public void Run(IDictionary <string, string> args, ILogger logger)
{
var dataDirectory = args["dataDirectory"];
var fileName = args["fileName"];
var collection = args["collection"];
var skip = args.ContainsKey("skip") ? int.Parse(args["skip"]) : 0;
var take = args.ContainsKey("take") ? int.Parse(args["take"]) : int.MaxValue;
var sampleSize = args.ContainsKey("sampleSize") ? int.Parse(args["sampleSize"]) : 1000;
var pageSize = args.ContainsKey("pageSize") ? int.Parse(args["pageSize"]) : 100000;
var collectionId = collection.ToHash();
var fieldsToStore = new HashSet <string> {
"language", "wikibase_item", "title", "text", "url"
};
var fieldsToIndex = new HashSet <string> {
"title", "text"
};
if (take == 0)
{
take = int.MaxValue;
}
var model = new BagOfCharsModel();
var payload = WikipediaHelper.ReadWP(fileName, skip, take, fieldsToStore, fieldsToIndex);
using (var sessionFactory = new SessionFactory(dataDirectory, logger))
{
var debugger = new IndexDebugger(logger, sampleSize);
using (var writeSession = new WriteSession(new DocumentWriter(collectionId, sessionFactory)))
{
foreach (var page in payload.Batch(pageSize))
{
using (var indexStream = new WritableIndexStream(collectionId, sessionFactory, logger: logger))
using (var indexSession = new IndexSession <string>(model, model))
{
foreach (var document in page)
{
writeSession.Put(document);
foreach (var field in document.IndexableFields)
{
indexSession.Put(document.Id, field.KeyId, (string)field.Value);
}
debugger.Step(indexSession);
}
indexStream.Write(indexSession.GetInMemoryIndex());
//foreach (var column in indexSession.InMemoryIndex)
//{
// Print($"wikipedia.{column.Key}", column.Value);
//}
}
}
}
}
}
public void Can_search_filestreamed_with_multiple_pages()
{
var model = new BagOfCharsModel();
const string collection = "Can_search_streamed_with_one_page_per_document";
var collectionId = collection.ToHash();
const string fieldName = "description";
_sessionFactory.Truncate(collectionId);
using (var stream = new WritableIndexStream(collectionId, _sessionFactory))
using (var writeSession = new WriteSession(new DocumentWriter(collectionId, _sessionFactory)))
{
var keyId = writeSession.EnsureKeyExists(fieldName);
for (long i = 0; i < _data.Length; i++)
{
var data = _data[i];
using (var indexSession = new IndexSession <string>(model, model))
{
var doc = new Document(new Field[] { new Field(fieldName, data, index: true, store: true) });
writeSession.Put(doc);
indexSession.Put(doc.Id, keyId, data);
stream.Write(indexSession.GetInMemoryIndex());
}
}
}
var queryParser = new QueryParser <string>(_sessionFactory, model);
using (var searchSession = new SearchSession(_sessionFactory, model, new PostingsReader(_sessionFactory)))
{
Assert.DoesNotThrow(() =>
{
foreach (var word in _data)
{
var query = queryParser.Parse(collection, word, fieldName, fieldName, and: true, or: false);
var result = searchSession.Search(query, 0, 1);
var document = result.Documents.FirstOrDefault();
if (document == null)
{
throw new Exception($"unable to find {word}.");
}
if (document.Score < model.IdenticalAngle)
{
throw new Exception($"unable to score {word}.");
}
Debug.WriteLine($"{word} matched with {document.Score * 100}% certainty.");
}
});
}
}
public void Run(IDictionary <string, string> args, ILogger logger)
{
var dataDirectory = args["dataDirectory"];
var fileName = args["fileName"];
var collection = args["collection"];
var skip = args.ContainsKey("skip") ? int.Parse(args["skip"]) : 0;
var take = args.ContainsKey("take") ? int.Parse(args["take"]) : int.MaxValue;
var sampleSize = args.ContainsKey("sampleSize") ? int.Parse(args["sampleSize"]) : 1000;
var pageSize = args.ContainsKey("pageSize") ? int.Parse(args["pageSize"]) : 100000;
var collectionId = collection.ToHash();
var fieldsToStore = new HashSet <string> {
"language", "wikibase_item", "title", "text"
};
var fieldsToIndex = new HashSet <string> {
"language", "title", "text"
};
if (take == 0)
{
take = int.MaxValue;
}
var model = new BagOfCharsModel();
var payload = WikipediaHelper.ReadWP(fileName, skip, take, fieldsToStore, fieldsToIndex);
var debugger = new BatchDebugger(logger, sampleSize);
using (var sessionFactory = new SessionFactory(dataDirectory, logger))
{
using (var writeSession = new WriteSession(new DocumentWriter(collectionId, sessionFactory)))
{
foreach (var page in payload.Batch(pageSize))
{
using (var indexSession = new IndexSession <string>(model, model))
{
foreach (var document in page)
{
writeSession.Put(document);
foreach (var field in document.IndexableFields)
{
foreach (var token in model.Tokenize((string)field.Value))
{
debugger.Step();
}
}
}
}
}
logger.LogInformation($"tokenized {debugger.StepCount} in {debugger.Time}.");
}
}
}
public static void WriteWatSegment(
string dataDirectory,
string fileName,
string collection,
IModel <string> model,
ILogger logger,
string refFileName)
{
var time = Stopwatch.StartNew();
var collectionId = collection.ToHash();
var storeFields = new HashSet <string>
{
"title", "description", "url", "filename"
};
var indexFields = new HashSet <string>
{
"title", "description", "url"
};
using (var sessionFactory = new SessionFactory(dataDirectory, logger))
using (var writeSession = new WriteSession(new DocumentWriter(collectionId, sessionFactory)))
using (var indexSession = new IndexSession <string>(model, model))
{
using (var queue = new ProducerConsumerQueue <Document>(document =>
{
sessionFactory.Write(document, writeSession, indexSession);
}))
{
foreach (var document in ReadWatFile(fileName, refFileName).Select(dic =>
new Document(
dic.Select(kvp => new Field(
kvp.Key,
kvp.Value,
index: indexFields.Contains(kvp.Key),
store: storeFields.Contains(kvp.Key))).ToList())))
{
queue.Enqueue(document);
}
}
using (var stream = new WritableIndexStream(collectionId, sessionFactory, logger: logger))
{
stream.Write(indexSession.GetInMemoryIndex());
}
}
logger.LogInformation($"indexed {fileName} in {time.Elapsed}");
}
public void Run(IDictionary <string, string> args, ILogger logger)
{
var time = Stopwatch.StartNew();
var dataDirectory = args["dataDirectory"];
var collectionId = args["collection"].ToHash();
var images = new MnistReader(args["imageFileName"], args["labelFileName"]).Read();
VectorNode tree;
var debugger = new IndexDebugger(logger);
var model = new LinearClassifierImageModel();
using (var sessionFactory = new SessionFactory(dataDirectory, logger))
{
sessionFactory.Truncate(collectionId);
using (var writeSession = new WriteSession(new DocumentWriter(collectionId, sessionFactory)))
using (var indexSession = new IndexSession <IImage>(model, model))
{
var imageIndexId = writeSession.EnsureKeyExists("image");
foreach (var image in images)
{
var imageField = new Field("image", image.Pixels, index: true, store: true);
var labelField = new Field("label", image.Label, index: false, store: true);
var document = new Document(new Field[] { imageField, labelField });
writeSession.Put(document);
indexSession.Put(document.Id, imageField.KeyId, image);
debugger.Step(indexSession);
}
var indices = indexSession.GetInMemoryIndex();
tree = indices[imageIndexId];
using (var stream = new WritableIndexStream(collectionId, sessionFactory, logger: logger))
{
stream.Write(indices);
}
}
}
logger.LogInformation($"indexed {debugger.Steps} mnist images in {time.Elapsed}");
Print(tree);
}
public void Can_produce_traversable_in_memory_index()
{
var model = new BagOfCharsModel();
VectorNode tree;
using (var indexSession = new IndexSession <string>(model, model))
{
for (long i = 0; i < _data.Length; i++)
{
indexSession.Put(i, 0, _data[i]);
}
tree = indexSession.GetInMemoryIndex()[0];
}
Debug.WriteLine(PathFinder.Visualize(tree));
Assert.DoesNotThrow(() =>
{
foreach (var word in _data)
{
foreach (var queryVector in model.Tokenize(word))
{
var hit = PathFinder.ClosestMatch(tree, queryVector, model);
if (hit == null)
{
throw new Exception($"unable to find {word} in tree.");
}
if (hit.Score < model.IdenticalAngle)
{
throw new Exception($"unable to score {word}.");
}
Debug.WriteLine($"{word} matched with {hit.Node.Vector.Label} with {hit.Score * 100}% certainty.");
}
}
});
}
