From 680c616ec4c026d322460761c10c1f89d71a61c7 Mon Sep 17 00:00:00 2001 From: brandenchan Date: Mon, 20 Sep 2021 16:05:46 +0200 Subject: [PATCH 1/7] Document ODFE and OpenSearch --- docs/latest/components/document_store.mdx | 66 +++++++++++++++++++++-- 1 file changed, 62 insertions(+), 4 deletions(-) diff --git a/docs/latest/components/document_store.mdx b/docs/latest/components/document_store.mdx index c9b618662..91e4e7fe9 100644 --- a/docs/latest/components/document_store.mdx +++ b/docs/latest/components/document_store.mdx @@ -29,12 +29,48 @@ docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.9.2 Next you can initialize the Haystack object that will connect to this instance. ```python +from haystack.document_store import ElasticSearchDocumentStore + document_store = ElasticsearchDocumentStore() ``` -Note that we also support [Open Distro for Elasticsearch](https://opendistro.github.io/for-elasticsearch-docs/). -Follow [their documentation](https://opendistro.github.io/for-elasticsearch-docs/docs/install/) -to run it and connect to it using Haystack's `OpenDistroElasticsearchDocumentStore` class. +### Open Distro for Elasticsearch + +Learn how to get started [here](https://opendistro.github.io/for-elasticsearch-docs/#get-started) + +If you have Docker set up, we recommend pulling the Docker image and running it. + +```bash +docker pull amazon/opendistro-for-elasticsearch:1.13.2 +docker run -p 9200:9200 -p 9600:9600 -e "discovery.type=single-node" amazon/opendistro-for-elasticsearch:1.13.2 +``` + +Next you can initialize the Haystack object that will connect to this instance. + +```python +from haystack.document_store import OpenDistroElasticsearchDocumentStore + +document_store = OpenDistroElasticsearchDocumentStore() +``` + +### OpenSearch + +Learn how to get started [here](https://opensearch.org/docs/#docker-quickstart) + +If you have Docker set up, we recommend pulling the Docker image and running it. + +```bash +docker pull opensearchproject/opensearch:1.0.1 +docker run -p 9200:9200 -p 9600:9600 -e "discovery.type=single-node" opensearchproject/opensearch:1.0.1 +``` + +Next you can initialize the Haystack object that will connect to this instance. + +```bash +from haystack.document_store import OpenSearchDocumentStore + +document_store = OpenSearchDocumentStore() +```
@@ -210,12 +246,34 @@ The Document Stores have different characteristics. You should choose one depend - Fast & accurate sparse retrieval with many tuning options - Basic support for dense retrieval - Production-ready -- Support also for Open Distro **Cons:** - Slow for dense retrieval with more than ~ 1 Mio documents +### Open Distro for Elasticsearch + +**Pros:** + +- Fully open source (Apache 2.0 license) +- Essentially the same features as Elasticsearch + +**Cons:** + +- Slow for dense retrieval with more than ~ 1 Mio documents + +### OpenSearch + +**Pros:** + +- Fully open source (Apache 2.0 license) +- Essentially the same features as Elasticsearch +- Has more support for vector similarity comparisons and approximate nearest neighbours algorithms + +**Cons:** + +- Not as optimized as dedicated vector similarity options like Milvus and FAISS +
### Milvus From 98f00125132094a98336e4ac4e3dd03e2a6dafb8 Mon Sep 17 00:00:00 2001 From: brandenchan Date: Mon, 20 Sep 2021 16:54:15 +0200 Subject: [PATCH 2/7] Document new converters --- docs/latest/components/preprocessing.mdx | 33 +++++++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) diff --git a/docs/latest/components/preprocessing.mdx b/docs/latest/components/preprocessing.mdx index 4043b33f6..3c0c4acd1 100644 --- a/docs/latest/components/preprocessing.mdx +++ b/docs/latest/components/preprocessing.mdx @@ -50,6 +50,17 @@ Please refer to [the API docs](/reference/file-converters) to see which converte valid_languages=["de","en"]) doc = converter.convert(file_path=file, meta=None) + + # Alternatively, if you have a PDF containing images, Haystack uses tessaract under the hood to OCR image PDFs. + + + from haystack.file_converter import PDFToTextOCRConverter + + + converter = PDFToTextOCRConverter(remove_numeric_tables=False, + valid_languages=["deu","eng"]) + + doc = converter.convert(file_path=file, meta=None) ), }, @@ -71,7 +82,7 @@ Please refer to [the API docs](/reference/file-converters) to see which converte content: (

- Haystack also has a`convert_files_to_dicts()` utility function that + Haystack also has a `convert_files_to_dicts()` utility function that will convert all txt or pdf files in a given folder into this dictionary format.

@@ -84,6 +95,26 @@ Please refer to [the API docs](/reference/file-converters) to see which converte
), }, + { + title: "Image", + content: ( +
+

+ Haystack supports extraction of text from images using OCR. +

+ 
+            
+              from haystack.file_converter import ImageToTextConverter
+            
+            
+            converter = ImageToTextConverter(remove_numeric_tables=True,
+            valid_languages=["de","en"])
+          
+          doc = converter.convert(file_path=file, meta=None)
+
+
+ ), + }, ]} /> From a21bec7f580a7448d798b70dafc8179a0b6f7d1e Mon Sep 17 00:00:00 2001 From: brandenchan Date: Mon, 20 Sep 2021 17:21:25 +0200 Subject: [PATCH 3/7] Document most similar documents pipeline --- .../components/ready_made_pipelines.mdx | 33 +++++++++++++++++-- 1 file changed, 31 insertions(+), 2 deletions(-) diff --git a/docs/latest/components/ready_made_pipelines.mdx b/docs/latest/components/ready_made_pipelines.mdx index 8cd1438bf..c2a650773 100644 --- a/docs/latest/components/ready_made_pipelines.mdx +++ b/docs/latest/components/ready_made_pipelines.mdx @@ -43,7 +43,7 @@ We typically pass the output of the Retriever to another component such as the R `DocumentSearchPipeline` wraps the [Retriever](/components/retriever) into a pipeline. Note that this wrapper does not endow the Retrievers with additional functionality but instead allows them to be used consistently with other Haystack Pipeline objects and with the same familiar syntax. Creating this pipeline is as simple as passing the Retriever into the pipeline’s constructor: -```python +``` python pipeline = DocumentSearchPipeline(retriever=retriever) query = "Tell me something about that time when they play chess." @@ -128,7 +128,7 @@ result = pipeline.run(query=query, params={"retriever": {"top_k": 10}, "reader": You may access the answer and other information like the model’s confidence and original context via the `answers` key, in this manner: -```python +``` python result["answers"] >>> [{'answer': 'der Klang der Musik', 'score': 9.269367218017578, @@ -209,4 +209,33 @@ Output: ], ... } +``` + +## MostSimilarDocumentsPipeline + +This pipeline is used to find the most similar documents to a given document in your document store. + +You will need to first make sure that your indexed documents have attached embeddings. +You can generate and store their embeddings using the `DocumentStore.update_embeddings()` method. + +``` python +from haystack.pipeline import MostSimilarDocumentsPipeline + +msd_pipeline = MostSimilarDocumentsPipeline(document_store) +result = msd_pipeline.run(document_ids=[doc_id1, doc_id2, ...]) +print(result) +``` + +Output: + +``` python +[[ + {'text': "Southern California's economy is diver...", + 'score': 0.8605178832348279, + 'question': None, + 'meta': {'name': 'Southern_California'}, + 'embedding': ..., + 'id': '6e26b1b78c48efc6dd6c888e72d0970b'}, + ... +]] ``` \ No newline at end of file From e944e76fd816aa0462fba984ae8d93ae734b2c69 Mon Sep 17 00:00:00 2001 From: brandenchan Date: Tue, 21 Sep 2021 17:00:48 +0200 Subject: [PATCH 4/7] Add evaluation guide --- docs/latest/guides/evaluation.mdx | 92 +++++++++++++++++++++++++++++++ docs/latest/menu.json | 46 +++++----------- lib/constants.ts | 15 +++++ 3 files changed, 120 insertions(+), 33 deletions(-) create mode 100644 docs/latest/guides/evaluation.mdx diff --git a/docs/latest/guides/evaluation.mdx b/docs/latest/guides/evaluation.mdx new file mode 100644 index 000000000..9e45c3cbd --- /dev/null +++ b/docs/latest/guides/evaluation.mdx @@ -0,0 +1,92 @@ +# Evaluation + +Haystack has all the tools needed to evaluate Retrievers, Readers and Generators in both +open domain and closed domain modes. +Evaluation and the metrics that it generates are vital for: +- judging how well your system is performing on a given domain. +- comparing the performance of different models +- identifying underperforming components in your pipeline + +
+ +**Tutorial:** This documentation page is meant to give an in depth understanding of the concepts involved in evaluation. +To get started using Haystack for evaluation, we recommend having a look at our [evaluation tutorial](/tutorials/evaluation) + +
+ +## Datasets + +Annotated datasets are crucial for evaluating the retrieval as well as the question answering capabilities of your system. +Haystack is designed to work with question answering datasets that follow SQuAD format. +Please check out our [annotation tool](/guides/annotation) if you're interested in creating your own dataset. + +
+ +**Data Tool:** have a look at our `SquadData` object in `haystack/squad_data.py` if you'd like to manipulate SQuAD style data using Pandas dataframes. + +
+ +## Open vs Closed Domain + +There are two evaluation modes known as **open domain** and **closed domain.** + +**Closed domain** means single document QA. +In this setting, you want to make sure the correct instance of a string is highlighted as the answer. +So you compare the indexes of predicted against labeled answers. +Even if the two strings have identical content, if they occur in different documents, +or in different positions in the same document, they count as wrong. + +**Open domain** means multiple-document QA (typically over the entire database). +Here, you only look for a match or overlap between the two answer strings. +Even if the predicted answer is extracted from a different position than the correct answer, +that's fine as long as the strings match. + +## Metrics: Retrieval + +### Recall + +Recall measures how many times the correct document was among the retrieved documents over a set of queries. +For a single query, the output is binary: either a document is contained in the selection, or it is not. +Over the entire dataset, the recall score amounts to a number between zero (no query retrieved the right document) and one (all queries retrieved the right documents). + +Note that recall is affected by the number of documents that the retriever returns. +If the retriever returns only one or a few documents, it is a tougher task to retrieve correct documents. +Make sure to set the Retriever's `top_k` to an appropriate value and to also define the `top_k` in `Retriever.eval()` or `EvalDocuments` + +### Mean Reciprocal Rank (MRR) + +In contrast to the recall metric, mean reciprocal rank takes the position of the top correctly retrieved document (the “rank”) into account. +It does this to account for the fact that a query elicits multiple responses of varying relevance. +Like recall, MRR can be a value between zero (no matches) and one (the system retrieved a correct document for all queries as the top result). +For more details, check out [this page](https://en.wikipedia.org/wiki/Mean_reciprocal_rank) + +### Mean Average Precision (mAP) + +Mean average precision is similar to mean reciprocal rank but takes into account the position of every correctly retrieved document. +Like MRR, mAP can be a value between zero (no matches) and one (the system retrieved correct documents for all top results). +mAP is particularly useful in cases where there are more than one correct document to be retrieved. +For more details, check out [this page](https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Mean_average_precision) + + +## Metrics: Question Answering + +### Exact Match (EM) + +Exact match measures the proportion of cases where the predicted answer is identical to the correct answer. +For example, for the annotated question answer pair “What is Haystack?" + "A question answering library in Python”, +even a predicted answer like “A Python question answering library” would yield a zero score because it does not match the expected answer 100 percent. + +### F1 + +The F1 score is more forgiving and measures the word overlap between the labeled and the predicted answer. +Whenever the EM is 1, F1 will also be 1. +To learn more about the F1 score, check out this guide + +### Semantic Answer Similarity (SAS) + +Semantic Answer Similarity uses a transformer-based cross-encoder architecture to evaluate the semantic similarity of two answers rather than their lexical overlap. +While F1 and EM would both score “one hundred percent” as sharing zero similarity with “100 %", SAS is trained to assign this a high score. + +SAS is particularly useful to seek out cases where F1 doesn't give a good indication of the validity of a predicted answer. + +You can read more about SAS in [this paper](https://arxiv.org/abs/2108.06130). \ No newline at end of file diff --git a/docs/latest/menu.json b/docs/latest/menu.json index fbdbefc34..efa4b7ccd 100644 --- a/docs/latest/menu.json +++ b/docs/latest/menu.json @@ -39,6 +39,7 @@ {"slug": "languages", "title": "Languages Other Than English"}, {"slug": "domain-adaptation","title": "Domain Adaptation"}, {"slug": "optimization", "title": "Optimization"}, + {"slug": "evaluation", "title": "Evaluation"}, {"slug": "annotation", "title": "Annotation Tool"}, {"slug": "rest-api", "title": "REST API"}, {"slug": "chatbots", "title": "Chatbot Integration"} @@ -107,43 +108,22 @@ "subMenuTitle": "API Reference", "pathPrefix": "/reference/", "items": [ - { - "slug": "document-store", - "title": "Document Store" - }, + {"slug": "document-store", "title": "Document Store"}, { "slug": "retriever", "title": "Retriever" }, { "slug": "reader", "title": "Reader" }, { "slug": "generator", "title": "Generator" }, - { - "slug": "summarizer", - "title": "Summarizer" - }, - { - "slug": "translator", - "title": "Translator" - }, - { - "slug": "preprocessor", - "title": "Preprocessor" - }, - { - "slug": "file-converters", - "title": "File Converters" - }, - { "slug": "crawler", "title": "Crawler" }, - { - "slug": "evaluation", - "title": "Evaluation" - }, + {"slug": "summarizer", "title": "Summarizer"}, + {"slug": "translator", "title": "Translator"}, + {"slug": "preprocessor", "title": "Preprocessor"}, + {"slug": "file-converters", "title": "File Converters"}, + {"slug": "crawler", "title": "Crawler" }, + {"slug": "evaluation", "title": "Evaluation"}, { "slug": "pipelines", "title": "Pipelines" }, - { - "slug": "knowledge-graph", - "title": "Knowledge Graph" - }, - { - "slug": "graph-retriever", - "title": "Graph Retriever" - } + {"slug": "knowledge-graph", "title": "Knowledge Graph"}, + {"slug": "graph-retriever", "title": "Graph Retriever"}, + {"slug": "classifier", "title": "Classifier"}, + {"slug": "question-generator", "title": "Question Generator"}, + {"slug": "ranker", "title": "Ranker"} ] } ] diff --git a/lib/constants.ts b/lib/constants.ts index 2b5f01b6c..8d6257411 100644 --- a/lib/constants.ts +++ b/lib/constants.ts @@ -75,6 +75,21 @@ export const referenceFiles: Meta = { filename: "graph_retriever.md", title: "Graph Retriever", }, + { + slug: "question-generator", + filename: "question_generator.md", + title: "Question Generator", + }, + { + slug: "classifier", + filename: "classifier.md", + title: "Classifier", + }, + { + slug: "ranker", + filename: "ranker.md", + title: "Ranker", + }, ], }; From 81876a4efe4ba8794231604ba0c2d9351881f6b6 Mon Sep 17 00:00:00 2001 From: brandenchan Date: Tue, 21 Sep 2021 17:02:28 +0200 Subject: [PATCH 5/7] Change order of sections --- docs/latest/guides/evaluation.mdx | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/docs/latest/guides/evaluation.mdx b/docs/latest/guides/evaluation.mdx index 9e45c3cbd..ea93dbf8e 100644 --- a/docs/latest/guides/evaluation.mdx +++ b/docs/latest/guides/evaluation.mdx @@ -14,18 +14,6 @@ To get started using Haystack for evaluation, we recommend having a look at our
-## Datasets - -Annotated datasets are crucial for evaluating the retrieval as well as the question answering capabilities of your system. -Haystack is designed to work with question answering datasets that follow SQuAD format. -Please check out our [annotation tool](/guides/annotation) if you're interested in creating your own dataset. - -
- -**Data Tool:** have a look at our `SquadData` object in `haystack/squad_data.py` if you'd like to manipulate SQuAD style data using Pandas dataframes. - -
- ## Open vs Closed Domain There are two evaluation modes known as **open domain** and **closed domain.** @@ -89,4 +77,16 @@ While F1 and EM would both score “one hundred percent” as sharing zero simil SAS is particularly useful to seek out cases where F1 doesn't give a good indication of the validity of a predicted answer. -You can read more about SAS in [this paper](https://arxiv.org/abs/2108.06130). \ No newline at end of file +You can read more about SAS in [this paper](https://arxiv.org/abs/2108.06130). + +## Datasets + +Annotated datasets are crucial for evaluating the retrieval as well as the question answering capabilities of your system. +Haystack is designed to work with question answering datasets that follow SQuAD format. +Please check out our [annotation tool](/guides/annotation) if you're interested in creating your own dataset. + +
+ +**Data Tool:** have a look at our `SquadData` object in `haystack/squad_data.py` if you'd like to manipulate SQuAD style data using Pandas dataframes. + +
\ No newline at end of file From ebbbe1860a36ed183657225490812570015e7dc5 Mon Sep 17 00:00:00 2001 From: brandenchan Date: Tue, 21 Sep 2021 17:36:35 +0200 Subject: [PATCH 6/7] Add generator page --- docs/latest/components/generator.mdx | 44 ++++++++++++++++++++++++---- docs/latest/menu.json | 1 + 2 files changed, 40 insertions(+), 5 deletions(-) diff --git a/docs/latest/components/generator.mdx b/docs/latest/components/generator.mdx index a0b1c639a..73bab5c44 100644 --- a/docs/latest/components/generator.mdx +++ b/docs/latest/components/generator.mdx @@ -2,27 +2,61 @@ While extractive QA highlights the span of text that answers a query, generative QA can return a novel text answer that it has composed. + The best current approaches, such as [Retriever-Augmented Generation](https://arxiv.org/abs/2005.11401) and [LFQA](https://yjernite.github.io/lfqa.html), can draw upon both the knowledge it gained during language model pretraining (parametric memory) as well as passages provided to it with a retriever (non-parametric memory). + With the advent of Transformer based retrieval methods such as [Dense Passage Retrieval](https://arxiv.org/abs/2004.04906), retriever and generator can be trained concurrently from the one loss signal.
-**Tutorial** - -Checkout our tutorial notebooks for a guide on how to build your own generative QA system with RAG ([here](/tutorials/retrieval-augmented-generation)) +**Tutorial:** Checkout our tutorial notebooks for a guide on how to build your own generative QA system with RAG ([here](/tutorials/retrieval-augmented-generation)) or with LFQA ([here](/tutorials/pipelines)).
-Pros +**Pros** - More appropriately phrased answers - Able to synthesize information from different texts - Can draw on latent knowledge stored in language model -Cons +**Cons** - Not easy to track what piece of information the generator is basing its response off of + +## Usage + +Initialize a Generator as follows: + +``` python +from haystack.generator.transformers import RAGenerator + +generator = RAGenerator( + model_name_or_path="facebook/rag-sequence-nq", + retriever=dpr_retriever, + top_k=1, + min_length=2 +) +``` + +Running a Generator in a pipeline: + +``` python +from haystack.pipeline import GenerativeQAPipeline + +pipeline = GenerativeQAPipeline(generator=generator, retriever=dpr_retriever) +result = pipelines.run(query='What are the best party games for adults?', top_k_retriever=20) +``` + +Running a stand-alone Generator: + +``` python +result = generator.predict( + query='What are the best party games for adults?', + documents=[doc1, doc2, doc3...], + top_k=top_k +) +``` diff --git a/docs/latest/menu.json b/docs/latest/menu.json index efa4b7ccd..4b8768109 100644 --- a/docs/latest/menu.json +++ b/docs/latest/menu.json @@ -26,6 +26,7 @@ {"slug": "reader", "title": "Reader"}, {"slug": "generator", "title": "Generator" }, {"slug": "summarizer", "title": "Summarizer"}, + {"slug": "classifier", "title": "Classifier"}, {"slug": "translator", "title": "Translator"}, {"slug": "knowledge-graph", "title": "Knowledge Graph"}, {"slug": "ranker", "title": "Ranker"}, From e4f8323da6651a1f62d0813341df9af2076353e4 Mon Sep 17 00:00:00 2001 From: brandenchan Date: Tue, 21 Sep 2021 17:47:57 +0200 Subject: [PATCH 7/7] Add classifier page --- docs/latest/components/classifier.mdx | 43 ++++++++++++++++++++++++++ public/img/classifier.png | Bin 0 -> 91169 bytes 2 files changed, 43 insertions(+) create mode 100644 docs/latest/components/classifier.mdx create mode 100644 public/img/classifier.png diff --git a/docs/latest/components/classifier.mdx b/docs/latest/components/classifier.mdx new file mode 100644 index 000000000..55c793f40 --- /dev/null +++ b/docs/latest/components/classifier.mdx @@ -0,0 +1,43 @@ +# Classifier + +The Classifier Node is a transformer based classification model used to create predictions that can be attached to retrieved documents as metadata. +For example, by using a sentiment model, you can label each document as being either positive or negative in sentiment. +Through a tight integration with the HuggingFace model hub, you can easily load any classification model by simply supplying the model name. + +![image](/img/classifier.png) + +
+ +Note that the Classifier is different from the Query Classifier. +While the Query Classifier categorizes incoming queries in order to route them to different parts of the pipeline, +the Classifier is used to create classification labels that can be attached to retrieved documents as metadata. + +
+ +## Usage + +Initialize it as follows: + +``` python +from haystack.classifier import FARMClassifier + +classifier_model = 'textattack/bert-base-uncased-imdb' +classifier = FARMClassifier(model_name_or_path=classifier_model) +``` + +It slotted into a pipeline as follows: + +``` python +pipeline = Pipeline() +pipeline.add_node(component=retriever, name="Retriever", inputs=["Query"]) +pipeline.add_node(component=classifier, name='Classifier', inputs=['Retriever']) +``` + +It can also be run in isolation: + +``` python +documents = classifier.predict( + query="", + documents = [doc1, doc2, doc3, ...] +): +``` \ No newline at end of file diff --git a/public/img/classifier.png b/public/img/classifier.png new file mode 100644 index 0000000000000000000000000000000000000000..6897f5d7a55dabb454e078ef0f172d603eaa78f3 GIT binary patch literal 91169 zcmbrmbySpH)HXhfiYTC>A|j442uMf>NDT1^I+TD&NJxV;NXXC_fJls>bPGs>lG3Gg zmvl&XgY>s=ed7Jr`>o$UKbFgd2_Qo^E*T6!RP3S8s27dov_neMuwxOqBZ`&FO-_OahNh zf7(o*=}qBa+T4wJKh(Z?L03UhPw#$}&l~5`{hG5O5djx3hg^I%dUnXc{PSaGY8Vdu zl$q?BAR~hx`MdZdzkhLkdz|q57nJOoku!gOW4w<2^ETmouW^4~N_?|){LgRa&!7+e z{w?}?1?JC7@}mF$FZS_j*q@8xAAgCHWeC2IT5Q@O$b3m;8-HO@qW`aO< zxds6)k zCM!{b{7# zM5>Hzj_#J@U~%3q^&z}FpL%ki@?zk8Z9V0L)8|ln6iRknOEO~ROWEC?iiSs1%FU1M zit3z6mvV16@r?T6(0=3cjGY}F@|9j4T$&s1WS5!*Hq`_@?=#RuYkzgS^<&o~w>Y}$ zLe!l*scSN8X>JbUI8-i?=4w&~Z(`c}8>~M&e|$00JGZJgV}#Bak-@IY?RFn38>tvs zI*2-S)L81p%n3rB`@{=3eP72TxQN%y&-fS58sRf@WJc5MX7i);Z{dul2JV-ZGa*zGFz}v+C9k609c+otrKcB8{6Ph>^ z@OLQt-uvg}{}H!TtN(=eL%aWvjiKG}&nV83d+3!fEG0f>u6x-$IjZyLi!A6&18`f8 z>Xh@7(9vzf}i{J(DDz{L3P>oV^J;G{ZF z{r=bRV{28IVfYcg_up^w|8C3k=&il|P>t7r@7S`im*vsSEv96@KY|(${^z&z>=XZN zcew^Q3FQ*?{QcNGXC!*^513R zO#hiSo~T?iW76P1Lm77T{&%kbU-M7JcInSJ&bR(IX04x_LCKgOO}*hl?n{m&b|zR9ioA9woCwS@2eH~Rn9 z?!Q%lK=|)Q|68X2e*5pO{|(^3LH!dL(bxa`t&oYZvW2~L`S(4RvHbU~|9p|_+yCb5 z-$1)w_3!5RZ^Zv?ow7$x|>|XyTv)r9c$8{&^ zC8n?vbHew}MZ*c~|Ad>LT|;cB*NcBfTT~l9(xK+~ZyXHjl6xDWH{Se{Jj~lP{=P$0 z^?xHXOu@>DT>fV_xax!uVg8x)VX}sc`Z3Q+^bW9JZx-bpOC@ReD|-5)g*I4b`o3Rc%Z7_iXWlIl82_`*H5pZgZ<4Yiwcql z4;~EeG*fzgm!f+woiMJLw6efnO=GpcHM!#`Sc1&#=NC70(4FVe?7z;S%GkV1x57&8 zreZvtx7(wr-Ti8HOCD)E-__6_eq&%@AhV^lb!2lVhvqMOI=U*UPoF+f zcz%WkPvQJR~Nvv)eDy|i~N+KlFNm?e)}*gm_@zlSER#2NO*YB?n?b) z(P~qOOm~j)kNLbI^~Tv$=4tquoAdsz4Z|L>;5mK#+zB_eTR|Dp++h)sky z@F^j|lqV4Kk|rA!Lr7?-9N|3;WWF-)iL%5Ab^;QPpYLPQrK7(5+F!bn1DYScsauEO zw7Xn=#c8dz8tq`x94XLQRH0ehE+aJWLng;?jw6Ir135#cOg|*+Az{vQmm}l@b*G4= z;IHyi$wZxl#SS1HWcvfYh& zcY}3-DQP4_H|`p-S^5uk2cJdcPUcL~Q^rFpF)k72q$r*mA8&6phxr^Ob-Vcf9B%B2 zf4pQ+*KIN6FqI{eEJ#B(`3!~EcX@NJidf!Mm7KRe@L#B~klW#*th!z2(XB3IsMDUJ zq?NKdPbA-p;yKUWUW2q&<_sr~-^D|y{R1M;p&@_{4CmUWvy-u*yqesBtQt-C;VQ0; z$g)@*v06Se(8V}CKd*;@SIhK~Y-4fc;s!&ARPqu76=|k&9kXh#_n&W`Ud+9Jz=d+v zqG~mecf)ZUzY>7Eos*YWU$M5v`TMxoh=v?qb|!8`jN?o}n!?Z3EnO`Oi+8^dLI>oS z8a_TzhpC8_kk4OeX*vDpGn#NfxngAPvaeQFR+=DZYqy9?9+w;~$~775za)m7T5};C ze_2^ufB3(ywDhJf*3QSr=L?cbG*2P>n&w4a?9s~tunW48!{oCEFE%-!3ds?|#feyz zcS%V%xe@$3R~`(I0ry7qx;(m3@-pXH&m;f$4KM^q^jtxAI_|cM$sxP4?AV{FNy19R z%jK)3KW0Am>aVEgrH1$rgmrQc`@bs7p(b9Cu}vLP{YDHIy#&ER)n>3Vk~k$ zC{)8hm&q;ug7q%d&4qy-*Z}$3Oi4=&?`xML4xxA^m)3?~o!a02QL4`&Tgg_k65``oq;pIA~3lg_o_QNlTpU?N&^ymd`?Kxja%bm&9m%)E%)W2dk=GQLG zKi{dO!9_aSB67+`a#k%(NW%u@DzdvwaM`Ud-Ic6l+jTw&Df;1(Mv`OWnJZaUn|uxQFY0jlXUc_7pwF3bi&YT($^Faoc@-tcuZ-BAY$s3_FRl5CttOjlYikd7Tn-lB%LpnnoZDAhX^-0f z;Io=WQOO1%+I!J?vk+EDGpww5O$l~YYUki!$H)W)y$xodkd`LX1jAk}8xFmm>ZewAfZ_1Pli!Fi=TJZFLjc!!wm_Ll7vEj(q zs&vPp?Os1%&-1wA{42N5Y{$dp7cuwI{(66Q0&Ptl)Hxfq$63ar#%3JVGfBz6B9xF6g4<;$0d8~o4bx(&4t zZfdbndEbxyLex-ydofVpde9K0n}YS7J>IjOoeT9E`zu7JuonH#s-31KHG023i-oy6 z*#436_2-EaHVZBQE;Bxc`-fd~?&ENJBafoCJEBg#ySc=2t@5I1Y%3&iKFzLiyC_Zd z!ED2JvOE12bvF~#o_#-mLCvE4=XWocjjX7~9S<(T?3$;H<4?B>>hFGj5+UdP#&YEr z@dK2rmE)qrkR&Zd)UR*!CHw0c&CA((^?{cmD)mlHtZq$TluPQMIR%B3hG&PY z{F_hm$=vR8n*<(g9_$YxgP%!CFzf!T)fmdvmut#D8E!RaXB_=3+s96H^#Y3d4i}fr zfX$EEtAg%Yj_WBJ5$lF&h9k^y{$+18sRTP)HtT07?0Qo8JU=}?Kq7uF|33XlYfa4? zEvgj7G|PI$pV1^J3v#Z;djbK+2SvuuV8u%sDA1j1r1JInvw&+{$6?1SKaE&#vMp&Z zW?>0qBE0shz}T$C!QDhh@M8Td?<8VhDkas$+E@HkslcR192S1{)%q8zvn_LyIwId^ zHqdUH*B-+C-pZMfiGaG;kO$O^gy9>#6&OH{(SL=8g$;b|9kJ+cs>7T|!v;=Izh!$l zH@?n0gAXN~Aw?I9yiuRq@Hl*^?L+`wL8O_nn&?rKael!HWWa7UyN(|}CO_WN{GDdW zt;Wxcv^LQB^fJH3EukQ@yZ-oE+!xvyfOB$ky45Q&GbIiHQkWm!H_&x12@5L}=EhDK zK4es8qRK>gd&9&DWpE_VRy8YOMJ(r85Ua+*$26pvaJ3Q>r0}_)gbnS9WB0(xqsxib z}NZ|^zZXM9bJfNu5C88hQt1$2_mV5VNZ znr$~75^*oi{Wy&dcsO!K54S{Ng^x=>0*fUHZ9CA^HMt7c(Ebt(OFfr+AK;|0>2ZSo zJ^T?F4qMgsM|Ps;DNv6k1MbT_c#v=082`(_-oCe~E;cd}{o`_ONZwJ@@R8TJEPoyA zTS4!uj#ZwkxP)Y(jE6DUk=CO5CzWi0%*sFECWF`yp`?ZJoJ{A@wGYEYxQqU(kc~hf zzj2b&+}K!LVV7D3#-SjMjGR1lsN;80!sGE8=O^}hsU|gb1#0?1VOR!y;4C9y%z!ok z=aasSzH&ur{>g=hwb3R_C={oE8646au!yt>hRb;7kjLe~Nch$kFFAnk{si|te1wAP zp`HBwUN)$$eR~0Tn@7XT*!uVY+;ex_0yRq*oZhanr+?2atd}GZ2peuV`Adgf$&M2y z#tW9jE+>;f(a?Ja-RNwu#>fTeAFY#)&CKWv<^6PB>@WnGZL%{jb$q*FpNfjt1=i<3 z=;VkW*!rxJp_XVXVwG!;e!d$6cT)5kzG^8|^!iBp`P|Z#8k-S6K6#`V%En+r_tw5z zACiJEMwee=CLvH__30r*sK(frYBc{uuujRZ@2qH--DPe^2ItaJS%w=j6>=lIUUUOg5LphJWD2$1@#(g({TwqPs zVNfiTT@HEK{9Reif|)Vo4dKzR#lLrFSKG|ooHfkah~FB3t>HHqU;IYT(-cEAXTa;q zkxy_Ob$W?j&$NIS>EGARW|Z#Fq-3aP={E+I?v7jYML-?WW5N+4MN46s(on=NAYeL~ zYhTSjsP$CGyAFf*glqr{Dm;8{q1UlKg0|NtBn}tfhC89SA#?fi#wt_3FM(1Ily{Qc+o&D zZw`0SIt`J&(vX4Qn`-w59dCY_EK~)DYd*3&(j2~nU}B>$ zwu^%&S&L)`fDe6=W7G+-Kq(|uEx&Ew_n~hxwqSG60pnK#`CE^w%C=-R0!oH)BxYbT zo(hEr7o>5maYPsmSB*Zib(%$X`UiG;49E3I8gWzpkPjmMsqXJ{a&mfheXUhr1wH1_ z-;*Pxp>S|LU6z41t=rGfFW=#9?M#oX9Va{cwUY~` z%?ROk+P}Bv;^mQ68lZViopHSLH?S}7uqge#yLO`TGS8NBHN) zd>{Q8f1|rM-q+6G*vJ@!CDn7`H|)A^%0}e~mGA@^!Q>K~>bP-wY_cZ)6iPvw z>8Z7)v8pd7!N&hFFE>{1p$ao)))xLj9v7YjiZ$VRdj|)y%CDub)7U*61xuax?3Iz4 z@mcldJ3AwB)|~!b3Do&XXXMBI9Da|Q0$t1Afwhk&8iWN6R&3-|7h-Y&9!k!-E+K)V*!~V)dN}52LM**Ol>&tuq=+OTs2wS4cv(KuW%j z^t3biNoTBf4kBp4$D&a`}>!!eWdwcr{hQUBCdTjCpN&>PE z5*X|qju)#UMf{s5v+Sbd-Q<4fe8{j4kYRVZ@|+Hp%?I9kVnkKt#ahC%Dcka#!voj5 zna79?waqrknP@w1HibImDOwRfbO#g*;fuAR+19F7_s0D~Z-^+31MO6PY3(%*h`Hg5 z6HAG(7$hmQJJyDs@p3omQyX4ZU@kEd{A*!0&1ZcjwP^vm?`I@no_AZJaCLnI?2v>&;2^gOIlQVu5K;Gc{Fd@ zN)1URS{(hMdGfM?-Pr5aXd{G63bxaZCl8(ZoDx4j((n^=(WbR&fmJU6?&?+01~-zX zeHupmL>LJ)=g};QeWBS-Fpf6$cM7*RtJG1hRS``AZa8jRRVzun==d4ggF*MDwN>sMS^57C>MPXD^gH z9zwe#QgrPPGYwilDx0f9E@0wr?KW+6jElU87C6{l^|<`}F{JSg8BTQ74T6A))Dq`=(e1@{@hlYz&eIUzBc~-Na z!}qu6H-}tG<;8k3QdbNXmTGpzxz--N*9z!I*npO)_#pM|&J1U+#daunv1oG^sL@&t zI2Es-gt}K9hyklBr|W-`i2Kb6^)a3#>@Rr;g=Jj;TAL(m>F!EU$y$74*q!EQZ@JXv4*0hnOQ1aM=+G_UVy$_`{k~Ssvi63b z*jO!n6jbj|g?4(TrVQeYfcMI^VVcbilup=O91_A}cjvO2sokHA`rL7z)48IWrSCH+ zV`^kn=Qv|b1V~>T*+bfSgZ8r>9UUv<^+W6Fwh9MIAyWG0p5wEt#S}Qxk7Bfzz)0BG z+Rm!HI7${QkR{%Gb%E@61zb!nw7Yc5dFfTy9dm>*(XgMsvC0*f&&9U`z8XFtKR2mzfJoakZTE$*&(lQ^8 zDpE4Ci|${HI+HWs7xFKB9ql#F=-ly3TB?Dqg@h0AKfF~jkfO-me*cNYpf%s#TDwff zlfsQ=m;Ed)s;A-4?sw2qc#f@Zjf{-MP$=^x((wa)vxO=%L2&)%6S4~mcJgqH+y8&Etp-^1sbjC) zh*q{7%V{H|=_yG1(!a?B%=KhFoe%MMF3YVkIO#-h{5k8lvjwNHC{_CJwT1qD@UecGzx>MO|9NIb#9Pylef@F@1K z;#sMNj{bf}vD$u_`fUg3vymjcT_}sbuBVmEGZ!8ow&^f}VVO z8?0Miu~F6C-JK5g_u^o&R^_ouHXx4t3l|EQ{KUR#e~F2S!Ci5Z7g)85+UbXL>es-c z^PO`irniga@w7?=>ehwO905q=1i!_srbAgI{-_QqpN_9B%hr^+1SUv30g<0A$0bMz zw~ChTdNg!hIpNk|RnXTWoh&+Nf3UyT=_ougDl4L@^7mevr^U)e#4wW_2WDDLL1pTS zw3Jj$>jv#hip)xq?l5j`EeGM2o+wb)IpG=bApE zm)7jorK?;+gKjLOo`@9@VIuT93{;S;AD}GW85seDn+kYDxG-}`%=adlfC+CPwXm=oZUqk&@6qY6 z3D3yl=NPDck`;1R2S|W&SXoWV>Qi|#a^^1z*HB*Irw`+CH^cfxe*`KUjQfE=1ywKn zNB&BUegF_7*=>XO{ndBBJZ@+ryNB>8GI?BwF{rLKRlBjWrNq>gnrl!=UwKy`D~^{Z zflxM}e=o!Xmo>-7al}dl;K)e`*Zn{}6OI7&)72}$Mzx*magLRJ%FQMqD{@uHM*N8K zlFtU;0TSaR388~VT{;mfm!QO*(NGkC>m^Ji$xGc+yCgPBV{sB_-&-lXJj8_HpzQj5cM=peob>iggfXYin*og@598}Flh&&5fhWgA zJ0l7FCE}$2?8jxFG*seQNRxVP3x#-z*ZNx9V4s*c`r?S0?gaF_i z?@5Ke3{L_2uFRAHeQm(9vFGk%Seprv3R>g173w;OHfGku_fZ)-7`QncCO z5(XUCQd2wP*T@vh5u4d*(G-kh#6oSI_`bKeO>hAHLqn z&KC!ulo--pfr)BXVmLQODyf8*Pg)TJx`8DCXP;0;L@he6Zjy%m>2U;{7AD*wQto%J zU3?`hS=eaU7ni!ifF{hUya!GzE~ckt;!|HG=Kfh?l+VZ3#84u22fVJd#hk`P-p7-c zyw!?maqsiw9wgSTeA5}&pTwlm(g2#9rbEV0H_hd8o5O?|T|D7NJBG)=hyxh3o2}1Uj&vAZG+RF+d`an)G6RJa81$Fl9 zS^LUZ(d*YvK6>;>?-kj!=?_L07M6l8(H!c@`K(XWKjHXq9q2-8#rZ){CKKe6PTu+& zJ@HCDMz&_$a?4S0^M0q$$Mlau15Hkw>pwJr#>C@)0e8K+v$w*RQL>%ta&!^<9U*c969`|hx zHt3-eU%bXqJLrS{;U^}4)~`D{bw$j`E?btpvB;@Q{sM{wa6@RG!U@zuj!rfCbn-5o z+LFg~eWm?t#|HfcZa9F3QtDFr8q>6#Ahb6P*-ydtN1E!7H?}=rpN+!Vz6#@z8u9Dz z<(A4>Ekxur#fJlIrhkV(ZzYCn=7oe`c39r4U{gwYJ6&=pbe8mTIRG1u2WbexL8-e1 z-)Xz(LQ6yLarYYUHmgPnmWG`8@C)rP3|3L+$r&iUUI<_15n(tNy7i`cvT$dikgSqt z|JSGt-*nt5rzw%uDDUYMtbfWQ;2TB;x*!!KMgcB?K%JAKqo=knk3L;GsVDcDm zTbkfM)Q_7O;{ugs@wDtTkjwCGh(^+@M{(_j=jD?|vWTW8p zk25CkR_`w*{*GPpl{$>``Vs{Y$X^Gc)Ey-G9)1N4mDv*={gm9rqJb`eM5|kloYVn2 zq9T_{3O;fgpBh9B--24Wq+nzz1UVU+UpThUf+MFifE)1W|G%6ESP{yeE32IM)EZJP*<*yjUtp%jleEa zS7K~zzD(462Qr1Nu|!DEo**P=VEs>jj*q|BYbTF>Ms;+UJ?KNt)gB6Tz=OqMO3QZv zj&Ua5*P@y)Fec?6`~p%+m;7q77CAv)QM>umJFOG2SjuWY!{2ux;C#j14N4M%SzYhuKb) zbb;ji(uHroe_TZERA7K~)$CN4 zBKt7<+RV6#grKj8o>n>Axsk?!1)(V?W@L85%(#pE*9jE&qgPUeoY)h`UOAU)94pAh zV&yI+Y%hrg92lfxAxfZh)FkPA+DJp_TH>>D!=Yp+)uk#hPX;m@Li}_pF`X1S$cViu z&k(9j4Sp3ABlNIq3P=i>u>oF*E#Fp_#7x6ybS?nb9lv6rpU2gC-BN7)AsbbPoZUCV zWiGg+mPvOE~g&kpZ^F2N}hj&=W84wQI)Q8 z8nj`9fxcI>*jHqBfZOL5MYZd*TR z+o0(K!K-lIad4fdFI_qYNAFUFx|Dw1kMrn|!en5@+P-Q7LR1iJZ(6FEPhdg~8yO9A z!&Swp{Fn@7^K$GN2__4o7h{ zbbhj2wE8nsQ}4Xe$>x|iq2W_RclKK+ah&Oo*8^}X4Up#4ZcVOkQQP~q6eln(Jh4JC@JHFM z3qDXli$4-wbnTHVi;wRM-AX%e7|gQf(t1^ekq~S_HmIX8{_qdJ>rK=iHsQWRdd#NN z|Lb+E<-PGgTG=wb=rOZ<<%m1?odLnpQ>d@hKuQ`VV*QWcW}K{j>~x)5;A~FDW^Sdn zW6O`y7cHRx7@lE9r99Wt9xM;ORNb#Bm!>ser{a^Ih^0oO1D9&t-yo8KMa7x^VjZ+N z=ey>~@q@NEhq9iWLnk>KSxQVdz?NTsr~{U&a>*T2*p9;spFq5m(i`!rI-gRqQUfdT z_k9N29kDgPnqGmar(aFZUlF^0DFI-k^p9Xfa?okeL9-)K!Y#1wj+;YuxS5)9Bj2ZojV;yHfxJ z*4~hcrF$9fa$AX2GCx+mpiAL-I!ZtP9hxfKxXna&bOch9F&~}hNlQnk4`#Q zw6wg_qJmPCyPo%^`mpfi0QO8~{A zEUl1jUfGqigEr}z3Z8Q|w@rJnF1Eb9R0?M_he zo#|rG0Kj#T^$hj(0^s`R-(T9RBpu4)QbdQyFoP=E8+L!f?mJQ!M z-w%{vW1p$1rY1Papp@@3FMWar=0PT+vF9-}U0&vC(qm4b{v zQkOyEtR54-h$foo6E7*PRhcQRtm^c$GTFGf=?=YP!oDfxG>a+A{1Q@Nk+ zr-3D@@uA(=yNQo`16{N-A8ZpAd9hZ1hNC<9-oV7E;5xRgM0*wT4L%#JYZ3gmCE#$` zA0`$8$abH^DKs>c*?a~ZJ>rozKc3jy*x2Mx86VWjIIjR2+5S67;5=tijC<=&~uLO~TH1U*&)JF2?0P%Ra%+p@qC!F-EKbTM_2k2V%5OFb506;o8mpZw!v*#xUrhQrplBbI3qy{J@VTR?3`F-rA zw@}fMPlEG%;@r06!c1BGjd)^K5&<&&usaDbvb&iG1mQ5#G6$V14S`zC; z`xdapa$vyPZraEOspsNvPLezcfhS|>>;&tGJV0`;dmUKqh0AP(&%kJIt6Cq}J%lna zGkXqO9!@AEl|EIcM%>NDXV3~pIl+BjHy-^dg&O)v+D%PAQR_boux-}Qf`jFEb*y*g zL;!Ay6i!nM_SHyHodBQLxXe>IY4_yFaBetXpfJD7kf2OHF%m#otvv(EEoWjkJ%y+I zf$n*jy8PD8uu6>8I2d^`7h5D8h=d_BR}>~+ zUqA~(jzY&gYdK2mT4o)V*$;RX)uQI)|VXQfT)ij0WcfK4F#U%TZN~0#oViKj!{6efCrmzdyvHqQGgtOD9C+=Sgzy%<&kiKfC^eDnV zbv{>7b=;3{s-9+xJ4h$2FwLz}3%9P^jz)v27u{zLDo@$c{Y%m(dt$^#0l{!EhCVaF z$NgjhrzHKg^oc|>V?CfG0&vTCHI|emvF%h|7=jLd?1ghE3WA+rAqMHF4l_UU@KzA) z;!G*vBN+HemW3-TSG3%XK(92%%XgVR041i`jC=7SD{w=530QEulm#?2o|(TPl&>be z_;nzl*AFJOu|c2$(c9FG4(>2uEgi-|;BGHhvy*40tgfnp(rN=*O|aWjvF=*^j;oFC zT8J1Bq;5YAuU490S3JhT?gDKh);)e@$%k02m+E?FsL4g@f#GLNj-^0sWF2)gPRiZ-C%w;?=%L3 zUuTL^hH}>Zfpqoag4kG$;z~}2)BI}G!LHq8gf09@fBfa8mx2X9QYZ)KKZ$n|yKLO& zyIw!Jx;bDQRp78#yeS9*TE5z?u*M0Aj=Mc8AmD?Z1a!3k2glD5Rs!{h5k&cc>*aye z=nP(^8IVYNuQ~Tqfmw^ak_}jgt~AvIbbqdC7_*wo-uh~~U5)5@>Sp`AwzDVPUK%xf z`}icOT;oF7VWebaN*{AnTP+l=N>{QyU+B+w-rMdoD@@DCAPkYK)g>cY0!34#yLdBA z0PvtKm{Q>4odb(biw@iw!jhh(j+#_eVMDtUvA%rC)2?E)a=EhOmhmj%&6AsRJ-1<+ zu~;l8CrDUB3PE=iW_Qnmp%4mv9|-YU-l)qce54L>MG9cSw)4Gk<`$@695&`WMXxX8 z9W5qY2dpIWKr$v>noh|0n5MzR`u{k{YC)Op-H0G{h0mScozZ{cVqy*nZ ziEgbwSb~`S4jq13KAcKqIrO~G6wD7_m_Zkd2YZL!=zi%7Va8I zV@-~VOmGVI+O9pLp@3EvobY3+h^3Fj|IBWee$OicCO@W(qK^f&1Yh5}Q9z95v=n}f z*9>`&mkzO3o>#CcjLv!cUEAx_)WSl^-bSBQet%xcdK!EtGd^B#wkr)f39M`e5R0Ps?WT9ylb=4()rF4$nnKPW%l05*x)9~my-`hb*1 z;rY6kdg7o4#vfGupZJpBd_p%0$y($karGm%9r`ky*O@BW(m^@8Q@Yg%Tj9E2uni2E+%g#>#}^T4TXX5Dk@fgPfq|~< z?E4_^tZn$CwScS7lmN{eZ)}{BP7p6>I{NA_-bHXk8R;vG`<3S2Ip%5~Y^x5emDK~E z*oxQ?Q?)Rpc|c!}oS?R?!H|N@Og<#L8jPzT!LZfwH3%fWYaGf3NJnnj-pKY`!Jt`GDZ+4V| zvrl*rxmE3nh@cNQjDhkBS9HkW_6wAktiTvnP{26ODOijF3BvnRMoJ0_CJP>uPOWs{ zRV?lGOs^F5@)+$a=OEXK0P4*qg73@CL*30qt8Mt{w*J%4=E}%adZFv3jm_JD6LF0M z+es{f$se$!@u^H3^5A(WV`UkDof^I|p}~nDO7DE$pYN8$UQ!U7A#JR#R*7%E6$9qI zMSba#u`FI5fWQpLr)Tv%j99t$W|%3PHbpy8%_Gsy6BBQ$3T&0+(T2+G@X0cogf)8) z=G6Bz1JDf$Z0%S$;MkY9=^dqvUIDtW0ae1)i+sWR@@TW*wpoL4~L0WwDM%Q@hW z$=0Kkczc-k>$AHE&1HQTzPEZZQS77zLd*T0v9WV1JfKipO}*w5ANS+TKx~Y+Psx~N zz$NpzV)B3zcccM>1{DsW_yM&3{#4b6Z;t1gD7|=%&Q3>szfU0mV->#G{2EtLQ8L1n z_cV?{cnXve>MI`KIRBPT;2RO`Wi-7yd)5E?1r~V`S}IaLUG3iOzG%u4en2qM-pLo9 zo8xs$)O5AtWf}uICYHoN%1DJKTBwtkD1Ns@d_* zE^8N;8@I6}7NNyQk9~VT_V2I*dlxc2m#i4s+1X=^xx#^fONG1G4sUsW1keqbi){Ll zF9k;A(p|U#%KhwFfmb?@4u6hno_w^`3XZJwPFaOnplhZ3L^*2T1omFxDn*pRXz$-2 zHveW~e6pZKo#Cc&AbvtV3k?@heD)2xP-J%oC(HJYp-A|_2M7|JIM=CtdJjH9i-YP3 z6hrK%{OFa$xpU&ak&xNEY3v$ly)jVF_53w`Wfh?|=N+yPzs5QBXjX!~g>2=I$C8Fg@TyoW(<4g!b6>FmPaRzmTc5vN{0^8 z%$*fS($LWEy#z!W7JC_aN?ux*2=im zm42c1BpIi9V~w34m#rY<^!%#DO@g=gm_%Gyn93iq3h!#CJP@(YY2)a)3(XFL0QYuT z$DUhQ$nLWOd3Gv*%KeM>_Y;Obh?mt*Ocz9--EG&I9gsM6W3897n4u>X!eB?<0n>(y z`c1oaXi%~n4}jZB_S|iRUUYCNf`k(o6nue_Rb6ehhR#w;CSn0xtICBvU|?y&t~P{d zBqPWu#dEiMT!Xb>>Uym|bUJ0TI*yzK{5$*QJ9SOK+w|Mv z)Mr=L&tLYKnNlMf#6B?IwDh#Ex#Juq;WUo~;`f6m5pf;Q&Bk20xHq;DiedNlLGg8KWkMs zMetu>QHdLW>ldQd4SpZrxo$} zu8*{WNET=#`=BZh2xs8^UI>>hZr=H+$O2g8gLsTA16nqB7dS>h7)a~f*n1i2=63)% z0&ct)H{NW9j-YL5AmdND)qEJGW~8lMaq0GCVoda93N-KQbC`P0r>}9YPsO@qfP|>* zB>Mn<^^5N8P$ueKh6y6@P6W#{+6*~ua%m6O+Kb_{p;-)>Umu4cM?oP$K|v@h9m^mc zK=N7x^IAw$E_Ap(-$9|KYx)NV1^M`9X6m40vZt9;^1=qxRfIp&w3L{Y^$7Tf)vf(s z)k1t0qcs*p?pn;&uJdjnEO|))Gnj&&1Drqg(Q9zCWYcMsjNmDjw zW4t&^J{pD6^Pndos5zbIMHA{@Dx4jme=iPsR~)d#!Y;}PRO;V>0BIUFY-y>Bre<_r zUQna9Swzb9t6>pVO}t=;s}ZM<_(psP03jZ@;5dIjc}Bh3lc9xa`L^}Jl^CO=sOHbV z(T^+I=sp3QPv8@@T~yB%SDsb-(J5Ivgm}P{iirU_UApuLCPQpWr*jq>9>n;T#iOgH z(SL|8GfNw$uttgUR%mMMrqfHadV(~_G(5wD@U;B_LWw2;QlKM=84IVw zg+9@1zdr^(^_i@R=Nxp|Tdu0~<_)=YBYwLYiRV$+SI}JlHEb&u{>5wnG_zp` zOwUNex#LIRNge%)mF`c$cLYzPVGmngUiLfzb;bMIF&gDv#KjA>W*l%R&@3*DF_%YQ zwye1ZWum$B{%X{aMJ3rv&1KvEYptF%J#Isb&AJ@Z*3g#=0OPk&Q+(C5XTV6`^!@Y< zTr1{69kvUkCK^&D~ z{AERRgY4cjYv9K596P>*g22R=r-22f{lOiZ=g$?X4@ynUA2FQMm49i%J-cz>U~azn z5O>*T3?6dh{A&a41MC_3G z)Re=GO+Ij>qoae~@SH+X%8mhWGCcIl58J^cM_M z?aKk1177#~4vU|XFmnNTAcEjF!+5jn)8J}sGVa|265&__&g}~>%NL0I4hnK|))W0h z4zi(d37qHVz1fCJNdzl$;;jOnw%zq>FFdTGs`>=}80603 zzv?9h_c!vgZlX}W*GC>gaEe3+g7SXEhSpx-C9TQU-P(F>AoDok+5x=vb6xwK2Rw9AV7tNO1%-gH&3A)Evuz@ zX!qvS%zPdusB2CLTtzVG)dBWj`kEg0=yi5w4grCI{{DUdp}oDm;P{Vb(I+RwrQVmLKqh7+bp}Zz@Fb{s z$YMaXL(2j*f@JoA6}vm1`#jH}f3@UA~>c z(Lk~Q7N0M8^B&aD796GnT}O~3xy%`YdKcm9!u$bA?;oQbt&*Z#uP+}i)WQa%z9y>TRRQmao4BJ!`p?UJXnQ>cy)UlI^${)WG zMIHV@o_7BD5qYoPB8ASC1~-tCp)o#PB`3MKSRg>+fs~Y(Rz$&NDk=ttASk1p3%{{> z6P(v#uRx;;f#>HNxAYl+^AcA1q; zMn+LcGNU3p84=lgCnDSTykDQs?{|EUZ^!$1|IsV%=lvMhb)M&SUe_~6=gh0dE8Ot! zouoyf!GX7N{ZYOzg2ahEyB6pDTeK(UP1ucy(H^Ze>MakNX&!P7@NgQ*CX z4YvQyeR6I~4V83-94-Nay8Q=44s2{l%`lu4;1>uUf>aeqU(8bp}k3}`0 z0#tnPId8+BoKsK2rzV@R+Gf%c~v8=aiO5dW&O4Qc(=sj*nWCY^eI3n9B zHi12QvrR|^4fD?EWMM`I{ZWx#+#GwVV0DPT!ns4&eAK>JmMzb?CWKxE^)k-lXLnWj zV|ujh&(37eRcw2{-*|AqeW0^qYYD5#zC1{PgW|`-2Xg~|c}uk1I|dy{2uH2^%2q+~ zJ)vWy)lM+|cCJAK?zyvPZQ)s{%O3Q*g|5l+QSQ4g(|Y~HT|+gfZgzI|i4JW{e6Z@H z1&3=o8up)d($^{!0+#wKLs%ej`OK@#Q#9ZcclTp4s=-`#~6_P;u5Sl{TLJC>2va@un1ILg*p5wdKU zB{a+l82HCI{@oeE5(5K6(O9h2$L-DE^m~0_(SpP?h3O*Yt7CCLOgBJ{i4z;3lz%qnBr;_h zo9dmHHmDlmC9@RQdYyFWI@i&$7C<){5JX2! zO(mCiRE_%J)jZbB0G&_FkJYNYqpR*d{wkZA=R>it;{}5XUAqpITy86!EZIrr->rki zJY%bk?<=}Ybqn*}mEH7mX6opQ>2fo7GM^I}m)u9s!^meuX0-GB@t~lf#aOhq)HgM8 zb8`B!k|&wj+1c&bwSQniANH5zY*RWWwCAV z9#*qc=jeSlZ4C|fm@H}jT*d?6&>;ic60tE|sKEw7@|-0J&XAZ`X&^1vmDuxHfD0=n z1cGCdl$SxiiITIkunbj5NC-3%ioJWg-QH)lM!vDJM@Vo8B=lXMI>|27BnVCFi=*Z~k?gPw^(dU1bROI*W^rm-kLbd^Q)_?xOSUs zOGl+V%bq}{qWgEsFa-*x28~Jk@DP6PCg(U`5EI^{0^{4TK_z2ug*ZtL^hLPRMlnicDAtl z#WPN$->L1b6+UOa8#Ir)jrxovo<4oLt+iEMLqp1aMS3$XCT6wd@cU?AU*Gcba)1At zmQ;m(d-t-jv3>sX<#%uCzgJyAx!P|H%FzLgO&e+;d+|`rW33=6>e`JP38=f3mv6#k zpscK{si}!rIx#%_ke`jfW_0)NO%086_a?69?^=%LGcLT-dxVP0JA8~lP-NZ{D{j

|z}uLKK(L#VE<(fd>2@wJR30ZmYyRpW(a3c-SBrlzi6 zyXK#ZMJC}0?N&h7VSIeghkj3$t&?0_^oI^*M%%#z&MdhHD!e^>Ynn zjEw;o#sTezhlU!Pnv9K%4)N#*3=jgkA~;n*F60XD_LV6m%wdvN>U+I~O!fnMdZmus zR&P_RI^X>0>}-w~p>=2`u{Ad{gLCp@c6N5ELMSDp@L4f2x3=`lh*rd5J1=A&LdZP8 zB$}3-YzM|dvSZgHU*BU!1SS~|kMhz|H!N>eAT6{`oNWv~J#_G(shOGI0HGilIkBj? zIDD2}A!GwtFklaX(DvS;IW;6bJNuHnycr~?kt76fOACu=L7O9NJ1WIYO-*@u-{QOX zBccse1#)t7N^NhvLzCmDhr7$m%4le5GcLubYigdi>d1a_R7mjrc~&Drs||7%qEAO> zXH;Y)`#PhqzrQh!y!Y}!I=c1@RhInO>FJ}?)DNNwgtM`-`}L+-B|;4xl*X5 zrKO^*ENK1HTj>O0MbK`jiYP3*EL45__9i#i7T@@y-~QFVlWq7kM9b6{FLJFqdFbh* zUcEY1wO9H&mdO&SD?dLUmkEEaX=zdH+t=3KE^9*Qj*5-##`pr97+EUWixAdJy;Oz=JM%ehYm6#%lv@n4AU+^0{c(kKyF1JqLGkrIF7ctMJRPWvG*uM6Tk zP7}UH^2?V?%~IRKVq%@`?W|V`tr^M558~bI8l(Ay?)E(m4;Qf;B1?}Wgi(F*qu57I z&iZTT&YdQ=_#f>gcqjjK5T0=1aEEIqN!*3MD+HHruJpI<2nl&-dzJ=H(#FPyiHV8l z=2B-}B-a5(;rYQY{=Fsl_Z>I@zpKavGFA^t7Re1vnVRb@H8(K<>%h>v`X4_I-XaNO z^z`&Z_%~U{W@%1h;-?{FyEskpu|MT=44kYqzB-xx)T(w@w&n866000+`8#G_y4O>mRQ9IkU#ILPAom ze^1CzA&044zAWcXNc#%KJ2)ul@6r+s);NPWOK;NB9_11gW5d}Mc=U_HtgmTmuB@%G z9VZYjRmFZe=6`JQ+9%x?x2)6~!K$noC0ywocI^o8Pfi~6+TJubG^9FwxY%~^GiTH|jk3`TSiUA*)zh zpE1Ih5~kt9r~pxqwZFKVu zk{)aK(EmzK=}q^u!pqCl^kr=9b@bcfAR<>#(Grxl)G+ad;{I&j{b}tgGJ>~Vx698X zZhDq!85z9S3j(aAoaYJtN4%qTuV44`^K+bPzg-i;g#0Pu`Zx3O<3?4fO&qZY1BA2J zp%-4gN+lu~&l-TRIz2VTc7_mPjYHR7*@{K@hutLJ`FAhI!!tjN+6~F(Gchy20kpYD zA-!?@Qgkl%AMgyuR@D5VAz#zhesi;eIh@3>fLJRU9SzszfN!($)?U_CL?U@DNdT9< z`}fOUf>t2M7NOAExp+cYcm)@aOh#B#6iWWE@D2;!zP>&4(ln2c3Vmy6c%7d9wZGpB zYfmOK(^E`)i^MPm0Yu#f#FC86TP8jsAuDp*{{RacD}k`zCsyIYG5}iS6BKT;GBXFr zcZ5;lc{*ELg8~93e!M2eEx=708XCl^uf<;5l0OppjCYZeR1_on2Se)Id38qzMcP zN=mnZLQK|i3GjFXjTeT6-}Or!pIA30NwMrcjML!;5EbcW*j@Vju4lXVY8wi#-Ldn@VIaw;+fFQ&rrvARDC=6Us z=lAa2xsX!?B?L3cJ3U3ffkQ(^*Eeqv!c2bmS8yFYb?OvWS{^I%3dEv4yGqmP0)Y@_ zo9V*A*go>9h#q&l_VCBWg$14I-=yA;R8&>Dj^aW`MMWv#hm_^zzkdGgV@-eXAaPF~ zKkVvQA({{KWv&d85h5PP50++UKV}U8Vqj;*myd~263nzLB~)H*uc|tD@E~#N*ojMb za>9k~&>?w2C`Y!MB#Of4&YfdtXU|WFjMUK435X^rer!&Xillhe%|kd#r**8Lq~u5o z@)ud)iRh+#K$s?9kmCI52<$(*y3}T|+_*>)p|76s17^$a@NNN)t&>MkNg*LrlAnm? zCR0*UGBL&p5Kbd7Ja{jDSj(55z|n=cRx1x_k(z=apDJ06F!0|K;&4!1jof|U$a(Vr zKI7-ltFr|K=dtW30a|@n31<^9s;}$n)vMCy5mu5DA{APX2?#v1-p_EBK>Ufn$f1a) zCSvUDJ4i?jgQ(f%S;+~PGSCYt2zEdPc~M^all8HHCr>z!O1S>5Jhrp4NacHW_8C0< zNo*}O86jQc+BGhukR$*9qNSJO=7(?)z9w7Fd9E((Y{AB%@5AKF=g*`q1Vd3|%KR44 zH`yJ;G-C7fNShn;j~_ph2dD|dS2yW#pzy)z6Td$2`VcDbX=btC`?Z^3xEH0a^U;c4N@A@uv$BpIAzV_poUg5>_P_!`??Neso^iRR@n6%U?-!Bd zYW|Q_vciQhqGjRXv59@BrXnb5>+3&OxlG8CxpBA5bMt?8?yoqFrl~3C{bNDHd%Shc ziRNjOqjl_rJ;>y;g7pmz*n=!;7M4()tFu0`mwq%)Bq|`Jne5oPO<<@&tE)A@4eJp? z#D0lTLYlg|I*kwXOLD-BtFSznT-4J$T0={ChsnlaPuA94Wp_wN9CIOKeb!guMf{!j zPLOhjGW0G6#ER?OxWQCILU_k;`0z1u!eeR9(9D)00_)D!)>bMq>`Yj|Evv`!O^1?G zQhbOHyMO=wT0h*h6cX0DFmci_e$f8oEXl+(xnGY~Gvi7^tzSe0O$ZmkY_cVF z_y3*^4qYXsJzdfZ6K$T{=VAz|Kdh$9fhd52S2zW>?*DsW(|h;t zWBUoaWSZG!)KY(B4k;iCgJGFm1TV*)nVC1WNT0Fb+>;=k?)uU}h~9_) zrRw1zBSdpTmfFg3tNvvAOpC7zoolvKiHGsvtZP982TPNx)H{oHS&&kmc-| zeNs!9lspy!mMpRW7vH``T*OII--Ck;4#5%5!MoxY^XA11LJbvR7)OznoRtpYsx3p+ z>+b~h>>n5%TrhX?=TF79O&Y%G=sNZVU$|HX352Yw?q*6XS5mHDPVzXl8p0qJ-VP3O zYBH+5DD5scQp-?f#KQ7|o~v(PzU=U4WWfUQSy)&U!C`}31u6^C#`WKvgyW13!~LCugFETSHYRmu&9G#q!B-|W3 z-`t=UVty7CMN5VPg&OV~M60C?)2EP3%GQU%6B1^}#+Fz97jp^mstRE>pSxi~@z z(XWZ|b_e|~R&o`UhC)eB=r+J%_)AJkN*Ng$6e7uuRcaZP6%?6U){Hc@v<7?WAHI0; zVq#+A`PPSmg1^KD4FRh|eUyV2#BnrUCS?5g=m3bsM^}dUVy&sI&32^n!`+XoX5gOg z;>3Zi5Lhs=84{eJpddyNC@3rM9a6eTsn>?N18d1V1?|y3KBRpfYNvU4mKPVT?&pC! zb#-;!a-=&JF}G$)0AIw_ByLaEt5-vcq=qMSb#*&lYY;v{*{JX8ius)(Bq%5^C%3w` z78?_ztD|GxQ}hRrbZKdc+!{J~i@}X3o*VOX!Pj9m41NDHB{j7I5;Hgxo5h7cOViiL zNQmj6*if)-u!mB5?~8G2az!PjlW5W4;NUoWc7C!g!zX*{#yfz|+m@El9B%`CKw@mC zYM=~YER_EycD%YS)z5F&^^&luBYRRD{>|O!qxXbdu!3p4lI|8JCXc=kn$uUMmz8b6 z@&$J1u`r^9S_!hdtDv#X#N_13lP8a96H>l(PiEjqn7cN!$NwTxBmf(zPnF7nxu*FKS^hxv@E zun_3}2D|D1U9l~B?p)-O*ExQE)GQ%0X2KURv1hAzmxb2l%TIE0W^vH#q}pQb%H9;< z&n8R$yHf;mHL*_d2uel!9f~COqFjVzhO>s1(4bD8J`E3-KPA(tr#GjeYQy?}*U`}t zl|4vP&^K_hZupnfA}4XYmQ`BmGcz^K(S0`tABCm0wRIb9SENFb80Rj?iP>Hl5(xdE z86`d?rMl|sQHVrPbW$ZyEol1oZEkVAIUtes>~%)xdn2{fryQy_{rQg_8-y%|gAWLI z(VueZ-#^Gt5AGds+(pH8>xcN%cL@}25882%oMj3X;3YU`WM`KZ>FVq2U%mPrYvQ}$ zb`8SYyTdY6UqyZ+EEg}YWEH^jzH+=q3!?J_Ry|yuJV;NG}gjbai zN2s>~cHVMY10?e1`ClmYy?RBQodn4SQW0CV)Kh|yU#x=rR8R4_5N1MrHEBUref<=E z6;bH%+H%Jiz_i&%@5i7!Fj-;ZhY&CwfB1R(5k(i5thF1*-I#Gx>g(#leIIoH+u8Bs z$Ie~5&`)ztOf0U!C_^;~G|(K;%SE4?+a>MA3s%$LsbuxFwT7D-Sp8qW;tC<$^+0lT zU7mqAuX)W`S-Bpqw9xsa+NFlp9Ug71SMDPz6?qyJwcVQy@f>Q1EyAgz<5!a|f_qp% z)>=&8BKf&VQpKhpC%q>ip_)Pt0$>e&@?>dYfo3is4U3Lzu({B{3!M7EewCh>=H=#w zW~TSv@)_t@R@NkRF1Oik0r`J9sSye$s_~AOm}fD#v#&4lpQQ#;!`Fu_sv?l=pSGBdxJm!%W@xSJ}y4!7z z4{D-!P?5n$8-&KFJ$Al;e&*le4~scwKYR8Adi78L{X3*zcjq9CxiKeS6Y-{d(Yv^t zO5pzeB>}6Bs_JSSQSR#nin7K+$W#ZI#f6Q(?nf#?B}Hm$g{p?FciJOI zl*Rt2T)K46*46?~>~rM&_fv9m)!=q4qW3JS0;oUVaDXxoX%H9=;$M7Zh^S+<}3<^f2)_%|RS^OxC*f?v~ zqN1*Dak(oqF7g(_>`dtE%uJ%40kS3%2jB;zm9doPMs!WR0)mk+szS@Vp$y{pVIshF z57E$znTO8k5cMl<`eBb!CS4V|bb_$DmR5p@gBczctQG4@EHqUGE#Y3i$Mj8perPL( zYfC4+PZa1ivD-(swk&4eARb*DiS{JqH3oi@fRd860BUxu#kUEzEIbQ7Vv|@Zkn{)= zI#QK>V~wAXOC)ev8X5WT-hm1m17Ek=fOVC-ucGzwXLIviWK(1f2EkHV+>jlu@gyC|BRCnTp1%5FMfIe?manXQ; zz;JO|YwC_+J_mE**Am*sA3uMd*m|uWLhJJM*|X}ZD%9z_V4*~i2OWYm*4g@32L|X9Ie?y)7Wkm|^tCq|uuB;8WExdsslv#qGkMEWTai%AJ zPwGL<Ddp?_-7ItFTRVVWl+wKs;t@3h8Nx5q(h6ej zP}@|sJVZ^6@91T5`*tw^a&a-_rI6t!zCC|7Rvj$sA40N0*DL@NMG}O~BJQX=Z&;sV+K&p&azUxHoNF7QXZhy!e)U-?pT zy`)fTg1MrF*=v0{CohlhpG}L|#x0}fRIJ~X1kpgF@0`D_9RGv&`}gdDiAl)nWJ+0C z8EEeH9INgmn^S7Caa*@BO%|IwZdWQz5g}(H9bjT=>MywX?;r2!(?m}_(RGFyBCCt$ zA3I%M#e|2enV6(KctGN_pD896E4x~*dM9*JBDBh@b^Fh7O`+sVPy&m0;HQwb+)%(p z;ppjxp}edtK+nCop2JC6R05Avb}q;7jOn?(bhAV1tx^jr2={A4RTj+SmOKZ~6I?fr$z6WE#*u;@@ioIW~57 z)Y8oF++n7oGPJV##bb3`Y~LA!GKk8e*j5RN8xZ#JU1nVNUhjV-$Qeqc9~f|DW@17` z#)kp}9zXKHLjH{x5ehy|2S^4KahzZ3B5vRuyf>Q>SaN}Ae#^o_@muobz<~Gsg&L0R zSOumvg>dab_5G;ct$O0n%snsz^x8y024(ZpYEr)nrXk$N;)?EFa3yhmG!lOA^_w^8 zG!og9%sgKC|AZ!4IcBfFVTx2h@ijxj$kGzLr(Qx9+!k0qCE@5i|K7Og;z&=CjqGV3 zKW@QngqJK(?%S7o`MzqRu9*;u9zmg@h12kN>5XU#?^SA3c#+?gW!82(Huq zaGDuB9pWIIiXaO(TByt6!FKK1g~Nsg$3+JrnSFm3vO>hF?+$erAEB}?6ddr;Z}p4q zR~Jk}c1k-BPE06LRbg*&l+G40h-NP{7JmJks4$c?SPI08YeL%B?}?2^VHtH`{1t?I zL>#<=&S_`CePAT%w}akqMicY*yw6S3ZU*N1)I3=es>G#SNQwCmWLd zmJem-zI|8dnK+HP|0v{639qgA=N3hdv$5!r{m11v*;@Ve>lhX!`v8>Ff)csw3yMV! zYWu1o(V@bHH-?PjleFo9h_|rQC=h=CXk`|KqWwr6EtEwxgUCcEjNB5rKc$(Rq=iEY zNboWv1C`!aSy_0&7Pu2;)E7)DFCj`{bM-03v35WQ@Vo5Ut6Nz;5fRu!)C$YCuX%cYq}#vSm<)T*JpSx%)JV9n8#9xzt@m;O$|#`{x1d!it1zED zabn(6oZ1R-Li)H9c{S(=NT=;(<+RRC+6PbOAccs%scX(_4g@5vWY-bdN- z*7i*;tshA3YpbhEn92J?OHM%nJVB zCs3u} zx$bdj_TQVFy40snpMqk7!eASK1~7H1@(X zqgT&JO{S)=&&bRS>0WrjnKlC{&1-%3gTAN%bQl(K=XnH4$o9zWqxr9pmriP=+cq=i zm7uVRl+=!uHhWEUP9_46X} z>eg0!lp(<{h_Y6U+>mp(=Cx}dFrFwaEe&|)HPhRaE-`wXd|?rh@L7myIGdmtm>z(f zMf4O22;kBJh#P!sSgka;_BJKub9FT*q6kF$e%97Ux)Pzdjg~lqWR;2gQfr z_Yx4Ya6(OoM#tP=z=cB)G3n(dGtn~?#mth1YO6C^(M9YZl@VYq#M{MGUN@)QcN^R+ z7X}+aRk@hX=UI4o45%h*;sB%FbpEK-z89!yd#OUdWCm&vctX1!TRbc|s(A4tarSt8 zygtx6w{CPyOdwAdOeG%-Kly;6F67UHY$Bi0Qd8eDHLbRLfPu5lJLUOv+D{`!A-nfnoNm${ATt z%)Z4dQ9UXx6_cTQXWo3rC?BSyKlEghw(-Xn5Ch8I9upFB!lUZ@HK`_uLgV~c?o(%w zpb72k%NOIdzms6{`d2Jan=;VTJNfzMIz{-1adhZws*v>1g?p|-Y+;!HoYZ3 z>g!)YmV(p;Wn13mE|uu+-Mj0=x9a}fMDD^ajb!E_IS09>?mKkmMK#x{Q?GJPpXA~B zh-e8^;j-ak@91dOJ48Ii9?IYw_q|?fAuE{2Elct7<^6aY3&RvR#uS_>@c)3Ri6=@0 z+-b^LgAZkO#{A-rSLs+gWpPLE;Nj z-Pf<_FJ4IRPnyQ363|qzLY3_Epy^o^pn6 zfs57kl*28@szsI}Xgg5(^{af{26>o3G$RvJb&NRwxpT%(Fg@Z5&P%2cv9R9pYCvP1jJ zTr_l&Vv&M&@7($9>C^fA8zBG^wdZb0qOm8+!$BQ-lJd z-($)lMEL6G;zPoFC~IiRnKcw0#u{(kOwP`>HZVw>FsLRXRnLtTMEs*dcjATZhR|CQ zEqebI2%2|)HF(v*YP^@jGhx`h5;$aKS9tnJOY%j34%>ccL28EkP5=e9e*ZqvlIjn6 zF*p6h6sqkkUK_{qVp4$u@(t=9A3j(0TpVL}Uz%FepV49UyQchlIP&@Pucm{b5s~Az zHUjn|wXhBvESn?2aOYGtHa`Bx@l?3;QgVqMWFfTiyxZ!;MXy_%zJT|L(9+O+ZgA|X z=z~OBZzDPo;pRd z)Ux&EfK=G861*ofKrrui+wl!cZ<7+o*&|mf9<j**ALp}2Vb(jNptn9BG+aot0nh&^Xc`TIVjOB!Z~*w}CfG5z1@ z-s!>x(j{j&=HTdZ-o=e#D-2X#u+E`=$*4*pV;iCSc^*ArII-)t+`X3TKuStvc|&qknGXnxrQo1w+UlOi1^?kn@h-0n={yfKDyy11<3!;thg z`X0ac;aZ>Jyr5un<(@$3Rj#8P+IkJmB#Pw3>hRyLjw%tRXox5k6;iV@kdb6MCWnVF zsxXH;6a=!l&>cA_$_m zlvL8jA|ws&-q#~<^79wqxCIuPgiQ+0%B!zKcfYXho!s&)S0{Dqchj0UB9`9}YC-B(% z__x&8g@ulvAMMFL$l`(NkMZ#_tO#ps@)o z8eJ7(xc{!$FKwsWNYo&Jjt&l1C&OX5`7u8~PwWfmDGGOsxL|E%1}C<`)vJ|!$A*T7 z9VWgX>p1xno^1SlH*Ay#4?Y8=f~*S|%yInqN9R9m5)$PvFXOcf1-DDFPz6;nuwx^a zAp&#}3pNrt%!g|FY~f)kP}p^VxzSkQ4#FMGRKbf5mpMZp36MthLJ7Z;B>Ci~YS?)s zs1&golB&paMD4FwBK@B@TeFo&;z-~G#c;HEiQO>KZt^Tv1Th(BMFiK1$ExAmH|U1V zUO#fy4vLp-qF0;`(2e4d7ZhIz3JwOB&9@(|7qr=glBhWr zz#0I=4-oj&`0&n8|8z+oHt$>t4{sa^_wo4*M}~yBc=S3&7O;bvyP{^CHA;UcPM%cj zVi|P&d8+`VlnD?UFWK{XtE{Kz2K^UgjD6oOK1!=*cK%XhMRXz+p!Cx_ETD!K zIBfn2p-{q?8kE(#i$7!O^a!Qg`@q)55NT| zu6&S}`aY}=eA=-*XbrLoL!3K=e1r+E%du&2aq;pJWgO%Q2;A}I7f7{iBR@-$-9>Dt zSuaj6*%>%@ir6=RA!wn(HMz9w{McD0MgYS9b zW!=~U2mQcTucq)GT_{^RNk+GBwRBvZ212G6zXPr`D$zAKHugLziMPt>NvO0MX1X?R zJ;m!`lXK*vy7pbjJG4SKC#j-2?{c#L${kTQwhKlTCW+!M7Yt$oG_W@)T=~Q0Y+C;m zXBvm2wX@R+W?)q7_EA#e4le-SqC6TI5rL0~XjgTXUP}QcF=}*#J=IKyo$}EhTG$VH z{Q@Ipr(o($^W1j{?+IZ(l+t%1hwdrrMhJPb3T4l?)ViK|Gx$f(;OY>qe2+4$1)=v&;blXqaD=aRd=4Gj6@U*ZZO<#=K-Jwew0OsAHPjv{3_a1@o* zC+m)?gKdGg+=Zv9j%wpxA-ys#Dj$e@f14I6#VQuoMJb+0Aq%eXWy52GvWPR{GH!`_ zU=Cgc1PiTw&qUa8^oETQf z>86X>*EP^_++A=0%MSO!!P!sn>jhdQadNGcy>Mk#T98guYeJGYw6VGT?M~%BC{#$r zpnX5gOdc+C)$IHpS8@Y>HnbMQTdpv#br=n>rrl}Av6lI6ysLW->NT^Qn~!4(-BdG5 z&uDwj%{|+)^w`SU+Jw91^B;kHr)$J&22|a8(**4B@IqTIy7PD{cAkeQbKlnXCyKbh z$JdYK!cFQ`*p;Rfxv{y4o|9izO<-KcOfz+oX~7p7e(V)afB$1P5u6Za$TF5y?%r$0 z6ClM0$?9G>DfHajuj2 zLMQb1mvUL)z&s7;t(lpbZ{ARTP+&*~*Fsphuu2L8@bdh85wP#mCDQ*Gg%2w_#9Cpd zr_@(gRi$5&uW(=e>^W@&U<&eCdewU6C=S z5kzP$)Xw>^qPK9hF^N0^uXA@ur{f5%YNHO9_;7(r780O_25mp>vWAcd&;#@aJ&DrM zycM?7NQ}tpmYitXD84^^sHR0}=3f^wxq=Ma*I}FH=g6{vcJoe$S)E@f!p8oBJN@n3 z=JiC#tG*r=h3@oFO-g>aJOArZ>{lK^!9!AeSid=RKo{OO`V=-2biFNu3`5XWKm6L; zo8ZIhiS|;EAbYV4^PIkUKT<#?+P9*$%H0-Iuu2RRh6)&&fUE9ccC4E3Gp%Jq@+E z*0?%S8;*I*m4eQ4vE*SPs2kQ8k&w8?*S`Xp@>zdKNinJoJH}X11^7^-;LkMolZ$D6 zesIS$>CDAI775q()@~qW#71Yv5gjcp&f3KBu`zUQK3Fd}hPw+KMl&AU#WiquQrzEh zEiEObRXsM4zUPA*pj!9#7G8wXz9%s_cwdxVIM6FV|ET>yp3{=Y%6xK_=bNlM_dp;% zsqP543I1kbF#$8SoajgRdw+fTl3G;6K(c#M+A$pLEvw=BDJr!VwjzL|Ga4|!Xgah?wqCqqga_aZhV$p@O*!o%=>{tb>g(qOChIr{i_3l@Dl@pQ?Ra~;TMemQIph=Gcf)2RG|QZb6nw1C<&kJHfQ!Q%~dS) z*!9mrRkNt5EubvKK?my?M`ni60ugHM-)|`}PN;^!d)_q;uG&53UoCEGnrZL9@t7pD ztrtQE&{r>Xo_hyB6}*4HgOn6~fX)>y`znh@s73BQ93Ek2Q(_KJHAaiq34yv)2(`ht z^w27HR|TR#BYjTA`6qbj;NS-Es|qM15KL^}K9fciuCYrX$|@CQ=1F~Ws;Z$i6yQgp zOM&1$0qJ?lg7%k-n;R_|v3iD^ zfNaQj)5ddU{(;J@APx(h6uH`S#+SuWjPdjJO#qe#bcNp99(rHJ;@Lyq8zoh170BfU zIZjjUlfaq<SwcnR@bYvG%~3} ztm?oHAQI>^8_bhB?geFYYku%zy6WA%l}sXxAE7%-L@M;~R>wq$)2Q|j;QYVu7Wjf$ z0WW_OrRJ?yqpA)|S@&6t$*&(zJ*3=`zo4G0%V zv7FgTpHPg!ff2Q)Cc0l}$m~|;Z*}My4D~z!+l%-zT9ga@V0jt&{SQQ9kRFkcf{4gS zxbBDYk6m|4MndQ&W!zVHIYAVSfyB3j0#DmahE4za@EfSI?%5-Me}}SB_)m#wytR}6 z@G5qS=mK#SlaP|?he^Pg3Pl){w}S69!9tS3Juo}lWS3yl?{@1+)cD66*Xta7Se*~= zYi3_|^1|5wBbe*2kSdPkes-CR>MBe%@2xs^I2J(Ay#)1`7({@JDe8hGA9An>7&k1$VGdb;D{*yUxWREp}Uwv1-b*6S44 z9UYeuDY}~FYd@&BTcv;e)%AK`Ug~`w+L7JC>yBVqm)36{>ti1jE>bs=>cp`hxw2Qs zexG+b3c6r8#FT3Km9XXP>WY}p*8ZI&JRu5R9vH~#Tq0NWLZyYSiEIt<$=$wv@<*1! z{y2DD483ySya`zjz?s99?c%cGY%b2Y99M|RTYW-nu*MPfv}HqM<6+cy5KTOw9>B+p zApo>we-OJ7LC_KUwA^8@Ti4k0@5lA6EG!0iwIR6`D431 z^{&@M5W%Nka_=PfJ#1dQgj=EWvTnd9dVz%>;i=5zgir2U8(rVN@o;j=g*^iRjuc6) z*xveu?mFX|N=o02u!t_(eD0F#2f15XTSdLLHc)ONJCoB5-viHR5+az+_T!Kcp3|p| zJxt*9W%TMQW)QlgfA8LT8+9K5Z3Li%;S(z`&RmJ%=j7uHZ@A`aX*mV@2<)U|VDQbf z4MHSpl}6jWfdK(jUt%ft@3*MVa0A$F-J2CtQM6ID7_jcx6`FL;{F|*;o=?0AOmZkg zsf>-6=Rdf}+tb5PS9S@BVF+8cixX;kt-{OMBpZc(|0P z{&5$J%~hMBsVQZuD!g2jEIXiQE|ZVf<`qkoP>8u3ATA@Z0DW9V&X!2`4enR~f6fc@ zh57hgP}v;@dWh|Faeq=Ul?K^rqdLf=V>KXv{3E;S(aWgO?d0423T|}mOe1hG?pbaT z5$e;>JeTWo1nW5|<~+ya#&pADKM(fi6_w-eJ2>NbR)nY(CUUN}tt z5v5wxB;Imgrm9N-WrZE*o9XDx&EA;?DHyB&4Tine&KX*`*suOx%!d_nfo_xu3)y7= zh~>Y3%kUV;G`O@2i;IvcktAR_F!<>1GCv?=??`t9=z}ji=KU+mNu+CT-yvLu0J&h`5Xm3pdJ~!#$c1au1Eiokn^@=MA0LVOxSDur(B6B33mT*hOzr{Rbx~?8ZUp zF&u-}N8vgH*&UM%7Dp>259DgbY0G!6n13)E2*2t< zQx76Ru@p?@Z-BcPD_DS-i>_H6L0!iC<|oiC>o&$p#Gh$3*w*5AM2|cBlX2lf`RP&mmoA zU8XO=Et2`}T*iS+b6(f;3V4D2GVxhaNY%p+oxSCQSzlIx6iiGBcEevWHggc(wlc2W znJLchn9<=Hnfd*B!n)%KE})2d|f=CG3Waf z-yWa)g;!g0V#+?qCa67T5_|4;0+PB?b$rz7rQ(5)H>MQ)&EN^KDWTBS(pmz8Kpa7H z_%!h~cMxdd!q2H~K=S2S#(eLAFV`67|F?UY<_@Pd01RFqjJrDQNGYw7*%~=?g*n{g zMD0im(s9huPbed*J^WCeE3%yUL;Onhbto<>Ub)LxDv)Yw%+XKZTH<}`GF{Uw35=()AOVTBVHnYG$>qo60hU0rQ$@eIQRlJ_%CY}1$7$aKF%l^^@p1u zxN+&ys zFRnB;IjKZdh4KO1ShY`s+ABtY?Lf-^^gOkU#G+n_KKtNzZ~<2NgC?u`J*(%^FemDp z>E`9ME3!}8%X!?ZNqz%( z-Gys3yCA%{c$GLRQpJZ(*{U8AiMU{VSkcED(8!te?3?4y^$t6XaW>!)DqH4jJg8zdPS1 z#l&=BS5}y5swM9gLluV&$D^*ONVr4lHq%PicxQ8CsYj#<*ssiTlh~5=Pjsev;yxX`5tu-rayfn9PjC@?YljSL^_kvap%3kNvl~t;N=BPK>eeUnfuZm6z(Bdvrl3hK`@J z*DgYV<_`Q}#P`PlkfMwL+Yzy<9v7#FhSM82a?rHYa|wA`RrM{fAliAEG1VyaH(Xn1 zZd=>f5PM-zb{!cT+iU6z@t{#zRi(+75Kse)3CJ|x^H`*3)g9qR#s&(Pyh ze0$8n^-P|=ynm}k6d}SpOmE$i(IyOf<{U(4z?Jl$G~sjaY#;KPvfLtAOuy%op+Y(^ z(T)|cJ5Xr>y$A*mU^hsr6G$_0iN7!q0;(Ov6 z;R1wgVM%?9wwt-kXsmMJ9@)3y|q zxr&MkU*Bp(aMcEpJKY6=jNO&4(rxHz%sr4X6b{?(>PA$Z=$uqq=`Z;PqUTcO~X0AX?gZ1YuARj7#Xvf(`Ynq$`0yR$Naj9@%EM z&kV^a5S!wV7kc=dUxn7m)2FxL9fvlACZA;HY+~0Y;B-L-wiit;eJEt-e5iFM&gH_y zn4_2OB_EE;$ku)5=V;Q<(9qf0*=E4oUz(GahMF4u^+%_uJBCJ$C^ZoPP*B3y@CJjo zH!w|#gTti$>B*covj2VG;HS}6eLJM}P-nyIK@^!E(6x1TZkOvKl32cjl@Go(=o_G5 zh~a4O){CX|M6Z(*j4VhAV|KT|@1&%NM(*ckk~;~X-jC-B$kTd#qPreND?^;-K+ zxgbW)KotdG(g@96HolH=Ui}|EZmWXc=m0@Y7#QMyz|q#Uj)K_&j`+*_6=;=BzoXtxR8L4(4$2?2NMO1y9jOhGc*KM|^GYip4W8U!8e z?Nhc?WMppP%gw05kW17LyZ?4WrH7_^4>fN{iFbQT1kRj!-0&HthT*)Ug%J_0$hgp> zy0)zV&kK5j&|1s-0PB##?*x`6z|U{lFHLcne+1QERK<~13*Wt4{PPD6g@bNhDbE9t zc9`8;WYBB<8EuRaEL!LaP4hExl;2ecap>OoHzrJWTJC1zK#?3F6n}@OP1QK+%|I^d znUnRTR^)Nr8%E6OyZl(L=?>4>EPTlLM^B$W-)olup&q+G6WSar5Ih{$hkEqDzyS0? zo@wMaabzka`nsSDI=ECYaDhj)r{uoXYlYsYCpFMTQIyTW&mWLf&9@r%rav$IJ-~^kS5u!URqeej>}q)Q2_-CQ&-6kCnt~5gIdIu}5IY z3~H15=*X`n1dzDwfY5&m(+60|f%UdJ*xNe#E@HGS`olhB@|4fRho}c*n%<`#ePI{g zr!AWqMGdvJxdjEl79%SaxLFY3U@px?7d>k9FkYb79L{H`S+HNgDy8zeslMKb`>5<9 zbOwZA7!7(}F`Sd}y<$)sVpG9(k^nwDS+X=^k( zqR$6_5?*QGKY*&mtxY${4)*5$opE%m=M-9dFSgd!PQaXjgUP6|->&J=M>QB&P}4%2 z1~EjQxt$HFGBHtP_=yzW#y&4DsQQpbSgRjIjCdR}pWi8aNQ;E9hgdg1eOgaPM@LQV z(?7CQ0;}59Rc5mGoWnze?>yZZhJ|#Eh4K+=4!Th3!8)%6^tb#U*4{gw>p$)n{wNhn zSq&rEiH59@mGwmlMP@`+MlvEI5(&vnND(S2Ss`T`|I+5Ggs4Kv7?zwQt&JDpU^#Z z>T{XzWjnhJn1~~^4~mPw$jOPkeOompr->(EkkOxS0$~G37(jU-WPq@3B=9MzvtpJN z@&bNRr(KL}Ec(CSJ9ZB9$vA{BdwYZJ+>dUnscFmA=og9?e;qQ=(xNZlKrn?RSp;(! zXg_VyJGDm}fcvQK8753(g zB)Ioz)7J__2x7G@-~Bqk=X-Tr%K1DF{R=nNs6PX{t?T4&NW5v((~&tplD&zsxc%O~rG6uYzb3Ly;t=!j zxpk3iuz8=|;X&927u*hK;G3sL?REsl%8BI^Lo^6;~N#dDHqB(yYZ2tv8Q zwux7*GU+XGZ1tNmQXP&!hTxCP7RX!?BUFA}sVPk|ZP<^BxPIj~S!^z3Y#D$rRT(KRZ(Fl*o^;f&BTbtZ?$@@te=kK~W5GYs@50Y+f#+zI5Lu$$mz)gtVz4*RfXqSCc<}-0~*~!Hvh_ zKWzd{2XnsQ6C72oinMF}zn#tRMst}aNFuJO8NhY81&G><11^AtTMKh@$C@9kMZ#Q4 z`p*$p_QFT)_^}H!gy0iY6Riw1)Hl}qe`A-rz@@6|)bj$ffxv{JdbmOBpZ5o)GK_5r zt91m%X6bmftMf$BP`3C-fBStYT^Avjc7$LV@EskNt-E&pgv5@41|@_0gizgfUgwzA~$?#Yt zvLwApQEqsh(HX+}Ae>mCsW^9MyKR%v;o#S8JV>SELn0^| z8r+j8VftYGJ4i+7oFLzz`xK3s?OH)V+M9CytO%V8=<6_WN>81O zbS|Kea=#3-3s$4QXmpdkgOD=fbHzt%P7;bnCEl~4;$4oCsZ=Q}=!Qv9ih%~J;Sg+r zut~w~!M&FV-3tT@QiJrY$Bz#|{r!%bP%IV^zCu2}y`2hvhfjxXBx=8bTA^9v-)aPg z^KQdtdir}{Nr3ADmxD+wB8L!shDsJHndrjo>}=fUdW! zeM}UVtp_?U(d$Ij~{QujjVCE^6OsnG7j=#*k{Pv3&n~Mg4^Yj zwNH(DdU&Xb6R~vwQyc2*p0c3_)LE_={ zLEV!N0;v>>n>zwv`N0BmJt!syyf9U898o7o3s!|Rj#Re*E>&5t_+SaRLHW|u#&eKe zICtexk)Oo_FXj@)cco?k7UEDNG$9tcFAfAAhvh>t=(2cu2+e&9kr%P55Rbmzak$x-DAgMz6iau zZxV}&4sGTl9x}7DA7z7)n}SED3k(zQ%NdZ-*NPy}0q6ot^735!Ub+B3I~^`c+Fv|*gJ^4ND#9H^#Q8UxWHZlIS_CLK`z}kaN@}CkWf*& zTiVZtx3`$o4n&l05ouT+YO77}y%7;@{K8IGO>O--60n2p_1}X#+ey^eHu73)bGCTi z$2})?pNOph@RfMcFAlj{hc9og;v-t-8lt0`Xnk0R+&XYPu$}B+X1?}CFsL>D zwEob#f})Kl&Q4!pul8fS%CP4i=v2Ii=sDnkqfC4_3Qp9ghs%!0p75CIXJTUuO`}uX z1NJ^f9JDDi)UV<2Ex>m~mYlh=^4a&HTX)L1?+~-u6!C#w+^!j(_Um5+<&RXQdHbE> z|8_5Ff4y(G;GQV7Q2;VxJOE9gz0e=aLwb~mg+~nR)lptlN(a&{+>AWaa`UY`aI-1DpvIN10K8;1S z0HdBk;=qG~U`T(VwkZ`6JACne9V6*xe&|`+tYY4?zj2qNI z27w|Q&P%Cyg2fd7*ofDEFXv7CnVV?&;heeDoN@&JZ-eKi2QfykqeJWH(doH4PK7Hz zpevNp{a;F+U*zo9_;E7oLgvhm#>wz|$a~z@ta{N-rDW z!{Q_w>?MO*5&VvuCkkKD0tcKPG$q{OqoZ~l+60|2A75aqATZ6)R8W{G7dmXto&`;@ zg!31t>_G$H+NP%d9E)rB?tKDBNlJdGH5o7LOQQGPtp0;n{j#CE1-B)?CpIZOTcj$! zH0Z;?KGPd_)?<~_?Do>jQrc+Syeoeb!UaDt`&)X$FpXQJhT|2BD&P%jeAf;5puGsO zF*fc&v4?N1?hKO$>6KGh7>oh_M9xxx8|21qpRt+v%=vDLW^vuhpRDg+sorVsN9C}Y zQx;w`uo4>1keFD)Zn>a`sb=l07dDt@UzgNPFL`TNxAw{u`XN zhy?cC!O9AK)OUa_SgyZkXQAXuArqR?@XukM&;pbNfECe+5DKH7o z74p_EJHeg)=vad2nK%AepAd5L`tMt%bGiBWe*LF5ie9AgPwS_VO^0PZJGSt~U2&j( z#$N(K9yX?n1c*+prmgDJ)6)=^ox~tQ24RA@$lEj+RjbFDYxoR5WDcMVB0+3|`r@sG zVwnn03*P>o&_txX;M7yoZ3h5}0Mm<)kOr+94v!&vV!YO$s60Tv06VHbKe$pnmkl5z zPvK0ip<{^Du4bnicK)SU*`39)JO(G}K1Vjk{o>s>GP})l#No#HJ$4hl`1{+wo&GZW zbbT1_&{@sZ@ATeC;L?|1Yq6{S+1ZV{5FhBh1f7G1*79!tkFcis9nz$0dOF9r}3`2$07`s4i0Z zcNRLCB(vd*f16%F|8@M>F_OpuG=fahd*vrkE59tr2s!=a)8}XzJDXLn74y=M{VEx3 z12|M$Jz;8LH%>!0WB8DTP;9X1GM;X3ZB110BU2Ax7eGE2Fa&cB6T)>KdkgRK1WN(1 z4{S1BmRV5qh^N)*e^TNQjw)=!`t{La=}=YSCL!Tv8F#^h2N`vvub$L?^26zgQ7Qjy z!i17sV``%U<>Q>Nx^O8OfM z_y}24VWq+pAs_j*T;^-8p21Criv;whHfU7K&u}%2deG>~8SRP6M4DdT%v)`tXQG z{WPt=_!llbNCYmCIx#fs-i19!Rf`54-lB&8DMr=e&ka-}0JMkS5 zNuY0FkZqJ-_4zaBTl~iTuY7IBUNzk5dptN?6}tTnxwrkd@lOmgP8;dxAxQ_rEuKO@ z-6|<2hKUfBF6+72A)G>C2kAN)?SmbR>6<`TkorG`m%=SY5Vh532G2Ll{fIswyzzJ= z!*H3>Gaw$uw`7GQ^d!7gblWkt6C)thdr(XO2$X6kjC;U5Bau2XF>(CIk1HruAXq^0 zq;77$ic^(i^M>E}uxf0MS_5N`yVu7>@3;N0)r~2B_de>%PGd528C(08_5m6nRUXD$(*DDMsm0M90F#3-Jyf;b}udxPH?C5a49s)*%R?$}@VAj~X6Fd+UYJ9Y411_lbWC zWjM&fAx{K9iT97zB-$B*blAxHK7GO@yn~VSDGz=cp%(`wN{%Ifdg0TC{eo%UE zNA`6b#-M%ij1Yozys)p_uOFkqkbHzpBA@vmJC&%p(Eb}isfv0FwhAvCVklT~OqO}i zW`dE1tQ$K-^Vl(*^3&LIWveTf3)r|f5-IXs>*K20n(T5U_-7ccWQAl8?aC$pGS6^x zE_ejdmzgeG@TKYT+~Senh=!FF;=g^^5-5{kX6As~59(c=8`L@`2;;^rn3a_UUn(*~ zJ8>?O70;eM=Ww8a#hJh#N1Xx+4Q@AMh-Bru(4+%G8O zJ_>pR+T;DhDrdm5LE7*RN}~<*Y?y}-T!et!i)sQ*jaz;{+tzV8?#Z!Y1D4<6Uq?%^ ztqA`qe%$Kx;@iSHN~<8>1YrbVEZK*Fo&9c96!&siXjed9`9V@-nJNe%6^<^%`hP@B z5L#_SCS~NWB?z_z+1J1NWNCeCIls5P@MsxZ=C2?ZXY6J1V*Ut4*OZ=*J0$E`x)H#j zYYLL{h^y-q*}KNXfc6|cB1h0wgJXa5CPkkV!cDQE2?aU}CEkpIO&YqQ#KavAc!aa9 z7iYI)_g%iEsFYmcHZA4!+BM6Ymptji+ul7~yesZV-*J#W^>k%lAIAe7{`%FR#j@5_ zsc`lw6!IxO@-ki?1vvqp?H?S!ChwIK3*3nR+EVn~FmsopygJ(r{cESMokW!|y>@C} zaHU0l_ug$AS8@LntDd4+39=?rnf5$pnr}mD^lsCT2T$aRhyJs?BOW!p0)S8PbBb-b z^n1f?lWvz}vaL3RrwS|lJK4jzl52T4nczVY{x<37rk}hWm5crx`?=dJ>H^)rb42hz z{=bBNO0Vbh=TFNA<=5HT(Do~+k~buq%KaMN|NH}DqC2*xcb6rnJM`C{`1~~&-~_n^ z{bvUA^%Ppw@YB&C9`siZTvgODD(_KvC^!z;h$CRY0NQcoYgy-Hmyi9pFQ1xPk3N6; zGfPs^`KcjQ&1*q7@cGrf&ItuST%=8pBouD}D}#9eBnbXD`?3I}`j8QL>{oNFqZ}4@ z3DXgDT-+5M8F^v2p0y+;<&qF8PB2LqA5S~= zd{Z!<^~sBcT!&l@{I57j9%E`~aIhVX4lt#F0KZ425Xm^S*bZKh0$R6FQ0DybyFi0; zZnxvMF#W$z2gC6oQ`u0gRv4VCI?=JfTQhq&fX!7w32vBo(Gmr2hv+ApoA+@XuRNZI z>?E8KXkPKDdNEJ$)~!aUNFw|^bWC3D&tJ#tVx5aHf<02wm5wdDIK(!POHWbMOCSPJ zIx+QaauO*Unz{&2MFd_EsuObJ&jTGpke=8wX4lMt>IJL_X{D7-R_2f7`R7iD6MK^;^BqagM%0jw*xmTGZrM5}?|bkMd7?5@JtH=mgSNCU&Mr8lDJ^$$b^{)EM?w2%LF=Mzi=TwX)ca2! zTn4zxq8e4Q1MiMSHFu_Whql69_N9ctntb8=_xBZ->ifN+`?rZb)&!@YAI$Gz!xl(E zTqI)xXF5^Hxn@ZWd&hmqk@pS~m;nhN;E-2nm3V^?&{O`zcRV%8Bv zbqNj|!`(<0aBSxOp6VOIrR5Z*bYse++aN>SjfM1O5&vEtVg`1CwmSzkKB(>-Zy=`x$?>3{3y7;_ATo^$g0 z!$m4Zmq^ubH5GGOtZz4_aDna0*nZbRc9x~LTCG)sd)nn|w=@lTMLXg0c>sQtg=){? z*0wf>Z!^vW-5PL#8h5f5v%+lyPwv%NWgCi!xbyi_xFKF4gZt*kA)d#`>EQfo0HJvwdlVrs(YUv3soZzsrW$%T`(M#%bd?Eo#r+L%Mc|V8P19* zP-NvEgxmu{j1t9W9XueP*yBxO-pTu7`hq>!Jzqqe*92AZa zaisr}=Ke~X$>!aizRyArw*0J?%76dcaqZE-qj8u3v5SdGNjLV09QtpUS5!lY|11JB zf{L)Wr-z3vMJ3Upiw#4F6+>EXQ}S2~w_&i0?JXRRq{TL?NoVb zWkaE--D$Wt4zjV&8*mF= z1c~GxDQAZsMYkOo3n{r-w^wN4r{C&bmsG`(|(c)G^Me0ak3GCpleeCvS~R zx@{dUd!BvNx^_7=?C7QsmMgs7M$l`9hif4MT~|(1!R{L~E*AjCz%nM;aK-UNsSC1n z&Tlsb8^e5qrpA$_HijT-5gxG?0^1aac-kzv0&z+BId+P$iK=px^k3{iFOAP z;U^qF&hJ~j7I*n{^>aGvH8r0gsO^BHVsR}eVa?Y>n;)gLhFD0AA=i+kMMa>$zyHVV z)rAYa{rw0+J18QelQ{D5q}a@PVK(nATixc{sl~i-DEqHBZ7KZWh{Nt@wIQu9T~p(5 z;bfK5$L(LE>8C>KU_(LtW=*)Vg#~xShq|Vw)lq}opgoC<2|W?--z>$wz{(M_7is#W zMTzEu04Xm-VR@w4?+b-}(1AI)uH$otd+`3d4K|Oyj#iMmD!vFs)*7$Ie|4NH*FY_= zC@Ym)9f}FG6W>NgcKh02Pt5?*1}!?`*DuCaEDH9W(-@95co&6Txj)=5$LTU!a zc*2E?O@TcsaOk3oOH2E^o80>VD-hw2K3U&f**>9}Dycpi<7XGU&`J&9HSnm-NbG^~ zB~cL(Og21x=#asp?pcn_kAW-CeD7&3x_fByxR5!72kM`L&qq#BBTA~)tL1Wni&JN=|K8-7jW-SLVFCE)i!GF;8#E~`_q z(c{D;qA8-8vOpi@HJ*tqBhR9hF9P`$qjP93ppkgDUmnu#WCg$d*-MPcO-&6A{F4|u zf>nVv6gry#K>&g#Len;5j|6qte=)Wg9ta)TK0cwNKXu>JC-q7H#n1-Tm&WbD497eW z-z7~!j1sh-0n+qqXys-6_X z9I~_z$LtUZPA#0-g=d&T14_l($jcpQP7nLU>q@|U#CwgM*GhEyNAgVm6QbaAUS9 z&%i!!olLDBTF>jJ_QE(8kv`{OUi{HhP*<}4PFT{t?~!3VKAT(Itxn7#5yAB$)WCxM z#lI^@4|+BJmmN)^1ea_roMG3yH#-M17mE;$p?%vU`qJM(PR*|@JQUyV0hJsSsvZav zo0>xO+g-%X4(NPIbO+$Wc?^2(r?mmTfbR|r{vXGuANRNv zs5XAy1L)MOoHGc2AlRm-Ph*&4j!gp_6yV6n^B8S@o3}py6@?0=7=%R4EF5B&an>zt z;2$d3cznUN$Nj7EtkO$kwjVE!tiw)35J7a)UKjOiQxzwc|Kq~U^UyeIzzNgaK`t&X z0f8!9-v$QrU?^wSsQ{^gTVc#p4b)~rIrcI_-@R8?e5U{%IDELt$~zQ8M;!o7E7+jn z!NJIV(A5LG#Px=`u`!0W^2rL&PXIWZ#GQ#FeOAuK%S(FC8-aed9XpWbK16o^pG(sO zGXZNNVy~CtydDBlb!6P+?9sjLnv6~y39)XlGQ)TpE!>x5p$yW~XfbdMV`Cs5e1_qR zfKCbu8ft2!^{@RngZ}{}fZUf~R{?Fn7p`>rfn^t);z>K{%?=XoM}Obm#Wz#48d0>+ zT5jlVLvv@-4}Bt*ZJ~?yHYFu}u%w6W_T2);%>Q+5mV`0}}IKd2P$j{tzCs8otmF?fR4+{H3ZhyhH zLP2m$w;j1}sBZqwPVAlwgEtffh1``TDc`Y5%UtYE2(Ll?$L%%Z8Tv5RKIIUxkmq7J=Q$yxhf*%m0o1UGwkB|4rm zVMci3DU2!W;n*CUFpCNTRy}eKX8FqO0huBt?a+5gvL*o-A7wKPyml5yG=x;{!GPXk z*fFKkLVyY^tsxA4xn4*NP0#{5S{3I|WF}WTD6OKhjcQGh*kWumEIlCRV zRu6qtYIQukbm&kh->Kb4+P+?)(!;@>qUl^6UUt^r3PrynWA)3a8)J}wz_grxr`H^a zF6GV?*h#K%$lUy9egne?fUIPDE9u@n%uhvdT3cJ&*!W|F$mG`C1_FP^(o(4hy9{4L zfcGy-q_2a6FRbgf$>g@8D8Uv0QPJ9V=8)YyP9IB4OC*<&MiK$Fya#_BVfZM!=;@ig zY*)YaH%2I*x3tW&sJxC1g_y^yif;yf!_@BrAOUuOTyw`yuL`hid9a_|*%H00%0?ul zdYBdgm*}qWZOXMDyS$ZnsP_D~$Oxn0M0ZXC=-EGNVSRT|x>!u{5|M|aWntTma~aab zh!Ri=qnW#qiPY1!r9*AYhP0E%nUDtxzW)9T;A1Sz%*v;pa!P7o?`8{A$tNm zX>aIne0xtagRmKjioK&j(5cJcrl4p?hj3CJ#YD9z*?NqFq8^bzEI)XC{d3!Xr3%~0 zdv`~nI(HdRTkJ)j1c+rlBna!*Z{8vD1uey+{zwOdehAY61mE0W@d}ef@&gqApO4~cvY~u7GvVyyhc{}ZC8 z`WK9M>Q2KvlBxfc#rB@W@KzxJ5e*B@LuH1MtmZ4D3bPk85@auj))+*wY+Uth*yP!h zLidw9ews#al%AEWL~f}MMePmF0+EL?7$GkJrGWu?e`@H?8;s@?kDtiXWkD*Cq~*IC z;A{I|T?{jgZDyv0cftL8SMEjm#O&;!Xh9|}KZn63#EEwE=D0{{%O7SJFBWvk_Q9}> zi8g;RE+>HQ%Fsx|=Lu$czpovKUraY~@`ez^l#>&>!=KvAUfo~Hn zGDpsnIFu_Yx_XwDJaSpuxUjG+;fl=8b_SCMD<-1lhkyPgMJiwuTc)W)?rp`;q}Qm1 zR2~09=Mo>*cC#LdSm-%O6lVt}J`9Oetr1aS;7Hh|ce*{4ow*PQyf2^^0 zmoW+#=o!>WrJ_+@`jPVU5Sp{$c|SZfb*Th*UsPdfsa%a`v-6Bqd_2ON@EK8BVH9K% zv|Jnnh%*}t9DM$b#$zG5IXNj!{Pl8tv~1qDvsoXF5hX@p=uO{K-A1}UvMtvs~1q}^ZdDjPb z9(En1NmzN3o~{zNK@ux{U!c<&Yth}BuDd?x(pJnjF+8MQM{kI#(SMwtzH=e*#HsW*zlc7NBXLf&(|(395hf}rX(=Aa^Dp`*E_1#V-6 zZet8^6`u0w&}96E9jdx0VnY7&4bvt{CbKOr2^4F8rpnHoS_A6TZqRKfzOhD2kvcXu z(H{FfhTNT4iz>&mids=T12Mwv zR=N=Q{VTxiqnoTL0*H-HezNUx1jf5qSq*}FiNA;R7QUZfaTY*RbJogA-1CR&i3exT zog?E9UxDwH)38^jP;4wvI_g@3z(UqeJ}DNeewm}Ty;|Hm8MP857^|@6`%KmB#!7%H41@L!nf&d-c@IAEX|Y41Oe8QX23a5X&kKo{83;sVJhl&s|Rvbzmn zTjBUbxpeF1P13%wuyy0cjnS1TopCLP0;%lEZ+gbPkbPIwx)%H=sUDUzJoot35Xlrvm?4>sF+-&6C9fL=TU2d=LY_QPR#|`4);dj zU3-Q@Z+K(`QM{P(5&bO>x)QRMKT=iM3Gmp?#qs&?nf&hF$*00}6B8#~rzy|%ohSF7 zWylh!Xjqd$4BIonW_WzOE}qfF7n zZoc;_s@LA(5X)x~W;!{iu< zj_vN!L^lJYG*J6kv?4JieymJn)Nwmv423!cDX#=uG z<*W1LX)!b{;^ooy@1*#TW`kr33V1)%@(!D%x`oGGrpspIi4sJ`gDfOv}KD4 zzoS!hI*8GbtLq+KBSDEzQG-z>LJ?Ls8*F+z-b^$iWpOrnbt(9Tg3QiK%DWQ8<^O&c zV_nsSI|7$f-WyAHE!w*7uR8jPC(Z9fXV9D0nf2VGa2tpFA9m8A!Y9}FR<0ru0grYe zFY1dh!;yh;g0+2pNyTUU+Q92XsX;6qur_QO0GB-c}SkNBQc^G9ySJJ@wab4z^J}zKx7TveoGyAAS zIZf^w9c5VNXH*5w2a+2sr(pf=h*1lNY91pvmbSmk?+P+MsVCT7W_gKmFxMoemF2)SCzNduXVedgU zylD5qay^ASbs*BDOTWcikMqf6EbFN2F)_^2oAlQDs%@tWx^$b$%E#A9{I^JNv$4{h zZxap+A*-qx>5pj+k&(9lQ(j2CnQMk}7;o6xPdTpg#PYvSdp+xkv`i5?%*$KlE2tFX z#KmWijl8#VzWw5b^@^Wb`~kn3gKmm1`4|9$bAJ#G$K?-mH+fXz0MN^1!sQHjq#LJ* znLs3`P=sQLw$r~i#F0Nm)86b)wN^b1pI*o(aIOE%+;AtYkYi-$ zQAPyUJW5VBhqRqk$e=+))1N8`T@9N0ZxR9Pj|`zCdRDleu&1l+Y$p?208^ov#<;!{ z|33Ac_~A_=9Hj2NAzyJ8rHS4m2 z^k&L4`xxJ;Q7l6;jPcPYAa=+jF-DTRygC)V`k$v0 z92ZzS73A`o?vONTYhNuKlwa6T8dRLk__Sw~-4P8#MP=n!eH2{UsKN&Z26i8d=W1|* zU?d94<~wzt@@*q5t4Q()*>ONjwzG$A5LHIG&2FJ%mUmx-9mBL0ADquZ)6zjMm0N~c zJ&F64lwFgPc5d#t&41(ej_&`C;?Z3cO$BKO4P$7kr1R&!0~(3xGySV8htL&);6PO) z^=eQF5>&6h^eHNXYYCk|69?3S1rZmB$OF$W?n=|3@XgmKjE2|bzps|}YKKzm z)iR`yk&%&V-<`0hFv*SQ!!$|Byc%3n*{SGNopgUS;=TE04!=78TS;dtNm0^IOFPWwlz9tv1(>9{dp0avAg1T`%u? z6WE=yG$?5yskDU%z2KmQn?rAwW&@#mM#P6=J+y)34lrPfy2Oao!V(5@8(30ZfDFkv%*L!M0(bC!kLZEo9*)`*RjmKyn~A zY^|(*E%;nx!vcKCW3Nl1??EIN{ppGR5M!sIt}?&?L56L;Rvztt=>?cx4esJNzcT4? zYi#$8I8%XolA!!=ghIS>2ef!kU<4}!t=NN@_&WwGH=0IBtac9YD`-CZWG>8-`wJlB z+ZzXIvb`R@SKLzB$(AguG)kpd*1p@z9q=lmxiLLtw}3!vn6G~#$5*UpLz$}Ke%Yq( zZX>1Sq{%s_wV@K?vyA((6i9vdoSNj_)3zhP3;Kak;>Aly%#zXAb#=_Ii<|@irszG|A&6S-D-dL|TQ|em#-{w|@8k2DkROiSG!%q44?8Vvp zcUw8PnVa|8|5B6K68l{^0;LlQ1AAu$mv8R??Za$ytpC92Qg8i=0b~ukwP)x;Fo+NW zRGrboPY^cLbnh^El>@y6G7rDCg!SSB6cd`&94Pz|!MaS!95CAw(+VbD(~?OJKs8}H z4-bdT@pkuw+3C~1XaVp~HnWLml#Kw-M$k?;8e%jq2tve|9&!S}=AANzoy~1+C+n5j z(40&$J(Z2F%#pB)LIu?Rc#6j1S;jQuW+GIr-+^8M>ertpvlj)67t(D0Ua^<=D#SN~ ziXZdG7?_#Q{>w(V3?u2{6BJ4i)~k<*0#mDA!dNyq(%Yb^jd8{fIc!tEW2`>|!!r=E zs7@ESCzOtc2A{zMzz|rMcQSBU0dJE>xD)y_G@j3X9z>@K;S`brtBU~OVvl*D;sXvK z7z!C;8>|lkUWlI{qfh_mSpOf4b;q&f*cBNaZHdUs&guVEyuWYDIh?$7Z|rfJcNTTp z-mb4s;adGgfMVN@gNZNAG5! zQ4hwcX&d#ORMI^k=A5wK%D=F@cvG@ZDF@Dnas{^2Mfy!#Ifx0Jb>wTAn#)L zM3p~pVasirx&QJ0g-jT#1O;!H=0AM&2-wUggxo*V)(uA!+iddR(ESC@Rz%H+CKu%N zun`%DL|;<{SL#jz;0ff;-KWhTTnZzC%&*$leHm>g)U^DGr;? zYYjHe-Fd3d{9s&BpSmIKRha(q(C**_X;=Y46*graDMaLp0vFdCs1oO1ztRy*560ZnEyC#n% z?m^DYuC+csU3b=HBY6_SIX1^2LAq z1TU60#XA4Hu}MT^-r{Y*y2Zs^HQ!o{?*AVE|7k7G{}uROWI<5P|8Ky*jDo5IV1KM$ zT(9d=z*)4%YM$bhsACh>ng=3)?nV2qWJ7y93nj4xR(k({wikc5SEc&XY&r1EwZs$qIFtEA(O1g~H#4|{C|y$??pETrPa50?OO0h89s2@?k>xtjJK zN6EX#54M--qWzWuK!Iz2!AD+I7b72j{r>HOc?igj`~*p}2M9`tBv}4BK0ET7@@R`E zrq6hbu)k^oj>XIFaNxLsv9b5>Z>p_;*=8YHdjI1fX0lMPf%9jF+*K9}XXo-O%ihJi zT4#jAgy?i+rpz;UWmEkt$kW}&O%fg&bsC;a1s#IX(a^AC60nMz+BfwW)KOG|@}IP~ z|5`26|2^aAyHoS0>Zc=cA$whIDa`Vd^V6=duu`EO6r6HdF^>%vW2!AH&5} z6~0=OXodlS%5_$f7Z&tSQ%7Z9-DX&7)Xp)u&Z(!v@E4pD0mOF>0l9Rh-4)e}^ zHqqBddfX6rca)}i)+pWGnF4*-+|#ait=(@@SdYPdbIc2W{y^sMHrD#EJ{&t4c{=km zGfSgCqO=?1|IaxCB&0tmx=`GQS_I=uH5ogNcv1BZb(r3!JX2FRh#E5TK=Bq6WPrWG zdO(1o+W-K5GMooDWQSLA5{6iSV?#(x*lnwr+eUHwPup8sy5cJU8QyQ0^99T_AQfz0 z^IXeyW1T7d9f*?x8Al4m!29Azm<2F~2v+cgn3oZ^6b9xYy#+2e4jWg%xBjkY+)VtE z_ZA;*^!Pho%Ty3b=gSmpJ!OYvdj#;zu!_mbnmX!Vp&hm8slV7Q7z)WJJM`qp!G|#g#%Jl5)Cq%CKHUX^{&fSM1hDnrC|LG~_`%(rO{}=d)=XQUf@f@3k z-3OybxnlTrSqeb<+jL(-dJkGkF*k9?t{jYyf!px3>+mR;GXQ)Ap3j1yL##b|zUDs* zFD~=JlUfP%Ca+O@fUOzT*8}7Ma!3OBIqGpZdfV<~K~4f8VhfNM28^z`GOgoa9S1I= zdE#;8Z6oyj-gDo*uxuc_!c?u$_N#~N!nc0I5EtYLkAoO_UkFbR%oM=n6)S$W?#Yy~ zeNCM})>1|}GxZH|K70p~nY)a#1e)TG4l%l283jnxQdz0I*pJES0`%TNCj&TBK^8-L z2+nOx_$sazjIQeHGQ_Bryf>-<>o95shD>kxsbwh1aAf%*v=8Y5VO?{ii+20tJ2{sv zFJoSiO5?26{OHplXq@0-x}lN^SD1jB{-W=&YcWuk^isQ)R7`#eO-0=*iMk1YBd-)} z11ulunEGUL;J^V~N1PFqLJWt-A4EGDB6|2^I|aA>ys>G_1^4Klx z9339hjkalfKJb@HZ-2Da%rsq|D#~|CaQE)uO}_<3%CODEM?ZqH)ffO6GNZ;B;fV;~ z*21Eq-3yx+@88(O>~`^@9G(EGSd6|xp@$d`z|E#3C#`I3kV$HRTYAt$@~&2d!{ybL z(#MZSu((I-V|U=xTjU8Y4(2No)F=!ybdxjIK9t$5v-N+5(aYK^N4-`0*-co0!91 zC$5*;cL>=eyin)!2EU3i74gm?u#bg=g{75mHA&!y9t9SiG~s*(dgHV2A%_zek9`5i z7owyqz&}B|lVhHM;m-7fP+w1bhl0jDs+;1)wGg?MuzoNgtNO9STq!tj!(XBM_fE#x zRbS~!3<;r?tJM%>5oO-ini507@PR>}d-3s72_|#yF(AYaiVe4o9<; zlvJ}r2Zk5mJI~?bytbY`)$JheL-^5Ywrt6s*7WBr_Lv&fkWV%@GrRKT4b`F}IUnVf zm=SCEnJdpanhkOo|H|T(5AQ{2^*PJ!lK*QAO_(4q!vJyl1r@dY5a^d!O0Ge&)nwX{=7$UQ2OKLsm}Gw4y2(6Tb}HHJ z#oqr7&PUrD6Hdn%cj3%_nysTwPRD`<#K?HD>Dl=INcyavjNZn~%nttKEwsLZ%Xil)u3mQLwrrhneEwiU?K@8qQ ztZ4y0J`OwxNQRIh!-nu1xyM`O&MK0mkDEW9pq~G;;4J)Kf+esdZruFkg$G?G{V4gz zo+dz|a|Jmg=~@R~nUB=&#pgG{*qPEt*@3r>__A}6Y@un;f}P!bCm!&j?R|DKgaI=S zhie9KO5XWpi^<5%yC$xh;%rXw5jRBV9n;5e8+=U$kk_4mUiDF#SWH5|Tsz#?ZD+gr z;E>Hhc-+06XBvgYJ-!>t7__3}@>}&AuS4)HRBZbT&tLD7UIY;W)DUez`XL!9DN-F{ zt-`T+x1iuCNk1L-Xm4#5J$Nty!;tPpRMme(=4I+%ov9Uk0-kGA9u-E!?kI7cIpm@VC-1jWHDiPv&$BR5KotxiTr_3A1U;> zIxs-Fjg0Ro{4z1Y&ct+LWGN8=Ab-E-H;qoH4%01y>VhK1CZTwQSMKEZDA+~NsRdfX*R=M3kPLetZIi-72(syZ z?(gp9r+;WMwxhu4u6B(_kPho>h|`c6TwN+B*PpxzvB8(TybcD(<##)x#K4>lQDf5p zpeA9~hcLyA^uURp8A2<#?}9e4AZtnL{t zp8t*KRzOJr_)MzCkQUb1+xq`twa$aWK&%UlgO7K|Wln28Qf+Dkumt1DrL?vDmbSL@ z=(?c4MU;{(B8%k2Mwp+#@YO& zRP-n|6WO&vk}!l{g7z3jLKFz#R|dxKn&;243A;~a;D66FOTTJNG95;eb#nqQSklE* zTjbaS<&Zpd$X5On?mH59(A6_F?E|I>W!S9&t*S1^LUq3Ap%1e}bJA+Xqn%ndryIBI zJmtJCr6fhf=lO#ss@n=$br5t}$$)33+~UwkW>w{BLCJ_b%E){7_APiqo5+lgo{1?s zIr-<;ub{g(4Sx_3M-D4oy(svxQ6t5Ej^7`69Oa~Bkpo=l48GvH&!4m0wIvt zq6)Cm*VdB61;(8*$$uHcLp{lZ$zG=v6v_pDh^gv=#}uGPCoeOyyS1ZZ(m;9f1E0Yj zYB!D~$CUl-u?Eb-5o}4aN{u*gMl$uxYIqTWfi>wj+hH)X4Zx;%;y6mM$SRZ@&|{CA zCcN+H&|iGa>+me{&}YyxAja(hve0SEa{L!kj$nhwm z&19omoZCr(W6!j#JAxjKyE_T09rSY&=dHI1;O+uWo0k0) zG1Rh1>%hQ9EU8_)l40hBXd3?FcbNR?i1gqeAG3G4@$K}T>_v0~!dp2U^L8&&GY}~` z$gg>KPWYVsna{Z)Lb6_kV9-VnKSYo%>9fN11B&zlL<11BCeH6c{KI9`M#!Am#4dSL zzNO9939lK1CybiEE|nm8!aie$L6w$F+Qc{*SF^joN8IH0mL7vk;$)ihHTJ8;vitp+ z-u^?PtjxVs46a}H@*+bFd;}?>Dw{c^6b;xJXJqDaFxmIj?zK!eJv)SZB-?oth7hk} zO;#sqf)N8hJo9~vEGM7`7ji7uLg)MUK`X-TbKVAJRtog8#0%>DAV2jVDHSocb=DQ% zdy4j1J(x!uUeG*NUsoqADw+pSY0sW9TxZCa0gR2401uIdhUO1W)%K-Ie`ujxd!;aZ zZ}Du-Bi^<(KJKN3HU=Z_VhK%^&VSZR2=I`FeWg#_sYm?8K@>O&`Fiz8aK^z8-?HV& zb0=74v;CIC^s3#YlB%-mc{~dC4Y3Ks_y(@OUR4`=`t|F}2;aPLK}5_H{W5?>%q2Qd z8q}x_c=QtjI_risPrlkhh5!(mN3NxHcw1i|lsB3P7t}9FdL$gt{?+Cx*v{^5gixkH z4hq;QuS81M3aZD|g(g4s%a8;#{H;zua$q+PI}S2yIo2B3b6nwv<1*kA2+Z{~=mHLkK{V_hY9lCn-Y1vv)$& z6c#c$2@GzoU%xI4uRWe1`Vecho|Gt{#=e5O0~@8W@4StT2OxRW`nc=R-68fEc|t{v zLF&d(F|e}(do0Es3E-=B=qOe@LX)9p-0H@|7s(IEBjdR7+c&N5E2l$7#B|?2Va@bT ztm^X`-wTHWMti0TLMS>Mu&f0lLWK>-$(3CW-_)KQ^- zyIyG9o6pSV-1av=?Ffoz$X)kZ_u5BJRoy5OV_MlIv>ZJb$ba zKH=dR*FT@#{q=kO_j*y=#{4Gf;56gRUgMMN>mLmi@`O?o6+E$2`Ite=W^Ra51p>vr zhb(Ni-Txh};CX8+5>Tp+neuRRA57iWvo;GnI%)6(aZ6D+i<}xLJJ?p~*ckYx)ncv^>S?6kBOqfLK24BVSG^ou0XSv!0i5;dzqrWZBn!xbePE?cmZ_=D~ra z9X^JMUief_%vt651n-BHxi3*$cKR=VH-5I~r^UJmf_Qo9=UN8q@AbS%^!OEL&twfR zZb)A9uqhyP7bnwQdp51>k`qs}%Xl_IyT){POVAg_3wlSbIW;70`!lu>iZ8Vv_BB9E zv&jl?WY_z$A5*6((|aO4hws+TjF_&*@0iXAzbj&)1Vj+E4T{>mlDp{M@_rGtSRWCI z_E70V2xA^q;`FQ?`{e2zZ|2H4^gOStKxBj*H>&iO z2}0>DvlOLBf}uS7PymJawj_AqY5Vzwqbt^Vs}eU~?_&7Wskp>7P4m)t;0%79;shK1 zLkXT2S5jsot1q9>rGqJ^~+&K>^E^(qb1C`8w-?Jp7|}{N(zTn+>6`mMN^U{gLy0&w;_d z5v5_q8@s3@{M!QiouG3uEK(r9+ln{uCF*9i(JbL1bFRtVTN`}^*<|fEVsSN#+HbV` zMe~1A_TKSazwi6-E2I<=m5>plQrSd8B_t9;Mpnumg^*cM8WJVhQf5X*_9`l?jBHui ztL$v|@$CJ*$M62#|J>g{K9A4$Q@vi#*Y&)v^E%JtJdWcGVaYptsUWpaW(&!utzb`N zubQ!OA2K8aurbV0v9Wl3isVk27`hR30+qnRV8C+FepnlF|F; z4;(7(tgZF+^`G!(xjYXH{CJLZp+)GZNN1QML-##FP*FzYxLpv}iGmAPH-vdFUatR5 z=!b|2m0XY42CMeAFq~d2!wcOl-*gx=1D@C>%6=}~a<)(J1?AEHYF^6|T`d4*U*OXP zYJon$5YL-GYwwv{(cXiNCY-s0L)lT>;rVtx&qy&gm_HgSY=+kCccjYbx_y1RPmubm zkr7y(P4c#fVP6IO79&_Jvv`SSIiQXTyBsIHnwZIF<6|kNe_5l+mO@@?7X=Jdf3?6j zkN6iHZSA+A_$GClVQ*gI(nl`rS`sgn;Os#OiFSy>VF==Mj}+eH+T| zKhSN!FGmVi)!3Lx70?Q}jFT|Vv%L55pR=tmYGT;-;ePL1(a%*O!-}VA%a2)l)(u<8 zyc@Sm>8qc7H%Soy?2YL@?m2ifu3{FVj!vlMQQSw4&h)f<1z3(hVwJL-@m4@YB&E?S z4K<+5LAtMpR(>l?yjEd1fWD_n$xdEwZtxqQ4dR9qb2$Vb7y0N?U`9KolK*|T@7ot? z!(Ri&9hY`hwKaa|%4FnRm*`N;{eUYN-<`Q{=g$4Syzu1zuJNU-YYlT?%Cz;Uanz10 zy$V%$g}v@`;o!$NWo7V2gv-L^tn1&PIU+PQS;TCvoM|d&opciEYW+F6Q}>8YJgX5eb+B$DO8yFgx0URNLY|^nf&EXW05e~ z%L8AS9?NNO#IP5MI9nvT&37Lj08#YEHQ)>YPd+Slnm)LI$L{F$eH;*A7QG3DjC=GU ziU3jvPy2;?!K_-0-IseV-nskE>e>mKmy^dY&e<9MYBEI$u~)!|RQSu*4_O9Ou&uhL z`lWjyxo`lqfV2}pP+#=fX97lI_AjOiQ$IvX!1GMx-j2IpZY43!tysg)|Q<-J*8Ku9Va_9^B!=ArSg-^tuF2JP-2Tt ztak0ZcqhfATHYMn7!-N*RO}Dk>4?aFVXz4~qvNBB!q4sj;Q~>b-fEU3k z1rZtLyzc1EL9_x<*nqSJk2{p6LFrlx_B(1+k=!#k;50K;do^Vy}{%|Cs$ zYZtB`I75F3r^LL!Zl%CVMo|Cz*ZKRgt`n!DlV!l40f`5Fhlya6#eXm)5<>>Y{cW&) zF#&)mouwX@p?iZrG%Rw-z>ERZ97s+O`KIujvH5!7E02j8Z`;gNlAs3YfnyNxFhXKj zP|(rYi6oX7%Y>R`XyXgvhhC5F$Liy)dyy2hSHHbhka|wcwTC#v`+uDirB@zm0IRW5 zuE-jBzxfO$2nz_2Bj+LJe`>G;RN}B2%zqWTIiMJIQ147zx#EWjCi9v753i4J#~weo znPzgyYqs$GuD3&5(kmv%@!N%mdtOObPj7s5w1Es+@wd9VKfqEI6)RCipkhB0za{P(vy6Qo8Wco0 zM_N6dQddc_)-oIo~>C~TFU6$KpBO5NaQ9w8wm*< z)LckQP?rb4q@tloEwt+Jm_SwUFL4x@GzRks)5ue^Nwt-iUxi5m!IIuP1jrtQNT+_K zCV1t+Qv=BVNtY?3Rp+hC(L0gxs(gx0j_xu-OcHoa{pyO{ST!yI`~dWJT6|JMF+FLa z292nje;>hNjhmD6*Qr!!x*Z&5@jIp1F>FJGgF{wNj}W5WmEvwl>JLPATkOKen`+jEK7cLRfX zTcCtj&EV4e9(D3vZ0u=Ic32|}_e{M`lJ$QW;_Bj3-~Ai6CGebhT3YBjS=NrAi&GQr zwXXAL4^|K%=&{$f*Ja-ANMR#vRLW9CN#3gT%15f8GK1{l_iZn)4T9|tBCi9jH>)K2 z5Br2n70(<%-2*x;S3ajdS{iwiURn|q_`iG$+pUAt~>vrSFSbQ8+czyQYC5PnfU*>+t=&k}Z5-H!j*G0(XXwVOAEu9WYHq~0rl*BL&16gb=xbWMagk%$QJH-t5~DPYAA z6l|n(31dzS#0UT|Oho&CBhs@oG6=Psk7p=s@=UH?ZNJjoW@d!8_=MJj=xBDGq2F>V z`)oWmd7C%R9_uyauu0u!YlyBktLj=F?kbEgSZ+mf7IaN=bRD!Gh@mZ(Kqr9}G~v5t zSP~)sbnkr~{EXt_8jGn!+uJj!ZDA?9h<4jdP>U5x1xXC-S~*1cql0N+9DQ>A!veO5 zOHjxu64L$%p4#>n63IvLAo+C+{lP&(SB~&ML6(ma9kaq>yX7GH2V0mZtXIS+9otlEd00wP62f!K@)`4KG;z<#2hE>N1)2NaI!ots~ z4;d_pzVhya55OueoF?F0o zJBfhAL~$;#f97p%fL9>?r~g%*I~Nlle_uIP6-K_M(BI+7IQcDRZWX$A>-!)z6Qn*P(~FKo`lXM+-L(YZ^3#&BByznf7XyQpM zRKPSHNVJB!IvF`Rp7Y$VBRG)wF8v&9rbEPxxCL2SkO7{6oBicA*Q@{-{f2AWq`Sj_# zwVn0opI};vl-oiI&=4`PTR?54&df~Lx3(rkPvcHZZ-%f4!<>2eB4H~1k6f4{0)*`ZsMy$e6D1QO4d=MUf`%1G`N<;e zzXJz+Q*^S;ce3`M1Mdv>JV?wA9yG@-4E78d0oe*{EBy?>6EUt(KupZG{midDBGjaq zV@LmnuWo;9H+ES=^VrXOyQVI!S>rMbM{eQ4{VXJecI(y{hA~`;k#Y{$UFecn{K-L& z3wOJ?u`^z9_HYMKTwE$(7B66UG>qt;1w;D7u^kz~Yt3$WT$C=4y~ADN8!rIk2o%qd z`?8(5RSWrc3hJ}$tSscP`?^@awM{m^Yeh$D>JNt^v0EQ1O+nVaYpamUi-?Gc-d@?O zocP(>XlOLm)syvRaQ_VKSJl;BCN3g*1E?Rtxm*A+DeW*sed5MRPz;a#9cj3C3|pbQ zvon174Wazz=B{XL4ElW#e`@;;fd*=>xubB>$4t!kSy|y|E*5)?>u&R)P*qmu7Oo&* zuM{Q$BUxB;;HLn}`tp=@1(@Aawv%C9!-PwAwxQlQD}C;NJC>$l<-;qlJK3T)Kb<_i z>1&C_Q7}o2gr2XVZKSwX=(XV)W+V56T_);C#n-RSqm8~|HZ5sa&KJJ-*t5L-8-PBH zkHIb1)zww?M;44F0J3$mmtsKsRcA?b_LO%s|GzFTk*| znZ_?)?t5-?+_F;uu@Zf~yA^YryKHEHOK>=zW)cX0vPLW-*G|+IK<@xL&hLqzjsd#@ z{$1h|GH2!GXB;bGdBs`$rZIs5^Q&(CrVj^?TsO*B6$$wEI3As66o!1U{R45$L?IPnP_rE3YNno-&mI6gGHhkD@KlY`Uy z=!cOYkjCt&Di3DsEkE%J)!#T2YrjIT?0pr;D0Mj;MZnCAP6e87;@oL=1W>3Oik0 zyh8vg7RHATC2b<7)W>8pM1_exFQG?3s{%GZoQ%!S7)Sg2&nQg6k_qk{Z|fYSz-p^Bil8o^l6wlcQtJ#ohxjQo{ALsLggQ1JDSmx!{>ktXs2pzC4n>|QO%fr3S$I$(}a?{wxY z93Un_aI@kQ$Ln90Bd=^%e;@~l9hU;-=FQEKIy#R3?YY&C49TvhFYLVk-_A8{=?Q)&bhDOrL}K z1nwpf&(2~+vEL^qCvmf~?%iA2-HV}p=mmc^c$1Q{>b3FL3#&~`wsk8;U@0y2LPCjJ zn*HSMOP4RZASuIaou-qD=yhWqwrtr#By{MkiFhO-Z9z(dt_eokF7nd}EYz}*_T9CG z{pSY5xt+@9;gQF(k#oWeDf4v@3QEUNR-0NTNC7$=FlcJofYd*x#rX(eXm_*wR6i5Qxb+bgj#(w75W*MU# z%(Xwm5Zl}t0qe&7HIN>I7l%&J00y7F24w&wYXlVpO4Ple+91WSh}3hI!xIx`D3}Xc ziYy+v(n-{#7-oQkn`P{6JRBq-gUm8hjjG}t%lV5tB?G=)*sJeb3BTVblQd+2Re>RY z&$*A);s~>uXz@G^S!GOdB8F#Gna&^^^OB?(}NITr3Lk z1^067Z61btQ!_KwP@&J|<@faYk(2|LAgsV~M-J$$!Zm8(DCkDneGL)O>8?8-x~#9d zV&aDlx8FJ=3n^xQSk^K3*r5&+0}Zt!i{$-U>3WFV$X;evIlwiOH~SI;$PJe$_^Opz zn7Pfj$4m$WwF+Qdjk^O zcO!Sn?~oN4cR>~zBRJO*JjMTPD6V4b>UGYFq~&lI=J`2n>;J4IwnlaL;P%D2gL^1c zw>Y(U_AeZ4HVg#r^Wjpi=;)AC)35wJBsm`=p#a3m>qy$%)G%pN8q-C;KTvN4bZp^y z(cbdM{u}78`*d71vmpz9Y!{`-@U!y^r<3Tqv8ju2Zt9PL!qcM zoUkoa4S|OQ&$6L4iR7s|)HpNw@~hdR=f?!)h1Y>$Jw97!5|2xEZ}ftAq6xGZFA)_X zvL|>QEl%#ElOo%p&2XBUt?xG#DOTQy{5qZ)Z~1xPd&nQq{REdQAM-GyHod0e=TAqt zB3u=U8!IYIK?Xz%;95G;7QJlq24HWauZYRru#N7(U=ma>sAE??e#YKIcG39uoYcJ& zmDSZCHSln77=c2IDA*@K_=j)*u=oBFthsU7fa3!;&gp{I!QtWI=4ueg@XYv6WL2uF zW?PCDMbBqu5w-{p!xyj!AV8qA{_Kfj0rCm>P~Kw39H9Y_ad2$RFVJNmT?5|}Hn_wT zp`y2tlwF6g6~Nv%S1n;t(QEAVr)G2T1Z_;`IJW{QFmEC+&Iw%}=+Ea(5WN(0Wc}v! zJ gCA;m5Ya2puVecqg_PyIIWk?-C(Hr}cc{VZsBbQVwnfFL`#^x!XQuQJq(@#4*$n3n+}4R5Osvx$!Q+d8oX-2c7GWh%*4!$xLA;v*K~pf1cXjBUfP}~ z;D#iaEyf_ezJO#8zZ@?dByzw~(hA3K{E`Dt2#PzzY9!Nup$ay|;J<@f`|_kPZiH)l zhd`Q!k)UH{I64Rr7>%GH0M6T5UoRjc;%ILl(=KZbtl-Jx$JlSaad8K9`iFTd+9-sihg5?&VPU8Hx&d`;#$JXGcQK3MSmp4w@9Nf(N; zgw5Z(DkRPO4ZDT{Tz!ac#ul;~DbDf{jD&Ik{}{J|DG-ZhsBVZ1Kr(9Q%6*A^t*bdd z>wnr%evt0}e(dPUf<07rpa1R2FDd6<6IonwFvEf4|-4Y(pE4tQ1=VEH6(pDxQfeyGvtYfBDT=nr)g#KEu=DwF^$zFih#ZMtA4g?INLmZ zd$*u$>>4SfLb!{K^dRg!x~Y8%YG?ouj|8<$(?8*e)45#Vx$52~rh|7TmG zRw~xH4_FKlgH7kKD;{5;N;E(MAp`&O{Cro`X}Hr|zFe>@inhEl{*E3cbclpw6&$oR z#n07@rzk7-SXU<~x0IaHo&6)~pdcE2&9}96cXvm{CC$~sbvs!_C~SN!=~>YViCp{dp3~-x#BEh1=p0n zNB6TFz8JOY+?V>TdVw#z&HB5WHnW3}%%T_akBH#eVV6l?wMNM+?ejv+J2<`sM3VFy z)UUFM#=Z9acW-|Cm*Y}PL$PgKR8=DFLYOd<0UXDeQR;i(oSZS zv$F1%Hz)#np97_T^5h9-3!}RC%i2UTQFgZ3vq8q*I@)|GSlsiTZ>wh#8n@bvjBhpH zexAwRiFkU$V}hc0o7yNe(j1Z>j{m+eHQMqb?KiSv3{kPOvU-fq`Snxx*~`|YzM~sL zp(j#UND%*y9zN;N;(n9xz%{Pc^p?7*7nH)ZH%{YHAJj^<}OkQU-6Faf;x$ zR5V%eDF3$8ts*zAGI{Pv>%>o7+gfNK-%M??+D28j`_sPJtuh=WZ@-@ug6Jt9aQsX1 ze#zP6lpVxcYEa-S-V*Zpf!KA8N5y7dnaZw>#bZJGYBf@aj~oG<`3toRas)V)6X`A^ z)SbXn2X(DlhPFe@1`JeZ7`Yv6P_AP4OB|pQUY13Xq7hj-s zIaC2R6Y%5W&`_N&mj|$NLDMA-E-?`i1Y=F!wjS6&WK@_0pMQJal4W&>sDDw@0S#c1 zc0axzfW`2BJjv5nmvcYjB#ACkci(Fpc-!ATm&9V>$?z#bR&)J9(QJ>030E@Z%}7eG zSovesM^p>=UT`cTX+`E%0x0%YTOv?kWIRtFJ_M_Xn6VC84a=%gG=PjoyYY!<-J~Pf9##frri#oB`6FW`TP+tO7y=c$WHgTDMd&0l?KYs z>$|Oo^cJR`^jaqzE|JSS`uqBxm6erMR}TsZ2!QP?Lr&e4D=UQIK12J;K>r&Ra&0uXvZrjr?zkYMx(x@GZrQ&k_c zeuy&*H`Bz2io_3hgi!iAvFu1q#uecxAu8HjSC@`1vTEi_a2&cGBQ0GANLn}P`WHDL z`aW8c4ExM&sYSqUYAxou^mo| z{p!WIgY^-`PkV?s?qW)`$7o+)26)TY#b(++a{XSH4Gs-$N7@DG8a<-!Lg#b~_R?#0 zvd?MH$#J+)=(|U8mwvbk(9`^l_$gDiJC{!b%r>{VdAcPw?ck$3x{XOdKSsQqmG8g^Ea)M&_|%or7nltSpi0e zEKa?q`oF%_((J$eO?26fh@*Uu2>2MH(^@q$8-DM|pW28=e`&|pG_KxG^A|ItV4{c? zw=;lGT3t1G93rmwz@r07G8dPYc1(dkJ10ksb^XxMqc$kmrPOlnzkd*D`Cnf)u!D&R z8pbm&lx!9b2y<7A)lwn1PwZ-1 zRg(@A+OmnbCYrz#qC3B+L@qm?bI^2`qo+%FNfKAI5PK^N*gC<_f(CWc+uW{uPAN&- ziTp4r$YJ>Dk<#51j3bMFA{x-CR64>a>nB2R`3?N`WkcIbPOjn}k*os|h<2lNk@f^U zA}i~ilSB9rZ+2CLQy*9FSZawc+P?iddq{b1401$>2a(0QOwnSm)q2l|2``#Khx64E zW3fx~2cnv{olF=uTo0^!YfW9A=Xmw>%;qNh_a8pcK+96t-P5yyBa7zp`p*-GPdIWO zIwWCQ`Ru+Dz8?}@SPZ;NOcc3ZgXaYkW5@RG?^04kpfz!HoPe1Z>4I9RuF45=I2V|0 zSRko~5a8?i0XtOhMq$X5ire~%t?a7N6PbX<`ID-zHhgPT`dG4=YpKRY(AW4*yoAPqIabp^I zKFD>W%(q5N0qTJBE@cM^$v0di>A1UKGX||F{x2;}8a**cYG*At;TCww(9jEgFd(fz z@v8w)`1^lD3z}mt{S?$#*49S~-!ZRHKVd!b@)*V(qW2qk4Rb+q^#AAGKV>G^DixcE zltDEv(SDPxz>5c~+&}IZk}0moT^pnlchx$)5c|XM_#)}pvZzzM$k3v(-S8OZ`7%l#l+B3r+Z3A4`Om*XXgn?$*^J}49G#p z6LSqPe%@3CjiJ8s2bj55Awa$6jc>XxV{#5XsQg9(yyR7}N;>mPZt}@Ndi#tnpQPnY zF)BS2JBbV%1&aAjG*Au%JPoc`)E?yVDlkc9l?wL?om1fv9o&d4HaM;O`?PEg@Db&4Wk&6;n`hq#Q%A*6z)}$8`<7gTbOYcM$~B_-s-pp^J%HJk z6TB#-0gpjo92YGP$8I=^Nk#i^rKW~009t}Kx&l<`1K}ZqwI`e;UY_JmYx-_ZM$Xu& z@`?6;a$!C@p|45~p6g_EAHQlKI><6;Y7cHnSL|sL6gRq_w$U`_|HXBTwM3$%b0I~X=yy2wknZ*y3 zq_9Sgy4zP4ewnPF?cU@_bSU?*Cl4(@h-kN$whI}TlKb3{z~QKJW^;8<_o2KTt!r&E zuj!aP&WG;s;Om}dd0H^gpti);zj<~VeMi=P`^=T!;)(HGX#Y5!XFG21?N}4^^5r@@ zBqd8jEc9o9^X|{!7D}Pr30}#_V%Xyywf)nlyCr{tNZ{DKl^tgd8rg6F72)_cK!)Do90jR&TtGS{iKtju6DgAum%>I< zu!LO}t>j?TdO^Xfo;Jzq9q+EGOfKy@NWJo5Zpn37y&2M$-~?o;tTynui%a#M}~up=oo zH8db-KM+N;XZP-gz#!cLP)0R7RbC~tlRD;9=!l|rFPbZ1F^5kG{%xX;Prkdc7RNni z|D`4EK!EDhp|~X{fenwreXM$+>k&uzE~^GU)9BMKD99X`zOgA{6V&vQcLP#Vtc{E~ zCUZd zTJGcc&aJPaqS$ou9ENuk@^)ZAj0>j%wt70tVR)+;mHxy4rKl(aG;-GA8S_$-I~fo& z5)w4#=ByY#{+Hd%NIfLNnE$6&3LIYu+O*2WeNKt-$wh((-fjQkpn9Q3y7zMEe8IYs zmb}{V1G#mUs0SqD>W#Vt5KdklI{iHc1G$y%iX5PFd z@L0@+0azv)ar0oAX2LjVK0Dj1@hdK#J^Tjx00T8OKVV!leqC%)(NFYYQ9XVBJV^8w zba3zd-%%Q*B@?Qvh*lc5tkLbFDHF}XwKgZWrLU=L=;;1US7&Aw-S~5OjFT??>+-JVNnIw~SkSw9Z%;e$ZG{K)tOlM)bEd<+WJ)P&01N|2eBb&^6=^1J8+iJVIibGjT z!o1@dVeBSI{U#Xn;x&4%tMGg;FzDiCJm-13yw+xPG7m)sQw4g9+ph57y@eAwW>7MgdS2t9jd~K2fvl|_o zxBMXgu`@7~C#>R{AH9h~LBkSA)qtK+rMCk94swoxqRqj zC;doNLTeJLJXrCgYjv(=b-vXQ5aGniDsp0Q7sPSn0lh^~M4Gd>xQJ$J>NLXw;LI0+VZ4&6isi|NPVQKA^Vw1YId$rJB& zPiuvywZ39{vG82z$CR*_ZJDwKtAlIJxgZei8O=mp%=oGg`j;N*adI4wBtbcgsboU* zx3sigBEyE|CgYLI7ys;=#YRCYi~hf~TO2?WBmwakBC8G1?#qRZ11@@SSe}dizxxPB z4*}=Fc-Y^+r7>v$J)Q-5^ekbW%s@x?9-UZmJC}QoFzjcv6r`SoQWAhMyhU#Tgmp~E zFr2z~!9cWdQR6@H_M)P`Uuf*(!jw)hCQu1D-nvx++zbCtRecu_R&)}~5_Te={QgR? z95x4LzgEFoS%B*YFRub3`oIAX5Cw?82w$AJoWJBM6L5EYyt?P-;zYRWHJv_xngePI zm-e;t&4qnbR9qK!S$HHq8m_WbB4J!Vo?@{D&wy zG^n^u_wpd9A%jP)aO%`YT;vc7)a?Xl*3jH63IP*Ny!3R(m^!NxEHfIg5(`x zyAAb+#t*M!gSDZ;7&71pJ62F*Ar#-s%nbQ@y){P;APMaDVwdSppq!*2Wgx;Y(~cww zY)<1>EaK}+-H&vmMFAL$PsS5{gx;Mb`$Zl_%NwT_bWm_YYjFqC0}vqV*uYc>kBC_P z%xL&ND}8C>VO>ki(Q@y0QvJTnNJ=XwX8owW2KJM~Gvci>eee&N)H*v&S_Tew|jai1I1Y9rw5e-7Ye8P+KGq zk3~=CL#+E9lxCEcStsHf+$IKe{o8MzamZzAT7J?Jl07$~%U!hlpUjU*j}H`RSehV0 zM$ZTCwrWr{h2MKxT2y%W`3I387;JB64VCp@Ghkp~fP37RiL(iN-25XWuRPWlTG$Vo zWJ+HA5j1eB;5nqC zA(3T#m`iMrPl{!tyJk!HCl+bww?d@1TTTsW>FCZ3Jc<1lp|#cv4hxzgAj(bs8yU>a z2W7k56mwN&7$#7MOii?Xc(xMX3L|PVnKgz49xERyj@b`&b5PYE? zsYN!+70^26a&A%fniNtIk18L?vNNA*+YuKL7|=;B6DT6MWKgVC4V2&t<{1a$9IpRo z)Z3Rfl3u57B3LuJk5ZC~Uw%!^T&%xaFRJqDNaIxf zigv4h$eVZahe1gVr%RRjBz4)*qoV&h1m5N1r#6|fO+T?;8mMBR|J|p<&cf1JueZ4R z(3YknFHF~oPoLz?^L*u#+IMT0s^{LHF4hqszq3a25k539j*q>swzu;?#4EEr(t2r^ z_@%tX?h6)2v#L|HX?7n}&C?A2YoL@`Y{a%Z1AJk_dSx1Ih%EpkSQozOgEEW2H<+E~U-BmG)h`kCC#C zJ|`JTBy}|oCttm#boj%{PiJjL&)V&CmCIM3h1E!}A6SxmYWe3zbDSXW4bhnO^mZ?G zH)~f_HGU{5N44Bo;>qNP4F#JhQ`ZfFY z5pBR>VS`_XWv_s(i?%pA6kv1i`dEIZh7m`QejTzd(&-|qZO2{ET7b=#8;__MX!rNK zo9m+&t!X-rpba7`@V2Sroxe>BkJkv3>T2&k{MBN1 z+m^5|5A69~$HIrEY_9>cXE%g%6n2Jzf^uQI-=oHE`iB};kVCG?!TGjHGSOeA-u6aHkuTqmZAp9L zRO`0>B~R1Ix{g~2aB*IXiT!DUA_pT+xD{k~A0wFNw*R_%@nc&^J9;PtJDCZEB}w+J z;<)3FSDrb)&RqJqJH0NdAv*mN3kz=F?c)x`U=yq@d;K}d&;JKnkGy()nf<^JC5T>* z!!TT?7I6Dye>wI{Y}3BazccOt*shQN*Cx1emrzMkRP9}8S(UtOwSE%e^=Mvu^&I1; zWw`+b7kZJ8^SkdcoPJu0IhrV!gKym-XiRFM$?VhGB)VV!P&ZGazT-UKl2|$v!KXX> zyV|3#jgzJJq=d?dcqeceeI3E$`^k=bX1m|{G08(9JLq7UARU&O%Qd`|2R4+c|D;H1 zeoAp~iF<3qx1X2JDv4-ZIrKvHPH$k@g-YZySR`&r#uvZIpgZ||ivf${Ew&A6*9rC> zyfczqsPY3(=B)2l{I^pOFC;n-QTK5E47b5Nue9sqkXvL(7}bLcQr& zNJYRbEa>ZUECl=^pwYRi5xpeUlkK*yMJ9%S<85AZm};t=&7iZ(!5PbYn(BNSRe>B^ zGZaL5o~PRX3oua9l5D^!metQF{HK~YeoD7f`P)1#GN4Cm!{U9D`3a>vR|S8p?n1tz z|FPx7eh_R-U*qc0FO30Q`IP>8t|;{@8WQJCh6%BVg(cj~-a$y>ow4x~+)sPRwGJyou+^2xMNr@uY18yTn{VIjTLy)!ZxKS zz_<(LJBU|PiUn0PVkDAU9xd43&w9d$CXk!Eg70?HKFzK zmo_sd#`{RNL97yJfKXx*Dl4${N^j3A!7dj#DFE`p*$Q<{&E0Hlp0I-&{s0UKR{-t` z7?6MZ%)nXMyHmmw2td@X1n42)+yjRZRNi6^jKsFE{EsL=2yuMe*x!aKq(<(b;n+*; zH~p`+9IIJJ4#i1-mT|~c(Y$r~dL%L^_)zy<&|5E2gQnZW*Zw<(n0-8nWaT1zt67p8eZF_EkCl7AG0qtTnV2u<)k-xw*MeN&%-O*#9Rzqwb}rr7g_PT74I6 zTKO#Rus5&GoRfl8>J!mpAghIFo(i<9-h%rDJe7zQ{u|oBr8knk#SLDPx3&(91p5rL zm~($e!?axAz~Dlo>cREmY!v0F29TXA@oaxn0tVFe)6mqlgGQEKaS{mpp#%RLU8Nm zXJ#6kn%-J1&Nafmj|C1jfYrxrhrzeUo|4&E&5GFfhocAR4@n}D;Z{Y{ze)Q-S9UV**Fto53y&lC$(COQHE*n3^1;ud$ zC&%O=CBdkia@j`Guiz!}qaxpT`r5Iv*r6#>VNV2bnX6akXo>rm`RB0 z1nTlztwrz+18l=Afe0sgbHFb+7{9fdIHkv~C{=?|4wu9`8y^8eJb=dF>{*EuCvuH8 z`wt1&XEexdbwE}^>g(P@fRvPV@i>3c6JZ?t5Adb_P*HlP@mTIDjsxnQ(ewYG-HfHA zhp~N>x6-On(ZJbAnd|^T%D)idva*m06ClWjA$PD)9)=i^->_srf*PGipePGvk_Zwo zAAALE80&JvF`Q?BXFoz{*`@X08=ZnJ02vt>#n19LpH&slyfK^xF%S}ts|li$-Ni)r z>XlsX z68G11*>|Z=?IhB%U(v2jzW|DjkeXqK4*kKKR4$OXVa?Rm-mVfQ)&Bka)3vmM{KM#F zf`x*?dZv5jAxZ(82(aku@B)Z)zP&*cY+nHzI|mU{IdE$r-c!{Zuy%r_&3S&+M+TkO80L=?1$2wRcnE9(=JA`U$y z(x+ltlVndiVT4&yE7frqIR$1Qgjy7ocW9QT=lDi{G2Y$5y`up_VJ*&E~La=eCx0_Kbd%|j1Z(vr3l)14qnfZr|p z(-*W+bx2A`%ujay#S@Yp_a`ohEH!oY8BjN|6VTuyTn>Sr#;ZnyOhO)(=mZu~=6lqz zFb62`#cM$)?LTSeDo}aHHH7}QlU-wMQhFd@B2sVHsyE+3{{Or_2#6&T)fAE_(^Kcq zzni8d7pazIxQni0)m=!i2aZg!O3l4Nz`DGzDH-7cL$!^a;c`oA%7l zAOl1DI?&*&f?2m%X{hEg*fT354ysRi3_mTT+K(ElcsmgDXN{AA^RK%_>LE3BI_g z$nX1?b#ySZmRC^l22}WtN1~8C03O59|Iw9!YY@>v9CB-1?!X8V#xb=Fy|@KXzCU#T z`#}l^>uPGYQ&SJLUxhC58}oB8+k%4$)F4y`u^kQIA`nOc*np@VpOW$u6I)?jSYKZc zUEY-PZ>_(*4`J#D-7qnv>SIp0>mytr$Ggmv3s>7aErv=q5&4Q7d~1Yp9!Y5osJnr& zIl4lzg_Gz%`bLi7F7681JtK1KxVWe1=1?0`Z{4~Vj59*ARCftM7lH*so*Aj&g<1)0 z0b@Hc%zuDm<8~^l*X;Vxjbo-57)+RZAlfwwAvFfOnG2XemW^cu&VRD&>6%pkSo=ZS+C!xzP(P^BTj#gy3C8A2MYe)@DQ z#2(b;=yCo5p9LS!hd!6Te-{EYt~nNmDI~FEC#(FKiVFjGi-eT}aS3*v`@^z4Z1T4W zGNY7DBsxlLtNX@CTq@~a0pR>Bu}RSWR|S=NIcA0Sfd|hB1y|#jPf(gsyjOr=A?l?% zkTHPt@U(!*-^D+A*XyEGw!u&)Bn9kpEp>Iv#D=%Au{U9`p+18bH8J%Oi2*V?#dW!_ z{un)IPVka0;D0!7*DbXyFBGSXWFS;d8J8p2UYVIe!&lJMfIy4Kj&NR^Rn^9*MeKTv z92CWk25;mH+~C0%6p=ZB+#u7=K}fk36%&(^zz#fog;3Of@#185;+*Cxh*~J(B7Y}7 zTNEkn5oQgF8qLN~*=Mu!LMfM@=MpTcGEGz{73%2mSn82_MSw#A^~Qcg9hBj`jAcY; z-G84%BQRWTZDJZ3{0w2i_J(0WPfzc;FPDrlgO;i{lI#|FW@}h z>E_M8Xblio{qf^2*p7&xHgOD`{QO>n)w>G@Z)5JFeyL|yMTL*?D|iUib2K43kF)9* z833LH1=Ms2HG`uEHr2Un$uTj#6B9pc)0vQlp(Gqq!>~Ay69jAPGKvPK`^3Zzkl&eh zkcKHL%Fty|q_HwJ^#MCqe-0tGrcuS+`v?YLCLh;$|}0EZ|&980zpXA0e4?X z{N}=hPC?X@tMD~Wo|u>@DC{hB=C9hzc~u5f7gTwxa2k~QQ7{d<=8numguiFWSv|cI zh%@NeK#mHs3j*5=#(>Vu5Y!L|7v%k!uVGxBt!-Ld+`NuT1pxL>W6o}*LWdPz#5&5e(ZOHI&H&DM0V4{d3hIA3>s4{OFqwMgewF_X`wugnd*9HW zhro}=aK{}4Q;~)GCbVmCs{J9ItF09g6tnxI@eO5`u61R$*Z3ow{iw=GGl*gq1dN9?*Y zx?0gPUea|+=wJa_l!DJ_YpXa8lAvHM+#x$yG&FF}HbNRLH0Oktv)R#n-Ru=M7Kvh{s}qlFh>YSICkAIze@bKYd z>7NB^zv`m>O(-Oile4J{w=gleLH7e$M`g@QY!v*#;qzxaM)3PE?J)?a0JFxSoPN~g z3|Z0;vOe%ah-C4Kj@ILZx2Pro?IKA&*t?DSn{n@+TV;t7JqCDMQW}8h3VO;~O5BNB zxuOj?1UNYB`F|QyHJ#TG26L_3)h}MG$HD@$L;Wr?nFw=xjECjRIMAyfhGmJqh$ht#D;24SI_ot(BKo+c|TW-Zrl(o7XU%)_EO%gbE7nu?GO7*}D*QL^aYgbw7al!gAp5n6K%Cz^q#GhocckQ; zL{uWabe#ChNwKkHyqoeAE1^OLhqN9LAriD-KHe^1fRp?*L<#i-4e|6;uC-jmCfS)! zyJy!f9koSVXpO~p&` z4}mUo!r#3NHgx$p(lL`QPkbaaWuP(ie*a!LfLM$>z<#~Q+djU5NH2npjuu}m%1q1_q@eAdc+-AuvZ^dAeUge~0-Yd~*Veik!n;g~o3^C*gYe?BFec zHCVO_1Vn!$Iev4r>m?nX(TNFnxc=Wa=24BprLzyNxEM8LYz#g634KO$A^hm4^fTV= z|6J9c%4L9NihqCO?!SEB{;L!wAVw<=_!+;#=uAk2i9rc~Pg)up1X1juEbrj27y0j5 zoI+&WPW<;Q_=PWmgKwOcmBLB_F8iIs(_8xnbv$9WQ-a~toL!v{k%}GF&#SMhB1n-T zRML!u0P~>|R>1rHGnSoIMB}JJ9&HC5V2PyTN`4>TATlC%SeJ48%gCjzvq9^=(DS~q z%f`8Lo09L{xZp?hT$7tAY@XcT&f=fMeV@WsT~x9-)MZoHHJQ^Dx^`g2XSVat@2EB5tVoCN!s)G_2SP)I zvwjRKV+e|S)NpNaZS?wF0#e~lOOuRCY-$&oR9{V6TFG9>xo6DR)Y zTraJEFgw_8R4}M2pHx`>Nb5OOD_mDJ-=ddZywIWQwfy5|{=TT&K%cM_Hmn(K{I4#= zKVP&8&qfl zf^2ljrIrSszixn+VSb&5Eefh1d-D39p730%F}%O9vHrIdj3BHD|B zhRHG#X2RAqm5+?AL>yklQ4&}38ZYj8(;F=7@_g1CP^(nY{--YTHLQf`kcDu5GZ;+Dmmn@o?_uTl( z>cGF+5a8+9v9M>)qUaB*U1}HrTee|WCZKzw{z$5Ep&DF7&4)1m|dvp*}aq$QPW+(DsHc@;^1fR6)N`E;n@s*y} z88C=cb^pcS!0!%XZoQ(?GowDq6DOR=HlMUmcclhLo7HW?cTnwe4$ji}&>qAL8{sl)%cKKOFF>qM7E^^+jaNuLZ{k zml7oHd*Os7Ywk7u*f2Zu@{zuszRyFdxyoC-Q?>^91>s+riyVy`sFY>d=I_s^Mv#Get@g8MI&+|Sb=!!bp99LG06CM85I*Lkg9?GpT(%e<{3Zu!o2UMWuLV5Ky6|zp(|NND2~cTyI>1Y6z8y)|Pkg58$@xopEAWqA?{@nWxx|AheRBZa{$Ph; zw`F~=6PA`<2b4ogO?~Nbu&)dbwp}sy0sxP{yQ_tX)%~AS7ev!g6ryz}RXat+@@o$wRC;{3yYny+fu_Dg){vc z(aV=kvU7#%99kTf*^v07@qr zU$eGhF+S5QY7!)gH$aUy<`QHg7dnv1n?_9l>2H}8jF;aSyh+GDW~D_ZTkm3UU{~%z zyC9u5GuGLyu0!f;vRN9ga_*mA`O3QBlx_`>S5RPCU`#LkT~N9#EHl0+*#LkTI5lg|N`F9&JrF=d)o$>2 z`Bl5~_8i_BA=1pm*1XuPDprDHp>s}54sGZr7P(v2V^g~hroEj>gptI@KDDGvW50iQ zOIxU=-yRIs4qCu_8e1X-NX=A&BZHo(W52&Rd>f7oB(d1$^&yjmRP8pTU8uS}d}oU8 ziBUZGq?2rHHGf}$q;qcU0l2yIl!2__$x>vT&(67A3OB9QMY;K*)j)5!5D^ZoABzmt zZ-VR$OYNstEFR#@`?QmEFke77JOdK^@y+kU+}4=3h-NO zYw9~hcZZG*m)F3?8?ATE_4aYX*W+gVepN53uRW5uQw#3;D22&MF-^ozZ6WJa**-PQ zhB61yPq4x+ZQL&yIhx5S#rf;1M&C~2uQdEujm!P8X}iB$9~S3Y**}WEYWZ zDnzqH45fN!nqD*%@Ao$|_4%})&%S@X^UrUd-{m>yxt(*KbDnd~mV=CUx7;{ZBV!(2 z6anEAz}~f&#C42Hf<);o&SB)jl_e_cmO`2-<@+e45Imr(eiyckiNPh1J3JK0A8_ zxh0RplyAs^0%%g|X}`%_aQu49MHw3>40OSXn_yEgGYd7h0)aNyh@s*D^L_bMj}i6o zWtLV+?l5xrW<7@s?Rq_J#_1tdLST?o3nGID++JIb!M~e9HGYGN0EAx!&d8wlo0wIW zDOa9w zf1IE#`)HxLJi2D0_4j14fM*=eR}4#7*>6Ref;40kL&;e>js3BwbRwNcOz$R>(3rJ) z?wi|PeQ2TX{fgb8-~hH;4GHvllJpLkDM0$LXZ>YKz{{+5cG*ytKiXogeb(v4o-&W)^kK|d&(KXLTZ*CxsM2|F;w#%Ai8O=(?YT1UE%$hB4r(Fp3Odh3Ezyv zSr^(UPYt)qgA?flCR_;VZV-+}a=Ic}EBZ?f8v8dl2USIwlvk7OegX5g4i(qa0t=2f zT!i+-wNch16xwPG}pGF^w{*x%bOUMoZ2^CFo}OAAwh_q zsqO4@p0;(DLaOFj?l)Pwg{}$hGxK3C*VEx_xiUKd@lT|aF_n{8OpeClZf^?cnRKGz z+u_AOngqaI@r8H{RmS^d6f@!h6pW37uy{yLr>#|b)eWw?0_!g8Kr#Q} z(9jTW_c@o%j^B21nSGOd`Q6y-78RqQ@w0?PbwTH%B!YMqI%*?T@uZa0TC&Hs2|42* zp#N-K_Z*FFpp1@=<|O09=zcS|$I%c(r^ZHIl^_1q4#U>ofIIT+z-_dYr0FERgU4$# zRh0Lv0KZk<8GtjzMA{DsS`8FpdC_*@gP;K7RWR zy=X%b&`WphU%4^2?vFc3AuHDFl|6ZI+6TUn^tDRk3;>Iy` zSM94|yC1%NTfHzE$(7`(4g|n^>e+I^+I8^p^5R1{I^p!BKIj2^Ze^=gN9(bUAidRP z{ibADek{i}u8kdd=+>dj$15GRA!t}p#gUMoFsdG*W8&;!ueVy}Rp0T}lPf-!*d?5JbS58OL3&25xD5**BzU9_r)5Psir(3lSq-OY~O+j`E{!wgIkIFnZ9xa`Qr3%CXmIG+x;E8@oaphSn#5Ohe_3U*#| z>sH=SSFOwDRG3;3tvP?NYj(In!G{~RtHOD$ic0A*O|N^i!?n^q{GU4g^uPHA=ko_( zcrxuVtb(^%7@MKOze_Et<+b^gF!O3^=G;`~5bc9-?R>TXJCsMD7v~IJH>i0dKssH*pT)%FmH$m+O6A%*$X-2ZC+qf?nsxQN*v9S>+ z3Faqv*B4#Wy0Wp6Qt!N&L<(X-mk1>U;_2uFzq8Yo(t1_GJB*xhGNXnGy_ z%IXJ&cFwC;?^}%KU3YC;!Z_pVGY*9W>0mD+Fh0uNUZ-t!LSejqXPikV>*tOJh9w&& zU#x&sjo9WM7E@)}A=}n%J>|_EHDHvk*Z0olZd05xILdsY_N(g#7`*4H4!gjMNhG<< z3AlM{q*gD=>PqnT>{&_f-E&io07bgX4A;#uuN#3Nf)a_Jat0a%65?AdB_dFep5`LO z8;Xts#7L@)HVSduUih@SHU_#R5-9+bnYb3BLvi`|md6^hdQoqYK2RWiel5H)=k@D= zOGJI=%$NlRMUL8i;wTu)2y3%ooZ*5M;!fl-Z0Lw5+9QH`Q&WZ51$=tDPzXIYun>m_ zfK;KNMIWJmC<+_oDCh~2f0*z#Jy}>~F8C?|(GeSN{4CT(b%>sR0Elcl_~rgXQHbT= zTmJ#RsLTF$QPc$<{fY#jLNGmC&RMdB{VFOJRa68o9m*Zwjl`5L62yy4MgZFmv5bR> z!a0{U@z~T7e|c7)erXBi649TH2n~!@UKNe1sL>yakdJn9HT3K41s@w{en5Gfux~!R z)*ze%;O|78MQ2MZ^91#&-V@>3mLcq}6~b1u2?mv9pQfXP4?cD^gh~P#(Oj}%yw0Es z$7n3=lUa6Qe}!`~ze4z|NvriN(l!VIjSVz4nR{YKnf}CF(7@00)d-1;fbu-4!$Q~x z4@HRnz42?HDuDs-XAc64G3#aq^;G!00Z~Y$H3yL^%FAQGz)_4;7L9vR?y%-4S6HpF zjEBMDO+*GnO!#=7?qO?Tr^$#q?R!+gL|$?9Q06}%uv2%!?&$l=*!GcnH2CW@%XTy^ zlq-ka?@7~~@YD;up?o5D-08Ylgmh#S*}regVc(?<6b4oQdgax+Ix9ccT^E(1_k59j*7~Yr$K%m$5!u%NBI%ui-^qE}7NwhpUnmw!R~E!h^lkO}wYfoM@5`H;96BBw zJ$$Thq#mQr;%XufE9W@pC|oi!LrQdelZ^T=)CZ=K|8r*vELV+*J9@~g%d6t)zO71A zwT(UWEv=gm8Gg-iSRU#tQ?+;?h6?k6WJ0|SZ@+Yw=7SS%8B?#_`Al}mjf{xjy{ckQ z=tMThg^D_18yIp1Vl`iM1@cf^FFb>_nFZXkD zv$CvB4b*Y-TWg8qLo?Vi!Vk&WaEGr_3cUNkhg{Bd`iDDhT=Kp@c6GpI7DmAQt5d9~ z!dG5Le2#ig4z|TNx3nCpeACArq$3Yx4dUSqRC5cQiw%*N4gR5!QmY0-Two3KH8WNS zCeoC(88t0zz*zCHS5c$=wa9Oa)5a!$99@#CqUFTZ?A)_Z0h6HLu~8-1)|dF zY~uLDw@<=&tzSO6?&R?1qNH%4^cG2@E|LYLEL>C>A_^p0H=?Effzte+A5JX*=E8z? WLEX%5C3*va2W`8Jd5P&yQU3)+3E+nS literal 0 HcmV?d00001