FSSCoding
50ddaa4b39
Root cause: Pre-trained model was loading successfully, causing CLI to skip
calibration entirely. System went straight to classification with 35% model.
Changes:
- config: Set calibration_model to qwen3:8b-q4_K_M (larger model for better instruction following)
- cli: Create separate calibration_llm provider with 8b model
- llm_analyzer: Improved prompt to force exact email ID copying
- workflow: Merge discovered categories with predefined ones
- workflow: Add detailed error logging for label mismatches
- ml_classifier: Fixed model path checking (was checking None parameter)
- ml_classifier: Add dual API support (sklearn predict_proba vs LightGBM predict)
- ollama: Fixed model list parsing (use m.model not m.get('name'))
- feature_extractor: Switch to Ollama embeddings (instant vs 90s load time)
Result: Calibration now runs and generates 16 categories + 50 labels correctly.
Next: Investigate calibration sampling to reduce overfitting on small samples.
82 lines
1.5 KiB
YAML
82 lines
1.5 KiB
YAML
version: "1.0.0"
|
|
|
|
calibration:
|
|
sample_size: 1500
|
|
sample_strategy: "stratified"
|
|
validation_size: 300
|
|
min_confidence: 0.6
|
|
|
|
processing:
|
|
batch_size: 100
|
|
llm_queue_size: 100
|
|
parallel_workers: 4
|
|
checkpoint_interval: 1000
|
|
checkpoint_dir: "checkpoints"
|
|
|
|
classification:
|
|
default_threshold: 0.75
|
|
min_threshold: 0.60
|
|
max_threshold: 0.90
|
|
adjustment_step: 0.05
|
|
adjustment_frequency: 1000
|
|
category_thresholds:
|
|
junk: 0.85
|
|
auth: 0.90
|
|
transactional: 0.80
|
|
newsletters: 0.75
|
|
conversational: 0.65
|
|
|
|
llm:
|
|
provider: "ollama"
|
|
fallback_enabled: true
|
|
|
|
ollama:
|
|
base_url: "http://localhost:11434"
|
|
calibration_model: "qwen3:8b-q4_K_M"
|
|
classification_model: "qwen3:1.7b"
|
|
temperature: 0.1
|
|
max_tokens: 2000
|
|
timeout: 30
|
|
retry_attempts: 3
|
|
|
|
openai:
|
|
base_url: "https://api.openai.com/v1"
|
|
api_key: "${OPENAI_API_KEY}"
|
|
calibration_model: "gpt-4o-mini"
|
|
classification_model: "gpt-4o-mini"
|
|
temperature: 0.1
|
|
max_tokens: 500
|
|
|
|
email_providers:
|
|
gmail:
|
|
batch_size: 100
|
|
microsoft:
|
|
batch_size: 100
|
|
imap:
|
|
timeout: 30
|
|
batch_size: 50
|
|
|
|
features:
|
|
text_features:
|
|
max_vocab_size: 10000
|
|
ngram_range: [1, 2]
|
|
min_df: 2
|
|
max_df: 0.95
|
|
embedding_model: "all-MiniLM-L6-v2"
|
|
embedding_batch_size: 32
|
|
|
|
export:
|
|
format: "json"
|
|
include_confidence: true
|
|
create_report: true
|
|
output_dir: "results"
|
|
|
|
logging:
|
|
level: "INFO"
|
|
file: "logs/email-sorter.log"
|
|
|
|
cleanup:
|
|
delete_temp_files: true
|
|
delete_repo_after: false
|
|
temp_dir: ".email-sorter-tmp"
|