assets/docs/preprocessing.md

@@ -1,6 +1,6 @@
 # Preprocessing Pipeline
 
-Declarative JSON-driven data preprocessing. One `SectionedMaskBuilder` handles all formats via `input.sections` (single-output) or `input.sources` (multi-output).
+Declarative JSON-driven data preprocessing. No code needed -- describe your input format and mask rules in a config file, the engine does the rest.
 
 ## Philosophy
 
@@ -9,57 +9,18 @@ Declarative JSON-driven data preprocessing. One `SectionedMaskBuilder` handles a
 | `tokenizer_config.json` (`chat_template`) | Formatting -- how roles become tokens |
 | `pipeline.json` (`mask`) | Masking -- which roles participate in training |
 
-A single config file captures the entire pipeline, reusable and version-controllable.
-
-## Config Structure
-
-```json
-{
-  "input":         {},   // sections (single) or sources (multi)
-  "mask":          {},   // role → "train" | "mask"
-  "mask_default":  "mask",
-  "preprocessing": {},
-  "output":        {}
-}
-```
-
-### Section Fields
-
-| Field | Type | Default | Description |
-|-------|------|---------|-------------|
-| `field` | str | -- | JSONL key to read |
-| `action` | str | -- | `"train"` / `"mask"` / `"$role"` |
-| `template` | bool | `false` | Apply `chat_template` per message |
-| `add_special_tokens` | bool | `true` for first non-template section | Add special tokens during encode |
-
-### Source Fields (multi-output mode)
-
-| Field | Type | Default | Description |
-|-------|------|---------|-------------|
-| `sections` | list[dict] | -- | Same as single-output section list |
-| `list_field` | bool | `false` | JSONL field holds a list; tokenise each element |
-| `mask_key` | str | `"{key}_mask"` | Explicit output key for loss mask |
-
----
+The two are fully decoupled. A single config file captures the entire pipeline, reusable and version-controllable. Extension is via factory registration (`@MaskBuilderFactory.register`) -- no need to touch existing code.
 
 ## Quick Start
 
 ### SFT Chat
 
-Input JSONL:
-
-```json
-{"messages": [{"role": "system", "content": "You are helpful."}, {"role": "user", "content": "Hi"}, {"role": "assistant", "content": "Hello!"}]}
-```
-
-Config:
-
 ```json
 {
+  "version": 1,
   "input": {
-    "sections": [
-      {"field": "messages", "action": "$role", "template": true}
-    ]
+    "type": "chat",
+    "messages_key": "messages"
   },
   "mask": {
     "system": "mask",
@@ -68,225 +29,172 @@ Config:
   },
   "mask_default": "mask",
   "preprocessing": {
-    "max_seq_len": 2048
+    "max_seq_len": 2048,
+    "deduplicate": true
   },
   "output": {
+    "domain_key": "source",
     "storage_format": "bin",
-    "dtype": {"loss_mask": "bool"}
+    "max_tokens_per_shard": 100000000
   }
 }
 ```
 
-Output keys: `sequence` (int32), `loss_mask` (bool)
+Three lines of mask rules cover the most common SFT case: train on assistant turns, mask everything else.
 
-### SFT Instruction
-
-Input JSONL:
-
-```json
-{"prompt": "Translate to French: Hello", "response": "Bonjour"}
-```
-
-Config:
+### Instruction Tuning
 
 ```json
 {
+  "version": 1,
   "input": {
-    "sections": [
-      {"field": "prompt",   "action": "mask", "add_special_tokens": true},
-      {"field": "response", "action": "train"}
-    ]
+    "type": "instruction",
+    "prompt_key": "instruction",
+    "response_key": "output"
+  },
+  "mask": {
+    "prompt": "mask",
+    "response": "train"
   },
   "mask_default": "mask",
   "preprocessing": {
     "max_seq_len": 2048
+  },
+  "output": {
+    "storage_format": "bin"
   }
 }
 ```
 
-Output keys: `sequence`, `loss_mask`
+Mask splits at the prompt/response field boundary.
 
-### Pretrain
-
-Input JSONL:
-
-```json
-{"text": "Artificial Intelligence is a field of computer science..."}
-```
-
-Config:
+### Pretraining
 
 ```json
 {
+  "version": 1,
   "input": {
-    "sections": [
-      {"field": "text", "action": "train"}
-    ]
+    "type": "text",
+    "text_key": "content"
   },
+  "mask": {},
   "preprocessing": {
-    "max_seq_len": 8192,
-    "min_chars": 100
+    "max_seq_len": 2048,
+    "min_chars": 50
+  },
+  "output": {
+    "storage_format": "bin"
   }
 }
 ```
 
-Output keys: `sequence` (no `loss_mask` — all tokens trained)
+No mask -- train on all tokens.
 
-### DPO
+### Run
 
-Input JSONL:
-
-```json
-{"chosen": [{"role": "user", "content": "What is 2+2?"}, {"role": "assistant", "content": "4"}], "rejected": [{"role": "user", "content": "What is 2+2?"}, {"role": "assistant", "content": "5"}]}
+```bash
+python scripts/tools/preprocess.py data/*.jsonl -o output/ -c sft.json
 ```
 
-Config:
-
-```json
-{
-  "input": {
-    "sources": {
-      "chosen": {
-        "sections": [
-          {"field": "chosen", "action": "$role", "template": true}
-        ]
-      },
-      "rejected": {
-        "sections": [
-          {"field": "rejected", "action": "$role", "template": true}
-        ]
-      }
-    }
-  },
-  "mask": {
-    "user": "mask",
-    "assistant": "train"
-  },
-  "mask_default": "mask"
-}
-```
-
-Output keys: `chosen`, `chosen_mask`, `rejected`, `rejected_mask`
-
-### GRPO
-
-Input JSONL:
-
-```json
-{"prompt": [{"role": "user", "content": "What is 2+2?"}], "responses": ["4", "Five", "Four"], "rewards": [1.0, 0.3, 0.8]}
-```
-
-Config:
-
-```json
-{
-  "input": {
-    "sources": {
-      "prompts": {
-        "sections": [
-          {"field": "prompt", "action": "mask", "template": true}
-        ]
-      },
-      "responses": {
-        "sections": [
-          {"field": "responses", "action": "train"}
-        ],
-        "list_field": true,
-        "mask_key": "masks"
-      },
-      "rewards": {
-        "sections": [
-          {"field": "rewards", "action": "value"}
-        ]
-      }
-    }
-  },
-  "mask": {
-    "user": "mask",
-    "assistant": "train"
-  },
-  "mask_default": "mask"
-}
-```
-
-Output keys: `prompts`, `responses`, `masks`, `rewards` (float32)
-
-- `action: "value"` — extract raw values from JSONL without tokenisation
-- `list_field: true` — tokenise each list element independently, then concatenate
-- `mask_key: "masks"` — rename the auto-generated mask key (default: `responses_mask`)
-
----
-
 ## Configuration Reference
 
 ### `input`
 
-| Field | Type | Default | Description |
-|-------|------|---------|-------------|
-| `sections` | list[dict] or null | `null` | Section specs for single-output mode |
-| `sources` | dict[str, dict] or null | `null` | Source specs for multi-output mode (DPO/GRPO) |
-
-When `sources` is set, `sections` is ignored.
+| Field | Type | Required | Default | Description |
+|-------|------|----------|---------|-------------|
+| `type` | string | yes | `"chat"` | Format: `"chat"`, `"instruction"`, or `"text"` |
+| `messages_key` | string | no | `"messages"` | JSON key for messages array (chat) |
+| `prompt_key` | string | no | `"prompt"` | JSON key for prompt field (instruction) |
+| `response_key` | string | no | `"response"` | JSON key for response field (instruction) |
+| `text_key` | string | no | `"text"` | JSON key for text field |
 
 ### `mask`
 
+A map of `{role_or_field: "mask" | "train"}`. The engine uses this to build `loss_mask`:
+
+- `"mask"` -- tokens in this span are ignored during training (`loss_mask=0`)
+- `"train"` -- tokens in this span contribute to the loss (`loss_mask=1`)
+
+For chat mode, keys are role names (`system`, `user`, `assistant`, ...).
+For instruction mode, keys are `"prompt"` and `"response"`.
+
 | Field | Type | Default | Description |
 |-------|------|---------|-------------|
-| `mask` | dict | `{}` | `{role: "train" \| "mask"}` |
-| `mask_default` | str | `"mask"` | Default action for unlisted roles |
+| `mask` | dict | `{}` | Role/field to action mapping |
+| `mask_default` | string | `"mask"` | Default action for unlisted roles |
 
 ### `preprocessing`
 
 | Field | Type | Default | Description |
 |-------|------|---------|-------------|
-| `max_seq_len` | int | `2048` | Truncate sequences to this length |
-| `min_chars` | int | `50` | Skip text-mode items shorter than this |
-| `max_chars` | int | `2000000` | Skip text-mode items longer than this |
-| `max_items` | int or null | `null` | Stop after N documents |
+| `max_seq_len` | int | `2048` | Maximum token length; truncated if exceeded |
+| `min_chars` | int | `50` | Minimum character length; dropped if shorter (text mode only) |
+| `max_chars` | int | `2000000` | Maximum character length; dropped if longer (text mode only) |
+| `deduplicate` | bool | `true` | Remove exact duplicates via MD5 of first 200 chars |
+| `max_items` | int or null | `null` | Maximum items to process; `null` = unlimited |
 
 ### `output`
 
 | Field | Type | Default | Description |
 |-------|------|---------|-------------|
-| `domain_key` | str or null | `null` | JSONL key for domain grouping |
-| `storage_format` | str | `"bin"` | `"bin"` (mmap) or `"h5"` |
-| `max_tokens_per_shard` | int | `100000000` | Flush threshold in cumulative tokens |
-| `dtype` | dict[str, str] | `{}` | Per-key tensor dtype override (e.g. `{"loss_mask": "bool"}`) |
-
----
+| `domain_key` | string or null | `null` | JSON key for domain grouping; `null` = all output to `__default__` |
+| `storage_format` | string | `"bin"` | `"bin"` (mmap, zero-copy) or `"h5"` (HDF5) |
+| `max_tokens_per_shard` | int | `100000000` | Max tokens per output shard |
 
 ## Mask Algorithm
 
-### Template mode (`template: true`)
+### Chat Mode (role-span tracking)
 
-For each message in the field's array:
+For each message in the `messages` array:
 
-1. Prepend BOS token (masked)
-2. Render through `chat_template` for that single message
-3. Encode rendered text
-4. Apply mask rule for the message's role
+1. Prepend BOS token (position 0, always masked)
+2. Render through the chat template for that single message
+3. Encode the rendered text, record token span `(start, end, role)`
+4. Concatenate all spans — special tokens from the chat template naturally prevent BPE merging across message boundaries
+5. Fill `loss_mask` from the mask rules
 
-### Non-template mode
+**Multi-turn example**:
 
-Encode the field value as text. Mask value is 1 (train) or 0 (mask) per the section's `action`.
+```
+Data:
+  [system: "You are helpful."]
+  [user: "What is 2+2?"]
+  [assistant: "4"]
+  [user: "What is 3+3?"]
+  [assistant: "6"]
 
-### Text config detection
+Config:
+  "mask": {"system": "mask", "user": "mask", "assistant": "train"}
 
-When no section uses `template` and all sections have `action: "train"`, the builder skips mask generation entirely — all tokens are trained.
+Result:
+  tokens:  <bos> [system span] [user span] [assistant:4 span] [user span] [assistant:6 span]
+  mask:      0       0            0              1               0             1
+```
 
----
+Both assistant turns are trained. All system and user tokens are masked.
+
+### Instruction Mode (field boundary)
+
+Encode the prompt and response fields independently, then split the mask at the field boundary.
+
+- `"prompt": "mask", "response": "train"` -- mask the left half, train the right half
+- `"prompt": "train", "response": "mask"` -- the reverse
+
+### Text Mode (no mask)
+
+Pure tokenization. No `loss_mask` is produced. Used for pretraining.
 
 ## Output Layout
 
 ### Single-Shard (`bin`)
 
 ```
-output/
-  __default__/
-    meta.json
-    sequence.bin
-    loss_mask.bin
-  wiki/
+output_dir/
+  __default__/              # when domain_key is null
+    meta.json               # {"sequence": {"shape": [N], "dtype": "int64"}, ...}
+    sequence.bin            # int64 raw bytes, mmap-able for zero-copy reads
+    loss_mask.bin           # int64 raw bytes
+  wiki/                     # when domain_key="source" and item["source"]="wiki"
     meta.json
     sequence.bin
     loss_mask.bin
@@ -294,10 +202,10 @@ output/
 
 ### Multi-Shard (`bin`)
 
-When `max_tokens_per_shard` is exceeded:
+When `max_tokens_per_shard` is exceeded, bin output is split into numbered shard subdirectories:
 
 ```
-output/
+output_dir/
   __default__/
     shard_0000/
       meta.json
@@ -309,38 +217,67 @@ output/
       loss_mask.bin
 ```
 
-`MmapStore` discovers all shards under the domain directory via `rglob("meta.json")`.
+`MmapStore` automatically discovers and merges all shards under the domain directory.
 
----
+### H5 Output
 
-## CLI
+HDF5 files are always named with a shard index, avoiding overwrite regardless of `max_tokens_per_shard`:
 
-```bash
-# SFT
-python scripts/tools/preprocess.py data/sft/*.jsonl -o output/sft/ -c configs/sft_chat.json
-
-# DPO
-python scripts/tools/preprocess.py data/dpo/*.jsonl -o output/dpo/ -c configs/dpo.json --tokenizer_path params
-
-# GRPO
-python scripts/tools/preprocess.py data/grpo/*.jsonl -o output/grpo/ -c configs/grpo.json
+```
+output_dir/
+  __default__/
+    data_0000.h5            # each H5 contains key→dataset groups
+    data_0001.h5
+  wiki/
+    data_0000.h5
 ```
 
----
-
-## Python API
+## Python API Usage
 
 ```python
 from astrai.preprocessing.pipeline import Pipeline
 from astrai.config.preprocess_config import PipelineConfig
 
-config = PipelineConfig.from_json("sft.json")
+config = PipelineConfig.from_json("sft_pipeline.json")
 Pipeline(
     config,
     ["data_part1.jsonl", "data_part2.jsonl"],
     output_dir="output/",
-    tokenizer_path="params",
+    tokenizer_path="params"
 ).run()
 ```
 
-> Document Update Time: 2026-06-03
+Or from the CLI:
+
+```bash
+python scripts/tools/preprocess.py data/*.jsonl -o output/ -c sft.json
+```
+
+## Extension
+
+Register a custom builder for new formats:
+
+```python
+from astrai.preprocessing.builder import BaseMaskBuilder, MaskBuilderFactory
+
+@MaskBuilderFactory.register("my_format")
+class MyFormatBuilder(BaseMaskBuilder):
+    def build(self, item: dict, config, tokenizer) -> dict | None:
+        # Return {"ids": [...], "loss_mask": [...], "domain": "..."}
+        # Return None to skip this item
+        ...
+```
+
+Then set `"input": {"type": "my_format"}` in your config.
+
+## Compared to Old Pipeline
+
+| Old (`astrai.preprocess.Pipeline`) | New (`astrai.preprocessing.pipeline.Pipeline`) |
+|---|---|
+| Configured via constructor arguments | Configured via JSON file |
+| Hardcoded `_transform_chat` / `_transform_text` | Factory-registered `Builder` with declarative mask rules |
+| Auto-detects format via magic key lists | Explicit `input.type` declaration |
+| Double-encodes (full + prompt), uses length diff for mask | Single-encode with role-span tracking |
+| Only trains the last assistant turn | Configurable: multi-turn, single-turn, or no mask |
+
+> Document Update Time: 2026-05-30

astrai/config/preprocess_config.py

@@ -1,9 +1,4 @@
-"""Pipeline configuration for JSONL preprocessing.
-
-Supports single-sequence (SFT/pretrain) and multi-output (DPO/GRPO)
-modes, both driven declaratively through ``input.sections`` or
-``input.sources``.
-"""
+"""Pipeline configuration for JSONL preprocessing."""
 
 from dataclasses import dataclass, field
 from typing import Dict, List, Optional
@@ -13,22 +8,7 @@ from astrai.config.base import BaseConfig
 
 @dataclass
 class InputConfig(BaseConfig):
-    """Declarative input mapping.
-
-    Single-output mode (backward-compatible)::
-
-        {"input": {"sections": [{"field": "messages", ...}]}}
-
-    Multi-output mode (DPO / GRPO)::
-
-        {"input": {"sources": {
-            "chosen": {"sections": [{"field": "chosen", ...}]},
-            "rejected": {"sections": [{"field": "rejected", ...}]},
-        }}}
-    """
-
     sections: Optional[List[Dict]] = None
-    sources: Optional[Dict[str, Dict]] = None
 
 
 @dataclass
@@ -45,13 +25,6 @@ class OutputConfig(BaseConfig):
     storage_format: str = "bin"
     max_tokens_per_shard: int = 100_000_000
     dtype: Dict[str, str] = field(default_factory=dict)
-    position_ids_mode: Optional[str] = None
-    """How to compute position_ids in packed sequences.
-
-    - ``None`` / ``"none"``: do not generate (backward compatible).
-    - ``"doc_reset"``: reset to 0 at each document boundary.
-    - ``"continuous"``: sequential 0, 1, 2, ... (pretrain, single doc).
-    """
 
 
 @dataclass

astrai/config/train_config.py

@@ -118,12 +118,6 @@ class TrainConfig(BaseConfig):
     val_dataset: Optional[Dataset] = field(
         default=None, metadata={"help": "Dataset for validation."}
     )
-    val_split: Optional[float] = field(
-        default=None,
-        metadata={
-            "help": "Ratio to split from training dataset for validation (e.g. 0.05). Ignored if val_dataset is set."
-        },
-    )
     val_step: int = field(
         default=1000,
         metadata={"help": "Number of optimizer steps between validation runs."},

astrai/dataset/dataset.py

@@ -223,7 +223,7 @@ class SFTDataset(BaseDataset):
 
     @property
     def required_keys(self) -> List[str]:
-        return ["sequence", "loss_mask", "position_ids"]
+        return ["sequence", "loss_mask"]
 
     def _fetch_data(self, begin_idx: int, end_idx: int, key: str) -> Tensor:
         return self.storage.fetch(begin_idx, end_idx, key)
@@ -231,17 +231,15 @@ class SFTDataset(BaseDataset):
     def __getitem__(self, index):
         begin_idx, end_idx = self.get_index(index)
 
-        x = self._fetch_data(begin_idx, end_idx, "sequence")
-        y = self._fetch_data(begin_idx + 1, end_idx + 1, "sequence")
-        position_ids = self._fetch_data(begin_idx, end_idx, "position_ids")
-        loss_mask = self._fetch_data(begin_idx + 1, end_idx + 1, "loss_mask")
+        x = self._fetch_data(begin_idx, end_idx, "sequence").to(dtype=torch.long)
+        y = self._fetch_data(begin_idx + 1, end_idx + 1, "sequence").to(
+            dtype=torch.long
+        )
+        loss_mask = self._fetch_data(begin_idx + 1, end_idx + 1, "loss_mask").to(
+            dtype=torch.bool
+        )
 
-        return {
-            "input_ids": x.to(dtype=torch.long),
-            "target_ids": y.to(dtype=torch.long),
-            "position_ids": position_ids.to(dtype=torch.long),
-            "loss_mask": loss_mask.to(dtype=torch.bool),
-        }
+        return {"input_ids": x, "target_ids": y, "loss_mask": loss_mask}
 
 
 @DatasetFactory.register("dpo")

astrai/dataset/storage.py

@@ -18,7 +18,6 @@ Key properties:
 """
 
 import bisect
-import glob
 import json
 import os
 from abc import ABC, abstractmethod
@@ -114,17 +113,13 @@ def detect_format(load_path: str) -> str:
             return "h5"
         raise ValueError(f"Unsupported file format: {suffix}")
 
-    h5_files = [
-        Path(p)
-        for pattern in ("*.h5", "*.hdf5")
-        for p in glob.glob(str(root / "**" / pattern), recursive=True)
-    ]
+    h5_files = list(root.rglob("*.h5")) + list(root.rglob("*.hdf5"))
     if h5_files:
         return "h5"
-    bin_files = [Path(p) for p in glob.glob(str(root / "**" / "*.bin"), recursive=True)]
+    bin_files = list(root.rglob("*.bin"))
     if bin_files:
         has_meta = (root / "meta.json").exists() or len(
-            [Path(p) for p in glob.glob(str(root / "**" / "meta.json"), recursive=True)]
+            list(root.rglob("meta.json"))
         ) > 0
         if has_meta:
             return "bin"
@@ -255,9 +250,7 @@ class MmapStore(Store):
         self._mmap_refs = []
         root = Path(path)
         all_raw: Dict[str, List[Tensor]] = {}
-        meta_paths = [
-            Path(p) for p in glob.glob(str(root / "**" / "meta.json"), recursive=True)
-        ]
+        meta_paths = list(root.rglob("meta.json"))
         for meta_path in meta_paths:
             raw = load_bin(str(meta_path.parent))
             for key, tensors in raw.items():

astrai/parallel/executor.py

@@ -2,7 +2,6 @@
 
 import contextlib
 import logging
-import os
 from contextlib import contextmanager
 from typing import Optional, Tuple
 
@@ -182,7 +181,7 @@ class DDPExecutor(BaseExecutor):
         if not self.use_distributed:
             logger.warning("DDP backend selected but world_size=1, model not wrapped")
             return model
-        local_rank = int(os.environ.get("LOCAL_RANK", get_rank()))
+        local_rank = get_rank()
         model = DDP(
             model,
             device_ids=[local_rank],

astrai/parallel/setup.py

@@ -1,5 +1,4 @@
 import os
-from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from functools import wraps
 from typing import Callable
@@ -31,7 +30,6 @@ def get_rank() -> int:
 def setup_parallel(
     rank: int,
     world_size: int,
-    local_rank: int,
     backend: str = "nccl",
     master_addr: str = "localhost",
     master_port: str = "29500",
@@ -43,18 +41,14 @@ def setup_parallel(
         return
 
     if world_size <= 1:
-        device_id = torch.device(device_type, local_rank)
-        os.environ["LOCAL_RANK"] = str(local_rank)
-        os.environ["WORLD_SIZE"] = "1"
-        os.environ["LOCAL_DEVICE"] = str(device_id)
         yield None
         return
 
-    device_id = torch.device(device_type, local_rank)
+    device_id = torch.device(device_type, rank)
 
     os.environ["MASTER_ADDR"] = master_addr
     os.environ["MASTER_PORT"] = master_port
-    os.environ["LOCAL_RANK"] = str(local_rank)
+    os.environ["LOCAL_RANK"] = str(rank)
     os.environ["WORLD_SIZE"] = str(world_size)
     os.environ["LOCAL_DEVICE"] = str(device_id)
 
@@ -96,7 +90,7 @@ def only_on_rank(rank, sync=False):
     return decorator
 
 
-def _run_single_rank(
+def wrapper_spawn_func(
     rank: int,
     world_size: int,
     backend: str,
@@ -106,10 +100,10 @@ def _run_single_rank(
     func: Callable,
     kwargs: dict,
 ):
+    try:
         with setup_parallel(
             rank=rank,
             world_size=world_size,
-        local_rank=rank,
             backend=backend,
             master_addr=master_addr,
             master_port=master_port,
@@ -117,99 +111,11 @@ def _run_single_rank(
         ):
             func(**kwargs)
 
-
-class LaunchStrategy(ABC):
-    """Strategy for launching a function in a distributed context."""
-
-    def __init__(
-        self,
-        world_size: int,
-        backend: str,
-        master_addr: str,
-        master_port: str,
-        device_type: str,
-        start_method: str,
-    ):
-        self.world_size = world_size
-        self.backend = backend
-        self.master_addr = master_addr
-        self.master_port = master_port
-        self.device_type = device_type
-        self.start_method = start_method
-
-    @abstractmethod
-    def launch(self, func: Callable, **kwargs):
-        raise NotImplementedError
-
-
-class TorchrunStrategy(LaunchStrategy):
-    """External orchestrator (torchrun, SLURM, K8s) — env vars pre-set."""
-
-    def launch(self, func: Callable, **kwargs):
-        rank = int(os.environ["RANK"])
-        world_size = int(os.environ["WORLD_SIZE"])
-        local_rank = int(os.environ.get("LOCAL_RANK", rank))
-        with setup_parallel(
-            rank=rank,
-            world_size=world_size,
-            local_rank=local_rank,
-            backend=self.backend,
-            master_addr=os.environ.get("MASTER_ADDR", self.master_addr),
-            master_port=os.environ.get("MASTER_PORT", self.master_port),
-            device_type=self.device_type,
-        ):
-            func(**kwargs)
-
-
-class LocalStrategy(LaunchStrategy):
-    """Local launcher — single-process or mp.start_processes."""
-
-    def launch(self, func: Callable, **kwargs):
-        args = (
-            self.world_size,
-            self.backend,
-            self.master_addr,
-            self.master_port,
-            self.device_type,
-            func,
-            kwargs,
-        )
-
-        if self.world_size == 1:
-            _run_single_rank(0, *args)
-            return
-
-        ctx = mp.start_processes(
-            _run_single_rank,
-            args=args,
-            nprocs=self.world_size,
-            start_method=self.start_method,
-            join=False,
-        )
-        try:
-            while not ctx.join():
-                pass
-        except BaseException:
-            for p in ctx.processes:
-                p.terminate()
-            ctx.join()
+    except Exception as e:
+        print(f"Error in rank {rank}: {e}")
         raise
 
 
-def _detect_launcher() -> str:
-    """Detect the distributed launcher from environment.
-
-    Returns one of: "torchelastic", "torchrun", "external", "local".
-    """
-    if dist.is_torchelastic_launched():
-        return "torchelastic"
-    if "LOCAL_WORLD_SIZE" in os.environ:
-        return "torchrun"
-    if "RANK" in os.environ and "WORLD_SIZE" in os.environ:
-        return "external"
-    return "local"
-
-
 def spawn_parallel_fn(
     func: Callable,
     world_size: int,
@@ -220,13 +126,41 @@ def spawn_parallel_fn(
     start_method: str = "spawn",
     **kwargs,
 ):
-    launcher = _detect_launcher()
-    if launcher in ("torchelastic", "torchrun", "external"):
-        strategy = TorchrunStrategy(
-            world_size, backend, master_addr, master_port, device_type, start_method
+    # clear environment variables
+    for key in [
+        "MASTER_ADDR",
+        "MASTER_PORT",
+        "RANK",
+        "WORLD_SIZE",
+        "LOCAL_RANK",
+        "LOCAL_DEVICE",
+    ]:
+        if key in os.environ:
+            del os.environ[key]
+
+    if world_size == 1:
+        device_id = torch.device(device_type, 0)
+        os.environ["LOCAL_RANK"] = "0"
+        os.environ["WORLD_SIZE"] = "1"
+        os.environ["LOCAL_DEVICE"] = str(device_id)
+
+        func(**kwargs)
+        return
+
+    wrapper_spawn_func_args = (
+        world_size,
+        backend,
+        master_addr,
+        master_port,
+        device_type,
+        func,
+        kwargs,
     )
-    else:
-        strategy = LocalStrategy(
-            world_size, backend, master_addr, master_port, device_type, start_method
+
+    mp.start_processes(
+        wrapper_spawn_func,
+        args=wrapper_spawn_func_args,
+        nprocs=world_size,
+        start_method=start_method,
+        join=True,
     )
-    strategy.launch(func, **kwargs)

astrai/preprocessing/builder.py

@@ -1,8 +1,7 @@
 """Mask building strategies for preprocessing pipeline.
 
 The single :class:`SectionedMaskBuilder` handles all input formats
-(single-sequence / DPO / GRPO) via declarative config: ``input.sections``
-for single-output or ``input.sources`` for multi-output.
+via declarative ``input.sections`` config.
 """
 
 from abc import ABC, abstractmethod
@@ -52,142 +51,43 @@ def _resolve_action(action: str, role: str, config) -> str:
 
 @MaskBuilderFactory.register("sectioned")
 class SectionedMaskBuilder(BaseMaskBuilder):
-    """Config-driven builder supporting single and multi-output modes.
+    """Config-driven builder: iterates over ``input.sections`` in order.
 
-    Single-output (backward-compatible)::
+    Each section specifies a JSONL field + mask action.
 
+    Section spec::
+
+        {
+            "field":    "messages",   # JSONL key
+            "action":   "$role",      # "train" | "mask" | "$role"
+            "template": true,         # apply chat_template per message (optional)
+            "add_special_tokens": false  # override encode flag (optional)
+        }
+
+    Example configs::
+
+        # Chat
         {"input": {"sections": [
             {"field": "messages", "action": "$role", "template": true}
         ]}}
-        → {"sequence": [...], "loss_mask": [...], "domain": "..."}
 
-    Multi-output (DPO / GRPO)::
+        # Instruction
+        {"input": {"sections": [
+            {"field": "prompt",   "action": "mask", "add_special_tokens": true},
+            {"field": "response", "action": "train"}
+        ]}}
 
-        {"input": {"sources": {
-            "chosen": {"sections": [
-                {"field": "chosen", "action": "$role", "template": true}
-            ]},
-            "rejected": {"sections": [
-                {"field": "rejected", "action": "$role", "template": true}
-            ]}
-        }}}
-        → {"chosen": [...], "chosen_mask": [...],
-           "rejected": [...], "rejected_mask": [...], "domain": "..."}
-
-    Output spec fields::
-
-        sections      – list of section specs (same format as single-output)
-        list_field    – True when the JSONL field holds a list of values to
-                        tokenise individually and concatenate (GRPO responses)
-        mask_key      – explicit output key for the loss mask
-                        (default: ``"{output_key}_mask"``)
-        dtype         – explicit tensor dtype for this output key
-                        (default: "int32")
+        # Text
+        {"input": {"sections": [
+            {"field": "text", "action": "train"}
+        ]}}
     """
 
     def build(self, item: dict, config, tokenizer) -> Optional[dict]:
-        sources_spec = getattr(config.input, "sources", None)
-        if sources_spec:
-            return self._build_multi(item, sources_spec, config, tokenizer)
-        return self._build_single(item, config, tokenizer)
-
-    def _build_single(self, item: dict, config, tokenizer) -> Optional[dict]:
         sections = config.input.sections
         if not sections:
             return None
 
-        ids, mask = self._process_sections(
-            item, sections, config, tokenizer, is_top_level=True
-        )
-        if ids is None:
-            return None
-
-        result: dict = {
-            "sequence": ids,
-            "domain": _extract_domain(item, config.output.domain_key),
-        }
-        if not all(m == 1 for m in mask):
-            result["loss_mask"] = mask
-        return result
-
-    def _build_multi(
-        self, item: dict, sources_spec: dict, config, tokenizer
-    ) -> Optional[dict]:
-        result: dict = {}
-        any_output = False
-
-        for output_key, spec in sources_spec.items():
-            sections = spec.get("sections", [])
-            if not sections:
-                continue
-
-            if self._is_value_section(sections):
-                ids = self._extract_raw_value(item, sections)
-                if ids is None:
-                    continue
-                result[output_key] = ids
-                any_output = True
-                continue
-
-            list_field = spec.get("list_field", False)
-            mask_key = spec.get("mask_key", f"{output_key}_mask")
-
-            if list_field:
-                ids, mask = self._process_list_field(item, sections, config, tokenizer)
-            else:
-                ids, mask = self._process_sections(
-                    item, sections, config, tokenizer, is_top_level=True
-                )
-
-            if ids is None:
-                continue
-
-            result[output_key] = ids
-            if not all(m == 1 for m in mask):
-                result[mask_key] = mask
-            elif "mask_key" in spec:
-                result[mask_key] = mask
-
-            any_output = True
-
-        if not any_output:
-            return None
-
-        result["domain"] = _extract_domain(item, config.output.domain_key)
-        return result
-
-    @staticmethod
-    def _is_value_section(sections: list) -> bool:
-        return len(sections) == 1 and sections[0].get("action") == "value"
-
-    @staticmethod
-    def _extract_raw_value(item: dict, sections: list):
-        """Extract a raw value from a JSONL field without tokenisation.
-
-        Used for GRPO rewards where the field contains float values.
-        """
-        sec = sections[0]
-        field = sec["field"]
-        raw = item.get(field)
-        if raw is None:
-            return None
-        if isinstance(raw, list):
-            return [float(v) for v in raw]
-        return [float(raw)]
-
-    def _process_sections(
-        self,
-        item: dict,
-        sections: list,
-        config,
-        tokenizer,
-        *,
-        is_top_level: bool = False,
-    ):
-        """Process a list of sections into ``(ids, loss_mask)``.
-
-        Returns ``(None, None)`` if the item should be skipped.
-        """
         all_ids: list[int] = []
         loss_mask: list[int] = []
 
@@ -196,7 +96,7 @@ class SectionedMaskBuilder(BaseMaskBuilder):
             s["action"] == "train" for s in sections
         )
 
-        if is_top_level and has_template and tokenizer.bos_token_id is not None:
+        if has_template and tokenizer.bos_token_id is not None:
             all_ids.append(tokenizer.bos_token_id)
             loss_mask.append(0)
 
@@ -210,46 +110,9 @@ class SectionedMaskBuilder(BaseMaskBuilder):
             )
 
             if use_template:
-                success = self._append_template_section(
-                    item, field, action, tokenizer, config, all_ids, loss_mask
-                )
-                if not success:
-                    continue
-            else:
-                success = self._append_text_section(
-                    item,
-                    field,
-                    action,
-                    tokenizer,
-                    add_special,
-                    is_text_config,
-                    config,
-                    all_ids,
-                    loss_mask,
-                )
-                if not success:
-                    continue
-
-            first_section = False
-
-        max_len = config.preprocessing.max_seq_len
-        all_ids = all_ids[:max_len]
-        loss_mask = loss_mask[: len(all_ids)]
-
-        if not all_ids:
-            return None, None
-
-        if is_top_level and has_template and len(all_ids) <= 1:
-            return None, None
-
-        return all_ids, loss_mask
-
-    def _append_template_section(
-        self, item, field, action, tokenizer, config, all_ids, loss_mask
-    ):
                 messages = item.get(field)
                 if not isinstance(messages, list) or not messages:
-            return False
+                    continue
                 for msg in messages:
                     role = msg.get("role", "")
                     act = _resolve_action(action, role, config)
@@ -260,79 +123,37 @@ class SectionedMaskBuilder(BaseMaskBuilder):
                     all_ids.extend(ids)
                     val = 1 if act == "train" else 0
                     loss_mask.extend([val] * len(ids))
-        return True
-
-    def _append_text_section(
-        self,
-        item,
-        field,
-        action,
-        tokenizer,
-        add_special,
-        is_text_config,
-        config,
-        all_ids,
-        loss_mask,
-    ):
+            else:
                 text = str(item.get(field, ""))
                 if not text.strip():
-            return False
+                    continue
                 if is_text_config:
                     pp = config.preprocessing
                     if pp.min_chars > 0 and len(text) < pp.min_chars:
-                return False
+                        continue
                     if len(text) > pp.max_chars:
-                return False
+                        continue
                 ids = tokenizer.encode(text, add_special_tokens=add_special)
                 all_ids.extend(ids)
                 val = 1 if action == "train" else 0
                 loss_mask.extend([val] * len(ids))
-        return True
 
-    def _process_list_field(self, item: dict, sections: list, config, tokenizer):
-        all_ids: list[int] = []
-        loss_mask: list[int] = []
-
-        for sec in sections:
-            field = sec["field"]
-            action = sec["action"]
-            use_template = sec.get("template", False)
-
-            values = item.get(field)
-            if not isinstance(values, list):
-                continue
-
-            for val in values:
-                if use_template:
-                    if isinstance(val, list):
-                        wrapper = {field: val}
-                        self._append_template_section(
-                            wrapper,
-                            field,
-                            action,
-                            tokenizer,
-                            config,
-                            all_ids,
-                            loss_mask,
-                        )
-                else:
-                    wrapper = {field: str(val)}
-                    self._append_text_section(
-                        wrapper,
-                        field,
-                        action,
-                        tokenizer,
-                        False,
-                        False,
-                        config,
-                        all_ids,
-                        loss_mask,
-                    )
+            first_section = False
 
         max_len = config.preprocessing.max_seq_len
         all_ids = all_ids[:max_len]
         loss_mask = loss_mask[: len(all_ids)]
 
         if not all_ids:
-            return None, None
-        return all_ids, loss_mask
+            return None
+
+        if has_template and len(all_ids) <= 1:
+            return None
+
+        result: dict = {
+            "sequence": all_ids,
+            "domain": _extract_domain(item, config.output.domain_key),
+        }
+        if not all(m == 1 for m in loss_mask):
+            result["loss_mask"] = loss_mask
+        return result

astrai/preprocessing/pipeline.py

@@ -81,20 +81,17 @@ class Pipeline:
             if result is None:
                 continue
 
-            domain = result.pop("domain", "__default__")
-
-            is_multi = bool(getattr(self.config.input, "sources", None))
-            if is_multi:
-                ids = self._primary_ids(result)
-            else:
             ids = result.pop("sequence")
-                result["sequence"] = ids
-
             if not ids:
                 continue
 
+            domain = result.pop("domain", "__default__")
+            result["sequence"] = ids
+
             bucket = domains[domain]
-            self._align_bucket(bucket, result, ids, is_multi)
+            for key in list(bucket.keys()):
+                if key not in result:
+                    bucket[key].append([1] * len(ids))
             for key, val in result.items():
                 bucket[key].append(val)
 
@@ -111,27 +108,6 @@ class Pipeline:
 
         print(f"Done. {count} documents tokenized.")
 
-    @staticmethod
-    def _primary_ids(result: dict) -> list:
-        """Return the first list-valued entry in *result* as the primary id
-        sequence for token counting."""
-        for val in result.values():
-            if isinstance(val, list) and val and isinstance(val[0], int):
-                return val
-        return []
-
-    @staticmethod
-    def _align_bucket(bucket: dict, result: dict, ids: list, is_multi: bool):
-        """Pad previously-accumulated keys that are missing from *result*."""
-        for key in list(bucket.keys()):
-            if key in result:
-                continue
-            if is_multi:
-                pad = bucket[key][-1] if bucket[key] else [1] * len(ids)
-                bucket[key].append(pad)
-            else:
-                bucket[key].append([1] * len(ids))
-
     def _iter_items(self):
         for path in self.paths:
             with open(path, "r", encoding="utf-8") as f:
@@ -144,7 +120,6 @@ class Pipeline:
     def _flush(self, domains, shard_idx):
         for domain, keys in domains.items():
             idx = shard_idx[domain]
-            chunk_dir = os.path.join(self.output_dir, domain)
             tensors = {}
             for key, ids_list in keys.items():
                 dt = _STR_TO_DTYPE.get(
@@ -153,27 +128,14 @@ class Pipeline:
                 tensors[key] = [
                     torch.tensor(list(chain.from_iterable(ids_list)), dtype=dt)
                 ]
-
-            pid_mode = self.config.output.position_ids_mode
-            if pid_mode and pid_mode != "none" and "sequence" in tensors:
-                pos_ids = []
-                if pid_mode == "doc_reset":
-                    for item in keys["sequence"]:
-                        pos_ids.extend(range(len(item)))
-                else:
-                    total = sum(len(item) for item in keys["sequence"])
-                    pos_ids = list(range(total))
-                tensors["position_ids"] = [torch.tensor(pos_ids, dtype=torch.int32)]
-
-            shard_path = os.path.join(chunk_dir, f"shard_{idx:04d}")
+            chunk_dir = os.path.join(self.output_dir, domain)
             fmt = self.config.output.storage_format
             if fmt == "bin":
-                save_bin(shard_path, tensors)
+                save_bin(os.path.join(chunk_dir, f"shard_{idx:04d}"), tensors)
             else:
                 save_h5(chunk_dir, f"data_{idx:04d}", tensors)
             shard_idx[domain] = idx + 1
-            first_key = "sequence" if "sequence" in tensors else next(iter(tensors))
             tqdm.tqdm.write(
                 f"  saved {domain}/shard_{idx:04d}  "
-                f"({tensors[first_key][0].numel():,} tokens)"
+                f"({tensors['sequence'][0].numel():,} tokens)"
             )

astrai/trainer/strategy.py

@@ -180,15 +180,14 @@ class SFTStrategy(BaseStrategy):
 
     def compute_loss(self, batch: Dict[str, Tensor]) -> Tensor:
         batch = move_to_device(batch, self.device)
-        input_ids, target_ids, position_ids, loss_mask = (
+        input_ids, target_ids, loss_mask = (
             batch["input_ids"],
             batch["target_ids"],
-            batch["position_ids"],
             batch["loss_mask"],
         )
 
         ignore_index = -100
-        logits = self.model(input_ids=input_ids, position_ids=position_ids)["logits"]
+        logits = self.model(input_ids=input_ids)["logits"]
         target_ids = target_ids.masked_fill(loss_mask == 0, ignore_index)
 
         loss = F.cross_entropy(

astrai/trainer/train_context.py

@@ -2,9 +2,8 @@ from dataclasses import dataclass, field
 from pathlib import Path
 from typing import Optional, Self
 
-import torch
 import torch.nn as nn
-from torch.utils.data import DataLoader, random_split
+from torch.utils.data import DataLoader
 
 from astrai.config.train_config import TrainConfig
 from astrai.dataset import ResumableDistributedSampler
@@ -109,27 +108,15 @@ class TrainContextBuilder:
         context.optimizer = cfg.optimizer_fn(model)
         context.scheduler = cfg.scheduler_fn(context.optimizer)
 
-        train_dataset = cfg.dataset
-        val_dataset = cfg.val_dataset
-
-        if val_dataset is None and cfg.val_split is not None:
-            n_total = len(cfg.dataset)
-            n_val = max(1, int(n_total * cfg.val_split))
-            n_train = n_total - n_val
-            generator = torch.Generator().manual_seed(cfg.random_seed)
-            train_dataset, val_dataset = random_split(
-                cfg.dataset, [n_train, n_val], generator=generator
-            )
-
         sampler_offset = context.iteration * cfg.batch_per_device
         sampler = ResumableDistributedSampler(
-            data_source=train_dataset,
+            data_source=cfg.dataset,
             start_epoch=context.epoch,
             start_iter=sampler_offset,
             seed=cfg.random_seed,
         )
         context.dataloader = DataLoader(
-            train_dataset,
+            cfg.dataset,
             batch_size=cfg.batch_per_device,
             sampler=sampler,
             num_workers=cfg.num_workers,
@@ -137,16 +124,16 @@ class TrainContextBuilder:
             prefetch_factor=cfg.prefetch_factor,
         )
 
-        if val_dataset is not None:
+        if cfg.val_dataset is not None:
             val_sampler = ResumableDistributedSampler(
-                data_source=val_dataset,
+                data_source=cfg.val_dataset,
                 start_epoch=0,
                 start_iter=0,
                 seed=cfg.random_seed,
                 shuffle=False,
             )
             context.val_dataloader = DataLoader(
-                val_dataset,
+                cfg.val_dataset,
                 batch_size=cfg.batch_per_device,
                 sampler=val_sampler,
                 num_workers=cfg.num_workers,

scripts/tools/evaluate_humaneval.py

@@ -1,336 +0,0 @@
-"""HumanEval code generation benchmark.
-
-Generates n completions per problem, extracts function bodies, executes
-against hidden tests, and computes pass@k.
-
-Usage::
-
-    python scripts/tools/evaluate_humaneval.py --param_path ./params \
-        --data_path HumanEval.jsonl.gz --output results.json \
-        --num_samples 200 --temperature 0.8 --max_tokens 512
-"""
-
-import argparse
-import json
-import os
-import re
-import signal
-import sys
-from math import prod
-from multiprocessing import Process, Queue
-from typing import Dict, List, Optional, Tuple
-
-import numpy as np
-import torch
-import tqdm
-
-from astrai.inference import InferenceEngine
-from astrai.model import AutoModel
-from astrai.tokenize import AutoTokenizer
-
-HUMANEVAL_URL = (
-    "https://github.com/openai/human-eval/raw/master/data/HumanEval.jsonl.gz"
-)
-
-_STOP_SEQUENCES = [
-    "\nclass ",
-    "\ndef ",
-    "\n# ",
-    "\nif __name__",
-    "\nprint(",
-    "\n\n\n",
-]
-
-
-def _download_humaneval(data_path: str):
-    if os.path.exists(data_path):
-        return
-    import gzip
-    import urllib.request
-
-    os.makedirs(os.path.dirname(data_path) or ".", exist_ok=True)
-    print(f"Downloading HumanEval from {HUMANEVAL_URL} ...")
-    tmp = data_path + ".tmp"
-    urllib.request.urlretrieve(HUMANEVAL_URL, tmp)
-    with gzip.open(tmp, "rb") as f_in:
-        with open(data_path, "wb") as f_out:
-            f_out.write(f_in.read())
-    os.remove(tmp)
-    print(f"  saved to {data_path}")
-
-
-def _load_problems(data_path: str) -> List[dict]:
-    problems = []
-    with open(data_path, "r", encoding="utf-8") as f:
-        for line in f:
-            line = line.strip()
-            if line:
-                problems.append(json.loads(line))
-    return problems
-
-
-def _extract_function_body(code: str, entry_point: str) -> Optional[str]:
-    """Extract the function body from a completion."""
-    pattern = rf"def\s+{re.escape(entry_point)}\b[^:]*:"
-    match = re.search(pattern, code)
-    if not match:
-        # Use the full code as-is if we can't find the function
-        return code
-
-    body_start = match.end()
-    lines = code[body_start:].split("\n")
-    body_lines = []
-    started = False
-
-    for line in lines:
-        stripped = line.rstrip()
-        if not stripped and not started:
-            continue
-        if not stripped and started:
-            body_lines.append("")
-            continue
-        if not started:
-            started = True
-        if stripped.lstrip() == stripped and started:
-            break
-        body_lines.append(stripped)
-
-    body = "\n".join(body_lines)
-    if not body.strip():
-        return None
-    return body
-
-
-def _trim_stop_sequences(text: str) -> str:
-    for stop in _STOP_SEQUENCES:
-        idx = text.find(stop)
-        if idx != -1:
-            text = text[:idx]
-    return text
-
-
-def _execute_code(problem: dict, completion: str, timeout: float = 3.0) -> bool:
-    """Run the completion against hidden tests in a subprocess."""
-
-    def _worker(queue, full_code):
-        try:
-            namespace = {}
-            exec(full_code, namespace)
-            check = namespace.get("check")
-            if check is None:
-                queue.put(False)
-                return
-            check(namespace.get(problem["entry_point"]))
-            queue.put(True)
-        except Exception:
-            queue.put(False)
-
-    full_code = problem["prompt"] + completion + "\n" + problem["test"]
-
-    queue: Queue = Queue()
-    proc = Process(target=_worker, args=(queue, full_code))
-    proc.start()
-    proc.join(timeout)
-
-    if proc.is_alive():
-        proc.terminate()
-        proc.join()
-        return False
-
-    try:
-        return queue.get_nowait()
-    except Exception:
-        return False
-
-
-def _pass_at_k(n: int, c: int, k: int) -> float:
-    """Unbiased estimator of pass@k."""
-    if n - c < k:
-        return 1.0
-    return 1.0 - float(prod(1.0 - k / np.arange(n - c + 1, n + 1)))
-
-
-def _deduplicate(completions: List[str]) -> List[str]:
-    seen = set()
-    unique = []
-    for c in completions:
-        if c not in seen:
-            seen.add(c)
-            unique.append(c)
-    return unique
-
-
-def _generate(
-    engine: InferenceEngine,
-    prompt: str,
-    num_samples: int,
-    max_tokens: int,
-    temperature: float,
-    top_p: float,
-    top_k: int,
-    batch_size: int,
-) -> List[str]:
-    batches = [prompt] * min(batch_size, num_samples)
-    completions = []
-    remaining = num_samples
-
-    while remaining > 0:
-        current = min(batch_size, remaining)
-        batch_prompts = batches[:current]
-        outputs = engine.generate(
-            prompt=batch_prompts,
-            stream=False,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            top_p=top_p,
-            top_k=top_k,
-        )
-        if isinstance(outputs, str):
-            outputs = [outputs]
-        completions.extend(outputs)
-        remaining -= current
-
-    return _deduplicate(completions)
-
-
-def evaluate(
-    engine: InferenceEngine,
-    problems: List[dict],
-    num_samples: int,
-    max_tokens: int,
-    temperature: float,
-    top_p: float,
-    top_k: int,
-    batch_size: int,
-    k_values: Tuple[int, ...] = (1, 10, 100),
-) -> Dict:
-    results = {}
-    all_pass_at_k = {k: [] for k in k_values}
-
-    for problem in tqdm.tqdm(problems, desc="HumanEval", unit="problem"):
-        task_id = problem["task_id"]
-        prompt = problem["prompt"]
-        entry_point = problem["entry_point"]
-
-        raw_completions = _generate(
-            engine,
-            prompt,
-            num_samples,
-            max_tokens,
-            temperature,
-            top_p,
-            top_k,
-            batch_size,
-        )
-
-        completions = []
-        for raw in raw_completions:
-            trimmed = _trim_stop_sequences(raw)
-            body = _extract_function_body(trimmed, entry_point)
-            if body:
-                completions.append(body)
-
-        passed = 0
-        for comp in completions:
-            if _execute_code(problem, comp):
-                passed += 1
-
-        n = len(completions)
-        c = passed
-        result = {"task_id": task_id, "n": n, "passed": c}
-        for k in k_values:
-            result[f"pass@{k}"] = round(_pass_at_k(n, c, k), 4)
-            all_pass_at_k[k].append(_pass_at_k(n, c, k))
-        results[task_id] = result
-
-    summary = {}
-    for k in k_values:
-        vals = all_pass_at_k[k]
-        summary[f"pass@{k}"] = round(float(np.mean(vals)), 4)
-    results["_summary"] = summary
-
-    return results
-
-
-def main():
-    parser = argparse.ArgumentParser(description="HumanEval benchmark")
-    parser.add_argument(
-        "--param_path", type=str, default="./params", help="Model directory"
-    )
-    parser.add_argument(
-        "--data_path",
-        type=str,
-        default="./humaneval/HumanEval.jsonl",
-        help="HumanEval JSONL file (auto-download if missing)",
-    )
-    parser.add_argument("--output", type=str, default=None, help="Output JSON path")
-    parser.add_argument(
-        "--num_samples",
-        type=int,
-        default=200,
-        help="Completions per problem",
-    )
-    parser.add_argument(
-        "--max_tokens", type=int, default=512, help="Max generation tokens"
-    )
-    parser.add_argument(
-        "--temperature", type=float, default=0.8, help="Sampling temperature"
-    )
-    parser.add_argument("--top_p", type=float, default=0.95, help="Top-p sampling")
-    parser.add_argument("--top_k", type=int, default=50, help="Top-k sampling")
-    parser.add_argument(
-        "--batch_size", type=int, default=1, help="Inference batch size"
-    )
-    parser.add_argument(
-        "--problems",
-        type=int,
-        nargs="+",
-        default=None,
-        help="Specific problem indices (0-based)",
-    )
-    args = parser.parse_args()
-
-    _download_humaneval(args.data_path)
-    problems = _load_problems(args.data_path)
-    if args.problems:
-        problems = [problems[i] for i in args.problems if i < len(problems)]
-
-    model = AutoModel.from_pretrained(args.param_path)
-    tokenizer = AutoTokenizer.from_pretrained(args.param_path)
-    model.to(device="cuda", dtype=torch.bfloat16)
-
-    engine = InferenceEngine(
-        model=model,
-        tokenizer=tokenizer,
-        max_batch_size=args.batch_size,
-    )
-
-    results = evaluate(
-        engine=engine,
-        problems=problems,
-        num_samples=args.num_samples,
-        max_tokens=args.max_tokens,
-        temperature=args.temperature,
-        top_p=args.top_p,
-        top_k=args.top_k,
-        batch_size=args.batch_size,
-        k_values=(1, 10, 100),
-    )
-
-    summary = results.pop("_summary")
-    print(f"\n{'=' * 60}")
-    for k, v in summary.items():
-        print(f"  {k}: {v:.2%}")
-    print(f"{'=' * 60}")
-
-    if args.output:
-        results["_summary"] = summary
-        with open(args.output, "w", encoding="utf-8") as f:
-            json.dump(results, f, indent=2, ensure_ascii=False)
-        print(f"Results saved to {args.output}")
-
-    engine.shutdown()
-
-
-if __name__ == "__main__":
-    main()

scripts/tools/evaluate_mmlu.py

@@ -157,32 +157,10 @@ def build_prompt(
     return prompt
 
 
-def apply_chat(
-    tokenizer, raw_prompt: str, n_shot: int, dev_data: list[dict] | None
-) -> str:
-    """Wrap raw MMLU prompt in the model's chat template format.
-
-    For few-shot, prepend example Q&A pairs as a second user/assistant exchange.
-    """
-    messages = []
-    if n_shot > 0 and dev_data:
-        for item in dev_data[:n_shot]:
-            q = f"Question: {item['question']}\n"
-            for k in ("A", "B", "C", "D"):
-                q += f"{k}. {item[k]}\n"
-            q += "Answer:"
-            messages.append({"role": "user", "content": q})
-            messages.append({"role": "assistant", "content": item["answer"]})
-    messages.append({"role": "user", "content": raw_prompt})
-    return tokenizer.apply_chat_template(
-        messages, tokenize=False, add_generation_prompt=True
-    )
-
-
 def choice_logprob(
     model, tokenizer, context_ids: list[int], choice_letter: str, device: str
 ) -> float:
-    choice_text = choice_letter
+    choice_text = f" {choice_letter}"
     choice_ids = tokenizer.encode(choice_text, add_special_tokens=False)
     input_ids = context_ids + choice_ids
     max_len = model.config.max_len
@@ -218,11 +196,8 @@ def evaluate_subject(
     correct = 0
     total = 0
     for item in tqdm.tqdm(test_data, desc=f"{subject:40s}", leave=False):
-        raw_prompt = build_prompt(
-            item["question"], item, subject, n_shot, dev_data or []
-        )
-        context = apply_chat(tokenizer, raw_prompt, n_shot, dev_data or [])
-        context_ids = tokenizer.encode(context)
+        prompt = build_prompt(item["question"], item, subject, n_shot, dev_data or [])
+        context_ids = tokenizer.encode(prompt)
         scores = {
             c: choice_logprob(model, tokenizer, context_ids, c, device)
             for c in ("A", "B", "C", "D")

scripts/tools/train.py

@@ -8,7 +8,6 @@ import torch.optim as optim
 from astrai.config import AutoRegressiveLMConfig, TrainConfig
 from astrai.dataset import DatasetFactory
 from astrai.model import AutoRegressiveLM
-from astrai.model.components.decoder_block import DecoderBlock
 from astrai.trainer import SchedulerFactory, Trainer
 
 
@@ -116,12 +115,6 @@ def parse_args() -> argparse.Namespace:
         default=0.05,
         help="cross_entropy function label smoothing parameter",
     )
-    parser.add_argument(
-        "--gradient_checkpointing",
-        action=argparse.BooleanOptionalAction,
-        default=False,
-        help="Enable activation checkpointing for DecoderBlock modules.",
-    )
 
     parser.add_argument(
         "--ckpt_interval",
@@ -135,36 +128,6 @@ def parse_args() -> argparse.Namespace:
         default="checkpoint",
         help="Directory to save checkpoints.",
     )
-    parser.add_argument(
-        "--val_split",
-        type=float,
-        default=None,
-        help="Ratio to split from training dataset for validation (e.g. 0.05).",
-    )
-    parser.add_argument(
-        "--val_step",
-        type=int,
-        default=1000,
-        help="Number of optimizer steps between validation runs.",
-    )
-    parser.add_argument(
-        "--metrics",
-        nargs="*",
-        default=["loss", "lr"],
-        help="Metrics to log (e.g. --metrics loss lr val_loss). Default: loss lr.",
-    )
-    parser.add_argument(
-        "--log_dir",
-        type=str,
-        default="checkpoint/logs",
-        help="Directory for metric logs.",
-    )
-    parser.add_argument(
-        "--log_interval",
-        type=int,
-        default=100,
-        help="Number of batch iterations between metric logs.",
-    )
     parser.add_argument(
         "--grpo_sync_interval",
         type=int,
@@ -178,24 +141,6 @@ def parse_args() -> argparse.Namespace:
         "--start_batch", type=int, default=0, help="Start batch for training."
     )
 
-    parser.add_argument(
-        "--master_addr",
-        type=str,
-        default="localhost",
-        help="Master node address for distributed training.",
-    )
-    parser.add_argument(
-        "--master_port",
-        type=str,
-        default="29500",
-        help="Master node port for distributed training.",
-    )
-    parser.add_argument(
-        "--backend",
-        type=str,
-        default="nccl",
-        help="Distributed training backend.",
-    )
     parser.add_argument("--nprocs", type=int, default=1, help="Number of GPUs to use.")
     parser.add_argument(
         "--parallel_mode",
@@ -264,11 +209,6 @@ def train(
     warmup_ratio: float,
     ckpt_interval: int,
     ckpt_dir: str,
-    val_split: float,
-    val_step: int,
-    metrics: list[str],
-    log_dir: str,
-    log_interval: int,
     dpo_beta: float,
     grpo_clip_eps: float,
     grpo_kl_coef: float,
@@ -282,15 +222,11 @@ def train(
     random_seed: int,
     num_workers: int,
     pin_memory: bool,
-    gradient_checkpointing: bool,
     window_size: int,
     stride: int,
     nprocs: int,
     parallel_mode: str,
     device_type: str,
-    backend: str,
-    master_addr: str,
-    master_port: str,
     start_method: str,
 ):
     assert train_type in ["seq", "sft", "dpo", "grpo"]
@@ -350,8 +286,6 @@ def train(
         },
     )
 
-    grad_ckpt_modules = [DecoderBlock] if gradient_checkpointing else []
-
     train_config = TrainConfig(
         model_fn=model_fn,
         strategy=train_type,
@@ -370,18 +304,9 @@ def train(
         num_workers=num_workers,
         pin_memory=pin_memory,
         nprocs=nprocs,
-        backend=backend,
-        master_addr=master_addr,
-        master_port=master_port,
         parallel_mode=parallel_mode,
         device_type=device_type,
         start_method=start_method,
-        val_split=val_split,
-        val_step=val_step,
-        metrics=metrics,
-        log_dir=log_dir,
-        log_interval=log_interval,
-        gradient_checkpointing_modules=grad_ckpt_modules,
         executor_kwargs=executor_kwargs,
         extra_kwargs=strategy_kwargs,
     )

tests/data/conftest.py

@@ -1,202 +0,0 @@
-import tempfile
-
-import pytest
-from tokenizers import Tokenizer, models, pre_tokenizers, trainers
-
-from astrai.config.preprocess_config import (
-    InputConfig,
-    PipelineConfig,
-    ProcessingConfig,
-)
-from astrai.tokenize import AutoTokenizer
-
-_SPECIAL_TOKENS_CONFIG = {
-    "bos_token": "<|begin_of_sentence|>",
-    "eos_token": "<|end_of_sentence|>",
-    "pad_token": "<|_pad_|>",
-    "unk_token": "<|_unk_|>",
-    "im_start": "<|im_start|>",
-    "im_end": "<|im_end|>",
-}
-
-_SPECIAL_TOKENS = list(_SPECIAL_TOKENS_CONFIG.values())
-
-_CHAT_TEMPLATE = (
-    "{% for message in messages %}"
-    "{% if message['role'] == 'system' %}"
-    "<|im_start|>system\n{{ message['content'] }}<|im_end|>\n"
-    "{% elif message['role'] == 'user' %}"
-    "<|im_start|>user\n{{ message['content'] }}<|im_end|>\n"
-    "{% elif message['role'] == 'assistant' %}"
-    "<|im_start|>assistant\n{{ message['content'] }}<|im_end|>\n"
-    "{% endif %}"
-    "{% endfor %}"
-    "{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
-)
-
-_CHAT_SECTIONS = [{"field": "messages", "action": "$role", "template": True}]
-
-_INSTRUCTION_SECTIONS = [
-    {"field": "prompt", "action": "mask", "add_special_tokens": True},
-    {"field": "response", "action": "train"},
-]
-
-_TEXT_SECTIONS = [{"field": "text", "action": "train"}]
-
-_GRPO_RESPONSE_SECTIONS = [{"field": "responses", "action": "train"}]
-
-
-def _build_chat_tokenizer():
-    tok = Tokenizer(models.BPE())
-    tok.pre_tokenizer = pre_tokenizers.ByteLevel(add_prefix_space=False)
-    tr = trainers.BpeTrainer(
-        vocab_size=512,
-        min_frequency=1,
-        special_tokens=_SPECIAL_TOKENS,
-    )
-    train_data = [
-        "hello world",
-        "Hi there!",
-        "You are helpful.",
-        "What is 2+2?",
-        "Tell me a story about dragons and knights.",
-        "Sure, here is a tale.",
-        "Translate to French: Hello",
-        "Bonjour",
-        "Artificial Intelligence is a field of computer science.",
-        "system",
-        "user",
-        "assistant",
-        "<|im_start|>",
-        "<|im_end|>",
-        *[chr(i) for i in range(32, 127)],
-    ]
-    tok.train_from_iterator(train_data, tr)
-
-    auto_tok = AutoTokenizer()
-    auto_tok._tokenizer = tok
-    auto_tok._special_token_map = {
-        "bos_token": "<|begin_of_sentence|>",
-        "eos_token": "<|end_of_sentence|>",
-        "pad_token": "<|_pad_|>",
-        "unk_token": "<|_unk_|>",
-    }
-    auto_tok.set_chat_template(_CHAT_TEMPLATE)
-    return auto_tok
-
-
-@pytest.fixture(scope="session")
-def chat_tokenizer():
-    return _build_chat_tokenizer()
-
-
-@pytest.fixture
-def temp_dir():
-    d = tempfile.mkdtemp()
-    yield d
-    import shutil
-
-    shutil.rmtree(d, ignore_errors=True)
-
-
-def make_chat_config():
-    return PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={"system": "mask", "user": "mask", "assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-
-
-def make_instruction_config():
-    return PipelineConfig(
-        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
-        mask={"prompt": "mask", "response": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-
-
-def make_text_config():
-    return PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        preprocessing=ProcessingConfig(
-            max_seq_len=2048, min_chars=1, max_chars=2_000_000
-        ),
-    )
-
-
-def make_dpo_chat_config():
-    return PipelineConfig(
-        input=InputConfig(
-            sources={
-                "chosen": {
-                    "sections": [
-                        {"field": "chosen", "action": "$role", "template": True}
-                    ]
-                },
-                "rejected": {
-                    "sections": [
-                        {"field": "rejected", "action": "$role", "template": True}
-                    ]
-                },
-            }
-        ),
-        mask={"user": "mask", "assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-
-
-def make_grpo_config():
-    return PipelineConfig(
-        input=InputConfig(
-            sources={
-                "prompts": {
-                    "sections": [
-                        {"field": "prompt", "action": "mask", "template": True}
-                    ]
-                },
-                "responses": {
-                    "sections": _GRPO_RESPONSE_SECTIONS,
-                    "list_field": True,
-                    "mask_key": "masks",
-                },
-                "rewards": {
-                    "sections": [{"field": "rewards", "action": "value"}],
-                },
-            }
-        ),
-        mask={"user": "mask", "assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-
-
-def make_grpo_no_template_config():
-    return PipelineConfig(
-        input=InputConfig(
-            sources={
-                "prompts": {
-                    "sections": [
-                        {
-                            "field": "prompt",
-                            "action": "mask",
-                            "add_special_tokens": True,
-                        }
-                    ]
-                },
-                "responses": {
-                    "sections": _GRPO_RESPONSE_SECTIONS,
-                    "list_field": True,
-                    "mask_key": "masks",
-                },
-                "rewards": {
-                    "sections": [{"field": "rewards", "action": "value"}],
-                },
-            }
-        ),
-        mask={"user": "mask", "assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )

tests/data/test_dataset.py

@@ -98,7 +98,6 @@ def test_sft_dataset_with_random_data(base_test_env):
     dummy_data = {
         "sequence": [torch.randint(0, 1000, (seq_length,), dtype=torch.int64)],
         "loss_mask": [torch.ones(seq_length, dtype=torch.bool)],
-        "position_ids": [torch.arange(seq_length, dtype=torch.int32)],
     }
 
     save_h5(test_dir, "sft_data", dummy_data)

tests/data/test_preprocess.py

@@ -0,0 +1,713 @@
+import json
+import os
+import tempfile
+
+import pytest
+from tokenizers import Tokenizer, models, pre_tokenizers, trainers
+
+from astrai.config.preprocess_config import (
+    InputConfig,
+    OutputConfig,
+    PipelineConfig,
+    ProcessingConfig,
+)
+from astrai.preprocessing.builder import (
+    MaskBuilderFactory,
+    SectionedMaskBuilder,
+)
+from astrai.preprocessing.pipeline import Pipeline, filter_by_length
+from astrai.tokenize import AutoTokenizer
+
+_SPECIAL_TOKENS_CONFIG = {
+    "bos_token": "<|begin_of_sentence|>",
+    "eos_token": "<|end_of_sentence|>",
+    "pad_token": "<|_pad_|>",
+    "unk_token": "<|_unk_|>",
+    "im_start": "<|im_start|>",
+    "im_end": "<|im_end|>",
+}
+
+_SPECIAL_TOKENS = list(_SPECIAL_TOKENS_CONFIG.values())
+
+_CHAT_TEMPLATE = (
+    "{% for message in messages %}"
+    "{% if message['role'] == 'system' %}"
+    "<|im_start|>system\n{{ message['content'] }}<|im_end|>\n"
+    "{% elif message['role'] == 'user' %}"
+    "<|im_start|>user\n{{ message['content'] }}<|im_end|>\n"
+    "{% elif message['role'] == 'assistant' %}"
+    "<|im_start|>assistant\n{{ message['content'] }}<|im_end|>\n"
+    "{% endif %}"
+    "{% endfor %}"
+    "{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+)
+
+
+def _build_chat_tokenizer() -> AutoTokenizer:
+    tok = Tokenizer(models.BPE())
+    tok.pre_tokenizer = pre_tokenizers.ByteLevel(add_prefix_space=False)
+    tr = trainers.BpeTrainer(
+        vocab_size=512,
+        min_frequency=1,
+        special_tokens=_SPECIAL_TOKENS,
+    )
+    train_data = [
+        "hello world",
+        "Hi there!",
+        "You are helpful.",
+        "What is 2+2?",
+        "Tell me a story about dragons and knights.",
+        "Sure, here is a tale.",
+        "Translate to French: Hello",
+        "Bonjour",
+        "Artificial Intelligence is a field of computer science.",
+        "system",
+        "user",
+        "assistant",
+        "<|im_start|>",
+        "<|im_end|>",
+        *[chr(i) for i in range(32, 127)],
+    ]
+    tok.train_from_iterator(train_data, tr)
+
+    auto_tok = AutoTokenizer()
+    auto_tok._tokenizer = tok
+    auto_tok._special_token_map = {
+        "bos_token": "<|begin_of_sentence|>",
+        "eos_token": "<|end_of_sentence|>",
+        "pad_token": "<|_pad_|>",
+        "unk_token": "<|_unk_|>",
+    }
+    auto_tok.set_chat_template(_CHAT_TEMPLATE)
+    return auto_tok
+
+
+@pytest.fixture(scope="session")
+def chat_tokenizer():
+    return _build_chat_tokenizer()
+
+
+@pytest.fixture
+def temp_dir():
+    d = tempfile.mkdtemp()
+    yield d
+    import shutil
+
+    shutil.rmtree(d, ignore_errors=True)
+
+
+_CHAT_SECTIONS = [{"field": "messages", "action": "$role", "template": True}]
+
+_INSTRUCTION_SECTIONS = [
+    {"field": "prompt", "action": "mask", "add_special_tokens": True},
+    {"field": "response", "action": "train"},
+]
+
+_TEXT_SECTIONS = [{"field": "text", "action": "train"}]
+
+
+def make_chat_config():
+    return PipelineConfig(
+        input=InputConfig(sections=_CHAT_SECTIONS),
+        mask={"system": "mask", "user": "mask", "assistant": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+def make_instruction_config():
+    return PipelineConfig(
+        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
+        mask={"prompt": "mask", "response": "train"},
+        mask_default="mask",
+        preprocessing=ProcessingConfig(max_seq_len=2048),
+    )
+
+
+def make_text_config():
+    return PipelineConfig(
+        input=InputConfig(sections=_TEXT_SECTIONS),
+        preprocessing=ProcessingConfig(
+            max_seq_len=2048, min_chars=1, max_chars=2_000_000
+        ),
+    )
+
+
+class TestPipelineConfig:
+    def test_default_values(self):
+        config = PipelineConfig()
+        assert config.version == 1
+        assert config.mask == {}
+        assert config.mask_default == "mask"
+        assert config.preprocessing.max_seq_len == 2048
+        assert config.output.storage_format == "bin"
+        assert config.input.sections is None
+
+    def test_from_dict_flat(self):
+        data = {
+            "version": 1,
+            "input": {
+                "sections": [{"field": "messages", "action": "$role", "template": True}]
+            },
+            "mask": {"system": "mask", "assistant": "train"},
+            "mask_default": "mask",
+            "preprocessing": {"max_seq_len": 1024},
+            "output": {"storage_format": "h5"},
+        }
+        config = PipelineConfig.from_dict(data)
+        assert config.input.sections == [
+            {"field": "messages", "action": "$role", "template": True}
+        ]
+        assert config.mask == {"system": "mask", "assistant": "train"}
+        assert config.preprocessing.max_seq_len == 1024
+        assert config.output.storage_format == "h5"
+
+    def test_to_dict_roundtrip(self):
+        config = PipelineConfig(
+            input=InputConfig(sections=_INSTRUCTION_SECTIONS),
+            mask={"prompt": "mask", "response": "train"},
+            mask_default="mask",
+        )
+        d = config.to_dict()
+        config2 = PipelineConfig.from_dict(d)
+        assert config2.input.sections == _INSTRUCTION_SECTIONS
+        assert config2.mask == {"prompt": "mask", "response": "train"}
+
+    def test_to_json_from_json(self, temp_dir):
+        config = PipelineConfig(
+            input=InputConfig(sections=_TEXT_SECTIONS),
+            mask={"text": "train"},
+            mask_default="mask",
+        )
+        path = os.path.join(temp_dir, "config.json")
+        config.to_json(path)
+        loaded = PipelineConfig.from_json(path)
+        assert loaded.input.sections == _TEXT_SECTIONS
+        assert loaded.mask == {"text": "train"}
+
+
+class TestChatMaskBuilder:
+    def test_simple_chat_mask(self, chat_tokenizer):
+        config = make_chat_config()
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {"role": "system", "content": "You are helpful."},
+                {"role": "user", "content": "Hello."},
+                {"role": "assistant", "content": "Hi there!"},
+            ]
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        assert result is not None
+        assert "sequence" in result
+        assert "loss_mask" in result
+        assert len(result["sequence"]) == len(result["loss_mask"])
+
+        ids = chat_tokenizer.decode(result["sequence"], skip_special_tokens=False)
+
+        assert "system" in ids.lower() or "<|im_start|>system" in ids
+        assert "assistant" in ids.lower() or "<|im_start|>assistant" in ids
+
+        total = len(result["sequence"])
+        trained = sum(result["loss_mask"])
+        assert trained > 0, "At least assistant tokens should be trained"
+        assert trained < total, "System and user tokens should be masked"
+
+    def test_mask_only_assistant_trained(self, chat_tokenizer):
+        config = make_chat_config()
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {"role": "user", "content": "What is 2+2?"},
+                {"role": "assistant", "content": "4"},
+            ]
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        mask = result["loss_mask"]
+        ids = result["sequence"]
+
+        assert len(ids) == len(mask)
+
+        trained_positions = [i for i, m in enumerate(mask) if m == 1]
+        assert len(trained_positions) > 0, "At least some tokens should be trained"
+
+        masked_positions = [i for i, m in enumerate(mask) if m == 0]
+        assert len(masked_positions) > 0, "User tokens should be masked"
+
+    def test_chat_all_masked(self, chat_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_CHAT_SECTIONS),
+            mask={"system": "mask", "user": "mask", "assistant": "mask"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+        )
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {"role": "system", "content": "You are helpful."},
+                {"role": "assistant", "content": "Hi there!"},
+            ]
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        assert sum(result["loss_mask"]) == 0
+
+    def test_chat_all_trained(self, chat_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_CHAT_SECTIONS),
+            mask={},
+            mask_default="train",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+        )
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {"role": "system", "content": "You are helpful."},
+                {"role": "assistant", "content": "Hi there!"},
+            ]
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        assert sum(result["loss_mask"]) == len(result["sequence"]) - 1
+
+    def test_empty_messages_returns_none(self, chat_tokenizer):
+        config = make_chat_config()
+        builder = SectionedMaskBuilder()
+        assert builder.build({"messages": []}, config, chat_tokenizer) is None
+        assert builder.build({}, config, chat_tokenizer) is None
+
+    def test_domain_extraction(self, chat_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_CHAT_SECTIONS),
+            mask={"assistant": "train"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+            output=OutputConfig(domain_key="source"),
+        )
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {"role": "user", "content": "Hi"},
+                {"role": "assistant", "content": "Hello"},
+            ],
+            "source": "wiki",
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        assert result["domain"] == "wiki"
+
+    def test_truncation_to_max_len(self, chat_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_CHAT_SECTIONS),
+            mask={"assistant": "train"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=10),
+        )
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": "Tell me a very long story about dragons and knights and magic.",
+                },
+                {"role": "assistant", "content": "Sure! Here is a tale..."},
+            ]
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        assert len(result["sequence"]) <= 10
+        assert len(result["loss_mask"]) == len(result["sequence"])
+
+
+class TestInstructionMaskBuilder:
+    def test_basic_instruction_mask(self, test_tokenizer):
+        config = make_instruction_config()
+        builder = SectionedMaskBuilder()
+        item = {"prompt": "Translate to French: Hello", "response": "Bonjour"}
+        result = builder.build(item, config, test_tokenizer)
+        assert result is not None
+        assert len(result["sequence"]) == len(result["loss_mask"])
+
+    def test_prompt_masked_response_trained(self, test_tokenizer):
+        config = make_instruction_config()
+        builder = SectionedMaskBuilder()
+        item = {"prompt": "hello", "response": "world"}
+        result = builder.build(item, config, test_tokenizer)
+        mask = result["loss_mask"]
+        ids = result["sequence"]
+
+        prompt_ids = test_tokenizer.encode("hello", add_special_tokens=True)
+        response_ids = test_tokenizer.encode("world", add_special_tokens=False)
+
+        p_len = min(len(prompt_ids), len(ids))
+        assert all(m == 0 for m in mask[:p_len])
+
+        if p_len < len(ids):
+            assert all(m == 1 for m in mask[p_len:])
+
+    def test_train_on_prompt(self, test_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(
+                sections=[
+                    {
+                        "field": "prompt",
+                        "action": "train",
+                        "add_special_tokens": True,
+                    },
+                    {"field": "response", "action": "mask"},
+                ]
+            ),
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+        )
+        builder = SectionedMaskBuilder()
+        item = {"prompt": "hello", "response": "world"}
+        result = builder.build(item, config, test_tokenizer)
+        mask = result["loss_mask"]
+        ids = result["sequence"]
+
+        prompt_ids = test_tokenizer.encode("hello", add_special_tokens=True)
+        p_len = min(len(prompt_ids), len(ids))
+        assert all(m == 1 for m in mask[:p_len])
+
+
+class TestTextMaskBuilder:
+    def test_basic_text(self, test_tokenizer):
+        config = make_text_config()
+        builder = SectionedMaskBuilder()
+        item = {"text": "Hello world. This is a test document."}
+        result = builder.build(item, config, test_tokenizer)
+        assert result is not None
+        assert "sequence" in result
+        assert len(result["sequence"]) > 0
+        assert "loss_mask" not in result
+
+    def test_empty_text_returns_none(self, test_tokenizer):
+        config = make_text_config()
+        builder = SectionedMaskBuilder()
+        assert builder.build({"text": ""}, config, test_tokenizer) is None
+        assert builder.build({"text": "   "}, config, test_tokenizer) is None
+
+    def test_too_short_text(self, test_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_TEXT_SECTIONS),
+            preprocessing=ProcessingConfig(min_chars=100),
+        )
+        builder = SectionedMaskBuilder()
+        assert builder.build({"text": "short"}, config, test_tokenizer) is None
+
+    def test_truncation(self, test_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_TEXT_SECTIONS),
+            preprocessing=ProcessingConfig(max_seq_len=3, min_chars=1),
+        )
+        builder = SectionedMaskBuilder()
+        item = {"text": "This is a very long text that should be truncated"}
+        result = builder.build(item, config, test_tokenizer)
+        assert len(result["sequence"]) <= 3
+
+
+class TestPipeline:
+    def test_full_chat_pipeline(self, temp_dir, chat_tokenizer):
+        tokenizer_dir = os.path.join(temp_dir, "tok")
+        os.makedirs(tokenizer_dir, exist_ok=True)
+        chat_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
+        with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
+            json.dump(
+                {
+                    "special_tokens": _SPECIAL_TOKENS_CONFIG,
+                    "chat_template": _CHAT_TEMPLATE,
+                },
+                f,
+            )
+
+        jsonl_path = os.path.join(temp_dir, "chat.jsonl")
+        with open(jsonl_path, "w", encoding="utf-8") as f:
+            f.write(
+                json.dumps(
+                    {
+                        "messages": [
+                            {"role": "system", "content": "You are helpful."},
+                            {"role": "user", "content": "Hi."},
+                            {"role": "assistant", "content": "Hello!"},
+                        ]
+                    }
+                )
+                + "\n"
+            )
+            f.write(
+                json.dumps(
+                    {
+                        "messages": [
+                            {"role": "user", "content": "What is 2+2?"},
+                            {"role": "assistant", "content": "4"},
+                        ]
+                    }
+                )
+                + "\n"
+            )
+
+        config = PipelineConfig(
+            input=InputConfig(sections=_CHAT_SECTIONS),
+            mask={"system": "mask", "user": "mask", "assistant": "train"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+            output=OutputConfig(storage_format="bin", domain_key=None),
+        )
+
+        out_dir = os.path.join(temp_dir, "output")
+        Pipeline(
+            config=config,
+            input_paths=[jsonl_path],
+            output_dir=out_dir,
+            tokenizer_path=tokenizer_dir,
+        ).run()
+
+        meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
+        assert os.path.exists(meta_path)
+        with open(meta_path, "r") as f:
+            meta = json.load(f)
+        assert "sequence" in meta
+        assert "loss_mask" in meta
+        assert meta["sequence"]["dtype"] == "int32"
+        assert meta["loss_mask"]["dtype"] == "int32"
+
+    def test_full_text_pipeline(self, temp_dir, test_tokenizer):
+
+        tokenizer_dir = os.path.join(temp_dir, "tok")
+        os.makedirs(tokenizer_dir, exist_ok=True)
+
+        test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
+        with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
+            json.dump(
+                {
+                    "special_tokens": {
+                        "pad_token": "<|_pad_|>",
+                        "unk_token": "<|_unk_|>",
+                    }
+                },
+                f,
+            )
+
+        jsonl_path = os.path.join(temp_dir, "text.jsonl")
+        with open(jsonl_path, "w", encoding="utf-8") as f:
+            f.write(
+                json.dumps(
+                    {
+                        "text": "Hello world this is a test document with enough characters to pass the minimum length filter."
+                    }
+                )
+                + "\n"
+            )
+            f.write(
+                json.dumps(
+                    {
+                        "text": "Another document for testing purposes with sufficient length to be processed."
+                    }
+                )
+                + "\n"
+            )
+
+        config = PipelineConfig(
+            input=InputConfig(sections=_TEXT_SECTIONS),
+            preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=10),
+            output=OutputConfig(storage_format="bin"),
+        )
+
+        out_dir = os.path.join(temp_dir, "output")
+        Pipeline(
+            config=config,
+            input_paths=[jsonl_path],
+            output_dir=out_dir,
+            tokenizer_path=tokenizer_dir,
+        ).run()
+
+        meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
+        assert os.path.exists(meta_path)
+        with open(meta_path, "r") as f:
+            meta = json.load(f)
+        assert "sequence" in meta
+        assert "loss_mask" not in meta
+        assert meta["sequence"]["dtype"] == "int32"
+
+    def test_full_instruction_pipeline(self, temp_dir, test_tokenizer):
+        tokenizer_dir = os.path.join(temp_dir, "tok")
+        os.makedirs(tokenizer_dir, exist_ok=True)
+        test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
+        with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
+            json.dump(
+                {
+                    "special_tokens": {
+                        "pad_token": "<|_pad_|>",
+                        "unk_token": "<|_unk_|>",
+                    }
+                },
+                f,
+            )
+
+        jsonl_path = os.path.join(temp_dir, "instruct.jsonl")
+        with open(jsonl_path, "w", encoding="utf-8") as f:
+            f.write(
+                json.dumps(
+                    {
+                        "prompt": "Tell me a joke",
+                        "response": "Why did the chicken cross the road?",
+                    }
+                )
+                + "\n"
+            )
+            f.write(
+                json.dumps(
+                    {
+                        "prompt": "What is AI?",
+                        "response": "Artificial Intelligence is a field of computer science.",
+                    }
+                )
+                + "\n"
+            )
+
+        config = PipelineConfig(
+            input=InputConfig(sections=_INSTRUCTION_SECTIONS),
+            mask={"prompt": "mask", "response": "train"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+            output=OutputConfig(storage_format="bin"),
+        )
+
+        out_dir = os.path.join(temp_dir, "output")
+        Pipeline(
+            config=config,
+            input_paths=[jsonl_path],
+            output_dir=out_dir,
+            tokenizer_path=tokenizer_dir,
+        ).run()
+
+        meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
+        assert os.path.exists(meta_path)
+        with open(meta_path, "r") as f:
+            meta = json.load(f)
+        assert "sequence" in meta
+        assert "loss_mask" in meta
+        assert meta["sequence"]["dtype"] == "int32"
+        assert meta["loss_mask"]["dtype"] == "int32"
+
+    def test_dtype_override(self, temp_dir, test_tokenizer):
+        tokenizer_dir = os.path.join(temp_dir, "tok")
+        os.makedirs(tokenizer_dir, exist_ok=True)
+        test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
+        with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
+            json.dump(
+                {
+                    "special_tokens": {
+                        "pad_token": "<|_pad_|>",
+                        "unk_token": "<|_unk_|>",
+                    }
+                },
+                f,
+            )
+
+        jsonl_path = os.path.join(temp_dir, "data.jsonl")
+        with open(jsonl_path, "w", encoding="utf-8") as f:
+            f.write(
+                json.dumps(
+                    {
+                        "prompt": "Q",
+                        "response": "A",
+                    }
+                )
+                + "\n"
+            )
+
+        config = PipelineConfig(
+            input=InputConfig(sections=_INSTRUCTION_SECTIONS),
+            mask={"prompt": "mask", "response": "train"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+            output=OutputConfig(
+                storage_format="bin",
+                dtype={"loss_mask": "bool"},
+            ),
+        )
+
+        out_dir = os.path.join(temp_dir, "output")
+        Pipeline(
+            config=config,
+            input_paths=[jsonl_path],
+            output_dir=out_dir,
+            tokenizer_path=tokenizer_dir,
+        ).run()
+
+        meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
+        with open(meta_path, "r") as f:
+            meta = json.load(f)
+        assert meta["sequence"]["dtype"] == "int32"
+        assert meta["loss_mask"]["dtype"] == "bool"
+
+
+class TestUtility:
+    def test_filter_by_length(self):
+        assert filter_by_length("hello world", min_len=5)
+        assert not filter_by_length("hi", min_len=5)
+        assert not filter_by_length("x" * 100, max_len=50)
+        assert filter_by_length("just right", min_len=5, max_len=20)
+
+
+class TestSectionedMaskBuilder:
+    def test_sectioned_chat(self, chat_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_CHAT_SECTIONS),
+            mask={"system": "mask", "user": "mask", "assistant": "train"},
+            mask_default="mask",
+            preprocessing=ProcessingConfig(max_seq_len=2048),
+        )
+        builder = SectionedMaskBuilder()
+        item = {
+            "messages": [
+                {"role": "user", "content": "What is 2+2?"},
+                {"role": "assistant", "content": "4"},
+            ]
+        }
+        result = builder.build(item, config, chat_tokenizer)
+        assert result is not None
+        assert len(result["sequence"]) == len(result["loss_mask"])
+        assert sum(result["loss_mask"]) > 0
+        assert 0 in result["loss_mask"]
+
+    def test_sectioned_instruction(self, test_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_INSTRUCTION_SECTIONS),
+            preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=0),
+        )
+        builder = SectionedMaskBuilder()
+        item = {"prompt": "Q: Why?", "response": "A: Because."}
+        result = builder.build(item, config, test_tokenizer)
+        assert result is not None
+        mask = result["loss_mask"]
+        assert mask[0] == 0
+        assert mask[-1] == 1
+
+    def test_sectioned_text(self, test_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_TEXT_SECTIONS),
+            preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=1),
+        )
+        builder = SectionedMaskBuilder()
+        item = {"text": "Hello world, this is a test."}
+        result = builder.build(item, config, test_tokenizer)
+        assert result is not None
+        assert "loss_mask" not in result
+
+    def test_sectioned_text_too_short(self, test_tokenizer):
+        config = PipelineConfig(
+            input=InputConfig(sections=_TEXT_SECTIONS),
+            preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=100),
+        )
+        builder = SectionedMaskBuilder()
+        item = {"text": "short"}
+        result = builder.build(item, config, test_tokenizer)
+        assert result is None
+
+
+class TestFactoryRegistration:
+    def test_registered_builders(self):
+        names = MaskBuilderFactory._registry.list_names()
+        assert "sectioned" in names
+
+    def test_create_sectioned_builder(self):
+        builder = MaskBuilderFactory.create("sectioned")
+        assert isinstance(builder, SectionedMaskBuilder)

tests/data/test_preprocess_builder.py

@@ -1,396 +0,0 @@
-from astrai.config.preprocess_config import (
-    InputConfig,
-    OutputConfig,
-    PipelineConfig,
-    ProcessingConfig,
-)
-from astrai.preprocessing.builder import (
-    MaskBuilderFactory,
-    SectionedMaskBuilder,
-)
-from tests.data.conftest import (
-    _CHAT_SECTIONS,
-    _INSTRUCTION_SECTIONS,
-    _TEXT_SECTIONS,
-    make_chat_config,
-    make_dpo_chat_config,
-    make_grpo_config,
-    make_instruction_config,
-    make_text_config,
-)
-
-
-def test_chat_simple(chat_tokenizer):
-    config = make_chat_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {"role": "system", "content": "You are helpful."},
-            {"role": "user", "content": "Hello."},
-            {"role": "assistant", "content": "Hi there!"},
-        ]
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert result is not None
-    assert "sequence" in result
-    assert "loss_mask" in result
-    assert len(result["sequence"]) == len(result["loss_mask"])
-
-    ids = chat_tokenizer.decode(result["sequence"], skip_special_tokens=False)
-    assert "system" in ids.lower() or "<|im_start|>system" in ids
-    assert "assistant" in ids.lower() or "<|im_start|>assistant" in ids
-
-    total = len(result["sequence"])
-    trained = sum(result["loss_mask"])
-    assert trained > 0
-    assert trained < total
-
-
-def test_chat_mask_only_assistant(chat_tokenizer):
-    config = make_chat_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {"role": "user", "content": "What is 2+2?"},
-            {"role": "assistant", "content": "4"},
-        ]
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    mask = result["loss_mask"]
-    ids = result["sequence"]
-    assert len(ids) == len(mask)
-
-    trained = [i for i, m in enumerate(mask) if m == 1]
-    masked = [i for i, m in enumerate(mask) if m == 0]
-    assert len(trained) > 0
-    assert len(masked) > 0
-
-
-def test_chat_all_masked(chat_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={"system": "mask", "user": "mask", "assistant": "mask"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {"role": "system", "content": "You are helpful."},
-            {"role": "assistant", "content": "Hi there!"},
-        ]
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert sum(result["loss_mask"]) == 0
-
-
-def test_chat_all_trained(chat_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={},
-        mask_default="train",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {"role": "system", "content": "You are helpful."},
-            {"role": "assistant", "content": "Hi there!"},
-        ]
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert sum(result["loss_mask"]) == len(result["sequence"]) - 1
-
-
-def test_chat_empty_messages(chat_tokenizer):
-    config = make_chat_config()
-    builder = SectionedMaskBuilder()
-    assert builder.build({"messages": []}, config, chat_tokenizer) is None
-    assert builder.build({}, config, chat_tokenizer) is None
-
-
-def test_chat_domain_extraction(chat_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={"assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-        output=OutputConfig(domain_key="source"),
-    )
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {"role": "user", "content": "Hi"},
-            {"role": "assistant", "content": "Hello"},
-        ],
-        "source": "wiki",
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert result["domain"] == "wiki"
-
-
-def test_chat_truncation(chat_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={"assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=10),
-    )
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {
-                "role": "user",
-                "content": "Tell me a very long story about dragons and knights and magic.",
-            },
-            {"role": "assistant", "content": "Sure! Here is a tale..."},
-        ]
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert len(result["sequence"]) <= 10
-    assert len(result["loss_mask"]) == len(result["sequence"])
-
-
-def test_instruction_basic(test_tokenizer):
-    config = make_instruction_config()
-    builder = SectionedMaskBuilder()
-    item = {"prompt": "Translate to French: Hello", "response": "Bonjour"}
-    result = builder.build(item, config, test_tokenizer)
-    assert result is not None
-    assert len(result["sequence"]) == len(result["loss_mask"])
-
-
-def test_instruction_prompt_masked(test_tokenizer):
-    config = make_instruction_config()
-    builder = SectionedMaskBuilder()
-    item = {"prompt": "hello", "response": "world"}
-    result = builder.build(item, config, test_tokenizer)
-    mask = result["loss_mask"]
-    ids = result["sequence"]
-
-    prompt_ids = test_tokenizer.encode("hello", add_special_tokens=True)
-    p_len = min(len(prompt_ids), len(ids))
-    assert all(m == 0 for m in mask[:p_len])
-    if p_len < len(ids):
-        assert all(m == 1 for m in mask[p_len:])
-
-
-def test_instruction_train_on_prompt(test_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(
-            sections=[
-                {"field": "prompt", "action": "train", "add_special_tokens": True},
-                {"field": "response", "action": "mask"},
-            ]
-        ),
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-    builder = SectionedMaskBuilder()
-    item = {"prompt": "hello", "response": "world"}
-    result = builder.build(item, config, test_tokenizer)
-    mask = result["loss_mask"]
-    ids = result["sequence"]
-
-    prompt_ids = test_tokenizer.encode("hello", add_special_tokens=True)
-    p_len = min(len(prompt_ids), len(ids))
-    assert all(m == 1 for m in mask[:p_len])
-
-
-def test_text_basic(test_tokenizer):
-    config = make_text_config()
-    builder = SectionedMaskBuilder()
-    item = {"text": "Hello world. This is a test document."}
-    result = builder.build(item, config, test_tokenizer)
-    assert result is not None
-    assert "sequence" in result
-    assert len(result["sequence"]) > 0
-    assert "loss_mask" not in result
-
-
-def test_text_empty(test_tokenizer):
-    config = make_text_config()
-    builder = SectionedMaskBuilder()
-    assert builder.build({"text": ""}, config, test_tokenizer) is None
-    assert builder.build({"text": "   "}, config, test_tokenizer) is None
-
-
-def test_text_too_short(test_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        preprocessing=ProcessingConfig(min_chars=100),
-    )
-    builder = SectionedMaskBuilder()
-    assert builder.build({"text": "short"}, config, test_tokenizer) is None
-
-
-def test_text_truncation(test_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        preprocessing=ProcessingConfig(max_seq_len=3, min_chars=1),
-    )
-    builder = SectionedMaskBuilder()
-    item = {"text": "This is a very long text that should be truncated"}
-    result = builder.build(item, config, test_tokenizer)
-    assert len(result["sequence"]) <= 3
-
-
-def test_sectioned_chat(chat_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={"system": "mask", "user": "mask", "assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-    )
-    builder = SectionedMaskBuilder()
-    item = {
-        "messages": [
-            {"role": "user", "content": "What is 2+2?"},
-            {"role": "assistant", "content": "4"},
-        ]
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert result is not None
-    assert len(result["sequence"]) == len(result["loss_mask"])
-    assert sum(result["loss_mask"]) > 0
-    assert 0 in result["loss_mask"]
-
-
-def test_sectioned_instruction(test_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
-        preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=0),
-    )
-    builder = SectionedMaskBuilder()
-    item = {"prompt": "Q: Why?", "response": "A: Because."}
-    result = builder.build(item, config, test_tokenizer)
-    assert result is not None
-    mask = result["loss_mask"]
-    assert mask[0] == 0
-    assert mask[-1] == 1
-
-
-def test_sectioned_text(test_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=1),
-    )
-    builder = SectionedMaskBuilder()
-    item = {"text": "Hello world, this is a test."}
-    result = builder.build(item, config, test_tokenizer)
-    assert result is not None
-    assert "loss_mask" not in result
-
-
-def test_sectioned_text_too_short(test_tokenizer):
-    config = PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=100),
-    )
-    builder = SectionedMaskBuilder()
-    assert builder.build({"text": "short"}, config, test_tokenizer) is None
-
-
-def test_factory_registered():
-    names = MaskBuilderFactory._registry.list_names()
-    assert "sectioned" in names
-
-
-def test_factory_create():
-    builder = MaskBuilderFactory.create("sectioned")
-    assert isinstance(builder, SectionedMaskBuilder)
-
-
-def test_dpo_chat_basic(chat_tokenizer):
-    config = make_dpo_chat_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "chosen": [
-            {"role": "user", "content": "What is 2+2?"},
-            {"role": "assistant", "content": "4"},
-        ],
-        "rejected": [
-            {"role": "user", "content": "What is 2+2?"},
-            {"role": "assistant", "content": "5"},
-        ],
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert result is not None
-    assert "chosen" in result
-    assert "rejected" in result
-    assert "chosen_mask" in result
-    assert "rejected_mask" in result
-    assert "domain" in result
-    assert len(result["chosen"]) == len(result["chosen_mask"])
-    assert len(result["rejected"]) == len(result["rejected_mask"])
-    assert sum(result["chosen_mask"]) > 0
-    assert sum(result["rejected_mask"]) > 0
-
-
-def test_dpo_chosen_only_trained(chat_tokenizer):
-    config = make_dpo_chat_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "chosen": [
-            {"role": "user", "content": "Hi"},
-            {"role": "assistant", "content": "Hello"},
-        ],
-        "rejected": [
-            {"role": "user", "content": "Hi"},
-            {"role": "assistant", "content": "Go away"},
-        ],
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert 0 in result["chosen_mask"]
-    assert 1 in result["chosen_mask"]
-    assert 0 in result["rejected_mask"]
-    assert 1 in result["rejected_mask"]
-
-
-def test_dpo_missing_field_is_none(chat_tokenizer):
-    config = make_dpo_chat_config()
-    builder = SectionedMaskBuilder()
-    assert builder.build({"chosen": [], "rejected": []}, config, chat_tokenizer) is None
-
-
-def test_grpo_basic(chat_tokenizer):
-    config = make_grpo_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "prompt": [{"role": "user", "content": "What is 2+2?"}],
-        "responses": ["4", "The answer is four", "Four", "2+2=4"],
-        "rewards": [1.0, 0.5, 0.8, 0.2],
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert result is not None
-    assert "prompts" in result
-    assert "responses" in result
-    assert "masks" in result
-    assert "rewards" in result
-    assert len(result["responses"]) == len(result["masks"])
-    assert result["rewards"] == [1.0, 0.5, 0.8, 0.2]
-
-
-def test_grpo_response_tokens_all_trained(chat_tokenizer):
-    config = make_grpo_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "prompt": [{"role": "user", "content": "Q"}],
-        "responses": ["A", "B"],
-        "rewards": [0.8, 0.2],
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    masks = result["masks"]
-    assert all(m == 1 for m in masks)
-    assert len(masks) == len(result["responses"])
-
-
-def test_grpo_single_reward(chat_tokenizer):
-    config = make_grpo_config()
-    builder = SectionedMaskBuilder()
-    item = {
-        "prompt": [{"role": "user", "content": "Q"}],
-        "responses": ["A"],
-        "rewards": 0.9,
-    }
-    result = builder.build(item, config, chat_tokenizer)
-    assert result["rewards"] == [0.9]

tests/data/test_preprocess_config.py

@@ -1,77 +0,0 @@
-import os
-
-from astrai.config.preprocess_config import (
-    InputConfig,
-    PipelineConfig,
-)
-from tests.data.conftest import (
-    _INSTRUCTION_SECTIONS,
-    _TEXT_SECTIONS,
-    make_dpo_chat_config,
-)
-
-
-def test_default_values():
-    config = PipelineConfig()
-    assert config.version == 1
-    assert config.mask == {}
-    assert config.mask_default == "mask"
-    assert config.preprocessing.max_seq_len == 2048
-    assert config.output.storage_format == "bin"
-    assert config.input.sections is None
-
-
-def test_from_dict_flat():
-    data = {
-        "version": 1,
-        "input": {
-            "sections": [{"field": "messages", "action": "$role", "template": True}]
-        },
-        "mask": {"system": "mask", "assistant": "train"},
-        "mask_default": "mask",
-        "preprocessing": {"max_seq_len": 1024},
-        "output": {"storage_format": "h5"},
-    }
-    config = PipelineConfig.from_dict(data)
-    assert config.input.sections == [
-        {"field": "messages", "action": "$role", "template": True}
-    ]
-    assert config.mask == {"system": "mask", "assistant": "train"}
-    assert config.preprocessing.max_seq_len == 1024
-    assert config.output.storage_format == "h5"
-
-
-def test_to_dict_roundtrip():
-    config = PipelineConfig(
-        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
-        mask={"prompt": "mask", "response": "train"},
-        mask_default="mask",
-    )
-    d = config.to_dict()
-    config2 = PipelineConfig.from_dict(d)
-    assert config2.input.sections == _INSTRUCTION_SECTIONS
-    assert config2.mask == {"prompt": "mask", "response": "train"}
-
-
-def test_to_json_from_json(temp_dir):
-    config = PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        mask={"text": "train"},
-        mask_default="mask",
-    )
-    path = os.path.join(temp_dir, "config.json")
-    config.to_json(path)
-    loaded = PipelineConfig.from_json(path)
-    assert loaded.input.sections == _TEXT_SECTIONS
-    assert loaded.mask == {"text": "train"}
-
-
-def test_dpo_config_roundtrip(temp_dir):
-    config = make_dpo_chat_config()
-    path = os.path.join(temp_dir, "config.json")
-    config.to_json(path)
-    loaded = PipelineConfig.from_json(path)
-    assert loaded.input.sources is not None
-    assert "chosen" in loaded.input.sources
-    assert "rejected" in loaded.input.sources
-    assert loaded.input.sections is None

tests/data/test_preprocess_pipeline.py

@@ -1,349 +0,0 @@
-import json
-import os
-
-from astrai.config.preprocess_config import (
-    InputConfig,
-    OutputConfig,
-    PipelineConfig,
-    ProcessingConfig,
-)
-from astrai.preprocessing.pipeline import Pipeline, filter_by_length
-from tests.data.conftest import (
-    _CHAT_SECTIONS,
-    _CHAT_TEMPLATE,
-    _INSTRUCTION_SECTIONS,
-    _SPECIAL_TOKENS_CONFIG,
-    _TEXT_SECTIONS,
-    make_dpo_chat_config,
-    make_grpo_no_template_config,
-)
-
-
-def test_filter_by_length():
-    assert filter_by_length("hello world", min_len=5)
-    assert not filter_by_length("hi", min_len=5)
-    assert not filter_by_length("x" * 100, max_len=50)
-    assert filter_by_length("just right", min_len=5, max_len=20)
-
-
-def test_full_chat_pipeline(temp_dir, chat_tokenizer):
-    tokenizer_dir = os.path.join(temp_dir, "tok")
-    os.makedirs(tokenizer_dir, exist_ok=True)
-    chat_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
-    with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
-        json.dump(
-            {
-                "special_tokens": _SPECIAL_TOKENS_CONFIG,
-                "chat_template": _CHAT_TEMPLATE,
-            },
-            f,
-        )
-
-    jsonl_path = os.path.join(temp_dir, "chat.jsonl")
-    with open(jsonl_path, "w", encoding="utf-8") as f:
-        f.write(
-            json.dumps(
-                {
-                    "messages": [
-                        {"role": "system", "content": "You are helpful."},
-                        {"role": "user", "content": "Hi."},
-                        {"role": "assistant", "content": "Hello!"},
-                    ]
-                }
-            )
-            + "\n"
-        )
-        f.write(
-            json.dumps(
-                {
-                    "messages": [
-                        {"role": "user", "content": "What is 2+2?"},
-                        {"role": "assistant", "content": "4"},
-                    ]
-                }
-            )
-            + "\n"
-        )
-
-    config = PipelineConfig(
-        input=InputConfig(sections=_CHAT_SECTIONS),
-        mask={"system": "mask", "user": "mask", "assistant": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-        output=OutputConfig(storage_format="bin", domain_key=None),
-    )
-
-    out_dir = os.path.join(temp_dir, "output")
-    Pipeline(
-        config=config,
-        input_paths=[jsonl_path],
-        output_dir=out_dir,
-        tokenizer_path=tokenizer_dir,
-    ).run()
-
-    meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
-    assert os.path.exists(meta_path)
-    with open(meta_path, "r") as f:
-        meta = json.load(f)
-    assert "sequence" in meta
-    assert "loss_mask" in meta
-    assert meta["sequence"]["dtype"] == "int32"
-    assert meta["loss_mask"]["dtype"] == "int32"
-
-
-def test_full_text_pipeline(temp_dir, test_tokenizer):
-    tokenizer_dir = os.path.join(temp_dir, "tok")
-    os.makedirs(tokenizer_dir, exist_ok=True)
-    test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
-    with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
-        json.dump(
-            {
-                "special_tokens": {
-                    "pad_token": "<|_pad_|>",
-                    "unk_token": "<|_unk_|>",
-                }
-            },
-            f,
-        )
-
-    jsonl_path = os.path.join(temp_dir, "text.jsonl")
-    with open(jsonl_path, "w", encoding="utf-8") as f:
-        f.write(
-            json.dumps(
-                {
-                    "text": "Hello world this is a test document with enough characters to pass the minimum length filter."
-                }
-            )
-            + "\n"
-        )
-        f.write(
-            json.dumps(
-                {
-                    "text": "Another document for testing purposes with sufficient length to be processed."
-                }
-            )
-            + "\n"
-        )
-
-    config = PipelineConfig(
-        input=InputConfig(sections=_TEXT_SECTIONS),
-        preprocessing=ProcessingConfig(max_seq_len=2048, min_chars=10),
-        output=OutputConfig(storage_format="bin"),
-    )
-
-    out_dir = os.path.join(temp_dir, "output")
-    Pipeline(
-        config=config,
-        input_paths=[jsonl_path],
-        output_dir=out_dir,
-        tokenizer_path=tokenizer_dir,
-    ).run()
-
-    meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
-    assert os.path.exists(meta_path)
-    with open(meta_path, "r") as f:
-        meta = json.load(f)
-    assert "sequence" in meta
-    assert "loss_mask" not in meta
-    assert meta["sequence"]["dtype"] == "int32"
-
-
-def test_full_instruction_pipeline(temp_dir, test_tokenizer):
-    tokenizer_dir = os.path.join(temp_dir, "tok")
-    os.makedirs(tokenizer_dir, exist_ok=True)
-    test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
-    with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
-        json.dump(
-            {
-                "special_tokens": {
-                    "pad_token": "<|_pad_|>",
-                    "unk_token": "<|_unk_|>",
-                }
-            },
-            f,
-        )
-
-    jsonl_path = os.path.join(temp_dir, "instruct.jsonl")
-    with open(jsonl_path, "w", encoding="utf-8") as f:
-        f.write(
-            json.dumps(
-                {
-                    "prompt": "Tell me a joke",
-                    "response": "Why did the chicken cross the road?",
-                }
-            )
-            + "\n"
-        )
-        f.write(
-            json.dumps(
-                {
-                    "prompt": "What is AI?",
-                    "response": "Artificial Intelligence is a field of computer science.",
-                }
-            )
-            + "\n"
-        )
-
-    config = PipelineConfig(
-        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
-        mask={"prompt": "mask", "response": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-        output=OutputConfig(storage_format="bin"),
-    )
-
-    out_dir = os.path.join(temp_dir, "output")
-    Pipeline(
-        config=config,
-        input_paths=[jsonl_path],
-        output_dir=out_dir,
-        tokenizer_path=tokenizer_dir,
-    ).run()
-
-    meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
-    assert os.path.exists(meta_path)
-    with open(meta_path, "r") as f:
-        meta = json.load(f)
-    assert "sequence" in meta
-    assert "loss_mask" in meta
-    assert meta["sequence"]["dtype"] == "int32"
-    assert meta["loss_mask"]["dtype"] == "int32"
-
-
-def test_dtype_override(temp_dir, test_tokenizer):
-    tokenizer_dir = os.path.join(temp_dir, "tok")
-    os.makedirs(tokenizer_dir, exist_ok=True)
-    test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
-    with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
-        json.dump(
-            {
-                "special_tokens": {
-                    "pad_token": "<|_pad_|>",
-                    "unk_token": "<|_unk_|>",
-                }
-            },
-            f,
-        )
-
-    jsonl_path = os.path.join(temp_dir, "data.jsonl")
-    with open(jsonl_path, "w", encoding="utf-8") as f:
-        f.write(json.dumps({"prompt": "Q", "response": "A"}) + "\n")
-
-    config = PipelineConfig(
-        input=InputConfig(sections=_INSTRUCTION_SECTIONS),
-        mask={"prompt": "mask", "response": "train"},
-        mask_default="mask",
-        preprocessing=ProcessingConfig(max_seq_len=2048),
-        output=OutputConfig(storage_format="bin", dtype={"loss_mask": "bool"}),
-    )
-
-    out_dir = os.path.join(temp_dir, "output")
-    Pipeline(
-        config=config,
-        input_paths=[jsonl_path],
-        output_dir=out_dir,
-        tokenizer_path=tokenizer_dir,
-    ).run()
-
-    meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
-    with open(meta_path, "r") as f:
-        meta = json.load(f)
-    assert meta["sequence"]["dtype"] == "int32"
-    assert meta["loss_mask"]["dtype"] == "bool"
-
-
-def test_dpo_pipeline(temp_dir, chat_tokenizer):
-    tokenizer_dir = os.path.join(temp_dir, "tok")
-    os.makedirs(tokenizer_dir, exist_ok=True)
-    chat_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
-    with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
-        json.dump(
-            {
-                "special_tokens": _SPECIAL_TOKENS_CONFIG,
-                "chat_template": _CHAT_TEMPLATE,
-            },
-            f,
-        )
-
-    jsonl_path = os.path.join(temp_dir, "dpo.jsonl")
-    with open(jsonl_path, "w", encoding="utf-8") as f:
-        f.write(
-            json.dumps(
-                {
-                    "chosen": [
-                        {"role": "user", "content": "Hi."},
-                        {"role": "assistant", "content": "Hello!"},
-                    ],
-                    "rejected": [
-                        {"role": "user", "content": "Hi."},
-                        {"role": "assistant", "content": "Go away."},
-                    ],
-                }
-            )
-            + "\n"
-        )
-
-    out_dir = os.path.join(temp_dir, "output")
-    Pipeline(
-        config=make_dpo_chat_config(),
-        input_paths=[jsonl_path],
-        output_dir=out_dir,
-        tokenizer_path=tokenizer_dir,
-    ).run()
-
-    meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
-    assert os.path.exists(meta_path)
-    with open(meta_path, "r") as f:
-        meta = json.load(f)
-    assert "chosen" in meta
-    assert "rejected" in meta
-    assert "chosen_mask" in meta
-    assert "rejected_mask" in meta
-    assert "sequence" not in meta
-
-
-def test_grpo_pipeline(temp_dir, test_tokenizer):
-    tokenizer_dir = os.path.join(temp_dir, "tok")
-    os.makedirs(tokenizer_dir, exist_ok=True)
-    test_tokenizer._tokenizer.save(os.path.join(tokenizer_dir, "tokenizer.json"))
-    with open(os.path.join(tokenizer_dir, "tokenizer_config.json"), "w") as f:
-        json.dump(
-            {
-                "special_tokens": {
-                    "pad_token": "<|_pad_|>",
-                    "unk_token": "<|_unk_|>",
-                }
-            },
-            f,
-        )
-
-    jsonl_path = os.path.join(temp_dir, "grpo.jsonl")
-    with open(jsonl_path, "w", encoding="utf-8") as f:
-        f.write(
-            json.dumps(
-                {
-                    "prompt": "Question?",
-                    "responses": ["Answer A", "Answer B"],
-                    "rewards": [0.8, 0.3],
-                }
-            )
-            + "\n"
-        )
-
-    out_dir = os.path.join(temp_dir, "output")
-    Pipeline(
-        config=make_grpo_no_template_config(),
-        input_paths=[jsonl_path],
-        output_dir=out_dir,
-        tokenizer_path=tokenizer_dir,
-    ).run()
-
-    meta_path = os.path.join(out_dir, "__default__", "shard_0000", "meta.json")
-    assert os.path.exists(meta_path)
-    with open(meta_path, "r") as f:
-        meta = json.load(f)
-    assert "prompts" in meta
-    assert "responses" in meta
-    assert "masks" in meta
-    assert "rewards" in meta
-    assert "sequence" not in meta