~>@X~W2aN){~K`#Jc{x+b&)m=Pl7|?kCw$rG!
z(`+1Kz6+yY)L~AD1+ym#TJep6HZekD+5dEYP9Y4$5V={+#7c+zY+Kz!405;FBOPG_t*>kM#jY3euza3O9m3
zDCn8e_&_lWMxDfbP13L$%=e^Dcv`=QFzKK`DVD3s)y0AY2s_9X;kJO;Fm9dCt|6EM{1eO;c8*Gvhz_yM4V07<w~}Om6A}%KR{UgyMhT+}|vKeQhIv&-lOF|1UT5fA>I>D&}y1ww}F2*n+WtorJ5R
zmW%Q1mig`!nH5C|?F?E5;2jLJ@eG-cThad3z1M$usD!94VcN9RyiFp$DQ>*>G~kH%
z>s4s~OShg~pQx@4A@8D;UfBTpSzGNi1LgnO=YGJ+`0vXH6Mrl~|Jhz~fyv^(FW>(e
z=l{F?KXK@MUk6XOL7%mg9
zKfiaPD{;QqNOyX5*xtS=Ug&=>Ai#UEEK-c;89q!9)MBx#@ANKvt#6{PrXE9evZ4g6
zw@pU6&}!`TBexZOSj}iBcl2N(U?+>e(KxW7EwRCvA1X?
zn^%U%FqGfVLS+=Tp8DKk8$vmH?!{e~$?<`sX>C$Dzb%Td9UwPl_U`UotwjtH(;
zF%9lmgxp4nsUVSm#J&r$Qf1cd`y_;bbo>59llbOC*j<_D02@)PCFIjfITOJfgGO(6
z>l#CJ8k4T7MP%R)$pU}r@gpPH_$7+h-L5XZ_+Fdu@T4cTNjMtmgPV4dmqsUut}CoQ
z^8>dC)f)a|<>N$D^ljVo^wRBku|V=pcI^(|PEn7V&FH($Zx_UjF$(Q#9E;rXfubBU
z-{S2M?on>OGLOBjMU}0UuWsM}Dt9y1;V3b6T|(lAV_lB<0gq@!h-i#TpnX7LsRdqW7M?^SKT>((EPb#Y>P6f}d$Hlb^TdqVX`cl{Un!2~+7i>`P?jbGEqiWS+G%Gsz;h!o3
zVM_p=7O}+2H!22K)OC2lIC=C|jG5=<$h8q{VRh?blR|GlKmLKWMifjAMG=S``!jye
zYZ6sP*NjNVr~d|=iYL0OKpwwSegtREq;MX3^P-W0eFcD#qf$AWuiGf>(hw1~|CQw_
zDXCXI3uFs{4<8JZkoYJ~05d=~7*0(oGIn-*jdA1C(<1XShxbABelNbCY=sv}xMMhO
zcDJM2KJ-4mPl^%^C}Jb2fBV5a-jx4F9n2$
z6c;Xef4>rUS3GWe%e;t{Z+5%~@10h?F@HGaFFrg2xkKUsY1J80vtS$ndw(2^5}-tZ
z&cDfVF}|t38ldLlUnOXf&C8FwgHe;LdZnx)^d~HJEx2dbB4%gAhi>v#iIc4bo0IAUSc&GQ^s>!3Mxj&p*XHP_ama(1Zh!V>IP7{qp|K!?^r4b3G?4g_Jth-
zSISnLEK0S2n|2TrB7($&N@ChrBG0V#QeFoue%SN)cvR6o{N=v@w`ll5Gu{f=tWdMz
zTa*P0i1kZUshi8-Oa_8QbE|Jgk5}4je>@AAsW>ti4qUoBRBCr*XfaynX8!x6HBNG6
zq4w=T%jXj|UZ$S=#T>|Jl$CfvQDqB;3dkMER@7f33&#w!g5cW;JYQ2!S~fhbY5n(A
zBLY7+UBT$K_q+a}=35)B*s)8k(q{hNGGZ4WXSP_x!WtBZ
z<Vp)sFghTs<2d$Ual`
zi~<`EI~No4I7WY*Th%RCMrCoW?yWU{RWe|GUH`qTekqxEJMIDAJ6ZB~Hj=z(DDkQT
zK9Gec&bne088WC+O|YMg1|^miztAZViCk?-%bg)o-r=ox27lI}^cmNBSQYD}8@)vR
zKC}rsp2Nw03L*FiMrWI{inrp3-5l*A@#)mH3t@72{$*wC+$cI@DdOFfapK}?vu+np
zp~kwaZLr?dx(`hR&a6T${pSMvL}w{9gA^a{LEPW5N3U{drILxt)4#Kh+3ly3^nYx3|H
z5nf*WzEOG1`h~!iaO+wbK
zH+k+of)p9aLq`2l2SYaHM_H$TW_Zg%#8ut#bNGfS{^z)vGmm}9?O8PS^Ve=8fY_VasE7)q!DC90fF;~Vd!D|6q?#$sC3BeMX1OCAuOk^Q%LNpCF1%gJz04TdF)9JtD`hij{g<1Ia4V}ncNf&
z?3hio!6pSazpk^;rh-d;%l*@_*;#aT!S%0qIn9FfFM7Ysr9wUfV+^-p{b{n-CbZ-{
z6+szl7Rm`7>g65V@3MtOoX=!q-{&(AR5GXcB
z3jNZ%F^N;7XED0DQJ7?xnyMGQRxo-J2^K9u9Sl_ZK&
z?z?6ArfzW^XL8O6Ey&0Pc4+Nv4A+5*?l%MFn|HH9JB6q_`Y`k5c8B-PIcGu!GA9Q>
z9Z#BxoBH^04qeIJ)-d&X{%l1mT}w)Unyu+Hq#KHUA4K~=+#BNWjQH;#6^EpjR=lbF
z=Fy5Z_woc+@6H_EpZ6gq_;JDy*AzZeb1;PlWfl6Y{p*7^n<4U!agGPzTW+=Qpmp%B
zm2{RE+d)obpr5)TUV!eGRt!1r1Ef+k8^WE$PNU_1#?mvukIX;qa7D{Xz{~UZ-pUgzO)vK
z(eOECe3MJ;{MDM~d_i9Ql_yyZX5ixW5_`g~`-Voa$(8tS%!ey^GB|;exP_hj>u?u)
z%46eI(3xd@l|0Ei}a{5{KFbd7P)^_i22?@)gz)aA?R6rJK)RCdJvOOQK
zWTcd6eeM*zuifVG0XrOk;4b)Siwr{8-NHIP9kq3)K_RFRFObLR1}M49di_;Q45
z#2}3l)s72IHLljXrMdcVB7;b}i!)-r<_fK!qf}pfuN*
z-b-6qz)1mYuNP)`_riRz>F{0|dN#+wcKHSXTnS?kcUz9b$@z58y-Y}sR)(PbuR}EOsGTCnU}kDqi_WTb}Xx{hS_)?$FoBU
zfyJ;JLU|%!!tZA|n1gJ7_3>AW0%AGKes5h)7V^5H$`RbYr(-rP6v2(I9X(gY`1HC1
z(U|W#D*vopQo}fG;--SaY_g8SyLb5LUxq3|4|oStFMypLWKina8Zl&Z$L2vx7n@Ci
zmO*c?>pQN()$RvfOv_WcVt#X#4U#!BzT=XKKGdT{2LGql@S4*$4^p+
zDyl`2c!Gi|CNq-mjJwC|BF<=uQ>S)E|E04`!{i2{l}~6#=tOM8_w>x4Mcw-#Cr>}k
z_YQ>4rwNki&@{B#=a+Z_`h~imri@5
zS69j^Q;J8G4!?I~0y7pjm4$ROlNi87!7
zb8eKjJnO2QiMn(GE4$IWJY{&|MEA4kYLTEW
zgh9k*MU(4u+|l^8PKwxmp)#PHxP;q+Q@lUp4-92$PtcRc?Y?Su6L9k9JEyMe!k6J5
za!r|dOw|j{2yk;~gQnideoG&9PK#k3jnGY@aqy@(P9@L1LcU+c$##n}wpIAo1hs|y
zDBC=&(wIDG@@q+cVEi6hl4bbnIN-_$#m5dqPWeDu|kpS>q{}#
zx9mS%S_LZYoncV-?sROjRE6jk
z3a2E6tKD-+(^nB@FOF#yd6`JLm}^A8{ix^NRT};M_@q(oYAC{8!LS~JP6_>Y1)6mUdZvAeTEDLb)1yOE
zO@1Y4c2~}t$IOhXiTXmD(8}&_joWrA;{aGh;RRcIX$j&uy3l?0ClMmvG}%5A$Kq0H
zyD*0jW2@eWz6fp3H!IjCKh-&B-rYH~BI6sK
zZP}!99Pn%bnKQRcGfCBzeKeS9h1AEHLTTDK0P++_+}MNTeI|Fr?(iv?z9{)HIo-~?`4{ICgCwY%FzKf6@Vlsingsd*7?z*izro4+Hds
zD`J{>d}xjO)tKV=(O5P(CT(q1S=lXeWnf2a+3HMg9-2&3wu~O3s$@6YJ_S(|zx-V-VO5*rY2);^nl9hpRq7GLUGke|qn^(3zClR%nJ5
zqLkcU$R8|RZl<*a+IZ{xb&>d?gX)Iq79fRUSl%yAvi)Z9cr$;0!a&cI^|WbgGmRnC
ze91Gwf-em}w;H_t+CA>;-)vK1i80}t*0Q%;FJ|+rp|jkqAQSRuf^p|_pMB|Typ%SR
zz%K8AQy*bRaqLYD*PC=1@gk<;O8|*y!4r9M3qbzczJ2VW800^)mnzpvw&Ge)j>DgU05Fic#@}tbO(-;zN%;KhM4ymZJrh|x%3ove3uPQx{M(@`Jt&mD5!!vleU#u^%knzf@-Pjug=YQTknd)lcHIQdikX2So
zj$~2ppsc_ff1*~`v4v3As8}zxPTkTinpNeI*D@
zAFq?Of}E+;kPnBO1XR*>QL)M!P87DYF
zjC0pW6AQ0%g}XWH7Ge%ruxu7F<%63KfvXq!)|8p!gmm^z5~UcmtmiX@I{L*lS*Pxy
z_yQ4^z&1dwClWiezc}y=?^c387ZGJ$;teln|M?Un76)C3%S1vlll%w=xDOxr0%o?Ku9$tgp
z65a70alI9FgxsjKsmiWCujlQds^amez#CQjm+;2Ca|bUE8ke#~gJ!dw^WmM5ghGmF
zvadKE2#Z}pty&M)@V(B!%7WS!ITELqDVLOh`*uY&2%LEw-j`%SWg&W-QVzYae98wu
ziS{V64@fz2MJkd0&(js^mp=`%8*P
zfXq}*+jGvNj#!O4;W&*>ZP2tkmV$U*W%^m;3Oh!L#)A(ir9Pqw^YsXX?srBodX!SR
z$NkS*K4b$)s2$Z$169+pU_{9Q*jvZ+sB?oJ;>cK@ib{AmJ$Ccy<84om-gG#Aeq(56yq-!
z5GilH*>cM{%ckVFhc%x{?9yzZ@N!MQ<_33i3=Dd2uf%|r?>))o6Lb}2
z{J_9aQw_98cF3I_KGEGH9bSwYa`O;z*hZ@lJ_4Jn!3wkJlNla+{Nbe-@zluy|C?V4DIG{
zX#ju`?pbE&d1gPb;yJP|R6>ujES}tE0r(}_U=z8d!misZ?sw1qa!uD!C3g)
za4q*fh4H4GFzoPN+5JLjz{Ce8kYI`FOE(kzkQ_3ue27z-3VD9w7;?U<^O=HU@%}}8
zqMi?wGO{*E8helY;WZi!i}Vaiupa*yAHp(w$bggwqFPrqrcOOLSyb!}
zqiIpIJ(G&?t~8S!(!r*$3sNYb+5jbDy7EgY|}5`Fb;+ACUb(UcSpeTz9UJ^BN++Spar
z>@O$a2f4W2i81#QomR@@;+w2`xw-0bDQ{$jG)qv?}3&GWa|TDvHe2Tob;x)gkI%AxelLm+`EfeT;%J^c#X
zNqJFQ#6>r}Y|u3MB3yG4x{jhjSV{qt#{cc*xGC}~N5)0PuEma&4Y#~s8M6R}`F+@+
za8G_zO^tD$An%v^V2hF&k^XP5AfZ^#f;P^lzdw!nJVuBrGV!Q
zUQ+}^ZBQBY8by{)9&#M6Iaci#3Pm$W$}j}VIqrxNcTz@6`I0RY
zefP}Bqd|wSPvmBpurF-oMagO72c&iX@Wm@%!W_L)1dnPes7q;EZ9f!AO&bHHtydH8
zE1q5c;o?&4DO~p2u_d}x)b#Sk_gAkp9~XJ@U!041Op|y({;u)`A<^OxRr9W#^z3r(
zuHMbvZlNcf6?%5!W0zph{*l-@q%z%4bEJ}5qb`4MUAv_#2f8YA5g_vcOVaFe?3t=E
z$kUf8n%6u|686{wZ=OZ8+>e)X?H?}RI(v0GvNT+_E_&ZH{VG$S*(a+^uaQZI-CMtz<DRO%rW2P4I0B|w$bp>>JR6Hs6Qv_
zX;lWK<8~wVew#^sqA12Rk=ve+Nw1Cm&%V?tcbD8kJ-xEx+;fJ^3kq-gB!na?C~vTv
z2U&_hd0uVc503gCTD92T0s^lT+)Lfj3bF9da|@ezTiN9DUi6C*oSWVjF^OVySnzW)
zeTF_fA80?AXKT_j&0`E~tqC@WEHL}(aq3=N6P~ARe!Ds@COt6KjF9UPK?f2!56Dd#
za(81n)Ba0LE(mo;-9z;BgW$^{w?M%-UdhLrGd1_f#UKsx4zwL^Urq(kShMeRd-_7Ye&ARV-M6YmjA
zClKoew<<6mRWuSuJo-2JzM`<`A;1k-20F)k>i(lIZYsevM7VWssp4c~l{ZRNF8h7z
zq3bPc2|jh&JMr;K=gwC0Hs3l}XWJStP555fKQX#vIvyNWy{nty_xEII>YLG`1Lmb1
z>@a_1nxVhm{BaftvmhP`|772n>?*TUviMmHrfPo
zA$NGN_uT%nU5>`%khseqy%2Lz4y`xa$3tRut|uUJotIN=O8_A
zXt>63g#y|^z-^-E#+NEYnR7u_$HB|>|1|q
z`C&jG@##~8)=v1D3(q%F$~FgW)Lqo<3%mvOZ+M^7EL4x?VjSG7W*eN6dq4g=W&S=47t>dB$`@^18mn_DcE!$@<;B7Gmuv<5e
zC|YA9duI+l6IJ!Oe_?;|V@z-S+!f^2vXb98;Wqqw^v)>8NL@liX;TVHr8E+US-h^_x98rJE0za7N5uvzRZ1v|Sm+_3w17zO(n|=6h=>A5K%@pi
z10)1gAV4S)6;SClkN}}5p@q;AAR)iY{{nwwZlH1aYM>=5(pK^ps|E<*B<<(
zMJ`@Rvo{!OMBuc-j)|PNP7wp~9#9=;duRytDtC^CN;LC)G-W}}r)wrv&kN@}>h100
z)Acp!aOVXL%Ke1}Z`zDr15MH$)+iM=-+yx4X+YP>UVCipL5JAk&G4|W9I`+AT@yTb
zhf$4eM@IDfQYPf;E3FEj&3aJ$fla39@1fKj^&N70Tkh>s`DtO={4aS@`_j2#mOl9n
zhpN)r(V-1Bh=jMyh|?(=>t;{6V!SrThOUpIAJ8f9Jj|-do@7B7#ob1r@>dERm|Oq$
ziO{RUs_iccZ^vyKE(mG_s4N1TB1FNlsZDd$HL#j-IS&R~Ym=LR9~)vMdm
z1+c7*#>jRxqk@2~rosA!Eu{k!Jtlh^3M#V&orG|3Kij1|2>gguO==1usKfGR&yea`
z8TfXTe6@K!pnTa5oiZA2oxVq9^A^q6R~&9JG=VPbpKiUqL|?3g-ET{g^v%xhD(`*kPAcAiC^n9FU5Z6p&3l}Q$f<$eWJe$d8?9IOJJ4l=s_@b&i3F=k3
z5>7@(+45Oss)dH>A%zXQ@0o1VhH=A_&r=vx&S~wms;N%-w>mb^BlHRUkAvN+c?`a2
zm~midR6ZQb^BTK?ks1|cF2zJfp4b>)ZwpQ$dc(P=!;oOp++|y(hRuI7XCUPbB@as+
zE89?4c9xtyrehO+F2h4YZYMyDRMYYOy2UuJ80~Uuhd=YoYG$W&<
z8@92mMFR?n@urgC|6jJc9~wT!sCR35v>p%M9(+qQ*{o~pw|_+6Qz1`h)@aRDCXd^$
zG`ZN#sja``jv`|TUngu}C&x;xXHB2_`snbL6@7^>Xt9u!zR&cAF`4b$P1wkNZ8P*RybSt=dY2@TMHVC
zI!4C<&|r|R&8FYXu%_f*qZ=5jCjGj!X#Q?YsV7CdggGa^bkw||N6S7Klcs8@bqcJ|1C#MrR>9GyoqWm1~q
zY_0Uq-kc9Z@=i%Y?c+02jlzjDsk+?9P4K7{K?h(p5jF`QiWz4XG-PF46o5GOnK8?3s}myGf2l@+17{4kTl%gqg*eb_}sAe?3@6wh=BQ)=1E6S8VRx}?{1$kZB1g3d!E8=;`g`$J~oJj
zJXje3=P-x@Gt)l4mvUGgDXd=nr@>YB*uqR=@KQ*sMs4*-TugLhYQ>3~mF7Vhfm-0k
zf8d+MFpEMdC-bcCZL83g=n4Len{1I3
zMw8xojl;?a#mgX@J4@XqN^_FU+be2G2<@D;F^17l#V9kkm9CvgNq=8TH%*!mNjc+9
zSWrnM9u?c%W`LIuuzmuaOGluYeVe#iB>sbaZXJA)v!N$gk4tT?eJ50*%@c2Uy)`ml
zUnyPD6YR5dPrz|s9@^0|c7MBIT6gzB?X7o#b_R77H!v7%4PvrAh0}PZnmmoDllZoH
zOM^2tEFabnC#%_;
z*l(jx0}Op1X7Zy&wQ@qWR(GX9W%TtgM^Ysn%+w=>D4zZcsRiLY$c>Usi@
zmHEB*QTtLGm_f&ZWz{S%Hfv6tV#>6?P^ZeuV+twakC#8t8eW=x*E8mgH6Pi>oGP24
z1w$+{8+4Jb*K=a#_U_5gB6+z6mXDU(0wKn&bk=;h0H5E228IBRw(%W4E&9BkoDkC2
znnmK`-LrBoPF=H?C~g7w#ZmZMKrmPj^X;A^l7Zg(U$Mz0XhEo*&&u^o5f>$E;-?hU
z0eM-G&vj?^hx5l@AejxxZ(B>gEo!+~OX$g8yI_C(c>bzTpcjYpw>Bu<+qqrz?|Ns4
zhvnVoP)Vuxt0k7!p&cdit!M(&ahc@x*h$a&3fR|VZ6leB>n_l+ZT-Bds_ML1pPpJx
z6s(jzrYJj+A+z(O#3R*PemvZn>P`leHtky!X37v8t6Ju2{OSs`p%(0+We@wvW7e+s
z^yz8vw(?7R(gyliRy&N@*D=p^f>Jml*F?pEW$w8~aJPh^qEXA!X0M$C0Acu%=22;?
zGnlsCA$g(^CdYIkX=|zc0X5%1Gd7%(g2f<)j#LKPATFxE3oVDe(;sv!kFfrlF(K4u
z-{fa(p2{E&^$=gGlgv^*AlSSMjP^l!sJuPLm52;!FM|^}gQn^rv&e1(f&)b*Bz7SS
zEmq!KyIgrJ=&ku(i80Bc02@T;*DSm(g$<{!;HHQ-jH$nH&|eWAjmj9DY>ta}Xg&f=
zEADxSFsP!yY~NSh&D*#GHC+fc*=#z&%whc3UY`7X9LWD2g3=R7*O4#yR&Pa$h&-O*
zdzu?%6c5vL#DC=)TsS3KU54Hfs+_)-IV-gi+&+*sRks%=`TX2utnbv=HMOcocgBZj
z{8q~bIA{m{){4(t^6}1NrbH}FwG}rUXLE-&r$Nmz>=P%U^~Dk^NmP7z_{brc+{niZ
zewr1|4tqYPXb-6_Er&WP_?6sDO`M?={5aMtobK^Y6|$j%Fo
zDXw3euVGT~nx6vXdBM(`!ONZgR4TgbOXXsBpI)s|iCMQ7d}FwKWN>xry`DWcOiSSR
z^rfufSS&z@{n^!ReO}6w&Cu<5r}CY
zJ)f%DUo9xZP}|ju$B#B1NcIWk$RSND!coJ>XKgbX|@A(c4N
zmE=emE0lT{C>54aStcmfOnOyEIQ0c|&zhg$lOCd)u)QiNQ+6>Z{?3qr=Q2g!hz2<(
zuP7`?ohKQ$)C;!_DIKneP%L|7Ux}eUCJ*io(#x_X=165pH1%V1aSu`s5+AQ-^jc0F
z*ogfg|IETw9*07ad+sBtE+Pg&IP7V3hlS2;NqOi6Ol+fyiwBsaj0R-@zSLi0joIP8
z(NoLhF|j!;=P58&>HhdJBwBZjfO!$n;I%#W_3xWG%lJcsdV@YsQjOc^S;=8qE0b5s
zl-xtBgytQj<7d#fR(1U30nH<5KoF{pep_n&eF=3Hco_BLujZ|Ug45lXL_Nk|hmk{Q
z>kE2a?~jLAv*%6-NLu(%U?R=!TeKq16FtE)(&=2f1Q@Zf6~kV%OLc6oO-;Tlm?0Fu
z#|UA;o1!D`Za6yt0HpSrqgeqF5UsE_qALBI&`nwUOL950_76*Q4YhUPVY%T;Lc?&S
za_f_oj~IqlM;8UFKX^}#8(8pePJ@9;)3t#rREcOC8)HpV`G(nM5(OoBhJp#xmMsX^
zXEi^CTj1BD)>!^zL&O9Ce9_sM#Y$SHvqMG0GU@H^`L(oz8j7pv>?2=NUWf*IPMQwv
zggzEH?`>uMqCvqr-o`TiU;jz7O3bH^hg3fmgDCR*`#=51w4_1*dt9O-63q9w#!a2S
zmgEsK_jylE6RG%~C%hjHM&xYWQM1~lUrNaH9|-@Z{R*u^l$Tk6(lAB)fi4Yb`s=y&
z?W_F$cv_`-a{!FOBywVLuhRPXlM@Z$^`&5}`)rVlw>8g7d*;8)ASC9tLABX;(bp;hbmB$v8TXp^^c4t4ZA%xt&Wwa={8Q_6z?csFGSy3
z5Ug!9$3{aE$tRN2TLAUq%Fy*OOqI+FE-
z$fT(*xR|y~J{w+#d@FBJ=Xkz%^Sl^Xtgq^wV}y8reV~NE>QU)mE>3&swxlWl(dX~w
zWOMRSrq`3Q4hV2FpK5xRH43VHK6`izua;<>x7lCtV#S)V{cSd84@iIB?)~H=C*iht
zO7~)#a5znVo$cP%R6EuUpGXXnIxJ^vQ6{_(Y9up+7_c5mLAj+mMQh=7%--;+*MbG&~lwmbGlAG;mUfr7sre!A&y*KfqfS&cSvi}}IqnZt-F0yE=tl$tSf=aZEI
zMz}#Op>~)$tU8=yW#2Qv8eVKDd{9kJ{0Au`+hBKC}T`vC08DUN4Bc5R~*p{ORQICZ0QW?Pq
zqvG#I`3La7l+uX`Zr%KQs!`y{0zs8%b3(C~h#ws-T{alHdge9}EcMi?#!@R2ZjlXn
zs=rg!OIildSDutMb6w#3prkDl-rf;T7KRvcf;goM_MEM@E4TlKz?0=8JQjdTYViuO
z(y}o!tWo!aUY8Y=_wGARkJZ=RbyK2uK4USVe7ic8LS$e%*oVW9Uw6yy+84L6zuod9
ziOlnL^7p?qP?GPI@z79jYa&=t>-7dOVLj;l_O`uRpniDU=6&QBI`O1#7|kpF<*=gi
zz66TW7RPIa6kCJ$%wKQf({y&+xffgnk6mZn9Y?(|mq=kF=3A{)SY?SdrvqNaEL`2V_lymy#wOrRr)=>jZh>eNa
zM({aq=RP&Js3|@GeABh7YOLRuh>m71RheEYgO$i9pz#V;1A+8WR^FABiBn`T8>7f^
z7$IY>eH(s)tSrJ>YemujvXiz~g>P*o^VH7#oREtt`KB(@li1yqaV~vGKAcv(`C!MW
zBOPUIJ_NqWaSLcGnV4o#1y!usB{NC_io|5OLoWhl{!@MD3dEI7<9~@n3=PReg5I$N1X=jPz-JBCvU6l@6iOToo(ZUm}Pvs
ze?)9k6FlJ1SM-9dL6j>k&>kx1l-PL4!3dwCBJ+F`WmO(%Q`h*wzI?9{ExjXZft$e+
z^N&4O`HXixK&#k+O#Ny3k6&GWfaY@Ln+a>ffwOx+anq|r*)e>mPi_0S!ktUwVPxgb
z06IIr3G2K=+&tfLG|kvFskB#BQtDfOs$qcWq;Ealo*8ReyZkvwG_9_sMAsgvW=iFb
zi07uXLa)hda~Ax478ve~z%vy-e#EfUTJwRtlUJ};G^f8_Nt%pl@8~f7a#thAVS8xH
zPU8|e&7QAzxoRV*z-ORAZI+>xfIkFuR|XF6&tYueOf_aq3Db&_l}(@1&0k+^(P^CK
zJ||x@Jsqo%_zB{I7XC5h0a2`o#j~u0Y{tbG7it^I1{tiT4&QI
zh6c%DibJ8;@Mxa~4E&>`xg*D|<{gVAI4ya7@Yjdty2W65Qm#FcOt@&rCWf~Sxpjxvh!=;C2lBWAqvhv3ln}ca{CnWkPg+F-Eg#}
zPB}(SXo0{U%#nl2c@*IuEDMp!YM!ze%1W8_EB_7Ja?accr;p-k+Fe*>)zkS%CvylD
zwiQqY`-14n;t;yVNv9o+c*kbi{$=Ky{~|QRT}8GYn~Abv0B%1^FkxnSdgy{ZE~i}{
zae4K%KkxkhisYM6MIePFir?lPlvdYLcYf27LYs(sBB^s{$@b~ANehZ40Fje4CT1;1}G$FY@PW=6nl%eLa<
z(5%fe_(bFyDubMo7kFRDOtffuKlJtDm~|$y;xZ)wncuY;60PINuG@pt4bbse>pP}L
zA?XMl0Biv7ZFZ&%qsQ?EHnv*na+9(trT&mjrBm)miQ@ABf3KvNiiV>W;z}%zyJ-~-
zom4z$NVMh6i%7?LOvB3O_HK-TUl}ty7^KlgfAHslx
zar~552ET5ucNs!N=tjEUd!pP9r!PKLPL^n8)H662cA86QnAGf-TRW&|PRLkmId_n!
zh_Fuir~eWE`r&wAmMfMl^>{!YD*DwlSA9rTqxSW^`8kt=6m9J)u^ED{SEH!ZFX>br
zk9fy~6s#<}i^pY+8pItH{zu}JH_J~$4R2`4PhhCU(}ErO-{Zt=4-;+0aD}u9^a#WJ-7Azkt714Ev&u8+~krc-oF-+uu?@w{8)yG
z@AGM53}tm`z>Vl`&9uW@nYj}OI9o{R&AYuEODH?%!Y3L`((EzWSYkXiR)v*{0o&)S
zIkHLBi(kG{RqU6GQX<p2c%8_nNhw-jxS)#?DCQqJpztHy1rb4eOAN{V)p%$=F)m+OIoTwsb^91$sAP
zG+v)9Hmg_sX!kFKEh8NxUv>-_?kVfe$YQ)kKPqN<1?Yvk-Xm?`6t3Cw
zIuj;i5sq+fpxOr%gu)^Rj1p~Nym(Qu>^DMM@y*SP|J-sMu>G5ZE9m9V@>egy8dxkg
zOD~DCytR7J$6Foo9vd5&!yH3~hu+Bh6J>U+$H}j#&>%;}_I>aPvFSG<6|`cyQ_dd!
zH@GIv$W}F`E|HDuX=&Nwm*h$8fPV!P@%a*|N8knG-#p0RVzcIv=34ygP?Ny(^?@qV
zH)xX2^qTZ;31-ebRrOo24&MHib&`wKBMG&s_2Q=Au)CJ-dgR{2e;C9h!uR~@W
zMwtPk`OcU#vN+k}3nv^YptV!>w_05hl7v#}1Tspua)gw<^gJd!1-qWeN(*ai(?|)J
zrNS4|njb2Iwy%z(j0W|8e4UKd+ZtB)x3A?DTPEW024`o`g^NA)xZ8s8KL4sD1#VK
z|0^-TT4i_cEMK$j*gD>eFTDM
zT9&TC#n7$5tTS#o>hFqGjue{aMnx&*lDq~T6Y2y$jWJoS%}t!1FiysS!zm_*i}hCQ
zam_9P8qYZ~wV{G4y36)W?%4cO{gePDao-AZ^FMO_XX9iJKQ8Lco9
znk82u5bi`~yuqs4WNsG?oC>%1u&bOI{`5+tvgu#1uC)S-xrQ?laM|49jIVs=)Q(&|q6l3iA1WaH52S1Ck%4Sn(xMOpwFipa#o&
zE=)CtsyiS@{k4M(izrLYe2>{<74pi5_&%(O9>n7^zT7*PjD}D`u
zobjH#HlShCMo%rG##FnekJ3N){`DdWy>3#3!}_sqSq>wZo4%E2AeTOH4+fR?0?O7Jk9cT&%BhpN$nkSBmgwn6#nFr_t
zqAwHlZfE}_gVE~YnTwD|3VON0(4nm-h!oi@0bw>ovm%!D^n|jc04PpIXCrA8G3g5V
z@!(j`K?nt3znxi#824u{=3s{ymE+-Yfk^FvJFz%d4yhJpD~TY?E9Ql9?hsNlP4&9y
z#W$7U%cWdAE2*_~HYH)Dr_KF?5UrizSInB(-pGW|^eFUG^~@&D-B*_xA2>4|CGfmb
zZkQ6KY0Xc*wz;W0=!Y8SV-D>*OnowgaFoWH@>GNv10cvGlz25r
z+&Y^kVqU-OnG>e*t46q;gWfQ)#+E;G7ZG8Vsmd@mNJ-jAV=Q*RF~q+ZA)7Yoa=
z_Wt46WChB+`zPyG{PESYy_YU}=3f)t@2)b5a(3t;s=h(L%#LWiiqbm{RN`;E;vPyV
z4pCehr(Y;b|87}a^0WHigDhpZy7~pZ9ws5w60KxrYJB^gpGbkSYtN3skCU%%NlZng
zj(6nD=*x{+4=;zm9?o{!4l7@~I`s}C1>2eKNK$*R5;XA8)!CwEtIbh9N6?Al-pt;x
zlTItKt8^NuY!Ftic8Wp>w*eW}TbaYe4jY8bH3RNW=p~65A1QWei;ZY!699`dBZc}?
z8w%?NPJ6PH59a!6z>l=E6LaPi8EQ1(ZXF+9`-
z8y6s9csIhAf<`ts#+f>f=WFfy}PZ>8!F1bgAJni!uCbp>KNZ
zh)PjHqUbz0|J|~!{&GPNDFOIZg5_O^F-7Q%7e3mt7ky^9=GrHK9AfO$iD$p?aD{L8
zK{_Q>tX~($T*5|)UBa?Hy?qwMKZ9O-QDm{xe{;r9?IN`4TPBKPUaJ&^>pA0Qf;kA9
z69pa%d#eu?U2^mibgzmuJL75>3J`=`fX4*XF#&XZYy>_z`_C5)IoGb$f)g_GBi6E9
zyF*s20SxA4?U|!DN-=NNq;I%g;{-x3d8PsTP6Be;qY(a^bG4fds+YD>84;1br8&X3
z%R|b{%xoznmk98s?+wI?`geNrDMau4pOrND?4cKN2SMLNRX#FhE{t5wl-Q>ON=yE6
zi@O2H0B9m163?-G`rUQ!OQ|;$qe{`wrlUdM_z{hNR(-hW*{iVk0qC10f2)V?bzx@j
zB>EQHSO%1q4~z$(9&Wce=j{i5w3WF3^88&=@Z%X>0KVFD1*D*YO|)u7sQfvOvvo{c
zXE`YCf|BA{k