New to Claude Skills? Learn how to install them →

orchestra-research on GitHub

ML Training Recipes

Free

Streamline your PyTorch training across diverse domains.

Get this skill

Free · Opens the source repo

What ML Training Recipes does

ML Training Recipes provides a comprehensive set of battle-tested patterns for training neural networks using PyTorch, applicable across various domains including natural language processing, computer vision, and biomedical applications. These recipes are drawn from production-level codebases and modern training practices, ensuring they meet the needs of developers and researchers alike. The skill includes detailed instructions on training loops, optimizer configurations, learning rate scheduling, and debugging techniques, making it a valuable resource for anyone involved in machine learning.

The skill is particularly useful for those looking to optimize their training processes, whether they are fine-tuning large language models, training models for image recognition, or working on complex biomedical data. With a focus on systematic experimentation and effective resource utilization, users can leverage these recipes to enhance their model performance and reduce training time. The included reference files cover a wide range of topics, from architecture selection to scaling laws, providing a solid foundation for both beginners and experienced practitioners.

Users will find specific guidance on choosing the right model architecture based on data type and scale, as well as practical advice on configuring optimizers and learning rates. The recipes also address common challenges such as debugging loss spikes and out-of-memory errors, enabling users to troubleshoot effectively. This skill is designed for data scientists, machine learning engineers, and researchers who want to streamline their workflow and improve their training outcomes.

When to use it

Use this skill when you need structured guidance on training neural networks across different domains, especially when optimizing performance or debugging issues.

When not to use it

This skill may not be suitable for users looking for a graphical interface or those who prefer a more hands-on approach without predefined recipes.

What you can build with it

Fine-tuning a Language Model

Utilize the recipes to effectively fine-tune a pre-trained language model like GPT or LLaMA with optimal learning rates and configurations.

Training Vision Models

Follow the guidelines for selecting and training vision models, ensuring efficient use of GPU resources and minimizing training time.

Biomedical Data Processing

Apply the recipes to train models on complex biomedical datasets, leveraging domain-specific optimizations and architectures.

How to install ML Training Recipes

View source

1. Install with the skills CLI

npx skills add orchestra-research/ai-research-skills/ml-training-recipes --agent claude-code

2. Or install it manually

Download the skill folder and drop it into ~/.claude/skills/ for all projects, or .claude/skills/ to scope it to one repo. Restart Claude Code so it picks up the new skill.

Anthropic's agentic coding CLI, and the reference implementation of Agent Skills. Drop a skill folder into ~/.claude/skills and Claude Code loads it automatically whenever a task matches the skill's description. Claude Code docs

Inside SKILL.md

Written by orchestra-research

ML Training Recipes

Battle-tested patterns for PyTorch training across domains. Drawn from production codebases (Karpathy's autoresearch/nanochat, torchvision, HuggingFace) and modern training practice.

Reference files (read when needed)

  • references/architecture.md — Transformer/LLM architecture code patterns, weight init
  • references/optimizers.md — Muon, AdamW hybrid, per-group LR, compiled optimizer steps
  • references/domain-specific.md — Vision, diffusion, contrastive, distributed, checkpointing, data loading
  • references/scaling-and-selection.md — Scaling laws, compute budget tables, decision trees, DGX Spark
  • references/biomedical.md — Drug discovery, protein models, medical imaging, genomics, clinical NLP
  • references/experiment-loop.md — Autonomous experiment loop (autoresearch keep/discard/revert)

Architecture Selection

Pick the right model by data type and data scale:

Data Type< 10K samples10K-100K> 100K
ImagesPretrained CNN + fine-tuneFine-tune ViT or CNNViT from scratch
Text (gen)Few-shot promptingFine-tune GPT/LLaMA (LoRA)Pretrain from scratch
TabularXGBoost/LightGBMStill XGBoostNeural viable
AudioPretrained WhisperFine-tune ASTTrain from scratch
MoleculesPretrained GNNFine-tune molecular LMTrain GNN from scratch
ProteinsESM-2 embeddings + headFine-tune ESM-2Train protein LM
Medical imgPretrained CNNnnU-Net (auto-config)Swin-UNETR / MedSAM

Key principle: architecture matters less than training recipe at equal compute. A well-tuned ResNet beats a poorly-tuned ViT (ref: "ResNet Strikes Back", Wightman 2021).

For biomedical domains, see references/biomedical.md. For sequence model selection and compute planning, see references/scaling-and-selection.md.


Scaling Laws

Chinchilla rule (Hoffmann et al., 2022)

Compute-optimal training: ~20 tokens per parameter.

Model SizeCompute-OptimalInference-Optimal (100×)
125M2.5B tokens12.5B tokens
1B20B tokens100B tokens
7B140B tokens700B tokens

FLOPs ≈ 6 × N × D (N=params, D=tokens). Data repetition limit: ~4 epochs before diminishing returns.


Training Loop

import gc, time, torch

torch.manual_seed(42)
torch.set_float32_matmul_precision("high")  # TF32 on Ampere+
autocast_ctx = torch.amp.autocast(device_type="cuda", dtype=torch.bfloat16)

grad_accum_steps = total_batch_size // (batch_size * seq_len)
step = 0

while not done:
    t0 = time.time()
    for micro_step in range(grad_accum_steps):
        with autocast_ctx:
            loss = model(x, y)
        (loss / grad_accum_steps).backward()
        x, y = next(train_loader)

    update_lr(optimizer, progress)
    optimizer.step()
    model.zero_grad(set_to_none=True)  # frees memory vs zeroing

    if loss.item() > 100:  # fast-fail on divergence
        print("FAIL: loss exploded"); exit(1)

    torch.cuda.synchronize()
    if step == 0:
        gc.collect(); gc.freeze(); gc.disable()  # avoid ~500ms GC stalls
    step += 1

Key principles

  • Gradient clipping: clip_grad_norm_(params, 1.0) — near-universal for Transformers. Exception: Muon optimizer normalizes updates via orthogonalization, so clipping is optional.
  • Tensor Core alignment: batch size, hidden dims should be multiples of 8 (bf16) or 64 (A100).
  • Time-based budgets make experiments comparable across hardware.
  • cudnn.benchmark = True for fixed-size vision inputs.

Optimizer Configuration

Modern LLM training uses different optimizers per parameter group:

Parameter TypeOptimizerLR (base)Weight Decay
2D weight matricesMuon0.040.2
Token embeddingsAdamW0.6 × scale0.0
Unembedding (lm_head)AdamW0.004 × scale0.0
Per-layer scalarsAdamW0.005 × scale0.0

LR scaling by dimension: lr * (d_model / 768)^(-0.5) — keeps dynamics stable across sizes.

Rules of thumb

  • Embeddings need higher LR (sparse updates). Never weight-decay embeddings.
  • Weight decay scheduling: linearly decay WD to 0 over training.
  • AdamW defaults: β1=0.9, β2=0.95, eps=1e-10 (not default 1e-8 — prevents stale updates in bf16).

For Muon details (polar express orthogonalization, NorMuon), see references/optimizers.md.


Learning Rate Scheduling

Time-based (autoresearch style)

def get_lr_multiplier(progress):  # progress = elapsed_time / time_budget
    if progress < warmup_ratio:
        return progress / warmup_ratio
    elif progress < 1.0 - warmdown_ratio:
        return 1.0
    else:
        cooldown = (1.0 - progress) / warmdown_ratio
        return cooldown + (1 - cooldown) * final_lr_frac

Cosine decay

def get_lr(step, total_steps, max_lr, min_lr, warmup_steps):
    if step < warmup_steps:
        return max_lr * step / warmup_steps
    progress = (step - warmup_steps) / (total_steps - warmup_steps)
    return min_lr + 0.5 * (max_lr - min_lr) * (1 + math.cos(math.pi * progress))

WSD (Warmup-Stable-Decay): gaining traction — easier to resume training mid-run.

Guidance

  • Warmup: 1-5% of training. Zero warmup valid with Muon (autoresearch uses WARMUP_RATIO=0.0).
  • Warmdown: 30-50% of training in LR decay. Matters more than warmup for final quality.
  • Final LR: 0 or ~10% of peak. Zero is simpler.

Mixed Precision & Compilation

import os
os.environ["PYTORCH_ALLOC_CONF"] = "expandable_segments:True"  # before torch import

import torch
torch.set_float32_matmul_precision("high")
autocast_ctx = torch.amp.autocast(device_type="cuda", dtype=torch.bfloat16)
model = torch.compile(model, dynamic=False)
  • bf16 (Ampere+): same exponent as fp32, no loss scaling needed. Preferred over fp16.
  • fp16: needs GradScaler. Use only on V100 or older.
  • dynamic=False enables max optimization. Add fullgraph=True if no graph breaks.
  • First steps are slow (JIT) — exclude from timing.

Memory & Performance

Meta device init (large models)

with torch.device("meta"):
    model = GPT(config)          # zero memory
model.to_empty(device="cuda")
model.init_weights()

MFU (Model FLOPs Utilization)

achieved_flops = model_flops_per_token * batch_tokens / step_time
mfu = achieved_flops / gpu_peak_flops
# H100 SXM: 989.5 TFLOPS | A100: 312 | RTX 4090: 165

Good targets: >30% decent, >40% good, >50% excellent (single-GPU).

OOM solutions (in order)

  1. Reduce DEVICE_BATCH_SIZE, increase grad_accum_steps
  2. PYTORCH_ALLOC_CONF=expandable_segments:True
  3. model.zero_grad(set_to_none=True)
  4. Meta device init → to_empty
  5. Activation checkpointing: torch.utils.checkpoint.checkpoint()
  6. 8-bit optimizer (bitsandbytes): ~30% savings on optimizer states

Hyperparameter Search

Priority order (tune first → last)

  1. Learning rate — most impactful. Always tune first.
  2. Batch size — largest that fits. Speed knob, not quality knob.
  3. Weight decay — 0.01-0.1 for AdamW.
  4. Warmup steps — 1-5% of training.

The 2025 default recipe

SettingValue
OptimizerAdamW (β1=0.9, β2=0.95, eps=1e-10)
Weight decay0.1
LR scheduleCosine decay or WSD
Peak LR3e-4 (scale down for larger models)
Precisionbf16
Grad clippingmax_norm=1.0
NormalizationRMSNorm (pre-norm)
ActivationSwiGLU
Position encodingRoPE
AttentionFlash Attention, optionally GQA

Debugging Checklist

Karpathy's recipe (still canonical)

  1. Become one with the data — visualize, check distributions, verify labels
  2. Get end-to-end running first — verify on a trivial case
  3. Overfit one batch — if you can't, you have a bug
  4. Then regularize — add regularization only after overfitting works
  5. Tune hyperparameters — start with known defaults

Loss exploding / NaN

  1. Reduce LR (3-10× smaller)
  2. Add gradient clipping: clip_grad_norm_(params, 1.0)
  3. Check for inf/nan in inputs
  4. Add logit soft capping: softcap * tanh(logits / softcap)
  5. Add QK-norm in attention
  6. Verify weight init (zero-init output projections?)
  7. Check loss reduction with gradient accumulation (loss / grad_accum_steps)

Slow training / Low MFU

  1. Verify torch.compile is active
  2. Check torch.set_float32_matmul_precision("high")
  3. Pin memory + non_blocking transfers
  4. Profile with torch.profiler
  5. GC stalls? gc.freeze(); gc.disable()
  6. Tensor Core alignment: dims multiples of 8/64

Loss plateau / Slow convergence

  1. LR too low — try 2-5× larger
  2. Warmup too long
  3. Weight decay too high
  4. Verify LR schedule is actually applied (print each step)
  5. Model too small for task

Silent failures

  1. Data leakage between train/val
  2. Wrong preprocessing at inference — augmentation mismatch
  3. Label errors — use cleanlab to detect
  4. Shuffling bugs — correlated batches
  5. Tokenizer mismatch with pretrained model

What to monitor

  • Gradient norms — spike precedes loss spike
  • Per-layer activation stats — reveals exploding/vanishing
  • Dead neurons — >50% zero ReLU = dying ReLU problem
  • Learning rate — verify schedule applied (common silent bug)

Experiment Management

Track experiments in TSV for easy comparison:

commit  val_bpb  memory_gb  status   description
a1b2c3d 0.9979   44.0       keep     baseline
b2c3d4e 0.9932   44.2       keep     increase matrix LR to 0.04
c3d4e5f 1.0050   44.0       discard  switch to GeLU (worse)

Simplicity criterion: all else equal, simpler is better. Removing something and getting equal results is a great outcome. For systematic agent-driven experimentation, see references/experiment-loop.md.

Evaluation metrics by domain

DomainPrimary MetricNotes
LLMBPB (bits per byte)Vocab-size-independent
ClassificationAccuracy / F1Macro-F1 for imbalanced
SegmentationmIoU / DicePer-class IoU reveals weak spots
GenerationFIDNeeds >10k samples
RegressionRMSE / MAELog-transform skewed targets

Frequently asked questions about ML Training Recipes

Similar skills