Update

LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

README.md

@@ -15,16 +15,16 @@ These results underscore HRM’s potential as a transformative advancement towar
 Ensure PyTorch and CUDA are installed. The repo needs CUDA extensions to be built. If not present, run the following commands:
 
 ```bash
-# Install CUDA 12.4
-CUDA_URL=https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run
+# Install CUDA 12.6
+CUDA_URL=https://developer.download.nvidia.com/compute/cuda/12.6.3/local_installers/cuda_12.6.3_560.35.05_linux.run
 
 wget -q --show-progress --progress=bar:force:noscroll -O cuda_installer.run $CUDA_URL
 sudo sh cuda_installer.run --silent --toolkit --override
 
-export CUDA_HOME=/usr/local/cuda-12.4 
+export CUDA_HOME=/usr/local/cuda-12.6
 
-# Install PyTorch with CUDA 12.4
-PYTORCH_INDEX_URL=https://download.pytorch.org/whl/cu124
+# Install PyTorch with CUDA 12.6
+PYTORCH_INDEX_URL=https://download.pytorch.org/whl/cu126
 
 pip3 install torch torchvision torchaudio --index-url $PYTORCH_INDEX_URL
 
@@ -32,6 +32,20 @@ pip3 install torch torchvision torchaudio --index-url $PYTORCH_INDEX_URL
 pip3 install packaging ninja wheel setuptools setuptools-scm
 ```
 
+Then install FlashAttention. For Hopper GPUs, install FlashAttention 3
+
+```bash
+git clone [email protected]:Dao-AILab/flash-attention.git
+cd flash-attention/hopper
+python setup.py install
+```
+
+For Ampere or earlier GPUs, install FlashAttenion 2
+
+```bash
+pip3 install flash-attn
+```
+
 ## Install Python Dependencies 🐍
 
 ```bash
@@ -62,6 +76,14 @@ OMP_NUM_THREADS=8 python pretrain.py data_path=data/sudoku-extreme-1k-aug-1000 e
 
 Runtime: ~10 hours on a RTX 4070 laptop GPU
 
+## Trained Checkpoints 🚧
+
+ - [ARC-AGI-2](https://huggingface.co/sapientinc/HRM-checkpoint-ARC-2)
+ - [Sudoku 9x9 Extreme (1000 examples)](https://huggingface.co/sapientinc/HRM-checkpoint-sudoku-extreme)
+ - [Maze 30x30 Hard (1000 examples)](https://huggingface.co/sapientinc/HRM-checkpoint-maze-30x30-hard)
+
+To use the checkpoints, see Evaluation section below.
+
 ## Full-scale Experiments 🔵
 
 Experiments below assume an 8-GPU setup.

dataset/build_maze_dataset.py

@@ -20,7 +20,7 @@ cli = ArgParser()
 
 
 class DataProcessConfig(BaseModel):
-    source_repo: str = "imone/small-sample-challenge-maze-30x30-hard"
+    source_repo: str = "sapientinc/maze-30x30-hard-1k"
     output_dir: str = "data/maze-30x30-hard-1k"
 
     subsample_size: Optional[int] = None

dataset/build_sudoku_dataset.py

@@ -16,7 +16,7 @@ cli = ArgParser()
 
 
 class DataProcessConfig(BaseModel):
-    source_repo: str = "imone/sudoku-hard-v2"
+    source_repo: str = "sapientinc/sudoku-extreme"
     output_dir: str = "data/sudoku-extreme-full"
 
     subsample_size: Optional[int] = None

evaluate.py

@@ -39,7 +39,7 @@ def launch():
 
     # Dataloader
     train_loader, train_metadata = create_dataloader(config, "train", test_set_mode=False, epochs_per_iter=1, global_batch_size=config.global_batch_size, rank=RANK, world_size=WORLD_SIZE)
-    eval_loader,  eval_metadata  = create_dataloader(config, "test", test_set_mode=True, epochs_per_iter=1, global_batch_size=config.global_batch_size, test_set_limit_examples=LIMIT_EXAMPLES, rank=RANK, world_size=WORLD_SIZE)
+    eval_loader,  eval_metadata  = create_dataloader(config, "test", test_set_mode=True, epochs_per_iter=1, global_batch_size=config.global_batch_size, rank=RANK, world_size=WORLD_SIZE)
 
     # Models
     train_state = init_train_state(config, train_metadata, world_size=WORLD_SIZE)

models/layers.py

@@ -4,6 +4,12 @@ import torch
 from torch import nn
 import torch.nn.functional as F
 
+try:
+    from flash_attn_interface import flash_attn_func  # type: ignore[import]
+except ImportError:
+    # Fallback to FlashAttention 2
+    from flash_attn import flash_attn_func  # type: ignore[import]
+
 from models.common import trunc_normal_init_
 
 
@@ -22,14 +28,14 @@ def rotate_half(x: torch.Tensor):
 
 
 def apply_rotary_pos_emb(q: torch.Tensor, k: torch.Tensor, cos: torch.Tensor, sin: torch.Tensor):
-    # q, k: [bs, num_heads, seq_len, head_dim]
+    # q, k: [bs, seq_len, num_heads, head_dim]
     # cos, sin: [seq_len, head_dim]
     orig_dtype = q.dtype
     q = q.to(cos.dtype)
     k = k.to(cos.dtype)
 
-    q_embed = (q * cos) + (rotate_half(q) * sin)
-    k_embed = (k * cos) + (rotate_half(k) * sin)
+    q_embed = (q * cos.unsqueeze(-2)) + (rotate_half(q) * sin.unsqueeze(-2))
+    k_embed = (k * cos.unsqueeze(-2)) + (rotate_half(k) * sin.unsqueeze(-2))
 
     return q_embed.to(orig_dtype), k_embed.to(orig_dtype)
 
@@ -110,10 +116,10 @@ class Attention(nn.Module):
         qkv = self.qkv_proj(hidden_states)
 
         # Split head
-        qkv = qkv.view(batch_size, seq_len, self.num_heads + 2 * self.num_key_value_heads, self.head_dim).transpose(-2, -3)
-        query = qkv[:, :self.num_heads]
-        key = qkv[:, self.num_heads: self.num_heads + self.num_key_value_heads]
-        value = qkv[:, self.num_heads + self.num_key_value_heads:]
+        qkv = qkv.view(batch_size, seq_len, self.num_heads + 2 * self.num_key_value_heads, self.head_dim)
+        query = qkv[:, :, :self.num_heads]
+        key = qkv[:, :, self.num_heads: self.num_heads + self.num_key_value_heads]
+        value = qkv[:, :, self.num_heads + self.num_key_value_heads:]
 
         # RoPE
         if cos_sin is not None:
@@ -121,10 +127,12 @@ class Attention(nn.Module):
             query, key = apply_rotary_pos_emb(query, key, cos, sin)
 
         # flash attn
-        attn_output = F.scaled_dot_product_attention(query=query, key=key, value=value, is_causal=self.causal)
+        attn_output = flash_attn_func(q=query, k=key, v=value, causal=self.causal)
+        if isinstance(attn_output, tuple):  # fa2 and fa3 compatibility
+            attn_output = attn_output[0]
 
         # attn_output: [batch_size, num_heads, seq_len, head_dim]
-        attn_output = attn_output.transpose(-2, -3).view(batch_size, seq_len, self.output_size)  # type: ignore
+        attn_output = attn_output.view(batch_size, seq_len, self.output_size)  # type: ignore
         return self.o_proj(attn_output)
 
 

pretrain.py

@@ -16,7 +16,6 @@ import coolname
 import hydra
 import pydantic
 from omegaconf import DictConfig
-from wandb.util import make_artifact_name_safe
 from adam_atan2 import AdamATan2
 
 from puzzle_dataset import PuzzleDataset, PuzzleDatasetConfig, PuzzleDatasetMetadata
@@ -126,7 +125,7 @@ def create_model(config: PretrainConfig, train_metadata: PuzzleDatasetMetadata,
         model: nn.Module = model_cls(model_cfg)
         model = loss_head_cls(model, **config.arch.loss.__pydantic_extra__)  # type: ignore
         if "DISABLE_COMPILE" not in os.environ:
-            model = torch.compile(model, dynamic=False, fullgraph=True)  # type: ignore
+            model = torch.compile(model, dynamic=False)  # type: ignore
 
         # Broadcast parameters from rank 0
         if world_size > 1: