Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/Tencent/digitalhuman
/ types & classes
Types & classes
810 in github.com/Tencent/digitalhuman
⨍
Functions
5,073
◇
Types & classes
810
↳
Endpoints
92
↓ 28 callers
Class
OutputParserError
Exception raised when parsing output from a command fails.
HATE/basic_hate/agentverse/output_parser/output_parser.py:22
↓ 27 callers
Class
DataProto
A DataProto is a data structure that aims to provide a standard protocol for data exchange between functions. It contains a batch (TensorDict
RLVER/code/verl/protocol.py:173
↓ 23 callers
Class
RayClassWithInitArgs
RLVER/code/verl/single_controller/ray/base.py:128
↓ 21 callers
Class
AgentFinish
Agent's return value.
HATE/basic_hate/agentverse/utils.py:15
↓ 19 callers
Class
DataProto
A DataProto is a data structure that aims to provide a standard protocol for data exchange between functions. It contains a batch (TensorDict
RLVMR/code/verl/protocol.py:173
↓ 19 callers
Class
RayWorkerGroup
RLVER/code/verl/single_controller/ray/base.py:176
↓ 16 callers
Class
RayResourcePool
RLVER/code/verl/single_controller/ray/base.py:49
↓ 16 callers
Class
StringFormatter
VISTA/TinyLLaVA-VISTA/tinyllava/data/template/formatter.py:24
↓ 14 callers
Class
EmptyFormatter
VISTA/TinyLLaVA-VISTA/tinyllava/data/template/formatter.py:18
↓ 13 callers
Class
Conversation
A class that keeps all conversation history.
VISTA/LLaVA-VISTA/llava/conversation.py:19
↓ 13 callers
Class
ScienceWorldEnv
Python wrapper for the simulator written in Scala. The methods that are being wrapped can be found in simulator/src/main/scala/scienceworld/runtim
RLVMR/code/agent_system/environments/env_package/sciworld/ScienceWorld/scienceworld/scienceworld.py:17
↓ 12 callers
Class
Message
VISTA/TinyLLaVA-VISTA/tinyllava/utils/message.py:8
↓ 12 callers
Class
Registry
Registry for storing and building classes.
HATE/basic_hate/agentverse/registry.py:6
↓ 11 callers
Class
LLM
An LLM for generating texts from given prompts and sampling parameters. This class includes a tokenizer, a language model (possibly distributed
RLVER/code/llm.py:52
↓ 10 callers
Class
ImagePreprocess
VISTA/TinyLLaVA-VISTA/tinyllava/data/image_preprocess.py:13
↓ 10 callers
Class
Response
SWF/src/base.py:27
↓ 10 callers
Class
SetWithGet
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/gen/utils/py_util.py:43
↓ 10 callers
Class
TextPreprocess
VISTA/TinyLLaVA-VISTA/tinyllava/data/text_preprocess.py:6
↓ 9 callers
Class
AgentCriticism
Agent's criticism.
HATE/basic_hate/agentverse/utils.py:22
↓ 9 callers
Class
NoisyLinear
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:877
↓ 9 callers
Class
Template
VISTA/TinyLLaVA-VISTA/tinyllava/data/template/base.py:15
↓ 8 callers
Class
EvaluatorMessage
HATE/basic_hate/agentverse/message.py:29
↓ 8 callers
Class
JsonRepairError
HATE/basic_hate/agentverse/llms/utils/jsonrepair.py:143
↓ 8 callers
Class
LLMResult
HATE/basic_hate/agentverse/llms/base.py:6
↓ 7 callers
Class
ParallelLlamaRMSNorm
RLVER/code/verl/models/llama/megatron/layers/parallel_rmsnorm.py:25
↓ 7 callers
Class
ParallelLlamaRMSNorm
RLVMR/code/verl/models/llama/megatron/layers/parallel_rmsnorm.py:25
↓ 7 callers
Class
ParallelQwen2RMSNorm
RLVER/code/verl/models/qwen2/megatron/layers/parallel_rmsnorm.py:25
↓ 7 callers
Class
ParallelQwen2RMSNorm
RLVMR/code/verl/models/qwen2/megatron/layers/parallel_rmsnorm.py:25
↓ 7 callers
Class
SelfAttention
From Multi-Head Attention module https://github.com/jadore801120/attention-is-all-you-need-pytorch
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:990
↓ 6 callers
Class
AgentAction
Agent's action to take.
HATE/basic_hate/agentverse/utils.py:7
↓ 6 callers
Class
BeamSearchNode
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/agents/modules/generic.py:280
↓ 6 callers
Class
ExecutorMessage
HATE/basic_hate/agentverse/message.py:24
↓ 6 callers
Class
FSDPUlyssesShardingManager
Sharding manager to support data resharding when using FSDP + Ulysses
RLVER/code/verl/workers/sharding_manager/fsdp_ulysses.py:31
↓ 6 callers
Class
RayClassWithInitArgs
RLVMR/code/verl/single_controller/ray/base.py:148
↓ 5 callers
Class
FSDPUlyssesShardingManager
Sharding manager to support data resharding when using FSDP + Ulysses
RLVMR/code/verl/workers/sharding_manager/fsdp_ulysses.py:32
↓ 5 callers
Class
FlopsCounter
Used to count mfu during training loop Example: flops_counter = FlopsCounter(config) flops_achieved, flops_promised = flops_
RLVMR/code/verl/utils/flops_counter.py:54
↓ 5 callers
Class
LlamaModel
Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`LlamaDecoderLayer`] Args: config: LlamaConfig
VISTA/LLaVA-VISTA/llava/train/modeling_llama.py:1073
↓ 5 callers
Class
Message
HATE/basic_hate/agentverse/message.py:7
↓ 5 callers
Class
OpenELMRMSNorm
VISTA/TinyLLaVA-VISTA/tinyllava/model/llm/openelm.py:342
↓ 5 callers
Class
RLHFDataset
We assume the dataset contains a column that contains prompts and other information
RLVER/code/verl/utils/dataset/rl_dataset.py:74
↓ 5 callers
Class
WebEnv
A wrapper of textEnv for models. Returns valid actions at each step of the game.
RLVMR/code/agent_system/environments/env_package/webshop/webshop/baseline_models/env.py:17
↓ 5 callers
Class
WordSegment
Word-level segment information
BatonVoice/audio_feature_extractor.py:62
↓ 4 callers
Class
BertModel
The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of cross-attention is added between
VISTA/TinyLLaVA-VISTA/tinyllava/model/connector/qformer.py:743
↓ 4 callers
Class
BiAttention
RLVMR/code/agent_system/environments/env_package/webshop/webshop/baseline_models/models/modules.py:120
↓ 4 callers
Class
EvalAIAnswerProcessor
Processes an answer similar to Eval AI copied from https://github.com/facebookresearch/mmf/blob/c46b3b3391275b4181567db80943473a8
VISTA/LLaVA-VISTA/llava/eval/m4c_evaluator.py:7
↓ 4 callers
Class
RLHFDataset
We assume the dataset contains a column that contains prompts and other information
RLVMR/code/verl/utils/dataset/rl_dataset.py:75
↓ 4 callers
Class
RayPPOTrainer
Note that this trainer runs on the driver process on a single CPU/GPU node.
RLVER/code/verl/trainer/ppo/ray_trainer.py:373
↓ 4 callers
Class
ResourcePoolManager
Define a resource pool specification. Resource pool will be initialized first. Mapping
RLVER/code/verl/trainer/ppo/ray_trainer.py:76
↓ 4 callers
Class
SFTDataset
This is an in-memory SFTDataset
RLVER/code/verl/utils/dataset/sft_dataset.py:34
↓ 4 callers
Class
Tracking
RLVER/code/verl/utils/tracking.py:24
↓ 4 callers
Class
Tracking
RLVMR/code/verl/utils/tracking.py:24
↓ 4 callers
Class
VirtualRLHFDataset
RLVER/code/verl/utils/dataset/rl_dataset.py:226
↓ 3 callers
Class
Agent
SWF/src/base.py:37
↓ 3 callers
Class
AlignedSegment
Aligned sentence segment
BatonVoice/audio_feature_extractor.py:71
↓ 3 callers
Class
AutoGptFormatter
Allows to handle custom placeholders 'title_color' and 'message_no_color'. To use this formatter, make sure to pass 'color', 'title' as log e
HATE/basic_hate/agentverse/logging.py:250
↓ 3 callers
Class
BertConfigForWebshop
RLVMR/code/agent_system/environments/env_package/webshop/webshop/baseline_models/models/bert.py:14
↓ 3 callers
Class
BertModelForWebshop
RLVMR/code/agent_system/environments/env_package/webshop/webshop/baseline_models/models/bert.py:30
↓ 3 callers
Class
DeterministicPlannerAgent
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/gen/agents/deterministic_planner_agent.py:4
↓ 3 callers
Class
EvalAIAnswerProcessor
Processes an answer similar to Eval AI copied from https://github.com/facebookresearch/mmf/blob/c46b3b3391275b4181567db80943473a8
VISTA/TinyLLaVA-VISTA/tinyllava/eval/m4c_evaluator.py:7
↓ 3 callers
Class
FSDPCheckpointManager
A checkpoint manager that saves and loads - model - optimizer - lr_scheduler - extra_states in a SPMD way. We save
RLVER/code/verl/utils/checkpoint/fsdp_checkpoint_manager.py:32
↓ 3 callers
Class
FSDPCheckpointManager
A checkpoint manager that saves and loads - model - optimizer - lr_scheduler - extra_states in a SPMD way. We save
RLVMR/code/verl/utils/checkpoint/fsdp_checkpoint_manager.py:32
↓ 3 callers
Class
LinearSchedule
Linear interpolation between initial_p and final_p over schedule_timesteps. After this many timesteps pass final_p is returned. :para
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/agents/modules/generic.py:152
↓ 3 callers
Class
LlamaRMSNorm
VISTA/LLaVA-VISTA/llava/train/modeling_llama.py:78
↓ 3 callers
Class
PhiModel
Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`PhiDecoderLayer`] Args: config: PhiConfig
VISTA/TinyLLaVA-VISTA/tinyllava/model/llm/modeling_phi.py:917
↓ 3 callers
Class
PlayerSimulator
RLVER/code/verl/workers/rollout/vllm_rollout/hard_player_simulator_dsv3.py:42
↓ 3 callers
Class
RayResourcePool
RLVMR/code/verl/single_controller/ray/base.py:69
↓ 3 callers
Class
ReferenceImageSpec
ScriptAgent/code/director_agent.py:375
↓ 3 callers
Class
ResourcePoolManager
Define a resource pool specification. Resource pool will be initialized first. Mapping
RLVMR/code/verl/trainer/ppo/ray_trainer.py:80
↓ 3 callers
Class
State
RLVMR/code/verl/utils/seqlen_balancing.py:49
↓ 3 callers
Class
StoryComponents
ScriptAgent/code/director_agent.py:427
↓ 3 callers
Class
TaskGameStateFullKnowledge
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/gen/game_states/task_game_state_full_knowledge.py:15
↓ 3 callers
Class
ThorEnv
an extension of ai2thor.controller.Controller for ALFRED tasks
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/env/thor_env.py:26
↓ 3 callers
Class
TinyLlavaForConditionalGeneration
VISTA/TinyLLaVA-VISTA/tinyllava/model/modeling_tinyllava.py:58
↓ 3 callers
Class
vLLMRollout
RLVER/code/verl/workers/rollout/vllm_rollout/vllm_rollout.py:85
↓ 3 callers
Class
vLLMRollout
RLVMR/code/verl/workers/rollout/vllm_rollout/vllm_rollout.py:57
↓ 2 callers
Class
AlfWorldEnvironmentManager
RLVMR/code/agent_system/environments/env_manager.py:29
↓ 2 callers
Class
AudioFeatureExtractor
Audio Feature Extractor Class Main Functions: 1. Load and process single audio files 2. Extract word-level features using Triton
BatonVoice/audio_feature_extractor.py:79
↓ 2 callers
Class
BertAttention
VISTA/TinyLLaVA-VISTA/tinyllava/model/connector/qformer.py:358
↓ 2 callers
Class
BertIntermediate
VISTA/TinyLLaVA-VISTA/tinyllava/model/connector/qformer.py:415
↓ 2 callers
Class
BertOnlyMLMHead
VISTA/TinyLLaVA-VISTA/tinyllava/model/connector/qformer.py:710
↓ 2 callers
Class
BertOutput
VISTA/TinyLLaVA-VISTA/tinyllava/model/connector/qformer.py:430
↓ 2 callers
Class
CriticMessage
HATE/basic_hate/agentverse/message.py:19
↓ 2 callers
Class
DataParallelPPOActor
RLVER/code/verl/workers/actor/dp_actor.py:39
↓ 2 callers
Class
DataParallelPPOActor
RLVMR/code/verl/workers/actor/dp_actor.py:39
↓ 2 callers
Class
DataProtoFuture
DataProtoFuture aims to eliminate actual data fetching on driver. By doing so, the driver doesn't have to wait for data so that asynchronous
RLVER/code/verl/protocol.py:606
↓ 2 callers
Class
DataProtoFuture
DataProtoFuture aims to eliminate actual data fetching on driver. By doing so, the driver doesn't have to wait for data so that asynchronous
RLVMR/code/verl/protocol.py:604
↓ 2 callers
Class
Demangler
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/agents/utils/misc.py:17
↓ 2 callers
Class
DigitCompletion
The implementation of a simple digit completion task. The prompt is a sequence of numbers with fixed difference. The task is to complete the
RLVER/code/tests/e2e/envs/digit_completion/task.py:19
↓ 2 callers
Class
EncoderRNN
RLVMR/code/agent_system/environments/env_package/webshop/webshop/baseline_models/models/modules.py:31
↓ 2 callers
Class
FSDPSFTTrainer
RLVER/code/verl/trainer/fsdp_sft_trainer.py:78
↓ 2 callers
Class
FlopsCounter
Used to count mfu during training loop Example: flops_counter = FlopsCounter(config) flops_achieved, flops_promised = flops_
RLVER/code/verl/utils/flops_counter.py:51
↓ 2 callers
Class
HumanOutputFormat
RLVMR/code/agent_system/environments/env_package/webshop/webshop/baseline_models/logger.py:30
↓ 2 callers
Class
LLaVATrainer
VISTA/TinyLLaVA-VISTA/tinyllava/train/tinyllava_trainer.py:120
↓ 2 callers
Class
LSTMCell
A basic LSTM cell.
RLVMR/code/agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:470
↓ 2 callers
Class
MegatronCheckpointManager
A checkpoint manager that saves and loads - model - optimizer - lr_scheduler - extra_states in a SPMD way. We save
RLVMR/code/verl/utils/checkpoint/megatron_checkpoint_manager.py:41
↓ 2 callers
Class
MegatronPPOActor
RLVER/code/verl/workers/actor/megatron_actor.py:54
↓ 2 callers
Class
MegatronPPOActor
RLVMR/code/verl/workers/actor/megatron_actor.py:53
↓ 2 callers
Class
MemoryBuffer
A memory buffer is a contiguous torch tensor that may combine multiple tensors sharing with the underlying memory. It must have a unique type
RLVER/code/verl/utils/memory_buffer.py:24
↓ 2 callers
Class
MemoryBuffer
A memory buffer is a contiguous torch tensor that may combine multiple tensors sharing with the underlying memory. It must have a unique type
RLVMR/code/verl/utils/memory_buffer.py:24
↓ 2 callers
Class
OpenRouterGeminiClient
A client class for interacting with Gemini 2.5 Pro model through OpenRouter API. This class handles the complete workflow of: 1. Acc
BatonVoice/openrouter_gemini_client.py:136
next →
1–100 of 810, ranked by callers