Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance.
Project price only 1 $
You can buy this project and download/modify it how often you want.
com.redislabs.redisai.Dag Maven / Gradle / Ivy
package com.redislabs.redisai;
import java.util.ArrayList;
import java.util.List;
import redis.clients.jedis.util.SafeEncoder;
public class Dag implements DagRunCommands {
private final List> commands = new ArrayList<>();
private final List tensorgetflag = new ArrayList<>();
/** Direct acyclic graph of operations to run within RedisAI */
public Dag() {}
protected List processDagReply(List reply) {
List outputList = new ArrayList<>(reply.size());
for (int i = 0; i < reply.size(); i++) {
Object obj = reply.get(i);
// TODO: Should encode 'OK', 'NA', etc. response
if (obj instanceof Exception) {
Exception ex = (Exception) obj;
outputList.add(new RedisAIException(ex.getMessage(), ex));
} else if (this.tensorgetflag.get(i)) {
outputList.add(Tensor.createTensorFromRespReply((List) obj));
} else {
outputList.add(obj);
}
}
return outputList;
}
@Override
public Dag setTensor(String key, Tensor tensor) {
List args = tensor.tensorSetFlatArgs(key, true);
this.commands.add(args);
this.tensorgetflag.add(false);
return this;
}
@Override
public Dag getTensor(String key) {
List args = Tensor.tensorGetFlatArgs(key, true);
this.commands.add(args);
this.tensorgetflag.add(true);
return this;
}
@Override
public Dag runModel(String key, String[] inputs, String[] outputs) {
List args = Model.modelRunFlatArgs(key, inputs, outputs, true);
this.commands.add(args);
this.tensorgetflag.add(false);
return this;
}
@Override
public Dag executeModel(String key, String[] inputs, String[] outputs) {
List args = Model.modelExecuteCommandArgs(key, inputs, outputs, -1L, true);
this.commands.add(args);
this.tensorgetflag.add(false);
return this;
}
@Override
public Dag runScript(String key, String function, String[] inputs, String[] outputs) {
List args = Script.scriptRunFlatArgs(key, function, inputs, outputs, true);
this.commands.add(args);
this.tensorgetflag.add(false);
return this;
}
@Override
public Dag executeScript(
String key,
String function,
List keys,
List inputs,
List args,
List outputs) {
List binary =
Script.scriptExecuteFlatArgs(key, function, keys, inputs, args, outputs, -1L, true);
this.commands.add(binary);
this.tensorgetflag.add(false);
return this;
}
List dagRunFlatArgs(String[] loadKeys, String[] persistKeys) {
List args = new ArrayList<>();
if (loadKeys != null && loadKeys.length > 0) {
args.add(Keyword.LOAD.getRaw());
args.add(SafeEncoder.encode(String.valueOf(loadKeys.length)));
for (String key : loadKeys) {
args.add(SafeEncoder.encode(key));
}
}
if (persistKeys != null && persistKeys.length > 0) {
args.add(Keyword.PERSIST.getRaw());
args.add(SafeEncoder.encode(String.valueOf(persistKeys.length)));
for (String key : persistKeys) {
args.add(SafeEncoder.encode(key));
}
}
for (List command : this.commands) {
args.add(Keyword.PIPE.getRaw());
args.addAll(command);
}
return args;
}
List dagExecuteFlatArgs(
String[] loadTensors, String[] persistTensors, String routingHint) {
List args = new ArrayList<>();
if (loadTensors != null && loadTensors.length > 0) {
args.add(Keyword.LOAD.getRaw());
args.add(SafeEncoder.encode(String.valueOf(loadTensors.length)));
for (String key : loadTensors) {
args.add(SafeEncoder.encode(key));
}
}
if (persistTensors != null && persistTensors.length > 0) {
args.add(Keyword.PERSIST.getRaw());
args.add(SafeEncoder.encode(String.valueOf(persistTensors.length)));
for (String key : persistTensors) {
args.add(SafeEncoder.encode(key));
}
}
if (routingHint != null) {
args.add(Keyword.ROUTING.getRaw());
args.add(SafeEncoder.encode(routingHint));
}
for (List command : this.commands) {
args.add(Keyword.PIPE.getRaw());
args.addAll(command);
}
return args;
}
}
