Compare commits
No commits in common. "8999ca89b84e2a0336b4357b7ef8bb4fe36b4cca" and "6715461a36c37356d259ccb15599af2075e617b7" have entirely different histories.
8999ca89b8
...
6715461a36
|
|
@ -5,10 +5,8 @@
|
|||
!*/
|
||||
|
||||
# Allow specific file types and root files
|
||||
!astrai/**/*.py
|
||||
!scripts/**/*.py
|
||||
!scripts/**/*.sh
|
||||
!tests/**/*.py
|
||||
!*.py
|
||||
!*.sh
|
||||
|
||||
# Allow GitHub files
|
||||
!/.github/**
|
||||
|
|
|
|||
|
|
@ -21,7 +21,6 @@ classDiagram
|
|||
|
||||
class BaseModelConfig {
|
||||
+Optional[str] model_type
|
||||
+float neftune_alpha
|
||||
+from_file(config_path) Self
|
||||
+to_file(config_path)
|
||||
}
|
||||
|
|
@ -59,12 +58,10 @@ classDiagram
|
|||
+Optional[int] dim_ffn
|
||||
+Optional[int] max_len
|
||||
+Optional[float] rope_theta
|
||||
+str attn_type
|
||||
+Optional[int] n_heads
|
||||
+Optional[int] n_kv_heads
|
||||
+Optional[bool] use_qk_norm
|
||||
+Optional[bool] use_gated_attention
|
||||
+str ffn_type
|
||||
+Optional[dict] rope_scaling
|
||||
+Optional[str] pooling_type
|
||||
+Optional[bool] normalize_embeddings
|
||||
|
|
@ -121,7 +118,7 @@ classDiagram
|
|||
+float max_grad_norm
|
||||
+list gradient_checkpointing_modules
|
||||
+int start_epoch
|
||||
+int start_samples
|
||||
+int start_batch
|
||||
+str ckpt_dir
|
||||
+int ckpt_interval
|
||||
+str log_dir
|
||||
|
|
@ -139,9 +136,7 @@ classDiagram
|
|||
+str start_method
|
||||
+str device_type
|
||||
+Optional[Dataset] val_dataset
|
||||
+Optional[float] val_split
|
||||
+int val_step
|
||||
+float neftune_alpha
|
||||
+str parallel_mode
|
||||
+dict executor_kwargs
|
||||
+dict extra_kwargs
|
||||
|
|
@ -220,13 +215,12 @@ classDiagram
|
|||
class Checkpoint {
|
||||
+dict state_dict
|
||||
+int epoch
|
||||
+int consumed_samples
|
||||
+int iteration
|
||||
+dict extra
|
||||
+dict meta
|
||||
+dict config
|
||||
+save(save_dir)
|
||||
+load(save_dir, broadcast) Checkpoint
|
||||
+load_any(save_dir, broadcast) Optional[Checkpoint]
|
||||
}
|
||||
}
|
||||
|
||||
|
|
@ -356,9 +350,7 @@ classDiagram
|
|||
|
||||
class Embedding {
|
||||
+Parameter weight
|
||||
+float neftune_noise_alpha
|
||||
+forward(x) Tensor
|
||||
+set_neftune_alpha(alpha)
|
||||
}
|
||||
}
|
||||
|
||||
|
|
@ -415,9 +407,7 @@ classDiagram
|
|||
+Dict _entries
|
||||
+register(name) decorator
|
||||
+create(name, *args, **kwargs) T
|
||||
+get_component_class(name) Type
|
||||
+list_registered() list
|
||||
+is_registered(name) bool
|
||||
}
|
||||
|
||||
class MaskBuilderFactory {
|
||||
|
|
@ -446,15 +436,13 @@ classDiagram
|
|||
+dict model_config
|
||||
+BaseExecutor executor
|
||||
+int epoch
|
||||
+int consumed_samples
|
||||
+int iteration
|
||||
+float loss
|
||||
+float grad_norm
|
||||
+DataLoader val_dataloader
|
||||
+float val_loss
|
||||
+int world_size
|
||||
+int rank
|
||||
+dict kwargs
|
||||
+optimizer_step() int
|
||||
}
|
||||
|
||||
class TrainContextBuilder {
|
||||
|
|
@ -606,6 +594,18 @@ classDiagram
|
|||
+create(name, **kwargs) TrainCallback
|
||||
}
|
||||
|
||||
class Muon {
|
||||
+float lr
|
||||
+float momentum
|
||||
+float weight_decay
|
||||
+bool nesterov
|
||||
+int ns_steps
|
||||
+Optional[float] adamw_lr
|
||||
+tuple adamw_betas
|
||||
+float adamw_eps
|
||||
+float adamw_wd
|
||||
+step(closure) Optional[float]
|
||||
}
|
||||
}
|
||||
|
||||
namespace inference {
|
||||
|
|
@ -810,9 +810,7 @@ classDiagram
|
|||
|
||||
class ChatMessage {
|
||||
+str role
|
||||
+Optional[str] content
|
||||
+Optional[List[Dict]] tool_calls
|
||||
+Optional[str] tool_call_id
|
||||
+str content
|
||||
}
|
||||
|
||||
class ChatCompletionRequest {
|
||||
|
|
@ -829,8 +827,6 @@ classDiagram
|
|||
+Optional[float] frequency_penalty
|
||||
+Optional[Dict[int, float]] logit_bias
|
||||
+Optional[str] user
|
||||
+Optional[List[ToolDef]] tools
|
||||
+Optional[Union[str, Dict]] tool_choice
|
||||
}
|
||||
|
||||
class AnthropicMessage {
|
||||
|
|
@ -854,7 +850,7 @@ classDiagram
|
|||
<<abstract>>
|
||||
+prepare(request, engine) Tuple[str, GenContext, List[str]]
|
||||
+format_stream_start(ctx) List[str]
|
||||
+format_chunk(token) List[str]
|
||||
+format_chunk(token) str
|
||||
+format_stream_end(ctx, stop) List[str]
|
||||
+format_response(ctx, content, stop) Dict
|
||||
}
|
||||
|
|
@ -862,7 +858,7 @@ classDiagram
|
|||
class OpenAIResponseBuilder {
|
||||
+prepare(request, engine) Tuple
|
||||
+format_stream_start(ctx) List[str]
|
||||
+format_chunk(token) List[str]
|
||||
+format_chunk(token) str
|
||||
+format_stream_end(ctx, stop) List[str]
|
||||
+format_response(ctx, content, stop) Dict
|
||||
}
|
||||
|
|
@ -870,7 +866,7 @@ classDiagram
|
|||
class AnthropicResponseBuilder {
|
||||
+prepare(request, engine) Tuple
|
||||
+format_stream_start(ctx) List[str]
|
||||
+format_chunk(token) List[str]
|
||||
+format_chunk(token) str
|
||||
+format_stream_end(ctx, stop) List[str]
|
||||
+format_response(ctx, content, stop) Dict
|
||||
}
|
||||
|
|
@ -1175,10 +1171,10 @@ classDiagram
|
|||
| **astrai.serialization** | Checkpoint | Model serialization |
|
||||
| **astrai.model** | AutoModel, AutoRegressiveLM, EmbeddingEncoder, DecoderBlock, GQA, MLA, MLP, DeepSeekMoE, AttnFactory, FFNFactory, RMSNorm, Linear, RotaryEmbedding, Embedding | Neural network model |
|
||||
| **astrai.tokenize** | AutoTokenizer, ChatTemplate | Tokenizer and chat template |
|
||||
| **astrai.trainer** | Trainer, TrainContext, TrainContextBuilder, BaseStrategy–GRPOStrategy, StrategyFactory, BaseScheduler–WSDScheduler, SchedulerFactory, TrainCallback(Protocol)–ValidationCallback, CallbackFactory | Training workflow |
|
||||
| **astrai.trainer** | Trainer, TrainContext, TrainContextBuilder, BaseStrategy–GRPOStrategy, StrategyFactory, BaseScheduler–WSDScheduler, SchedulerFactory, TrainCallback(Protocol)–ValidationCallback, CallbackFactory, Muon | Training workflow |
|
||||
| **astrai.inference** | InferenceEngine, InferenceScheduler, Executor, KVCache–KvcacheView, Allocator–Storage, Task, TaskManager, TaskStatus, GenerationRequest, GenerateResult, BaseSamplingStrategy–SamplingPipeline, ProtocolHandler, ResponseBuilder, OpenAIResponseBuilder, AnthropicResponseBuilder, StopChecker, GenContext, ChatMessage–MessagesRequest, app | Inference service |
|
||||
| **astrai.parallel** | spawn_parallel_fn, setup_parallel, get_rank/get_world_size/get_current_device, only_on_rank, BaseExecutor, ExecutorFactory, NoneExecutor, DDPExecutor, FSDPExecutor, GradientState, AccumOptimizer, AccumScheduler, ParallelModel, RowParallelLinear, ColumnParallelLinear | Distributed parallel & gradient accumulation |
|
||||
| **astrai.factory** | BaseFactory | Component registration |
|
||||
| **astrai.factory** | Registry, BaseFactory[T] | Component registration |
|
||||
| **astrai.protocols** | OptimizerProtocol, SchedulerProtocol | Structural subtyping for optimizer/scheduler wrappers |
|
||||
|
||||
## Design Patterns
|
||||
|
|
|
|||
|
|
@ -46,10 +46,10 @@ The output `meta.json` records the storage format, key names, dtype, total token
|
|||
|
||||
### Format Detection
|
||||
|
||||
`detect_format(load_path)` inspects the path:
|
||||
`detect_format(load_path)` inspects the directory:
|
||||
|
||||
- If `load_path` is a file: checks suffix — `.h5`/`.hdf5` → `"h5"`, unknown suffix raises `ValueError`
|
||||
- If `load_path` is a directory: recursively globs for `*.h5`/`*.hdf5` files → `"h5"`, or `*.bin` + `**/meta.json` → `"bin"`
|
||||
- If `*.h5` files exist → `"h5"` (HDF5 backend)
|
||||
- If `*.bin` + `meta.json` files exist → `"bin"` (memory-mapped backend)
|
||||
|
||||
### Store Backends
|
||||
|
||||
|
|
@ -83,7 +83,7 @@ DatasetFactory.load(train_type, load_path, window_size, stride=None, storage_typ
|
|||
→ detect_format(load_path)
|
||||
→ StoreFactory.create(storage_type)
|
||||
→ Store.load(load_path)
|
||||
→ _normalize(raw) # base Store, shared by both backends
|
||||
→ H5Store._normalize() / MmapStore._normalize()
|
||||
→ Store._data[Dict[str, List[Tensor]]] + _cum[Dict[str, List[int]]]
|
||||
→ BaseDataset.__getitem__(idx)
|
||||
→ get_index(idx) → [begin, end)
|
||||
|
|
|
|||
|
|
@ -23,7 +23,7 @@ RoPE is applied **before** KV cache write, not after — otherwise position enco
|
|||
|
||||
## KVCache System
|
||||
|
||||
Seven classes working together:
|
||||
Six classes (plus two helpers) working together:
|
||||
|
||||
```
|
||||
KVCache (facade)
|
||||
|
|
@ -152,13 +152,12 @@ Supports `stop_sequences` and streaming via `event: content_block_delta`.
|
|||
data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":...,"model":"astrai",
|
||||
"choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}
|
||||
|
||||
data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":0,"model":"astrai",
|
||||
data: {"id":"chatcmpl-...","object":"chat.completion.chunk",...,
|
||||
"choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}
|
||||
|
||||
data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":...,"model":"astrai",
|
||||
"choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}
|
||||
|
||||
data: {"prompt_tokens":5,"completion_tokens":1,"total_tokens":6}
|
||||
data: {"id":"chatcmpl-...","object":"chat.completion.chunk",...,
|
||||
"choices":[{"index":0,"delta":{},"finish_reason":"stop"}],
|
||||
"usage":{"prompt_tokens":5,"completion_tokens":1,"total_tokens":6}}
|
||||
|
||||
data: [DONE]
|
||||
```
|
||||
|
|
@ -168,7 +167,7 @@ data: [DONE]
|
|||
```
|
||||
event: message_start
|
||||
data: {"type":"message_start","message":{"id":"msg_...","model":"astrai","role":"assistant",
|
||||
"content":[],"usage":{"input_tokens":0}}}
|
||||
"content":[],"stop_reason":null,...}}
|
||||
|
||||
event: content_block_start
|
||||
data: {"type":"content_block_start","index":0,"content_block":{"type":"text","text":""}}
|
||||
|
|
@ -180,7 +179,7 @@ event: content_block_stop
|
|||
data: {"type":"content_block_stop","index":0}
|
||||
|
||||
event: message_delta
|
||||
data: {"type":"message_delta","delta":{"stop_reason":"end_turn","stop_sequence":null},"usage":{...}}
|
||||
data: {"type":"message_delta","delta":{"stop_reason":"end_turn"},"usage":{...}}
|
||||
|
||||
event: message_stop
|
||||
data: {"type":"message_stop"}
|
||||
|
|
@ -188,20 +187,26 @@ data: {"type":"message_stop"}
|
|||
|
||||
### Error Responses
|
||||
|
||||
The server returns standard HTTP status codes. Pydantic validation errors (e.g. missing required fields)
|
||||
are handled automatically by FastAPI with 422 status. The only application-level error is engine initialization:
|
||||
All endpoints use standard HTTP status codes:
|
||||
|
||||
| Status | Meaning |
|
||||
|--------|---------|
|
||||
| 200 | Success |
|
||||
| 400 | Invalid request (bad JSON, missing fields, validation error) |
|
||||
| 405 | Method not allowed |
|
||||
| 422 | Unprocessable entity (Pydantic validation) |
|
||||
| 500 | Internal server error (model crash, OOM, scheduler failure) |
|
||||
| 503 | Service unavailable (model not loaded, engine not ready) |
|
||||
|
||||
Error response body (503):
|
||||
Error response body:
|
||||
|
||||
```json
|
||||
{
|
||||
"detail": "Engine not initialized"
|
||||
"error": {
|
||||
"message": "Invalid request: max_tokens must be > 0",
|
||||
"type": "invalid_request_error",
|
||||
"code": 400
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
|
|
@ -215,13 +220,16 @@ Response:
|
|||
|
||||
```json
|
||||
{
|
||||
"total_tasks": 128,
|
||||
"total_tokens": 10240,
|
||||
"active_tasks": 3,
|
||||
"waiting_queue": 2
|
||||
"active_requests": 3,
|
||||
"waiting_requests": 2,
|
||||
"total_requests": 128,
|
||||
"cache_usage": 0.45,
|
||||
"tokens_generated": 10240
|
||||
}
|
||||
```
|
||||
|
||||
`cache_usage` is the fraction of KV cache pages currently in use (0.0–1.0).
|
||||
|
||||
## Engine API
|
||||
|
||||
```python
|
||||
|
|
|
|||
|
|
@ -53,7 +53,7 @@
|
|||
| `--ckpt_interval` | Iterations between checkpoints | 5000 |
|
||||
| `--ckpt_dir` | Checkpoint save directory | checkpoint |
|
||||
| `--start_epoch` | Resume from epoch (0 = from scratch) | 0 |
|
||||
| `--start_samples` | Resume from sample count per rank | 0 |
|
||||
| `--start_batch` | Resume from batch iteration | 0 |
|
||||
|
||||
### Validation
|
||||
|
||||
|
|
@ -67,8 +67,8 @@
|
|||
| Parameter | Description | Default |
|
||||
|-----------|-------------|---------|
|
||||
| `--log_dir` | Directory for metric logs | checkpoint/logs |
|
||||
| `--log_interval` | Number of optimizer steps between metric logs | 1 |
|
||||
| `--metrics` | Metrics to log (e.g. --metrics loss lr val_loss) | ["loss", "lr", "grad_norm"] |
|
||||
| `--log_interval` | Number of batch iterations between metric logs | 100 |
|
||||
| `--metrics` | Metrics to log (e.g. --metrics loss lr val_loss) | ["loss", "lr"] |
|
||||
|
||||
### Gradient Checkpointing
|
||||
|
||||
|
|
@ -100,17 +100,6 @@
|
|||
| `--grpo_sync_interval` | GRPO ref_model sync interval (steps) | 200 | `grpo` |
|
||||
| `--neftune_alpha` | NEFTune noise alpha (0=disabled, typical: 5.0) | 0.0 | `sft` |
|
||||
|
||||
### Scheduler
|
||||
|
||||
| Parameter | Description | Default |
|
||||
|-----------|-------------|---------|
|
||||
| `--schedule_type` | LR scheduler type (`cosine`, `sgdr`, `wsd`) | cosine |
|
||||
| `--min_rate` | Minimum LR as fraction of base LR | None (scheduler default) |
|
||||
| `--cycle_length` | SGDR first cycle length in steps | None (total_steps - warmup_steps) |
|
||||
| `--t_mult` | SGDR cycle length multiplier per restart | 2 |
|
||||
| `--stable_steps` | WSD stable plateau steps | None (required for wsd) |
|
||||
| `--decay_steps` | WSD decay steps | None (total_steps - warmup_steps - stable_steps) |
|
||||
|
||||
### Usage Example
|
||||
|
||||
```bash
|
||||
|
|
@ -189,7 +178,7 @@ python scripts/tools/generate.py \
|
|||
| `input_files` | path(s) | required | Input JSONL file(s), supports glob (`data/*.jsonl`) |
|
||||
| `--output_dir`, `-o` | path | required | Output directory for processed data |
|
||||
| `--config`, `-c` | path | required | Preprocessing pipeline config (JSON) |
|
||||
| `--tokenizer_path` | str | `params` | Path to tokenizer directory |
|
||||
| `--num_workers` | int | `4` | Number of parallel workers |
|
||||
|
||||
Usage:
|
||||
```bash
|
||||
|
|
|
|||
|
|
@ -26,9 +26,8 @@ A single config file captures the entire pipeline, reusable and version-controll
|
|||
|
||||
```json
|
||||
{
|
||||
"version": 1,
|
||||
"input": {}, // sections (single) or sources (multi)
|
||||
"mask": {}, // role -> "train" | "mask"
|
||||
"mask": {}, // role → "train" | "mask"
|
||||
"mask_default": "mask",
|
||||
"preprocessing": {},
|
||||
"output": {}
|
||||
|
|
@ -221,12 +220,11 @@ Config:
|
|||
}
|
||||
```
|
||||
|
||||
Output keys: `prompts`, `prompts_mask`, `responses`, `masks`, `rewards` (float32)
|
||||
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`)
|
||||
- `prompts_mask` is auto-generated (all masked) and unused by GRPOStrategy
|
||||
|
||||
---
|
||||
|
||||
|
|
@ -276,11 +274,12 @@ When `sources` is set, `sections` is ignored.
|
|||
|
||||
### Template mode (`template: true`)
|
||||
|
||||
For each message in the field's array:
|
||||
|
||||
1. Prepend BOS token (masked)
|
||||
2. For each message in the field's array:
|
||||
1. Render through `chat_template` for that single message
|
||||
2. Encode rendered text
|
||||
3. Apply mask rule for the message's role
|
||||
2. Render through `chat_template` for that single message
|
||||
3. Encode rendered text
|
||||
4. Apply mask rule for the message's role
|
||||
|
||||
### Non-template mode
|
||||
|
||||
|
|
@ -288,7 +287,7 @@ Encode the field value as text. Mask value is 1 (train) or 0 (mask) per the sect
|
|||
|
||||
### Text config detection
|
||||
|
||||
When no section uses `template` and all sections have `action: "train"`, the builder omits `loss_mask` from the output — all tokens are trained.
|
||||
When no section uses `template` and all sections have `action: "train"`, the builder skips mask generation entirely — all tokens are trained.
|
||||
|
||||
---
|
||||
|
||||
|
|
@ -299,15 +298,13 @@ When no section uses `template` and all sections have `action: "train"`, the bui
|
|||
```
|
||||
output/
|
||||
__default__/
|
||||
shard_0000/
|
||||
meta.json
|
||||
sequence.bin
|
||||
loss_mask.bin
|
||||
meta.json
|
||||
sequence.bin
|
||||
loss_mask.bin
|
||||
wiki/
|
||||
shard_0000/
|
||||
meta.json
|
||||
sequence.bin
|
||||
loss_mask.bin
|
||||
meta.json
|
||||
sequence.bin
|
||||
loss_mask.bin
|
||||
```
|
||||
|
||||
### Multi-Shard (`bin`)
|
||||
|
|
@ -327,7 +324,7 @@ output/
|
|||
loss_mask.bin
|
||||
```
|
||||
|
||||
For `bin` format, `MmapStore` discovers all shards under the domain directory via `rglob("meta.json")`. For `h5` format, `H5Store` discovers `.h5`/`.hdf5` files via recursive glob.
|
||||
`MmapStore` discovers all shards under the domain directory via `rglob("meta.json")`.
|
||||
|
||||
---
|
||||
|
||||
|
|
@ -352,7 +349,7 @@ python scripts/tools/preprocess.py data/grpo/*.jsonl -o output/grpo/ -c configs/
|
|||
from astrai.preprocessing.pipeline import Pipeline
|
||||
from astrai.config.preprocess_config import PipelineConfig
|
||||
|
||||
config = PipelineConfig.from_file("sft.json")
|
||||
config = PipelineConfig.from_json("sft.json")
|
||||
Pipeline(
|
||||
config,
|
||||
["data_part1.jsonl", "data_part2.jsonl"],
|
||||
|
|
|
|||
|
|
@ -58,9 +58,7 @@ on_train_begin
|
|||
context.loss = loss.item()
|
||||
stand_loss = loss / executor.grad_accum_steps
|
||||
executor.backward(stand_loss)
|
||||
context.consumed_samples += (
|
||||
context.config.batch_per_device * context.world_size
|
||||
)
|
||||
context.iteration += 1
|
||||
on_batch_end
|
||||
|
||||
if executor.sync_gradients:
|
||||
|
|
@ -80,13 +78,13 @@ on_train_end
|
|||
| `on_train_begin` | Before training starts | `GradientCheckpointingCallback` |
|
||||
| `on_epoch_begin` | Start of each epoch | `ProgressBarCallback` |
|
||||
| `on_batch_begin` | Every batch | — |
|
||||
| `on_optimizer_step` | Every accumulation window | `GradientClippingCallback`, `MetricLoggerCallback`, `ValidationCallback` |
|
||||
| `on_optimizer_step` | Every accumulation window | `GradientClippingCallback`, `ValidationCallback` |
|
||||
| `on_batch_end` | Every batch | `CheckpointCallback`, `MetricLoggerCallback`, `ProgressBarCallback` |
|
||||
| `on_epoch_end` | End of each epoch | `ProgressBarCallback` |
|
||||
| `on_error` | On exception during training | `CheckpointCallback`, `MetricLoggerCallback` |
|
||||
| `on_train_end` | Training ends (always via finally) | `CheckpointCallback`, `MetricLoggerCallback`, `GradientCheckpointingCallback` |
|
||||
|
||||
Default callbacks (in order): `gradient_checkpointing` (activation checkpointing, optional), `checkpoint` (safetensors, rank-0), `validation` (periodic validation on val_dataset), `metric_logger` (JSONL, rank-0), `progress_bar` (tqdm), `gradient_clipping`.
|
||||
Default callbacks (in order): `gradient_checkpointing` (activation checkpointing, optional), `checkpoint` (safetensors, rank-0), `metric_logger` (JSONL, rank-0), `progress_bar` (tqdm), `gradient_clipping`, `validation` (periodic validation on val_dataset).
|
||||
|
||||
## Strategies
|
||||
|
||||
|
|
@ -160,8 +158,8 @@ Callback wraps each `DecoderBlock.forward` with `torch.utils.checkpoint.checkpoi
|
|||
## Checkpoint
|
||||
|
||||
```
|
||||
Checkpoint(state_dict, epoch, consumed_samples, extra, meta, config)
|
||||
├── save(save_dir) rank-0 only: meta.json (epoch/consumed_samples/timestamp) + config.json (model config) + model.safetensors + optional {key}.pt (optimizer.pt, scheduler.pt)
|
||||
Checkpoint(state_dict, epoch, iteration, extra, meta, config)
|
||||
├── save(save_dir) rank-0 only: meta.json (epoch/iteration/timestamp) + config.json (model config) + model.safetensors + optional {key}.pt (optimizer.pt, scheduler.pt)
|
||||
└── load(save_dir, broadcast=False) loads from local disk; set broadcast=True to broadcast metadata from rank-0
|
||||
```
|
||||
|
||||
|
|
|
|||
|
|
@ -47,18 +47,14 @@ class TrainConfig(BaseConfig):
|
|||
|
||||
# checkpoint setting
|
||||
start_epoch: int = field(default=0, metadata={"help": "Start epoch for training."})
|
||||
start_samples: int = field(
|
||||
default=0,
|
||||
metadata={
|
||||
"help": "Start samples count (per rank). Superseded by checkpoint consumed_samples."
|
||||
},
|
||||
start_batch: int = field(
|
||||
default=0, metadata={"help": "Start batch iteration for training."}
|
||||
)
|
||||
ckpt_dir: str = field(
|
||||
default="./checkpoint", metadata={"help": "Checkpoint directory."}
|
||||
)
|
||||
ckpt_interval: int = field(
|
||||
default=5000,
|
||||
metadata={"help": "Number of optimizer steps between checkpoints."},
|
||||
default=5000, metadata={"help": "Number of iterations between checkpoints."}
|
||||
)
|
||||
|
||||
# lora setting
|
||||
|
|
@ -71,8 +67,12 @@ class TrainConfig(BaseConfig):
|
|||
log_dir: str = field(
|
||||
default="./checkpoint/logs", metadata={"help": "Directory for metric logs."}
|
||||
)
|
||||
log_interval: int = field(
|
||||
default=100,
|
||||
metadata={"help": "Number of batch iterations between metric logs."},
|
||||
)
|
||||
metrics: List[str] = field(
|
||||
default_factory=lambda: ["loss", "lr", "grad_norm"],
|
||||
default_factory=lambda: ["loss", "lr"],
|
||||
metadata={"help": "Metrics to record during training."},
|
||||
)
|
||||
|
||||
|
|
|
|||
|
|
@ -5,13 +5,10 @@ from astrai.dataset.dataset import (
|
|||
from astrai.dataset.sampler import ResumableDistributedSampler
|
||||
from astrai.dataset.storage import (
|
||||
H5Store,
|
||||
JsonlStore,
|
||||
MmapStore,
|
||||
Store,
|
||||
StoreFactory,
|
||||
detect_format,
|
||||
)
|
||||
from astrai.serialization import (
|
||||
load_bin,
|
||||
load_h5,
|
||||
save_bin,
|
||||
|
|
@ -25,7 +22,6 @@ __all__ = [
|
|||
"StoreFactory",
|
||||
"H5Store",
|
||||
"MmapStore",
|
||||
"JsonlStore",
|
||||
"detect_format",
|
||||
"save_h5",
|
||||
"load_h5",
|
||||
|
|
|
|||
|
|
@ -48,26 +48,24 @@ class BaseDataset(Dataset, ABC):
|
|||
f"Missing: {missing}"
|
||||
)
|
||||
|
||||
def load(self, load_path: str, storage_type: Optional[str] = None, **kwargs):
|
||||
def load(self, load_path: str, storage_type: Optional[str] = None):
|
||||
"""Load dataset from the given path.
|
||||
|
||||
Auto-detects the storage format if not specified.
|
||||
|
||||
Args:
|
||||
load_path: Path to the data directory or file
|
||||
storage_type: Force a specific storage type ("h5", "bin", "jsonl"),
|
||||
storage_type: Force a specific storage type ("h5", "bin"),
|
||||
or None for auto-detection
|
||||
**kwargs: Extra arguments forwarded to the store constructor and
|
||||
to ``store.load()``.
|
||||
|
||||
Raises:
|
||||
KeyError: If the loaded storage is missing required keys.
|
||||
"""
|
||||
if storage_type is None:
|
||||
storage_type = detect_format(load_path)
|
||||
self.storage = StoreFactory.create(storage_type, **kwargs)
|
||||
self.storage = StoreFactory.create(storage_type)
|
||||
self._load_path = load_path
|
||||
self.storage.load(load_path, **kwargs)
|
||||
self.storage.load(load_path)
|
||||
self._validate_keys()
|
||||
|
||||
@property
|
||||
|
|
@ -146,7 +144,6 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
|
|||
window_size: int,
|
||||
stride: Optional[int] = None,
|
||||
storage_type: Optional[str] = None,
|
||||
**kwargs,
|
||||
) -> "BaseDataset":
|
||||
"""Create and load a dataset in one step.
|
||||
|
||||
|
|
@ -155,8 +152,7 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
|
|||
load_path: Path to the data file
|
||||
window_size: Window size for data sampling
|
||||
stride: Stride between consecutive samples (default: same as window_size)
|
||||
storage_type: Storage type ("h5", "bin", "jsonl") or None for auto-detection
|
||||
**kwargs: Extra arguments forwarded to ``dataset.load()``.
|
||||
storage_type: Storage type ("h5", "bin") or None for auto-detection
|
||||
|
||||
Returns:
|
||||
Loaded dataset instance
|
||||
|
|
@ -165,7 +161,7 @@ class DatasetFactory(BaseFactory["BaseDataset"]):
|
|||
stride = window_size
|
||||
|
||||
dataset = cls.create(train_type, window_size, stride)
|
||||
dataset.load(load_path, storage_type=storage_type, **kwargs)
|
||||
dataset.load(load_path, storage_type=storage_type)
|
||||
|
||||
return dataset
|
||||
|
||||
|
|
|
|||
|
|
@ -74,7 +74,6 @@ class ResumableDistributedSampler(Sampler[int]):
|
|||
|
||||
self.epoch += 1
|
||||
self._indices = None
|
||||
self.iter = self.iter % self.num_samples_per_replica
|
||||
|
||||
@property
|
||||
def _remaining(self):
|
||||
|
|
|
|||
|
|
@ -14,31 +14,85 @@ Key properties:
|
|||
- Explicit length: _length = min(total elements across keys), set at load,
|
||||
__len__ returns O(1)
|
||||
- Zero-copy mmap: MmapStore wraps np.memmap(mode="r"), all DataLoader
|
||||
workers share OS page-cache pages
|
||||
workers share OS page-cache pages
|
||||
"""
|
||||
|
||||
import bisect
|
||||
import glob
|
||||
import json
|
||||
import logging
|
||||
import os
|
||||
from abc import ABC, abstractmethod
|
||||
from pathlib import Path
|
||||
from typing import Dict, List, Union
|
||||
|
||||
import h5py
|
||||
import numpy as np
|
||||
import torch
|
||||
from torch import Tensor
|
||||
|
||||
from astrai.config.preprocess_config import PipelineConfig
|
||||
from astrai.factory import BaseFactory
|
||||
from astrai.preprocessing.builder import MaskBuilderFactory
|
||||
from astrai.preprocessing.position_id import PositionIdStrategyFactory
|
||||
from astrai.serialization import (
|
||||
load_bin,
|
||||
load_h5,
|
||||
)
|
||||
from astrai.tokenize import AutoTokenizer
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
def save_h5(file_path: str, file_name: str, tensor_group: Dict[str, List[Tensor]]):
|
||||
os.makedirs(file_path, exist_ok=True)
|
||||
full_file_path = os.path.join(file_path, f"{file_name}.h5")
|
||||
with h5py.File(full_file_path, "w") as f:
|
||||
for key, tensors in tensor_group.items():
|
||||
grp = f.create_group(key)
|
||||
for idx, tensor in enumerate(tensors):
|
||||
arr = tensor.cpu().numpy()
|
||||
grp.create_dataset(f"data_{idx}", data=arr)
|
||||
|
||||
|
||||
def load_h5(file_path: str, share_memory=True) -> Dict[str, List[Tensor]]:
|
||||
tensor_group: Dict[str, List[Tensor]] = {}
|
||||
|
||||
root_path = Path(file_path)
|
||||
h5_files = list(root_path.rglob("*.h5")) + list(root_path.rglob("*.hdf5"))
|
||||
|
||||
for h5_file in h5_files:
|
||||
with h5py.File(h5_file, "r") as f:
|
||||
for key in f.keys():
|
||||
grp = f[key]
|
||||
dsets = []
|
||||
for dset_name in grp.keys():
|
||||
dset = grp[dset_name]
|
||||
tensor = torch.from_numpy(dset[:])
|
||||
if share_memory:
|
||||
tensor = tensor.share_memory_()
|
||||
dsets.append(tensor)
|
||||
|
||||
if tensor_group.get(key) is None:
|
||||
tensor_group[key] = []
|
||||
tensor_group[key].extend(dsets)
|
||||
|
||||
return tensor_group
|
||||
|
||||
|
||||
def save_bin(file_path: str, tensor_group: Dict[str, List[Tensor]]):
|
||||
os.makedirs(file_path, exist_ok=True)
|
||||
meta = {}
|
||||
for key, tensors in tensor_group.items():
|
||||
cat = torch.cat(tensors, dim=0)
|
||||
meta[key] = {"shape": list(cat.shape), "dtype": str(cat.dtype).split(".")[-1]}
|
||||
np.asarray(cat.cpu().numpy()).tofile(os.path.join(file_path, f"{key}.bin"))
|
||||
with open(os.path.join(file_path, "meta.json"), "w") as f:
|
||||
json.dump(meta, f)
|
||||
|
||||
|
||||
def load_bin(file_path: str) -> Dict[str, List[Tensor]]:
|
||||
with open(os.path.join(file_path, "meta.json"), "r") as f:
|
||||
meta = json.load(f)
|
||||
segments: Dict[str, List[Tensor]] = {}
|
||||
for key, info in meta.items():
|
||||
arr = np.memmap(
|
||||
os.path.join(file_path, f"{key}.bin"),
|
||||
dtype=info["dtype"],
|
||||
mode="r+",
|
||||
shape=tuple(info["shape"]),
|
||||
)
|
||||
segments[key] = [torch.from_numpy(arr)]
|
||||
return segments
|
||||
|
||||
|
||||
def detect_format(load_path: str) -> str:
|
||||
|
|
@ -48,7 +102,7 @@ def detect_format(load_path: str) -> str:
|
|||
load_path: Directory or file path
|
||||
|
||||
Returns:
|
||||
Format string ("h5", "bin", or "jsonl")
|
||||
Format string ("h5" or "bin")
|
||||
|
||||
Raises:
|
||||
FileNotFoundError: If no supported data files are found
|
||||
|
|
@ -58,8 +112,6 @@ def detect_format(load_path: str) -> str:
|
|||
suffix = root.suffix.lower()
|
||||
if suffix in (".h5", ".hdf5"):
|
||||
return "h5"
|
||||
if suffix == ".jsonl":
|
||||
return "jsonl"
|
||||
raise ValueError(f"Unsupported file format: {suffix}")
|
||||
|
||||
h5_files = [
|
||||
|
|
@ -76,11 +128,6 @@ def detect_format(load_path: str) -> str:
|
|||
) > 0
|
||||
if has_meta:
|
||||
return "bin"
|
||||
jsonl_files = [
|
||||
Path(p) for p in glob.glob(str(root / "**" / "*.jsonl"), recursive=True)
|
||||
]
|
||||
if jsonl_files:
|
||||
return "jsonl"
|
||||
raise FileNotFoundError(f"No supported data files found at {load_path}")
|
||||
|
||||
|
||||
|
|
@ -217,96 +264,3 @@ class MmapStore(Store):
|
|||
self._normalize(all_raw)
|
||||
for tensors in self._data.values():
|
||||
self._mmap_refs.extend(tensors)
|
||||
|
||||
|
||||
@StoreFactory.register("jsonl")
|
||||
class JsonlStore(Store):
|
||||
"""On-the-fly tokenization store for raw JSONL files.
|
||||
|
||||
A JSONL dataset directory contains ``*.jsonl`` files plus a
|
||||
``dataset_config.json`` file that follows the same schema as
|
||||
:class:`PipelineConfig` with an additional ``tokenizer_path`` field.
|
||||
Records are tokenized when the store is loaded and concatenated into
|
||||
segmented tensors matching the key layout expected by the dataset
|
||||
classes (``sequence``, ``loss_mask``, ``position_ids``, ...).
|
||||
"""
|
||||
|
||||
CONFIG_NAME = "dataset_config.json"
|
||||
|
||||
def load(self, path: str):
|
||||
root = Path(path)
|
||||
config_path = root / self.CONFIG_NAME
|
||||
if not config_path.exists():
|
||||
raise FileNotFoundError(
|
||||
f"JSONL dataset config not found: {config_path}. "
|
||||
f"Expected {self.CONFIG_NAME} alongside *.jsonl files."
|
||||
)
|
||||
|
||||
with open(config_path, "r", encoding="utf-8") as f:
|
||||
raw_config = json.load(f)
|
||||
|
||||
tokenizer_path = raw_config.pop("tokenizer_path", None)
|
||||
if tokenizer_path is None:
|
||||
raise ValueError(
|
||||
f"JSONL dataset config must specify 'tokenizer_path': {config_path}"
|
||||
)
|
||||
|
||||
self.config = PipelineConfig.from_dict(raw_config)
|
||||
tokenizer = AutoTokenizer.from_pretrained(tokenizer_path)
|
||||
mask_builder = MaskBuilderFactory.create("sectioned")
|
||||
position_strategy = PositionIdStrategyFactory.create(
|
||||
self.config.output.position_ids_mode
|
||||
)
|
||||
|
||||
raw: Dict[str, List[Tensor]] = {}
|
||||
doc_sequences: List[List[int]] = []
|
||||
|
||||
for jsonl_path in sorted(root.glob("*.jsonl")):
|
||||
with open(jsonl_path, "r", encoding="utf-8") as f:
|
||||
for line in f:
|
||||
line = line.strip()
|
||||
if not line:
|
||||
continue
|
||||
try:
|
||||
item = json.loads(line)
|
||||
except json.JSONDecodeError:
|
||||
logger.warning(
|
||||
"Failed to parse JSON line in %s, skipping", jsonl_path
|
||||
)
|
||||
continue
|
||||
|
||||
result = mask_builder.build(item, self.config, tokenizer)
|
||||
if result is None:
|
||||
continue
|
||||
|
||||
result.pop("domain", None)
|
||||
primary_ids = self._primary_ids(result)
|
||||
if not primary_ids:
|
||||
continue
|
||||
|
||||
doc_sequences.append(primary_ids)
|
||||
for key, ids in result.items():
|
||||
if key not in raw:
|
||||
raw[key] = []
|
||||
raw[key].append(torch.tensor(ids, dtype=self._infer_dtype(ids)))
|
||||
|
||||
pos_ids = position_strategy.generate(doc_sequences)
|
||||
if pos_ids:
|
||||
raw["position_ids"] = [torch.tensor(pos_ids, dtype=torch.int32)]
|
||||
|
||||
self._normalize(raw)
|
||||
|
||||
@staticmethod
|
||||
def _primary_ids(result: dict) -> List[int]:
|
||||
"""Return the first integer list in *result* as the primary id sequence."""
|
||||
for val in result.values():
|
||||
if isinstance(val, list) and val and isinstance(val[0], int):
|
||||
return val
|
||||
return []
|
||||
|
||||
@staticmethod
|
||||
def _infer_dtype(ids: List) -> torch.dtype:
|
||||
"""Infer tensor dtype from the first element of a token/value list."""
|
||||
if ids and isinstance(ids[0], float):
|
||||
return torch.float32
|
||||
return torch.int32
|
||||
|
|
|
|||
|
|
@ -37,7 +37,7 @@ def _resolve_type(
|
|||
ns = vars(mod)
|
||||
|
||||
if isinstance(arg, ForwardRef):
|
||||
return arg._evaluate(ns, None, recursive_guard=frozenset())
|
||||
return arg._evaluate(ns, None, frozenset(), recursive_guard=frozenset())
|
||||
|
||||
return ns.get(name)
|
||||
|
||||
|
|
|
|||
|
|
@ -132,12 +132,6 @@ class BaseExecutor:
|
|||
def grad_accum_steps(self) -> int:
|
||||
return self.gradient_state.num_steps
|
||||
|
||||
def clip_grad_norm(self, model: nn.Module, max_norm: float) -> float:
|
||||
total_norm = torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm)
|
||||
if isinstance(total_norm, torch.Tensor):
|
||||
return total_norm.item()
|
||||
return total_norm
|
||||
|
||||
|
||||
class ExecutorFactory(BaseFactory[BaseExecutor]):
|
||||
pass
|
||||
|
|
@ -266,14 +260,6 @@ class FSDPExecutor(BaseExecutor):
|
|||
return model.no_sync()
|
||||
return contextlib.nullcontext()
|
||||
|
||||
def clip_grad_norm(self, model: nn.Module, max_norm: float) -> float:
|
||||
if isinstance(model, FSDP) and self.use_distributed:
|
||||
total_norm = model.clip_grad_norm_(max_norm)
|
||||
if isinstance(total_norm, torch.Tensor):
|
||||
return total_norm.item()
|
||||
return total_norm
|
||||
return super().clip_grad_norm(model, max_norm)
|
||||
|
||||
def unwrap_model(self, model: nn.Module):
|
||||
if isinstance(model, FSDP) and self.use_distributed:
|
||||
with FSDP.state_dict_type(
|
||||
|
|
|
|||
|
|
@ -14,8 +14,8 @@ from typing import Dict, List
|
|||
|
||||
import torch
|
||||
|
||||
from astrai.dataset.storage import save_bin, save_h5
|
||||
from astrai.factory import BaseFactory
|
||||
from astrai.serialization import save_bin, save_h5
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
|
|
|||
|
|
@ -1,8 +1,5 @@
|
|||
"""Model checkpoint serialization helpers."""
|
||||
|
||||
import io
|
||||
import json
|
||||
import os
|
||||
import time
|
||||
from dataclasses import dataclass, field
|
||||
from pathlib import Path
|
||||
|
|
@ -139,7 +136,7 @@ def load_state_dict(path: Union[str, Path], broadcast: bool = False) -> dict:
|
|||
class Checkpoint:
|
||||
state_dict: Dict[str, Any] = field(default_factory=dict)
|
||||
epoch: int = 0
|
||||
consumed_samples: int = 0
|
||||
iteration: int = 0
|
||||
extra: Dict[str, Any] = field(default_factory=dict)
|
||||
meta: Dict[str, Any] = field(default_factory=dict)
|
||||
config: Dict[str, Any] = field(default_factory=dict)
|
||||
|
|
@ -153,7 +150,7 @@ class Checkpoint:
|
|||
|
||||
meta = {
|
||||
"epoch": self.epoch,
|
||||
"consumed_samples": self.consumed_samples,
|
||||
"iteration": self.iteration,
|
||||
"timestamp": time.strftime("%Y-%m-%dT%H:%M:%S"),
|
||||
**self.meta,
|
||||
}
|
||||
|
|
@ -179,7 +176,7 @@ class Checkpoint:
|
|||
return cls(
|
||||
state_dict=state_dict,
|
||||
epoch=meta.get("epoch", 0),
|
||||
consumed_samples=meta.get("consumed_samples", 0),
|
||||
iteration=meta.get("iteration", 0),
|
||||
extra=extra,
|
||||
config=config,
|
||||
)
|
||||
|
|
@ -1,43 +0,0 @@
|
|||
"""Serialization utilities for models and datasets.
|
||||
|
||||
This package re-exports checkpoint helpers and dataset storage helpers so
|
||||
that existing imports from ``astrai.serialization`` continue to work.
|
||||
"""
|
||||
|
||||
from astrai.serialization.checkpoint import (
|
||||
Checkpoint,
|
||||
load_json,
|
||||
load_model_config,
|
||||
load_model_weights,
|
||||
load_safetensors,
|
||||
load_state_dict,
|
||||
load_torch,
|
||||
save_json,
|
||||
save_model,
|
||||
save_safetensors,
|
||||
save_torch,
|
||||
)
|
||||
from astrai.serialization.dataset import (
|
||||
load_bin,
|
||||
load_h5,
|
||||
save_bin,
|
||||
save_h5,
|
||||
)
|
||||
|
||||
__all__ = [
|
||||
"Checkpoint",
|
||||
"load_json",
|
||||
"load_model_config",
|
||||
"load_model_weights",
|
||||
"load_safetensors",
|
||||
"load_state_dict",
|
||||
"load_torch",
|
||||
"save_json",
|
||||
"save_model",
|
||||
"save_safetensors",
|
||||
"save_torch",
|
||||
"load_bin",
|
||||
"load_h5",
|
||||
"save_bin",
|
||||
"save_h5",
|
||||
]
|
||||
|
|
@ -1,73 +0,0 @@
|
|||
"""Dataset storage serialization helpers (HDF5 / memory-mapped binary)."""
|
||||
|
||||
import json
|
||||
import os
|
||||
from pathlib import Path
|
||||
from typing import Dict, List
|
||||
|
||||
import h5py
|
||||
import numpy as np
|
||||
import torch
|
||||
from torch import Tensor
|
||||
|
||||
|
||||
def save_h5(file_path: str, file_name: str, tensor_group: Dict[str, List[Tensor]]):
|
||||
os.makedirs(file_path, exist_ok=True)
|
||||
full_file_path = os.path.join(file_path, f"{file_name}.h5")
|
||||
with h5py.File(full_file_path, "w") as f:
|
||||
for key, tensors in tensor_group.items():
|
||||
grp = f.create_group(key)
|
||||
for idx, tensor in enumerate(tensors):
|
||||
arr = tensor.cpu().numpy()
|
||||
grp.create_dataset(f"data_{idx}", data=arr)
|
||||
|
||||
|
||||
def load_h5(file_path: str, share_memory=True) -> Dict[str, List[Tensor]]:
|
||||
tensor_group: Dict[str, List[Tensor]] = {}
|
||||
|
||||
root_path = Path(file_path)
|
||||
h5_files = list(root_path.rglob("*.h5")) + list(root_path.rglob("*.hdf5"))
|
||||
|
||||
for h5_file in h5_files:
|
||||
with h5py.File(h5_file, "r") as f:
|
||||
for key in f.keys():
|
||||
grp = f[key]
|
||||
dsets = []
|
||||
for dset_name in grp.keys():
|
||||
dset = grp[dset_name]
|
||||
tensor = torch.from_numpy(dset[:])
|
||||
if share_memory:
|
||||
tensor = tensor.share_memory_()
|
||||
dsets.append(tensor)
|
||||
|
||||
if tensor_group.get(key) is None:
|
||||
tensor_group[key] = []
|
||||
tensor_group[key].extend(dsets)
|
||||
|
||||
return tensor_group
|
||||
|
||||
|
||||
def save_bin(file_path: str, tensor_group: Dict[str, List[Tensor]]):
|
||||
os.makedirs(file_path, exist_ok=True)
|
||||
meta = {}
|
||||
for key, tensors in tensor_group.items():
|
||||
cat = torch.cat(tensors, dim=0)
|
||||
meta[key] = {"shape": list(cat.shape), "dtype": str(cat.dtype).split(".")[-1]}
|
||||
np.asarray(cat.cpu().numpy()).tofile(os.path.join(file_path, f"{key}.bin"))
|
||||
with open(os.path.join(file_path, "meta.json"), "w") as f:
|
||||
json.dump(meta, f)
|
||||
|
||||
|
||||
def load_bin(file_path: str) -> Dict[str, List[Tensor]]:
|
||||
with open(os.path.join(file_path, "meta.json"), "r") as f:
|
||||
meta = json.load(f)
|
||||
segments: Dict[str, List[Tensor]] = {}
|
||||
for key, info in meta.items():
|
||||
arr = np.memmap(
|
||||
os.path.join(file_path, f"{key}.bin"),
|
||||
dtype=info["dtype"],
|
||||
mode="r+",
|
||||
shape=tuple(info["shape"]),
|
||||
)
|
||||
segments[key] = [torch.from_numpy(arr)]
|
||||
return segments
|
||||
|
|
@ -1,25 +1,42 @@
|
|||
from typing import Dict
|
||||
from typing import Any, Callable, Dict
|
||||
|
||||
import torch
|
||||
import torch.nn as nn
|
||||
|
||||
|
||||
def grad_norm(model: nn.Module, per_param: bool = False) -> float | Dict[str, float]:
|
||||
grads = [p.grad.detach() for p in model.parameters() if p.grad is not None]
|
||||
if not grads:
|
||||
return 0.0
|
||||
def _grad_stat(
|
||||
model: nn.Module, fn: Callable[[torch.Tensor], Any], default: Any
|
||||
) -> dict:
|
||||
results = {}
|
||||
for name, param in model.named_parameters():
|
||||
results[name] = default
|
||||
if param.grad is not None:
|
||||
results[name] = fn(param.grad.data)
|
||||
return results
|
||||
|
||||
total_sq = torch.stack([g.pow(2).sum() for g in grads]).sum()
|
||||
if per_param:
|
||||
norms = {}
|
||||
for name, param in model.named_parameters():
|
||||
if param.grad is not None:
|
||||
norms[name] = param.grad.norm(2).item()
|
||||
else:
|
||||
norms[name] = 0.0
|
||||
norms["total"] = total_sq.sqrt().item()
|
||||
return norms
|
||||
return total_sq.sqrt().item()
|
||||
|
||||
def grad_norm(model: nn.Module, norm_type: int = 2) -> Dict[str, float]:
|
||||
return _grad_stat(model, lambda g: g.norm(norm_type).item(), 0.0)
|
||||
|
||||
|
||||
def grad_std(model: nn.Module) -> Dict[str, float]:
|
||||
return _grad_stat(model, lambda g: g.std().item(), 0.0)
|
||||
|
||||
|
||||
def grad_max(model: nn.Module) -> Dict[str, float]:
|
||||
return _grad_stat(model, lambda g: g.max().item(), -float("inf"))
|
||||
|
||||
|
||||
def grad_min(model: nn.Module) -> Dict[str, float]:
|
||||
return _grad_stat(model, lambda g: g.min().item(), float("inf"))
|
||||
|
||||
|
||||
def grad_mean(model: nn.Module) -> Dict[str, float]:
|
||||
return _grad_stat(model, lambda g: g.mean().item(), 0.0)
|
||||
|
||||
|
||||
def grad_nan_num(model: nn.Module) -> Dict[str, int]:
|
||||
return _grad_stat(model, lambda g: g.isnan().sum().item(), 0)
|
||||
|
||||
|
||||
def ctx_get_loss(ctx):
|
||||
|
|
@ -35,4 +52,24 @@ def ctx_get_val_loss(ctx):
|
|||
|
||||
|
||||
def ctx_get_grad_norm(ctx):
|
||||
return ctx.grad_norm
|
||||
return grad_norm(ctx.model)
|
||||
|
||||
|
||||
def ctx_get_grad_std(ctx):
|
||||
return grad_std(ctx.model)
|
||||
|
||||
|
||||
def ctx_get_grad_max(ctx):
|
||||
return grad_max(ctx.model)
|
||||
|
||||
|
||||
def ctx_get_grad_min(ctx):
|
||||
return grad_min(ctx.model)
|
||||
|
||||
|
||||
def ctx_get_grad_mean(ctx):
|
||||
return grad_mean(ctx.model)
|
||||
|
||||
|
||||
def ctx_get_grad_nan_num(ctx):
|
||||
return grad_nan_num(ctx.model)
|
||||
|
|
|
|||
|
|
@ -53,7 +53,7 @@ class CosineScheduler(BaseScheduler):
|
|||
optimizer,
|
||||
warmup_steps: int,
|
||||
lr_decay_steps: int,
|
||||
min_rate: float = 0.01,
|
||||
min_rate: float = 0.05,
|
||||
last_epoch: int = -1,
|
||||
):
|
||||
self.warmup_steps = warmup_steps
|
||||
|
|
@ -65,15 +65,11 @@ class CosineScheduler(BaseScheduler):
|
|||
def get_lr(self) -> List[float]:
|
||||
# warmup
|
||||
if self.last_epoch < self.warmup_steps:
|
||||
warmup_factor = max(
|
||||
self.min_rate, self.last_epoch / max(self.warmup_steps, 1)
|
||||
)
|
||||
warmup_factor = max(self.min_rate, self.last_epoch / self.warmup_steps)
|
||||
return [base_lr * warmup_factor for base_lr in self.base_lrs]
|
||||
|
||||
# cosine decay
|
||||
decay_progress = (self.last_epoch - self.warmup_steps) / max(
|
||||
self.lr_decay_steps, 1
|
||||
)
|
||||
decay_progress = (self.last_epoch - self.warmup_steps) / self.lr_decay_steps
|
||||
decay_progress = min(decay_progress, 1.0)
|
||||
cosine_decay = 0.5 * (1.0 + math.cos(math.pi * decay_progress))
|
||||
decay_factor = max(self.min_rate, cosine_decay)
|
||||
|
|
@ -108,7 +104,7 @@ class SGDRScheduler(BaseScheduler):
|
|||
optimizer,
|
||||
warmup_steps: int,
|
||||
cycle_length: int,
|
||||
min_rate: float = 0.01,
|
||||
min_rate: float = 0.05,
|
||||
t_mult: int = 2,
|
||||
last_epoch: int = -1,
|
||||
):
|
||||
|
|
@ -122,9 +118,7 @@ class SGDRScheduler(BaseScheduler):
|
|||
def get_lr(self):
|
||||
# warmup
|
||||
if self.last_epoch < self.warmup_steps:
|
||||
warmup_factor = max(
|
||||
self.min_rate, self.last_epoch / max(self.warmup_steps, 1)
|
||||
)
|
||||
warmup_factor = max(self.min_rate, self.last_epoch / self.warmup_steps)
|
||||
return [base_lr * warmup_factor for base_lr in self.base_lrs]
|
||||
|
||||
# SGDR
|
||||
|
|
@ -188,7 +182,7 @@ class WSDScheduler(BaseScheduler):
|
|||
warmup_steps: int,
|
||||
stable_steps: int,
|
||||
decay_steps: int,
|
||||
min_rate: float = 0.01,
|
||||
min_rate: float = 0.0,
|
||||
last_epoch: int = -1,
|
||||
):
|
||||
self.warmup_steps = warmup_steps
|
||||
|
|
@ -200,7 +194,7 @@ class WSDScheduler(BaseScheduler):
|
|||
|
||||
def get_lr(self) -> List[float]:
|
||||
if self.last_epoch < self.warmup_steps:
|
||||
factor = max(self.min_rate, self.last_epoch / max(self.warmup_steps, 1))
|
||||
factor = self.last_epoch / max(self.warmup_steps, 1)
|
||||
return [base_lr * factor for base_lr in self.base_lrs]
|
||||
|
||||
offset = self.last_epoch - self.warmup_steps
|
||||
|
|
|
|||
|
|
@ -196,7 +196,7 @@ class SFTStrategy(BaseStrategy):
|
|||
|
||||
ignore_index = -100
|
||||
input_mask = make_doc_boundary_mask(position_ids)
|
||||
target_ids = target_ids.masked_fill(~loss_mask, ignore_index)
|
||||
target_ids = target_ids.masked_fill(loss_mask == 0, ignore_index)
|
||||
logits = self.model(
|
||||
input_ids=input_ids, position_ids=position_ids, input_mask=input_mask
|
||||
)["logits"]
|
||||
|
|
|
|||
|
|
@ -9,6 +9,7 @@ from typing import IO, Callable, List, Optional, Protocol, runtime_checkable
|
|||
import torch
|
||||
import torch.distributed as dist
|
||||
import torch.nn as nn
|
||||
from torch.nn.utils import clip_grad_norm_
|
||||
from torch.utils.checkpoint import checkpoint as torch_checkpoint
|
||||
from tqdm import tqdm
|
||||
|
||||
|
|
@ -17,7 +18,12 @@ from astrai.parallel import only_on_rank
|
|||
from astrai.parallel.setup import get_current_device, get_rank
|
||||
from astrai.serialization import Checkpoint
|
||||
from astrai.trainer.metric_util import (
|
||||
ctx_get_grad_max,
|
||||
ctx_get_grad_mean,
|
||||
ctx_get_grad_min,
|
||||
ctx_get_grad_nan_num,
|
||||
ctx_get_grad_norm,
|
||||
ctx_get_grad_std,
|
||||
ctx_get_loss,
|
||||
ctx_get_lr,
|
||||
ctx_get_val_loss,
|
||||
|
|
@ -80,9 +86,7 @@ class GradientClippingCallback(TrainCallback):
|
|||
self.max_grad_norm = max_grad_norm
|
||||
|
||||
def on_optimizer_step(self, context: TrainContext):
|
||||
context.grad_norm = context.executor.clip_grad_norm(
|
||||
context.model, self.max_grad_norm
|
||||
)
|
||||
clip_grad_norm_(context.model.parameters(), self.max_grad_norm)
|
||||
|
||||
|
||||
@CallbackFactory.register("gradient_checkpointing")
|
||||
|
|
@ -139,35 +143,34 @@ class CheckpointCallback(TrainCallback):
|
|||
self.interval = interval
|
||||
self.weight_only = weight_only
|
||||
self.save_extra_fn = save_extra_fn or CheckpointCallback.save_extra
|
||||
self.last_ckpt_step = 0
|
||||
self.last_ckpt_iter = 0
|
||||
|
||||
def _save_checkpoint(self, context: TrainContext):
|
||||
state_dict = context.executor.unwrap_model(context.model)
|
||||
self.last_ckpt_step = context.optimizer_step
|
||||
self.last_ckpt_iter = context.iteration
|
||||
|
||||
if get_rank() == 0:
|
||||
save_path = os.path.join(
|
||||
self.save_dir,
|
||||
f"epoch_{context.epoch}_step_{context.optimizer_step}",
|
||||
self.save_dir, f"epoch_{context.epoch}_iter_{context.iteration}"
|
||||
)
|
||||
extra = self.save_extra_fn(context)
|
||||
meta = context.config.to_dict()
|
||||
context.checkpoint = Checkpoint(
|
||||
state_dict=state_dict,
|
||||
epoch=context.epoch,
|
||||
consumed_samples=context.consumed_samples,
|
||||
config=context.model_config,
|
||||
iteration=context.iteration,
|
||||
extra=extra,
|
||||
meta=meta,
|
||||
config=context.model_config,
|
||||
)
|
||||
context.checkpoint.save(save_path)
|
||||
|
||||
def on_batch_end(self, context: TrainContext):
|
||||
if context.optimizer_step - self.last_ckpt_step >= self.interval:
|
||||
if context.iteration - self.last_ckpt_iter >= self.interval:
|
||||
self._save_checkpoint(context)
|
||||
|
||||
def on_train_end(self, context: TrainContext):
|
||||
if context.optimizer_step != self.last_ckpt_step:
|
||||
if context.iteration != self.last_ckpt_iter:
|
||||
self._save_checkpoint(context)
|
||||
|
||||
def on_error(self, context: TrainContext):
|
||||
|
|
@ -199,27 +202,23 @@ class ProgressBarCallback(TrainCallback):
|
|||
|
||||
@only_on_rank(0)
|
||||
def on_epoch_begin(self, context: TrainContext):
|
||||
total_steps = len(context.dataloader) // context.executor.grad_accum_steps
|
||||
self.progress_bar = tqdm(
|
||||
total=total_steps,
|
||||
context.dataloader,
|
||||
desc=f"Epoch {context.epoch + 1}/{self.num_epoch}",
|
||||
dynamic_ncols=True,
|
||||
file=self.file or sys.stdout,
|
||||
)
|
||||
|
||||
@only_on_rank(0)
|
||||
def on_optimizer_step(self, context: TrainContext):
|
||||
self.progress_bar.update(1)
|
||||
def on_batch_end(self, context: TrainContext):
|
||||
postfix = {
|
||||
"step": context.optimizer_step,
|
||||
"loss": f"{context.loss:.4f}",
|
||||
"lr": f"{context.optimizer.param_groups[-1]['lr']:.2e}",
|
||||
}
|
||||
if context.grad_norm is not None:
|
||||
postfix["grad_norm"] = f"{context.grad_norm:.2f}"
|
||||
if context.val_loss is not None:
|
||||
postfix["val_loss"] = f"{context.val_loss:.4f}"
|
||||
self.progress_bar.set_postfix(postfix)
|
||||
self.progress_bar.update(1)
|
||||
|
||||
@only_on_rank(0)
|
||||
def on_epoch_end(self, context: TrainContext):
|
||||
|
|
@ -228,20 +227,20 @@ class ProgressBarCallback(TrainCallback):
|
|||
self.progress_bar.close()
|
||||
|
||||
|
||||
@CallbackFactory.register("metric")
|
||||
class MetricCallback(TrainCallback):
|
||||
@CallbackFactory.register("metric_logger")
|
||||
class MetricLoggerCallback(TrainCallback):
|
||||
def __init__(
|
||||
self,
|
||||
log_dir: str,
|
||||
save_interval: int,
|
||||
log_interval: int = 10,
|
||||
metrics: List[str] = None,
|
||||
val_step: int = 0,
|
||||
):
|
||||
self.last_log_flush_step = 0
|
||||
self.last_log_iter = 0
|
||||
self._last_val_loss = None
|
||||
self.save_interval = save_interval
|
||||
self.log_interval = log_interval
|
||||
self.metrics = metrics or ["loss", "lr"]
|
||||
self.val_step = val_step
|
||||
self._next_val_step = 0
|
||||
|
||||
self.log_dir = Path(log_dir) if log_dir else Path.cwd() / "logs"
|
||||
self.log_dir.mkdir(parents=True, exist_ok=True)
|
||||
|
|
@ -253,6 +252,11 @@ class MetricCallback(TrainCallback):
|
|||
"lr": ctx_get_lr,
|
||||
"val_loss": ctx_get_val_loss,
|
||||
"grad_norm": ctx_get_grad_norm,
|
||||
"grad_std": ctx_get_grad_std,
|
||||
"grad_max": ctx_get_grad_max,
|
||||
"grad_min": ctx_get_grad_min,
|
||||
"grad_mean": ctx_get_grad_mean,
|
||||
"grad_nan_num": ctx_get_grad_nan_num,
|
||||
}
|
||||
|
||||
def _metrics(self, context: TrainContext, names):
|
||||
|
|
@ -268,13 +272,46 @@ class MetricCallback(TrainCallback):
|
|||
"type": event_type,
|
||||
"timestamp": time.strftime("%Y-%m-%dT%H:%M:%S"),
|
||||
"epoch": context.epoch,
|
||||
"step": context.optimizer_step,
|
||||
"consumed_samples": context.consumed_samples,
|
||||
"iter": context.iteration,
|
||||
**extra,
|
||||
}
|
||||
self.log_cache.append(entry)
|
||||
|
||||
def _run_validation(self, context: TrainContext) -> float:
|
||||
@only_on_rank(0)
|
||||
def _flush(self, epoch, iter):
|
||||
log_file = self.log_dir / f"epoch_{epoch}_iter_{iter}_metric.jsonl"
|
||||
log_file.parent.mkdir(parents=True, exist_ok=True)
|
||||
with open(log_file, "w") as f:
|
||||
for log in self.log_cache:
|
||||
f.write(json.dumps(log) + "\n")
|
||||
|
||||
def on_batch_end(self, context):
|
||||
if context.iteration % self.log_interval == 0:
|
||||
step_metrics = [m for m in self.metrics if m != "val_loss"]
|
||||
self._append("step", context, **self._metrics(context, step_metrics))
|
||||
if context.iteration - self.last_log_iter >= self.save_interval:
|
||||
self._flush(context.epoch, context.iteration)
|
||||
self.last_log_iter = context.iteration
|
||||
|
||||
def on_optimizer_step(self, context):
|
||||
if context.val_loss is not None and context.val_loss != self._last_val_loss:
|
||||
self._append("validation", context, val_loss=context.val_loss)
|
||||
self._last_val_loss = context.val_loss
|
||||
|
||||
def on_epoch_end(self, context):
|
||||
self._append("epoch", context)
|
||||
|
||||
def on_train_end(self, context):
|
||||
if context.iteration != self.last_log_iter:
|
||||
self._flush(context.epoch, context.iteration)
|
||||
|
||||
def on_error(self, context):
|
||||
self._flush(context.epoch, context.iteration)
|
||||
|
||||
|
||||
@CallbackFactory.register("validation")
|
||||
class ValidationCallback(TrainCallback):
|
||||
def _run_validation(self, context: TrainContext):
|
||||
context.model.eval()
|
||||
|
||||
total_loss = 0.0
|
||||
|
|
@ -286,49 +323,27 @@ class MetricCallback(TrainCallback):
|
|||
total_loss += loss.item()
|
||||
num_batches += 1
|
||||
|
||||
avg_loss = total_loss / max(num_batches, 1)
|
||||
|
||||
if context.world_size > 1 and dist.is_initialized():
|
||||
stats = torch.tensor(
|
||||
[total_loss, float(num_batches)], device=get_current_device()
|
||||
)
|
||||
dist.all_reduce(stats, op=dist.ReduceOp.SUM)
|
||||
avg_loss = (stats[0] / stats[1]).item()
|
||||
else:
|
||||
avg_loss = total_loss / max(num_batches, 1)
|
||||
loss_tensor = torch.tensor([avg_loss], device=get_current_device())
|
||||
dist.all_reduce(loss_tensor, op=dist.ReduceOp.AVG)
|
||||
avg_loss = loss_tensor.item()
|
||||
|
||||
context.val_loss = avg_loss
|
||||
context.model.train()
|
||||
return avg_loss
|
||||
|
||||
@only_on_rank(0)
|
||||
def _flush(self, epoch, step):
|
||||
log_file = self.log_dir / f"epoch_{epoch}_step_{step}_metric.jsonl"
|
||||
log_file.parent.mkdir(parents=True, exist_ok=True)
|
||||
with open(log_file, "w") as f:
|
||||
for log in self.log_cache:
|
||||
f.write(json.dumps(log) + "\n")
|
||||
step_count = context.iteration // context.config.grad_accum_steps
|
||||
logger.info(
|
||||
f"Epoch {context.epoch + 1}, Step {step_count}, Val Loss: {avg_loss:.4f}"
|
||||
)
|
||||
|
||||
def on_optimizer_step(self, context):
|
||||
if (
|
||||
context.val_dataloader is not None
|
||||
and self.val_step > 0
|
||||
and context.optimizer_step >= self._next_val_step
|
||||
):
|
||||
context.val_loss = self._run_validation(context)
|
||||
self._next_val_step = context.optimizer_step + self.val_step
|
||||
self._append("validation", context, val_loss=context.val_loss)
|
||||
|
||||
step_metrics = [m for m in self.metrics if m != "val_loss"]
|
||||
self._append("step", context, **self._metrics(context, step_metrics))
|
||||
|
||||
if context.optimizer_step - self.last_log_flush_step >= self.save_interval:
|
||||
self._flush(context.epoch, context.optimizer_step)
|
||||
self.last_log_flush_step = context.optimizer_step
|
||||
|
||||
def on_epoch_end(self, context):
|
||||
self._append("epoch", context)
|
||||
|
||||
def on_train_end(self, context):
|
||||
if context.optimizer_step != self.last_log_flush_step:
|
||||
self._flush(context.epoch, context.optimizer_step)
|
||||
|
||||
def on_error(self, context):
|
||||
self._flush(context.epoch, context.optimizer_step)
|
||||
def on_optimizer_step(self, context: TrainContext):
|
||||
if context.val_dataloader is None:
|
||||
return
|
||||
cfg = context.config
|
||||
if cfg.val_step <= 0:
|
||||
return
|
||||
step_count = context.iteration // cfg.grad_accum_steps
|
||||
if step_count % cfg.val_step == 0:
|
||||
self._run_validation(context)
|
||||
|
|
|
|||
|
|
@ -29,9 +29,8 @@ class TrainContext:
|
|||
executor: BaseExecutor = field(default=None)
|
||||
|
||||
epoch: int = field(default=0)
|
||||
consumed_samples: int = field(default=0)
|
||||
iteration: int = field(default=0)
|
||||
loss: float = field(default=0.0)
|
||||
grad_norm: Optional[float] = field(default=None)
|
||||
val_dataloader: Optional[DataLoader] = field(default=None)
|
||||
val_loss: Optional[float] = field(default=None)
|
||||
|
||||
|
|
@ -39,14 +38,6 @@ class TrainContext:
|
|||
rank: int = field(default=0)
|
||||
kwargs: Dict[str, Any] = field(default_factory=dict)
|
||||
|
||||
@property
|
||||
def optimizer_step(self) -> int:
|
||||
return self.consumed_samples // (
|
||||
self.config.batch_per_device
|
||||
* self.world_size
|
||||
* self.config.grad_accum_steps
|
||||
)
|
||||
|
||||
|
||||
class TrainContextBuilder:
|
||||
def __init__(
|
||||
|
|
@ -98,10 +89,7 @@ class TrainContextBuilder:
|
|||
if checkpoint.config:
|
||||
context.model_config = checkpoint.config
|
||||
context.epoch = checkpoint.epoch or cfg.start_epoch
|
||||
if checkpoint.consumed_samples > 0:
|
||||
context.consumed_samples = checkpoint.consumed_samples
|
||||
else:
|
||||
context.consumed_samples = cfg.start_samples * context.world_size
|
||||
context.iteration = checkpoint.iteration or cfg.start_batch
|
||||
context.checkpoint = checkpoint
|
||||
|
||||
if cfg.lora is not None:
|
||||
|
|
@ -127,7 +115,7 @@ class TrainContextBuilder:
|
|||
cfg.dataset, [n_train, n_val], generator=generator
|
||||
)
|
||||
|
||||
sampler_offset = context.consumed_samples // context.world_size
|
||||
sampler_offset = context.iteration * cfg.batch_per_device
|
||||
sampler = ResumableDistributedSampler(
|
||||
data_source=train_dataset,
|
||||
start_epoch=context.epoch,
|
||||
|
|
|
|||
|
|
@ -34,12 +34,13 @@ class Trainer:
|
|||
cfg.ckpt_dir,
|
||||
cfg.ckpt_interval,
|
||||
),
|
||||
CallbackFactory.create("validation"),
|
||||
CallbackFactory.create(
|
||||
"metric",
|
||||
"metric_logger",
|
||||
log_dir=cfg.log_dir,
|
||||
save_interval=cfg.ckpt_interval,
|
||||
log_interval=cfg.log_interval,
|
||||
metrics=cfg.metrics,
|
||||
val_step=cfg.val_step,
|
||||
),
|
||||
CallbackFactory.create("progress_bar", cfg.n_epoch),
|
||||
CallbackFactory.create("gradient_clipping", cfg.max_grad_norm),
|
||||
|
|
@ -73,9 +74,7 @@ class Trainer:
|
|||
context.loss = loss.item()
|
||||
stand_loss = loss / executor.grad_accum_steps
|
||||
executor.backward(stand_loss)
|
||||
context.consumed_samples += (
|
||||
context.config.batch_per_device * context.world_size
|
||||
)
|
||||
context.iteration += 1
|
||||
self._call_callbacks("on_batch_end", context)
|
||||
|
||||
if executor.sync_gradients:
|
||||
|
|
|
|||
|
|
@ -1,4 +1,3 @@
|
|||
from argparse import ArgumentParser
|
||||
from pathlib import Path
|
||||
|
||||
import torch
|
||||
|
|
@ -8,82 +7,42 @@ from astrai.model import AutoModel
|
|||
from astrai.tokenize import AutoTokenizer
|
||||
|
||||
PROJECT_ROOT = Path(__file__).resolve().parents[2]
|
||||
|
||||
|
||||
def parse_args():
|
||||
parser = ArgumentParser(description="Interactive streaming chat")
|
||||
parser.add_argument(
|
||||
"--model_path",
|
||||
type=Path,
|
||||
default=PROJECT_ROOT / "params",
|
||||
help="Path to model weights (params/ or checkpoint/epoch_N_step_M/)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--temperature",
|
||||
type=float,
|
||||
default=0.8,
|
||||
help="Sampling temperature (default: 0.8)",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--top_p",
|
||||
type=float,
|
||||
default=0.95,
|
||||
help="Top-p sampling threshold",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--top_k",
|
||||
type=int,
|
||||
default=50,
|
||||
help="Top-k sampling threshold",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--max_tokens",
|
||||
type=int,
|
||||
default=2048,
|
||||
help="Maximum tokens to generate",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--system_prompt",
|
||||
type=str,
|
||||
default="You are a helpful assistant.",
|
||||
help="Optional system prompt",
|
||||
)
|
||||
return parser.parse_args()
|
||||
PARAMETER_ROOT = Path(PROJECT_ROOT, "params")
|
||||
|
||||
|
||||
def chat():
|
||||
args = parse_args()
|
||||
model_path = args.model_path
|
||||
|
||||
model = AutoModel.from_pretrained(model_path)
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
||||
model = AutoModel.from_pretrained(PARAMETER_ROOT)
|
||||
tokenizer = AutoTokenizer.from_pretrained(PARAMETER_ROOT)
|
||||
model.to(device="cuda", dtype=torch.bfloat16)
|
||||
engine = InferenceEngine(model=model, tokenizer=tokenizer)
|
||||
|
||||
messages = [{"role": "system", "content": args.system_prompt}]
|
||||
messages = [{"role": "system", "content": "You are a helpful assistant."}]
|
||||
engine = InferenceEngine(model=model, tokenizer=tokenizer)
|
||||
|
||||
while True:
|
||||
query = input(">> ")
|
||||
if query == "!exit":
|
||||
break
|
||||
|
||||
# Add user message
|
||||
messages.append({"role": "user", "content": query})
|
||||
|
||||
# Generate response
|
||||
full_response = ""
|
||||
prompt = tokenizer.apply_chat_template(messages, tokenize=False)
|
||||
|
||||
for token in engine.generate(
|
||||
prompt=prompt,
|
||||
stream=True,
|
||||
max_tokens=args.max_tokens,
|
||||
temperature=args.temperature,
|
||||
top_p=args.top_p,
|
||||
top_k=args.top_k,
|
||||
max_tokens=2048,
|
||||
temperature=0.8,
|
||||
top_p=0.95,
|
||||
top_k=50,
|
||||
):
|
||||
print(token, end="", flush=True)
|
||||
full_response += token
|
||||
|
||||
print()
|
||||
# Add assistant response to messages
|
||||
messages.append({"role": "assistant", "content": full_response.strip()})
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -1,307 +0,0 @@
|
|||
"""SVD effective rank & weight statistics analysis for model checkpoints."""
|
||||
|
||||
import argparse
|
||||
import json
|
||||
from pathlib import Path
|
||||
|
||||
import safetensors.torch
|
||||
import torch
|
||||
|
||||
|
||||
def effective_rank_metrics(w: torch.Tensor) -> dict:
|
||||
if w.ndim == 1:
|
||||
return {"shape": tuple(w.shape), "is_1d": True}
|
||||
|
||||
w = w.float()
|
||||
s = torch.linalg.svdvals(w)
|
||||
s_sq = s**2
|
||||
total = s_sq.sum()
|
||||
cumsum = torch.cumsum(s_sq, dim=0) / total
|
||||
|
||||
min_dim = min(w.shape[0], w.shape[1])
|
||||
er_90 = (cumsum < 0.90).sum().item() + 1
|
||||
er_95 = (cumsum < 0.95).sum().item() + 1
|
||||
er_99 = (cumsum < 0.99).sum().item() + 1
|
||||
|
||||
p = s_sq / total
|
||||
p = p[p > 1e-30]
|
||||
entropy = -(p * torch.log(p)).sum()
|
||||
entropic_rank = torch.exp(entropy).item()
|
||||
|
||||
return {
|
||||
"shape": tuple(w.shape),
|
||||
"min_dim": min_dim,
|
||||
"er_90": er_90,
|
||||
"er_95": er_95,
|
||||
"er_99": er_99,
|
||||
"er_99_norm": er_99 / min_dim,
|
||||
"er_95_norm": er_95 / min_dim,
|
||||
"entropic_rank": entropic_rank,
|
||||
"entropic_rank_norm": entropic_rank / min_dim,
|
||||
"top1_ratio": s[0].item() / s.sum().item(),
|
||||
"top5_ratio": s[:5].sum().item() / s.sum().item(),
|
||||
"decay_ratio": s[-1].item() / s[0].item(),
|
||||
"condition_number": s[0].item() / s[-1].item(),
|
||||
"mean": w.mean().item(),
|
||||
"std": w.std().item(),
|
||||
"min": w.min().item(),
|
||||
"max": w.max().item(),
|
||||
}
|
||||
|
||||
|
||||
def format_header(headers: list[str], widths: list[int]) -> str:
|
||||
return "".join(h.ljust(w) for h, w in zip(headers, widths))
|
||||
|
||||
|
||||
def format_row(values: list[str], widths: list[int]) -> str:
|
||||
return "".join(v.ljust(w) for v, w in zip(values, widths))
|
||||
|
||||
|
||||
def group_by_component(results: dict[str, dict]) -> dict[str, list[dict]]:
|
||||
groups: dict[str, list[dict]] = {}
|
||||
for key, r in results.items():
|
||||
parts = key.split(".")
|
||||
if parts[0] == "layers" and len(parts) >= 3:
|
||||
sub = parts[2:]
|
||||
if sub[0] == "attention":
|
||||
comp = f"attn.{sub[1]}"
|
||||
elif sub[0] == "mlp":
|
||||
comp = f"mlp.{sub[1]}"
|
||||
elif sub[0] == "input_norm":
|
||||
comp = "input_norm"
|
||||
elif sub[0] == "post_attention_norm":
|
||||
comp = "post_attn_norm"
|
||||
else:
|
||||
comp = ".".join(sub)
|
||||
else:
|
||||
comp = key
|
||||
groups.setdefault(comp, []).append(r)
|
||||
return groups
|
||||
|
||||
|
||||
def print_component_summary(results: dict[str, dict], title: str):
|
||||
groups = group_by_component(results)
|
||||
matrix_groups = {
|
||||
k: [v for v in vs if not v.get("is_1d")]
|
||||
for k, vs in groups.items()
|
||||
if any(not v.get("is_1d") for v in vs)
|
||||
}
|
||||
|
||||
widths = [20, 12, 12, 12, 12, 12]
|
||||
print(f"\n{title}")
|
||||
print(
|
||||
format_header(
|
||||
["Component", "N", "ER@99%", "EntRank%", "Top1 σ(%)", "Cond. Num"], widths
|
||||
)
|
||||
)
|
||||
print("-" * sum(widths))
|
||||
|
||||
for name in sorted(matrix_groups.keys()):
|
||||
items = matrix_groups[name]
|
||||
n = len(items)
|
||||
print(
|
||||
format_row(
|
||||
[
|
||||
name,
|
||||
str(n),
|
||||
f"{sum(r['er_99_norm'] for r in items) / n:.4f}",
|
||||
f"{sum(r['entropic_rank_norm'] for r in items) / n:.4f}",
|
||||
f"{sum(r['top1_ratio'] for r in items) / n:.4f}",
|
||||
f"{sum(r['condition_number'] for r in items) / n:.1f}",
|
||||
],
|
||||
widths,
|
||||
)
|
||||
)
|
||||
|
||||
all_er = [
|
||||
r["er_99_norm"]
|
||||
for vs in matrix_groups.values()
|
||||
for r in vs
|
||||
if "_norm" not in r or not r.get("is_1d")
|
||||
]
|
||||
if all_er:
|
||||
m = sum(all_er) / len(all_er)
|
||||
print(f"\n Overall Mean ER@99: {m:.4f} ({m * 100:.1f}% of dimension)")
|
||||
if m > 0.85:
|
||||
print(" → HIGH utilization: model near capacity → need more params")
|
||||
elif m > 0.5:
|
||||
print(" → MODERATE utilization: some headroom left")
|
||||
else:
|
||||
print(" → LOW utilization: significant unused capacity")
|
||||
|
||||
|
||||
def print_layer_grid(results: dict[str, dict]):
|
||||
comps = [
|
||||
"attn.q_proj",
|
||||
"attn.k_proj",
|
||||
"attn.v_proj",
|
||||
"attn.o_proj",
|
||||
"mlp.up",
|
||||
"mlp.gate",
|
||||
"mlp.down",
|
||||
]
|
||||
widths = [6] + [10] * len(comps)
|
||||
metric = "er_99_norm"
|
||||
|
||||
print(f"\n--- Per-Layer Effective Rank (99% energy) ---")
|
||||
print(format_header(["Layer"] + comps, widths))
|
||||
print("-" * sum(widths))
|
||||
|
||||
layer_data: dict[int, dict[str, dict]] = {}
|
||||
for key, r in results.items():
|
||||
parts = key.split(".")
|
||||
if parts[0] != "layers":
|
||||
continue
|
||||
li = int(parts[1])
|
||||
sub = parts[2:]
|
||||
if sub[0] == "attention":
|
||||
cname = f"attn.{sub[1]}"
|
||||
elif sub[0] == "mlp":
|
||||
cname = f"mlp.{sub[1]}"
|
||||
else:
|
||||
continue
|
||||
layer_data.setdefault(li, {})[cname] = r
|
||||
|
||||
for li in sorted(layer_data):
|
||||
values = [str(li)]
|
||||
for c in comps:
|
||||
v = layer_data[li].get(c, {}).get(metric, 0)
|
||||
values.append(f"{v:.4f}")
|
||||
print(format_row(values, widths))
|
||||
|
||||
|
||||
def print_weight_stats(results: dict[str, dict]):
|
||||
groups = group_by_component(results)
|
||||
widths = [20, 12, 12, 12, 12]
|
||||
print(f"\n--- Weight Value Statistics ---")
|
||||
print(format_header(["Component", "Mean", "Std", "Min", "Max"], widths))
|
||||
print("-" * sum(widths))
|
||||
|
||||
for name in sorted(groups.keys()):
|
||||
items = groups[name]
|
||||
means = [r.get("mean", 0) for r in items]
|
||||
stds = [r.get("std", 0) for r in items]
|
||||
mins = [r.get("min", 0) for r in items]
|
||||
maxs = [r.get("max", 0) for r in items]
|
||||
g_mean = sum(means) / len(means)
|
||||
g_std = sum(stds) / len(stds)
|
||||
g_min = min(mins)
|
||||
g_max = max(maxs)
|
||||
print(
|
||||
format_row(
|
||||
[
|
||||
name,
|
||||
f"{g_mean:.6f}",
|
||||
f"{g_std:.6f}",
|
||||
f"{g_min:.6f}",
|
||||
f"{g_max:.6f}",
|
||||
],
|
||||
widths,
|
||||
)
|
||||
)
|
||||
|
||||
|
||||
def print_params_summary(results: dict[str, dict]):
|
||||
total_2d = sum(
|
||||
r["shape"][0] * r["shape"][1] for r in results.values() if not r.get("is_1d")
|
||||
)
|
||||
total_1d = sum(r["shape"][0] for r in results.values() if r.get("is_1d"))
|
||||
print(f"\n Total 2D params: {total_2d:,}")
|
||||
print(f" Total 1D params: {total_1d:,}")
|
||||
print(f" Total params: {total_2d + total_1d:,}")
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(
|
||||
description="SVD effective rank & weight statistics of a model checkpoint."
|
||||
)
|
||||
parser.add_argument(
|
||||
"--ckpt_dir",
|
||||
type=str,
|
||||
required=True,
|
||||
help="Path to checkpoint directory (containing model.safetensors + config.json).",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--compare",
|
||||
type=str,
|
||||
nargs="*",
|
||||
help="Additional checkpoint directories to compare against.",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--no_svd",
|
||||
action="store_true",
|
||||
help="Skip SVD analysis, only show weight statistics (mean/std/min/max).",
|
||||
)
|
||||
args = parser.parse_args()
|
||||
|
||||
def analyze_one(ckpt_dir: str, label: str):
|
||||
ckpt_dir = Path(ckpt_dir)
|
||||
weights_path = ckpt_dir / "model.safetensors"
|
||||
if not weights_path.exists():
|
||||
print(f"ERROR: {weights_path} not found")
|
||||
return {}
|
||||
|
||||
meta = {}
|
||||
meta_path = ckpt_dir / "meta.json"
|
||||
if meta_path.exists():
|
||||
with open(meta_path) as f:
|
||||
meta = json.load(f)
|
||||
|
||||
print(f"\n{'=' * 70}")
|
||||
print(f" {label}: {ckpt_dir}")
|
||||
if meta:
|
||||
print(
|
||||
f" Iteration: {meta.get('iteration', '?')}, "
|
||||
f"Strategy: {meta.get('strategy', '?')}, "
|
||||
f"nprocs={meta.get('nprocs', '?')}"
|
||||
)
|
||||
print(f"{'=' * 70}")
|
||||
|
||||
print(f"Loading weights...")
|
||||
sd = safetensors.torch.load_file(str(weights_path))
|
||||
print(f" {len(sd)} keys loaded")
|
||||
|
||||
weight_keys = [
|
||||
k
|
||||
for k in sd
|
||||
if ".weight" in k and "rotary_embedding" not in k and "freqs_cis" not in k
|
||||
]
|
||||
|
||||
results = {}
|
||||
if not args.no_svd:
|
||||
print(f"Computing SVD on {len(weight_keys)} tensors...")
|
||||
for i, k in enumerate(sorted(weight_keys)):
|
||||
print(f" [{i + 1}/{len(weight_keys)}] {k:<60s}", end="\r")
|
||||
results[k] = effective_rank_metrics(sd[k])
|
||||
print()
|
||||
else:
|
||||
print(f"Computing stats on {len(weight_keys)} tensors (no SVD)...")
|
||||
for i, k in enumerate(sorted(weight_keys)):
|
||||
t = sd[k]
|
||||
results[k] = {
|
||||
"shape": tuple(t.shape),
|
||||
"is_1d": t.ndim == 1,
|
||||
"mean": t.float().mean().item(),
|
||||
"std": t.float().std().item(),
|
||||
"min": t.float().min().item(),
|
||||
"max": t.float().max().item(),
|
||||
}
|
||||
|
||||
print_params_summary(results)
|
||||
if not args.no_svd:
|
||||
print_component_summary(
|
||||
results, "\n=== SVD Effective Rank by Component ==="
|
||||
)
|
||||
print_layer_grid(results)
|
||||
print_weight_stats(results)
|
||||
return results
|
||||
|
||||
analyze_one(args.ckpt_dir, "Primary")
|
||||
|
||||
if args.compare:
|
||||
for cdir in args.compare:
|
||||
analyze_one(cdir, "Compare")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
|
|
@ -52,11 +52,8 @@ def compute_ifd(
|
|||
|
||||
|
||||
def _compute_ifd_raw(model, tokenizer, instruction, response, device, max_len) -> dict:
|
||||
instr_ids = tokenizer.encode(instruction, add_special_tokens=False)
|
||||
resp_ids = tokenizer.encode(response, add_special_tokens=False)
|
||||
|
||||
if len(resp_ids) > max_len:
|
||||
resp_ids = resp_ids[:max_len]
|
||||
instr_ids = tokenizer.encode(instruction)
|
||||
resp_ids = tokenizer.encode(response)
|
||||
|
||||
if not resp_ids:
|
||||
return {
|
||||
|
|
@ -69,39 +66,28 @@ def _compute_ifd_raw(model, tokenizer, instruction, response, device, max_len) -
|
|||
qa_len = len(instr_ids) + len(resp_ids)
|
||||
if qa_len > max_len:
|
||||
overflow = qa_len - max_len
|
||||
if overflow >= len(instr_ids):
|
||||
resp_ids = resp_ids[:max_len]
|
||||
instr_ids = []
|
||||
else:
|
||||
instr_ids = instr_ids[overflow:]
|
||||
|
||||
if not instr_ids:
|
||||
return {
|
||||
"L_cond": None,
|
||||
"L_uncond": None,
|
||||
"ifd": None,
|
||||
"error": "response too long for context",
|
||||
}
|
||||
instr_ids = instr_ids[overflow:]
|
||||
|
||||
instr_len = len(instr_ids)
|
||||
resp_len = len(resp_ids)
|
||||
|
||||
qa_ids = instr_ids + resp_ids
|
||||
qa_tensor = torch.tensor([qa_ids], device=device, dtype=torch.long)
|
||||
|
||||
with torch.inference_mode():
|
||||
logits_qa = model(torch.tensor([qa_ids], device=device, dtype=torch.long))[
|
||||
"logits"
|
||||
][0]
|
||||
logits_resp = model(torch.tensor([resp_ids], device=device, dtype=torch.long))[
|
||||
"logits"
|
||||
][0]
|
||||
logits_qa = model(qa_tensor)["logits"][0]
|
||||
|
||||
resp_logits = logits_qa[instr_len - 1 : -1]
|
||||
resp_targets = logits_resp.new_tensor(resp_ids, dtype=torch.long)
|
||||
resp_targets = torch.tensor(resp_ids, device=device, dtype=torch.long)
|
||||
L_cond = F.cross_entropy(resp_logits, resp_targets, reduction="mean").item()
|
||||
|
||||
resp_tensor = torch.tensor([resp_ids], device=device, dtype=torch.long)
|
||||
|
||||
with torch.inference_mode():
|
||||
logits_resp = model(resp_tensor)["logits"][0]
|
||||
|
||||
unp_logits = logits_resp[:-1]
|
||||
unp_targets = logits_resp.new_tensor(resp_ids[1:], dtype=torch.long)
|
||||
unp_targets = resp_tensor[0, 1:]
|
||||
L_uncond = F.cross_entropy(unp_logits, unp_targets, reduction="mean").item()
|
||||
|
||||
ifd = L_cond / L_uncond if L_uncond > 0 else None
|
||||
|
|
@ -199,7 +185,7 @@ def process_file(
|
|||
output_file: str,
|
||||
instr_key: str,
|
||||
resp_key: str,
|
||||
max_len: int = 2048,
|
||||
max_len: int,
|
||||
use_chat_template: bool = False,
|
||||
):
|
||||
device = "cuda" if torch.cuda.is_available() else "cpu"
|
||||
|
|
|
|||
|
|
@ -150,8 +150,8 @@ def parse_args() -> argparse.Namespace:
|
|||
parser.add_argument(
|
||||
"--metrics",
|
||||
nargs="*",
|
||||
default=["loss", "lr", "grad_norm"],
|
||||
help="Metrics to log (e.g. --metrics loss lr val_loss). Default: loss lr grad_norm.",
|
||||
default=["loss", "lr"],
|
||||
help="Metrics to log (e.g. --metrics loss lr val_loss). Default: loss lr.",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--log_dir",
|
||||
|
|
@ -159,6 +159,12 @@ def parse_args() -> argparse.Namespace:
|
|||
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,
|
||||
|
|
@ -169,10 +175,7 @@ def parse_args() -> argparse.Namespace:
|
|||
"--start_epoch", type=int, default=0, help="Start epoch for training."
|
||||
)
|
||||
parser.add_argument(
|
||||
"--start_samples",
|
||||
type=int,
|
||||
default=0,
|
||||
help="Start samples (per rank) for training.",
|
||||
"--start_batch", type=int, default=0, help="Start batch for training."
|
||||
)
|
||||
|
||||
parser.add_argument(
|
||||
|
|
@ -266,20 +269,7 @@ def create_model(config):
|
|||
|
||||
|
||||
def create_optimizer(model, **kwargs) -> optim.Optimizer:
|
||||
decay_params = []
|
||||
no_decay_params = []
|
||||
for name, param in model.named_parameters():
|
||||
if not param.requires_grad:
|
||||
continue
|
||||
if param.dim() < 2 or "norm" in name or "bias" in name:
|
||||
no_decay_params.append(param)
|
||||
else:
|
||||
decay_params.append(param)
|
||||
param_groups = [
|
||||
{"params": decay_params, "weight_decay": kwargs.pop("weight_decay", 0.01)},
|
||||
{"params": no_decay_params, "weight_decay": 0.0},
|
||||
]
|
||||
return optim.AdamW(param_groups, fused=True, **kwargs)
|
||||
return optim.AdamW(model.parameters(), fused=True, **kwargs)
|
||||
|
||||
|
||||
def create_scheduler(
|
||||
|
|
@ -314,7 +304,7 @@ def train(
|
|||
n_epoch: int,
|
||||
batch_per_device: int,
|
||||
start_epoch: int,
|
||||
start_samples: int,
|
||||
start_batch: int,
|
||||
grad_accum_steps: int,
|
||||
warmup_ratio: float,
|
||||
ckpt_interval: int,
|
||||
|
|
@ -323,6 +313,7 @@ def train(
|
|||
val_step: int,
|
||||
metrics: list[str],
|
||||
log_dir: str,
|
||||
log_interval: int,
|
||||
dpo_beta: float,
|
||||
grpo_clip_eps: float,
|
||||
grpo_kl_coef: float,
|
||||
|
|
@ -440,7 +431,7 @@ def train(
|
|||
n_epoch=n_epoch,
|
||||
batch_per_device=batch_per_device,
|
||||
start_epoch=start_epoch,
|
||||
start_samples=start_samples,
|
||||
start_batch=start_batch,
|
||||
ckpt_interval=ckpt_interval,
|
||||
grad_accum_steps=grad_accum_steps,
|
||||
max_grad_norm=max_grad_norm,
|
||||
|
|
@ -458,6 +449,7 @@ def train(
|
|||
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,
|
||||
|
|
|
|||
|
|
@ -75,7 +75,7 @@ class MultiTurnDataset(Dataset):
|
|||
|
||||
|
||||
class EarlyStoppingDataset(Dataset):
|
||||
"""Dataset that triggers early stopping after consuming a specified number of samples."""
|
||||
"""Dataset that triggers early stopping after a specified number of iterations."""
|
||||
|
||||
def __init__(self, length=10, stop_after=5):
|
||||
self.length = length
|
||||
|
|
|
|||
|
|
@ -25,9 +25,7 @@ def test_single_process():
|
|||
|
||||
scheduler.step()
|
||||
|
||||
checkpoint = Checkpoint(
|
||||
state_dict=model.state_dict(), epoch=3, consumed_samples=120
|
||||
)
|
||||
checkpoint = Checkpoint(state_dict=model.state_dict(), epoch=3, iteration=30)
|
||||
|
||||
with tempfile.TemporaryDirectory() as tmpdir:
|
||||
checkpoint.save(tmpdir)
|
||||
|
|
@ -35,7 +33,7 @@ def test_single_process():
|
|||
loaded_checkpoint = Checkpoint.load(tmpdir)
|
||||
|
||||
assert loaded_checkpoint.epoch == 3
|
||||
assert loaded_checkpoint.consumed_samples == 120
|
||||
assert loaded_checkpoint.iteration == 30
|
||||
|
||||
|
||||
def test_checkpoint_with_extra():
|
||||
|
|
@ -48,10 +46,7 @@ def test_checkpoint_with_extra():
|
|||
"scheduler": {"last_epoch": 5},
|
||||
}
|
||||
checkpoint = Checkpoint(
|
||||
state_dict=model.state_dict(),
|
||||
epoch=1,
|
||||
consumed_samples=40,
|
||||
extra=extra,
|
||||
state_dict=model.state_dict(), epoch=1, iteration=10, extra=extra
|
||||
)
|
||||
|
||||
with tempfile.TemporaryDirectory() as tmpdir:
|
||||
|
|
@ -82,7 +77,7 @@ def simple_training():
|
|||
checkpoint = Checkpoint(
|
||||
state_dict=model.state_dict(),
|
||||
epoch=2,
|
||||
consumed_samples=40,
|
||||
iteration=10,
|
||||
)
|
||||
|
||||
rank = get_rank()
|
||||
|
|
|
|||
|
|
@ -1,18 +1,14 @@
|
|||
import json
|
||||
import os
|
||||
|
||||
import numpy as np
|
||||
import pytest
|
||||
import torch
|
||||
|
||||
from astrai.config.preprocess_config import PipelineConfig
|
||||
from astrai.dataset.dataset import DatasetFactory, SEQDataset
|
||||
from astrai.dataset.storage import (
|
||||
H5Store,
|
||||
StoreFactory,
|
||||
detect_format,
|
||||
)
|
||||
from astrai.serialization import (
|
||||
load_bin,
|
||||
save_bin,
|
||||
save_h5,
|
||||
|
|
@ -23,39 +19,6 @@ def _rand_seq(length, vocab=1000):
|
|||
return torch.randint(0, vocab, (length,), dtype=torch.int64)
|
||||
|
||||
|
||||
def _save_test_tokenizer(test_dir, tokenizer):
|
||||
tokenizer_path = os.path.join(test_dir, "tokenizer")
|
||||
os.makedirs(tokenizer_path, exist_ok=True)
|
||||
tokenizer.save_pretrained(tokenizer_path)
|
||||
return tokenizer_path
|
||||
|
||||
|
||||
def _write_jsonl_dataset(test_dir, tokenizer_path, records, config_overrides=None):
|
||||
data_dir = os.path.join(test_dir, "jsonl_data")
|
||||
os.makedirs(data_dir, exist_ok=True)
|
||||
|
||||
with open(os.path.join(data_dir, "data.jsonl"), "w", encoding="utf-8") as f:
|
||||
for record in records:
|
||||
f.write(json.dumps(record, ensure_ascii=False) + "\n")
|
||||
|
||||
config = {
|
||||
"tokenizer_path": tokenizer_path,
|
||||
"version": 1,
|
||||
"input": {"sections": [{"field": "text", "action": "train"}]},
|
||||
"preprocessing": {"max_seq_len": 128},
|
||||
"output": {"position_ids_mode": "continuous"},
|
||||
}
|
||||
if config_overrides:
|
||||
config.update(config_overrides)
|
||||
|
||||
with open(
|
||||
os.path.join(data_dir, "dataset_config.json"), "w", encoding="utf-8"
|
||||
) as f:
|
||||
json.dump(config, f, ensure_ascii=False, indent=2)
|
||||
|
||||
return data_dir
|
||||
|
||||
|
||||
def _make_seq_dataset(
|
||||
test_dir, name="data", seq_length=200, train_type="seq", data=None, **load_kwargs
|
||||
):
|
||||
|
|
@ -409,106 +372,3 @@ def test_dataset_load_explicit_storage_type(base_test_env):
|
|||
dataset = _make_seq_dataset(test_dir, "explicit", storage_type="h5")
|
||||
assert len(dataset) > 0
|
||||
assert dataset.count == 200
|
||||
|
||||
|
||||
def test_detect_format_jsonl_dir(base_test_env):
|
||||
test_dir = base_test_env["test_dir"]
|
||||
tokenizer_path = _save_test_tokenizer(test_dir, base_test_env["tokenizer"])
|
||||
data_dir = _write_jsonl_dataset(
|
||||
test_dir,
|
||||
tokenizer_path,
|
||||
[{"text": "hello world"}, {"text": "foo bar baz"}],
|
||||
)
|
||||
assert detect_format(data_dir) == "jsonl"
|
||||
|
||||
|
||||
def test_jsonl_store_seq(base_test_env):
|
||||
test_dir = base_test_env["test_dir"]
|
||||
tokenizer_path = _save_test_tokenizer(test_dir, base_test_env["tokenizer"])
|
||||
data_dir = _write_jsonl_dataset(
|
||||
test_dir,
|
||||
tokenizer_path,
|
||||
[{"text": "hello world"}, {"text": "foo bar baz qux"}],
|
||||
config_overrides={"preprocessing": {"max_seq_len": 128, "min_chars": 0}},
|
||||
)
|
||||
|
||||
store = StoreFactory.create("jsonl")
|
||||
store.load(data_dir)
|
||||
assert len(store) > 0
|
||||
assert "sequence" in store.keys
|
||||
|
||||
dataset = DatasetFactory.load("seq", data_dir, window_size=8)
|
||||
assert len(dataset) > 0
|
||||
item = dataset[0]
|
||||
assert "input_ids" in item
|
||||
assert "target_ids" in item
|
||||
assert item["input_ids"].dtype == torch.long
|
||||
|
||||
|
||||
def test_jsonl_store_sft(base_test_env):
|
||||
test_dir = base_test_env["test_dir"]
|
||||
tokenizer = base_test_env["tokenizer"]
|
||||
tokenizer.set_chat_template(
|
||||
"{% for message in messages %}{{ message['role'] }}:{{ message['content'] }}\n{% endfor %}"
|
||||
)
|
||||
tokenizer_path = _save_test_tokenizer(test_dir, tokenizer)
|
||||
data_dir = _write_jsonl_dataset(
|
||||
test_dir,
|
||||
tokenizer_path,
|
||||
[
|
||||
{
|
||||
"messages": [
|
||||
{"role": "system", "content": "sys"},
|
||||
{"role": "user", "content": "hi"},
|
||||
{"role": "assistant", "content": "hello"},
|
||||
]
|
||||
}
|
||||
],
|
||||
config_overrides={
|
||||
"input": {
|
||||
"sections": [{"field": "messages", "action": "$role", "template": True}]
|
||||
},
|
||||
"mask": {"system": "mask", "user": "mask", "assistant": "train"},
|
||||
"mask_default": "mask",
|
||||
},
|
||||
)
|
||||
|
||||
store = StoreFactory.create("jsonl")
|
||||
store.load(data_dir)
|
||||
assert "sequence" in store.keys
|
||||
assert "loss_mask" in store.keys
|
||||
assert "position_ids" in store.keys
|
||||
|
||||
dataset = DatasetFactory.load("sft", data_dir, window_size=8)
|
||||
item = dataset[0]
|
||||
assert "input_ids" in item
|
||||
assert "target_ids" in item
|
||||
assert "loss_mask" in item
|
||||
assert "position_ids" in item
|
||||
assert item["loss_mask"].dtype == torch.bool
|
||||
|
||||
|
||||
def test_jsonl_store_pipeline_config_roundtrip(base_test_env):
|
||||
test_dir = base_test_env["test_dir"]
|
||||
config_path = os.path.join(test_dir, "dataset_config.json")
|
||||
with open(config_path, "w", encoding="utf-8") as f:
|
||||
json.dump(
|
||||
{
|
||||
"tokenizer_path": os.path.join(test_dir, "tokenizer"),
|
||||
"version": 1,
|
||||
"input": {"sections": [{"field": "text", "action": "train"}]},
|
||||
"mask": {"assistant": "train"},
|
||||
"preprocessing": {"max_seq_len": 64},
|
||||
"output": {"position_ids_mode": "doc_reset"},
|
||||
},
|
||||
f,
|
||||
ensure_ascii=False,
|
||||
indent=2,
|
||||
)
|
||||
|
||||
with open(config_path, "r", encoding="utf-8") as f:
|
||||
raw = json.load(f)
|
||||
raw.pop("tokenizer_path")
|
||||
config = PipelineConfig.from_dict(raw)
|
||||
assert config.output.position_ids_mode == "doc_reset"
|
||||
assert config.preprocessing.max_seq_len == 64
|
||||
|
|
|
|||
|
|
@ -52,7 +52,7 @@ def create_train_config(
|
|||
batch_per_device: Batch size per device (default: 2)
|
||||
grad_accum_steps: Gradient accumulation steps (default: 1)
|
||||
max_grad_norm: Maximum gradient norm for clipping (default: 1.0)
|
||||
ckpt_interval: Checkpoint save interval in optimizer steps (default: 5)
|
||||
ckpt_interval: Checkpoint save interval in iterations (default: 5)
|
||||
random_seed: Random seed for reproducibility (default: 42)
|
||||
**kwargs: Additional arguments passed to TrainConfig
|
||||
|
||||
|
|
|
|||
|
|
@ -44,14 +44,14 @@ def test_early_stopping_simulation(base_test_env, early_stopping_dataset):
|
|||
pass
|
||||
|
||||
# Resume from latest checkpoint
|
||||
load_dir = os.path.join(base_test_env["test_dir"], "epoch_0_step_1")
|
||||
load_dir = os.path.join(base_test_env["test_dir"], "epoch_0_iter_2")
|
||||
trainer = Trainer(train_config)
|
||||
trainer.train(resume_dir=load_dir)
|
||||
|
||||
# Verify checkpoint was saved at expected step
|
||||
load_dir = os.path.join(base_test_env["test_dir"], "epoch_1_step_5")
|
||||
# Verify checkpoint was saved at expected iteration
|
||||
load_dir = os.path.join(base_test_env["test_dir"], "epoch_1_iter_10")
|
||||
import json
|
||||
|
||||
with open(os.path.join(load_dir, "meta.json")) as f:
|
||||
meta = json.load(f)
|
||||
assert meta["consumed_samples"] == 20
|
||||
assert meta["iteration"] == 10
|
||||
|
|
|
|||
Loading…
Reference in New Issue