Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/Gen-Verse/OpenClaw-RL
/ functions
Functions
10,561 in github.com/Gen-Verse/OpenClaw-RL
⨍
Functions
10,561
◇
Types & classes
1,405
↳
Endpoints
144
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/dino.py:135
Method
forward
(self, input)
Megatron-LM/megatron/legacy/model/vision/dino.py:269
Method
forward
(self, input)
Megatron-LM/megatron/legacy/model/vision/inpainting.py:51
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/inpainting.py:78
Method
forward
(self, input)
Megatron-LM/megatron/legacy/model/vision/inpainting.py:119
Method
forward
(self, hidden_states)
Megatron-LM/megatron/legacy/model/vision/vit_backbone.py:41
Method
forward
(self, input)
Megatron-LM/megatron/legacy/model/vision/vit_backbone.py:212
Method
forward
(self, x, H, W)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:46
Method
forward
(self, x, H, W)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:101
Method
forward
(self, x, H, W)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:158
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:195
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:335
Method
forward
(self, x, H, W)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:349
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:30
Method
forward
Args: x: input features with shape of (num_windows*B, N, C) mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:119
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:245
Method
forward
x: B, H*W, C
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:319
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:400
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:453
Method
forward
(self, x)
Megatron-LM/megatron/legacy/model/vision/swin_backbone.py:569
Method
forward
(self)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:169
Method
forward
(self)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:311
Method
forward
(self)
Megatron-LM/megatron/legacy/mpu/tests/commons.py:17
Method
forward
Runs the fwd pass of _kernel_make_viewless_tensor
Megatron-LM/megatron/core/utils.py:677
Method
forward
(self, input: Tensor)
Megatron-LM/megatron/core/fusions/fused_layer_norm.py:131
Method
forward
Forward pass for scaled upper-triangular masked softmax. Args: ctx: Autograd context used to stash tensors for backward.
Megatron-LM/megatron/core/fusions/fused_softmax.py:20
Method
forward
Forward pass for scaled masked softmax. Args: ctx: Autograd context used to stash tensors for backward. inputs (torch
Megatron-LM/megatron/core/fusions/fused_softmax.py:69
Method
forward
Forward pass for scaled softmax (no mask). Args: ctx: Autograd context used to stash tensors for backward. inputs (to
Megatron-LM/megatron/core/fusions/fused_softmax.py:116
Method
forward
Forward pass of softmax with masked input. In case attn_mask_type is causal the mask is generated and None can be passed. A user-defi
Megatron-LM/megatron/core/fusions/fused_softmax.py:219
Method
forward
Forward implementation for the cross entropy loss.
Megatron-LM/megatron/core/fusions/fused_cross_entropy.py:89
Method
forward
Forward function for IndicesToMultihot Convert indices to multihot representation. Args: indices: [num_of_tokens, topk]
Megatron-LM/megatron/core/fusions/fused_indices_converter.py:186
Method
forward
Forward function for ApplyMLARotaryEmbQ. Args: q: [seq_len, batch_size, head_num, qk_head_dim + emb_dim]
Megatron-LM/megatron/core/fusions/fused_mla_yarn_rope_apply.py:216
Method
forward
Forward function for ApplyMLARotaryEmbKV. Args: kv: [seq_len, batch_size, head_num, k_dim + v_dim] or [t
Megatron-LM/megatron/core/fusions/fused_mla_yarn_rope_apply.py:602
Method
forward
Forward pass of biased GEGLU activation. Args: ctx: Autograd context object for saving tensors for backward pass. inp
Megatron-LM/megatron/core/fusions/fused_bias_geglu.py:89
Method
forward
Forward pass of GEGLU activation. Args: ctx: Autograd context object for saving tensors for backward pass. input (tor
Megatron-LM/megatron/core/fusions/fused_bias_geglu.py:124
Method
forward
Forward pass of weighted Quick-GEGLU. Args: ctx: Autograd context object for saving tensors for backward pass. input
Megatron-LM/megatron/core/fusions/fused_bias_geglu.py:307
Method
forward
Forward pass of weighted Quick-GEGLU. Args: ctx: Autograd context object for saving tensors for backward pass. input
Megatron-LM/megatron/core/fusions/fused_bias_geglu.py:354
Method
forward
(ctx, input, bias)
Megatron-LM/megatron/core/fusions/fused_bias_gelu.py:39
Method
forward
The forward pass of the Linear Cross Entropy. If tp_group is not None, the weight tensor to each TP rank should be (global_vo
Megatron-LM/megatron/core/fusions/fused_linear_cross_entropy.py:66
Method
forward
forward method for `WeightedSquaredReLUFunction` Args: ctx : context object to store intermediate tensors. input (tor
Megatron-LM/megatron/core/fusions/fused_weighted_squared_relu.py:65
Method
forward
Forward pass of biased SwiGLU activation. Args: ctx: Autograd context object for saving tensors for backward pass. in
Megatron-LM/megatron/core/fusions/fused_bias_swiglu.py:105
Method
forward
Forward pass of SwiGLU activation. Args: ctx: Autograd context object for saving tensors for backward pass. input (to
Megatron-LM/megatron/core/fusions/fused_bias_swiglu.py:152
Method
forward
(ctx, input, weights, fp8_input_store)
Megatron-LM/megatron/core/fusions/fused_bias_swiglu.py:194
Method
forward
Forward.
Megatron-LM/megatron/core/post_training/modelopt/layers.py:168
Method
forward
Forward. Args: input_ (torch.Tensor): Input tensor.
Megatron-LM/megatron/core/tensor_parallel/layers.py:276
Method
forward
Forward with frozen weight.
Megatron-LM/megatron/core/tensor_parallel/layers.py:344
Method
forward
Forward.
Megatron-LM/megatron/core/tensor_parallel/layers.py:456
Method
forward
Forward of ColumnParallelLinear Args: input_: 3D tensor whose order of dimension is [sequence, batch, hidden]
Megatron-LM/megatron/core/tensor_parallel/layers.py:970
Method
forward
Forward of RowParallelLinear Args: input_: 3D tensor whose order of dimension is [sequence, batch, hidden] Returns:
Megatron-LM/megatron/core/tensor_parallel/layers.py:1262
Method
forward
Vocab parallel cross entropy forward function.
Megatron-LM/megatron/core/tensor_parallel/cross_entropy.py:124
Method
forward
Forward pass.
Megatron-LM/megatron/core/tensor_parallel/random.py:485
Method
forward
Forward pass.
Megatron-LM/megatron/core/tensor_parallel/random.py:558
Method
forward
Forward pass.
Megatron-LM/megatron/core/tensor_parallel/inference_layers.py:89
Method
forward
Forward pass.
Megatron-LM/megatron/core/tensor_parallel/inference_layers.py:144
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:206
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:226
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:245
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:265
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:285
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:312
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:360
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:389
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:409
Method
forward
Forward function.
Megatron-LM/megatron/core/tensor_parallel/mappings.py:422
Method
forward
Calls the wrapped module's forward() method.
Megatron-LM/megatron/core/distributed/data_parallel_base.py:18
Method
forward
Wrapped forward pass of the model managed by FSDP.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/megatron_fsdp.py:1229
Method
forward
Forward pass of the RegisterFSDPBackwardFunction function.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/megatron_fsdp.py:1248
Method
forward
Passes through inputs unchanged in the forward pass.
Megatron-LM/megatron/core/pipeline_parallel/utils.py:135
Method
forward
Schedule node forward
Megatron-LM/megatron/core/pipeline_parallel/utils.py:200
Method
forward
(ctx, *args)
Megatron-LM/megatron/core/pipeline_parallel/fine_grained_activation_offload.py:487
Method
forward
(ctx, tensor, cpu_offload_handler, name)
Megatron-LM/megatron/core/pipeline_parallel/fine_grained_activation_offload.py:532
Method
forward
Forward.
Megatron-LM/megatron/core/extensions/kitchen.py:667
Method
forward
Forward.
Megatron-LM/megatron/core/extensions/kitchen.py:993
Method
forward
Forward.
Megatron-LM/megatron/core/extensions/transformer_engine.py:622
Method
forward
Forward.
Megatron-LM/megatron/core/extensions/transformer_engine.py:1049
Method
forward
(ctx, x, group_size)
Megatron-LM/megatron/core/extensions/transformer_engine.py:1170
Method
forward
Forward.
Megatron-LM/megatron/core/extensions/transformer_engine.py:1405
Method
forward
Forward.
Megatron-LM/megatron/core/extensions/transformer_engine.py:1925
Method
forward
Forward pass through the multimodal model. Args: input_ids: Input token IDs [batch_size, seq_length] position_ids: Po
Megatron-LM/megatron/core/models/mimo/model/base.py:214
Method
forward
Process image data through encoding and projection. Args: encoder_inputs: Dictionary where keys match encoder names in self.encod
Megatron-LM/megatron/core/models/mimo/submodules/vision.py:163
Method
forward
Process data for this modality through encoding and projection. Args: encoder_inputs (Dict[str, Any]): Dictionary
Megatron-LM/megatron/core/models/mimo/submodules/base.py:182
Method
forward
Forward pass for audio modality submodules. Args: encoder_inputs: Dictionary where keys match encoder names in self.encoders
Megatron-LM/megatron/core/models/mimo/submodules/audio.py:129
Method
forward
Forward pass. Args: hidden_states (Tensor): output hidden states from decoder word_embeddings_weight (Tensor): word e
Megatron-LM/megatron/core/models/T5/t5_model.py:67
Method
forward
Forward pass. Args: encoder_input_ids (Tensor): input ids for encoder decoder_input_ids (Tensor): input ids for decod
Megatron-LM/megatron/core/models/T5/t5_model.py:275
Method
forward
(self, hidden_states: Tensor, sequence_index=0)
Megatron-LM/megatron/core/models/bert/pooler.py:38
Method
forward
Forward function of BERT model Forward function of the BERT Model This function passes the input tensors through the embedding layer,
Megatron-LM/megatron/core/models/bert/bert_model.py:294
Method
forward
forward pass
Megatron-LM/megatron/core/models/bert/bert_lm_head.py:44
Method
forward
Forward function of the GPT Model This function passes the input tensors through the embedding layer, and then the decoder and finally into th
Megatron-LM/megatron/core/models/gpt/gpt_model.py:435
Method
forward
Run multimodal projector. Args: hidden_states (torch.Tensor): Input. Returns: torch.Tensor: The projected ou
Megatron-LM/megatron/core/models/vision/multimodal_projector.py:64
Method
forward
Forward function of the CLIP ViT Model. This function passes the input tensors through the embedding layer and then the transformer.
Megatron-LM/megatron/core/models/vision/clip_vit_model.py:164
Method
forward
Forward function of the RADIO ViT Model. This function passes the input tensors through the embedding layer and then the transformer.
Megatron-LM/megatron/core/models/vision/radio.py:176
Method
forward
Forward pass of the embedding module. Args: input_ids (Tensor): The input tokens position_ids (Tensor): The position
Megatron-LM/megatron/core/models/common/embeddings/language_model_embedding.py:99
Method
forward
Forward pass of multimodal RoPE embedding. Args: position_ids (torch.Tensor): A postion_id tensor with shape [3, batchsize, seqle
Megatron-LM/megatron/core/models/common/embeddings/rotary_pos_embedding.py:306
Method
forward
Args: Returns:
Megatron-LM/megatron/core/models/common/embeddings/relative_pos_embedding.py:175
Method
forward
Forward function of the LLaVA model. Args: images (torch.Tensor): input images of shape [num_tiles, img_h, img_w].
Megatron-LM/megatron/core/models/multimodal/llava_model.py:790
Method
forward
Siglip forward.
Megatron-LM/megatron/core/models/huggingface/clip_model.py:37
Method
forward
Forward function
Megatron-LM/megatron/core/models/huggingface/module.py:57
Method
forward
Qwen forward.
Megatron-LM/megatron/core/models/huggingface/qwen_model.py:39
Method
forward
Cross attention for Retro encoder. Notation: ns : Sequence length. bs : Batch size. d : Hidden size.
Megatron-LM/megatron/core/models/retro/encoder_attention.py:34
Method
forward
Retro decoder bias-dropout-add. Args: training (bool): If training, then apply dropout. fused (bool): Fuse bias-dropo
Megatron-LM/megatron/core/models/retro/encoder_attention.py:172
← previous
next →
7,401–7,500 of 10,561, ranked by callers