opencode-skills-collection 3.1.12 → 3.1.14

package/bundled-skills/transformers-js/references/PIPELINE_OPTIONS.md

@@ -32,26 +32,26 @@ The third parameter, `options`, allows you to configure how the model is loaded
 interface PretrainedModelOptions {
   // Progress tracking
   progress_callback?: (info: ProgressInfo) => void;
-  
+
   // Model configuration
   config?: PretrainedConfig;
-  
+
   // Cache and loading
   cache_dir?: string;
   local_files_only?: boolean;
   revision?: string;
-  
+
   // Model-specific settings
   subfolder?: string;
   model_file_name?: string;
-  
+
   // Device and performance
   device?: DeviceType | Record<string, DeviceType>;
   dtype?: DataType | Record<string, DataType>;
-  
+
   // External data format (large models)
   use_external_data_format?: boolean | number | Record<string, boolean | number>;
-  
+
   // ONNX Runtime settings
   session_options?: InferenceSession.SessionOptions;
 }
@@ -61,18 +61,25 @@ interface PretrainedModelOptions {
 
 ### Progress Callback
 
-Track model download and loading progress. **Note:** Models consist of multiple files (model weights, config, tokenizer, etc.), and each file reports its own progress:
+Track model download and loading progress. **Recommended:** use `progress_total` for end-to-end progress, and optionally use `progress` for per-file details.
 
 ```javascript
 const fileProgress = {};
 
 const pipe = await pipeline('sentiment-analysis', null, {
   progress_callback: (info) => {
+    // Recommended: end-to-end loading progress
+    if (info.status === 'progress_total') {
+      console.log(`Total: ${info.progress.toFixed(1)}%`);
+      return;
+    }
+
+    // Optional: per-file progress
     if (info.status === 'progress') {
       fileProgress[info.file] = info.progress;
       console.log(`${info.file}: ${info.progress.toFixed(1)}%`);
     }
-    
+
     if (info.status === 'done') {
       console.log(`✓ ${info.file} complete`);
     }
@@ -84,10 +91,10 @@ const pipe = await pipeline('sentiment-analysis', null, {
 
 ```typescript
 type ProgressInfo = {
-  status: 'initiate' | 'download' | 'progress' | 'done' | 'ready';
+  status: 'initiate' | 'download' | 'progress' | 'progress_total' | 'done' | 'ready';
   name: string;       // Model id or path
-  file: string;       // File being processed
-  progress?: number;  // Percentage (0-100, only for 'progress' status)
+  file?: string;      // File being processed (per-file events)
+  progress?: number;  // Percentage (0-100, for 'progress' and 'progress_total')
   loaded?: number;    // Bytes downloaded (only for 'progress' status)
   total?: number;     // Total bytes (only for 'progress' status)
 };
@@ -102,6 +109,11 @@ const fileProgressBars = {};
 
 const pipe = await pipeline('image-classification', null, {
   progress_callback: (info) => {
+    if (info.status === 'progress_total') {
+      statusDiv.textContent = `Total: ${info.progress.toFixed(1)}%`;
+      return;
+    }
+
     if (info.status === 'progress') {
       // Create progress bar for each file if not exists
       if (!fileProgressBars[info.file]) {
@@ -115,15 +127,15 @@ const pipe = await pipeline('image-classification', null, {
         progressContainer.appendChild(fileDiv);
         fileProgressBars[info.file] = fileDiv.querySelector('.progress-fill');
       }
-      
+
       // Update progress bar
       fileProgressBars[info.file].style.width = `${info.progress}%`;
-      
+
       const mb = (info.loaded / 1024 / 1024).toFixed(2);
       const totalMb = (info.total / 1024 / 1024).toFixed(2);
       statusDiv.textContent = `${info.file}: ${mb}/${totalMb} MB`;
     }
-    
+
     if (info.status === 'ready') {
       statusDiv.textContent = 'Model ready!';
     }
@@ -289,7 +301,7 @@ const pipe = await pipeline('sentiment-analysis', 'model-id', {
 
 **Common devices:**
 - `'wasm'` - WebAssembly (CPU, most compatible)
-- `'webgpu'` - WebGPU (GPU, faster in browsers)
+- `'webgpu'` - WebGPU (GPU acceleration in supported runtimes)
 - `'cpu'` - CPU
 - `'gpu'` - Auto-detect GPU
 - `'cuda'` - NVIDIA CUDA (Node.js with GPU)
@@ -310,8 +322,8 @@ const pipe = await pipeline('automatic-speech-recognition', 'model-id', {
 ```
 
 **WebGPU Requirements:**
-- Chrome/Edge 113+
-- Enable chrome://flags/#enable-unsafe-webgpu (if needed)
+- Runtime with WebGPU support (browser, Node.js, Bun, or Deno)
+- Compatible hardware/driver stack
 - Adequate GPU memory
 
 
@@ -464,8 +476,8 @@ import { pipeline } from '@huggingface/transformers';
 
 const pipe = await pipeline('sentiment-analysis', null, {
   progress_callback: (info) => {
-    if (info.status === 'progress') {
-      console.log(`${info.file}: ${info.progress.toFixed(1)}%`);
+    if (info.status === 'progress_total') {
+      console.log(`Total: ${info.progress.toFixed(1)}%`);
     }
   }
 });

package/bundled-skills/transformers-js/references/TEXT_GENERATION.md

@@ -74,19 +74,19 @@ await generator('Tell me a story', {
   <div id="output"></div>
 
   <script type="module">
-    import { pipeline, TextStreamer } from 'https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.8.1';
-    
+    import { pipeline, TextStreamer } from 'https://cdn.jsdelivr.net/npm/@huggingface/transformers@4';
+
     const generator = await pipeline(
       'text-generation',
       'onnx-community/Qwen2.5-0.5B-Instruct',
       { dtype: 'q4' }
     );
-    
+
     window.generate = async function() {
       const prompt = document.getElementById('prompt').value;
       const outputDiv = document.getElementById('output');
       outputDiv.textContent = '';
-      
+
       const streamer = new TextStreamer(generator.tokenizer, {
         skip_prompt: true,
         skip_special_tokens: true,
@@ -94,7 +94,7 @@ await generator('Tell me a story', {
           outputDiv.textContent += token;
         },
       });
-      
+
       await generator(prompt, {
         max_new_tokens: 200,
         temperature: 0.7,
@@ -119,10 +119,10 @@ function StreamingGenerator() {
 
   const handleGenerate = async (prompt) => {
     if (!prompt) return;
-    
+
     setLoading(true);
     setOutput('');
-    
+
     // Load model on first generate
     if (!generatorRef.current) {
       generatorRef.current = await pipeline(
@@ -131,7 +131,7 @@ function StreamingGenerator() {
         { dtype: 'q4' }
       );
     }
-    
+
     const streamer = new TextStreamer(generatorRef.current.tokenizer, {
       skip_prompt: true,
       skip_special_tokens: true,
@@ -145,7 +145,7 @@ function StreamingGenerator() {
       temperature: 0.7,
       streamer,
     });
-    
+
     setLoading(false);
   };
 
@@ -225,17 +225,17 @@ await generator(prompt, {
   // Token limits
   max_new_tokens: 512,        // Maximum tokens to generate
   min_new_tokens: 0,          // Minimum tokens to generate
-  
+
   // Sampling
   temperature: 0.7,           // Randomness (0.0-2.0)
   top_k: 50,                  // Consider top K tokens
   top_p: 0.95,                // Nucleus sampling
   do_sample: true,            // Use random sampling (false = always pick most likely token)
-  
+
   // Repetition control
   repetition_penalty: 1.0,    // Penalty for repeating (1.0 = no penalty)
   no_repeat_ngram_size: 0,    // Prevent repeating n-grams
-  
+
   // Streaming
   streamer: streamer,         // TextStreamer instance
 });
@@ -260,23 +260,23 @@ await generator(prompt, { temperature: 1.2, max_new_tokens: 100 });
 
 ```javascript
 // Greedy (deterministic)
-await generator(prompt, { 
+await generator(prompt, {
   do_sample: false,
-  max_new_tokens: 100 
+  max_new_tokens: 100
 });
 
 // Top-k sampling
-await generator(prompt, { 
+await generator(prompt, {
   top_k: 50,
   temperature: 0.7,
-  max_new_tokens: 100 
+  max_new_tokens: 100
 });
 
 // Top-p (nucleus) sampling
-await generator(prompt, { 
+await generator(prompt, {
   top_p: 0.95,
   temperature: 0.7,
-  max_new_tokens: 100 
+  max_new_tokens: 100
 });
 ```
 

package/bundled-skills/trl-training/SKILL.md

@@ -0,0 +1,327 @@
+---
+name: trl-training
+description: Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.
+risk: unknown
+source: https://github.com/huggingface/skills/tree/main/skills/trl-training
+source_repo: huggingface/skills
+source_type: official
+date_added: 2026-07-01
+license: Apache-2.0
+license_source: https://github.com/huggingface/skills/blob/main/LICENSE
+---
+
+# TRL Training Skill
+## When to Use
+
+Use this skill when you need train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.
+
+
+You are an expert at using the TRL (Transformers Reinforcement Learning) library to train and fine-tune large language models.
+
+## Overview
+
+TRL provides CLI commands for post-training foundation models using state-of-the-art techniques:
+
+- **SFT** (Supervised Fine-Tuning): Fine-tune models on instruction-following or conversational datasets
+- **DPO** (Direct Preference Optimization): Align models using preference data
+- **GRPO** (Group Relative Policy Optimization): Train models by ranking multiple sampled outputs relative to each other and optimizing based on their comparative rewards.
+- **RLOO** (Reinforce Leave One Out): Online RL training with generation-based rewards
+- **Reward Model Training**: Train reward models for RLHF
+
+TRL is built on top of Hugging Face Transformers and Accelerate, providing seamless integration with the Hugging Face ecosystem.
+
+## Core Commands
+
+### trl sft - Supervised Fine-Tuning
+
+Fine-tune language models on instruction-following or conversational datasets.
+
+**Full training:**
+
+```bash
+trl sft \
+  --model_name_or_path Qwen/Qwen2-0.5B \
+  --dataset_name trl-lib/Capybara \
+  --learning_rate 2.0e-5 \
+  --num_train_epochs 1 \
+  --packing \
+  --per_device_train_batch_size 2 \
+  --gradient_accumulation_steps 8 \
+  --eos_token '<|im_end|>' \
+  --eval_strategy steps \
+  --eval_steps 100 \
+  --output_dir Qwen2-0.5B-SFT \
+  --push_to_hub
+```
+
+**Train with LoRA adapters:**
+
+```bash
+trl sft \
+  --model_name_or_path Qwen/Qwen2-0.5B \
+  --dataset_name trl-lib/Capybara \
+  --learning_rate 2.0e-4 \
+  --num_train_epochs 1 \
+  --packing \
+  --per_device_train_batch_size 2 \
+  --gradient_accumulation_steps 8 \
+  --eos_token '<|im_end|>' \
+  --eval_strategy steps \
+  --eval_steps 100 \
+  --use_peft \
+  --lora_r 32 \
+  --lora_alpha 16 \
+  --output_dir Qwen2-0.5B-SFT \
+  --push_to_hub
+```
+
+### trl dpo - Direct Preference Optimization
+
+Align models using preference data (chosen/rejected pairs).
+
+**Full training:**
+
+```bash
+trl dpo \
+  --dataset_name trl-lib/ultrafeedback_binarized \
+  --model_name_or_path Qwen/Qwen2-0.5B-Instruct \
+  --learning_rate 5.0e-7 \
+  --num_train_epochs 1 \
+  --per_device_train_batch_size 2 \
+  --max_steps 1000 \
+  --gradient_accumulation_steps 8 \
+  --eval_strategy steps \
+  --eval_steps 50 \
+  --output_dir Qwen2-0.5B-DPO \
+  --no_remove_unused_columns
+```
+
+**Train with LoRA adapters:**
+
+```bash
+trl dpo \
+  --dataset_name trl-lib/ultrafeedback_binarized \
+  --model_name_or_path Qwen/Qwen2-0.5B-Instruct \
+  --learning_rate 5.0e-6 \
+  --num_train_epochs 1 \
+  --per_device_train_batch_size 2 \
+  --max_steps 1000 \
+  --gradient_accumulation_steps 8 \
+  --eval_strategy steps \
+  --eval_steps 50 \
+  --output_dir Qwen2-0.5B-DPO \
+  --no_remove_unused_columns \
+  --use_peft \
+  --lora_r 32 \
+  --lora_alpha 16
+```
+
+### trl grpo - Group Relative Policy Optimization
+
+Train models using reward functions or LLM-as-a-judge for evaluating generations and providing rewards.
+
+**Basic usage:**
+
+```bash
+trl grpo \
+  --model_name_or_path Qwen/Qwen2.5-0.5B \
+  --dataset_name trl-lib/gsm8k \
+  --reward_funcs accuracy_reward \
+  --output_dir Qwen2-0.5B-GRPO \
+  --push_to_hub
+```
+
+### trl rloo - Reinforce Leave One Out
+
+Online RL training where the model generates text and receives rewards based on custom criteria.
+
+**Basic usage:**
+
+```bash
+trl rloo \
+  --model_name_or_path Qwen/Qwen2.5-0.5B \
+  --dataset_name trl-lib/tldr \
+  --reward_model_name_or_path sentiment-analysis:nlptown/bert-base-multilingual-uncased-sentiment \
+  --output_dir Qwen2-0.5B-RLOO \
+  --push_to_hub
+```
+
+### trl reward - Reward Model Training
+
+Train a reward model to score text quality for RLHF.
+
+**Full training:**
+
+```bash
+trl reward \
+  --model_name_or_path Qwen/Qwen2-0.5B-Instruct \
+  --dataset_name trl-lib/ultrafeedback_binarized \
+  --output_dir Qwen2-0.5B-Reward \
+  --per_device_train_batch_size 8 \
+  --num_train_epochs 1 \
+  --learning_rate 1.0e-5 \
+  --eval_strategy steps \
+  --eval_steps 50 \
+  --max_length 2048
+```
+
+**Train with LoRA adapters:**
+
+```bash
+trl reward \
+  --model_name_or_path Qwen/Qwen2-0.5B-Instruct \
+  --dataset_name trl-lib/ultrafeedback_binarized \
+  --output_dir Qwen2-0.5B-Reward-LoRA \
+  --per_device_train_batch_size 8 \
+  --num_train_epochs 1 \
+  --learning_rate 1.0e-4 \
+  --eval_strategy steps \
+  --eval_steps 50 \
+  --max_length 2048 \
+  --use_peft \
+  --lora_task_type SEQ_CLS \
+  --lora_r 32 \
+  --lora_alpha 16
+```
+
+## Configuration Files
+
+TRL supports YAML configuration files for reproducible training. All CLI arguments can be specified in a config file.
+
+**Example config (sft_config.yaml):**
+
+```yaml
+model_name_or_path: Qwen/Qwen2.5-0.5B
+dataset_name: trl-lib/Capybara
+learning_rate: 2.0e-5
+num_train_epochs: 1
+per_device_train_batch_size: 8
+gradient_accumulation_steps: 2
+output_dir: ./sft_output
+use_peft: true
+lora_r: 16
+lora_alpha: 16
+report_to: trackio
+```
+
+**Launch with config:**
+
+```bash
+trl sft --config sft_config.yaml
+```
+
+**Override config values:**
+
+```bash
+trl sft --config sft_config.yaml --learning_rate 1.0e-5
+```
+
+## Distributed Training
+
+TRL integrates with Accelerate for multi-GPU and multi-node training.
+
+**Multi-GPU training:**
+
+```bash
+trl sft \
+  --config sft_config.yaml \
+  --num_processes 4
+```
+
+**Use predefined Accelerate configs:**
+
+TRL provides predefined configs: `single_gpu`, `multi_gpu`, `fsdp1`, `fsdp2`, `zero1`, `zero2`, `zero3`
+
+```bash
+trl sft \
+  --config sft_config.yaml \
+  --accelerate_config zero2
+```
+
+**Custom Accelerate config:**
+
+```bash
+# Generate custom config
+accelerate config
+
+# Use custom config
+trl sft --config sft_config.yaml --config_file ~/.cache/huggingface/accelerate/default_config.yaml
+```
+
+**Fully Sharded Data Parallel (FSDP):**
+
+```bash
+trl sft --config sft_config.yaml --accelerate_config fsdp2
+```
+
+**DeepSpeed ZeRO:**
+
+```bash
+trl sft --config sft_config.yaml --accelerate_config zero3
+```
+
+## Troubleshooting
+
+### CUDA Out of Memory
+
+- Reduce `--per_device_train_batch_size` and increase `--gradient_accumulation_steps`
+- Enable `--use_peft` for LoRA training
+- Use `--gradient_checkpointing` to save memory
+- Try smaller model or longer sequence truncation
+
+### Dataset Loading Issues
+
+- Verify dataset exists: check Hugging Face Hub or local path
+- Check dataset format matches expected columns
+- Use `--dataset_config` for multi-config datasets
+- Inspect dataset: `from datasets import load_dataset; ds = load_dataset(name)`
+
+### Model Loading Issues
+
+- Verify model exists on Hugging Face Hub
+- Check if gated model requires authentication: `hf auth login`
+- For local models, provide absolute path
+- Ensure sufficient disk space and memory
+
+### Slow Training
+
+- Enable dataset `--packing` for short sequences
+- Use larger `--per_device_train_batch_size` if memory allows
+- Enable `--tf32` for faster computation on Ampere GPUs
+- Use `--bf16` on supported hardware
+- Consider multi-GPU training with `--num_processes`
+
+### Generation Issues (GRPO/RLOO)
+
+- Check prompt format in dataset
+- Adjust `--temperature` and `--top_p` for generation
+- Verify the reward function (for GRPO/RLOO)
+
+## Additional Resources
+
+- **Documentation**: https://huggingface.co/docs/trl
+- **GitHub**: https://github.com/huggingface/trl
+- **Examples**: https://github.com/huggingface/trl/tree/main/examples
+
+## Best Practices
+
+1. **Start with SFT**: Always fine-tune base models with SFT before preference alignment
+2. **Use LoRA for efficiency**: Enable `--use_peft` for faster training and lower memory
+3. **Monitor training**: Use `--report_to trackio` (or `--report_to wandb` or `--report_to tensorboard`) for tracking
+4. **Save checkpoints**: TRL automatically saves checkpoints in `--output_dir`
+5. **Test on small datasets first**: Verify pipeline works before full training
+6. **Use configuration files**: Create YAML configs for reproducibility
+7. **Leverage Accelerate**: Use multi-GPU training for faster iteration
+
+When helping users with TRL:
+- Always check which training method is appropriate for their use case
+- Verify dataset format matches the expected schema
+- Recommend starting with smaller models for testing
+- Suggest LoRA for resource-constrained environments
+- Point to specific documentation sections for advanced features
+
+## Limitations
+
+- Use this skill only when the task clearly matches its upstream product or API scope.
+- Verify commands, API behavior, pricing, quotas, credentials, and deployment effects against current official documentation before making changes.
+- Do not treat generated examples as a substitute for environment-specific tests, security review, or user approval for destructive or costly actions.