discourses 1.0.6

Official Python SDK for Discourses — Era-calibrated financial sentiment analysis

discourses.io Python SDK

Official Python SDK for discourses.io — the financial social media platform and financial language processing API

📄 Read the Whitepaper   ·   🔬 Methodology   ·   📖 Documentation   ·   🔑 Get API Key

---

Why discourses.io?

discourses.io is a financial social media platform and financial language processing API built for the modern market.

Traditional sentiment analysis treats language as static. But financial language evolves—"diamond hands" meant nothing in 2008, but signals strong conviction today. discourses.io solves this with era-calibrated lexicons built on peer-reviewed academic research.

Era	Period	What It Captures
primitive	Pre-2015	Traditional financial vocabulary, formal market language
meme	2015–2021	Rise of retail trading, WSB culture, crypto slang
present	2021–now	Modern market discourse, post-meme institutionalization

---

Installation

pip install discourses

---

Quick Start

from discourses import Discourses

client = Discourses(api_key="your-api-key")
result = client.analyze("Strong growth with excellent outlook ahead")

print(result.label)      # "very_bullish"
print(result.outlook)    # 0.99

---

API Reference

🔑 Initialize the Client

Every request requires authentication with your API key. Get yours at discourses.io/dashboard.

from discourses import Discourses

client = Discourses(api_key="your-api-key")

Parameter	Type	Default	Description
api_key	str	required	Your discourses.io API key
base_url	str	https://discourses.io/api/v1	API base URL
timeout	int	30	Request timeout in seconds

---

Endpoints

📊 Analyze — Single Text Sentiment

POST /analyze — Documentation

Analyze sentiment of any financial text. Returns classification, confidence, and detailed sentiment scores.

from discourses import Discourses

# Initialize client
client = Discourses(api_key="your-api-key")

# Analyze text
result = client.analyze("Strong growth with excellent outlook ahead")

# Access the results
print(f"Label:      {result.label}")
print(f"Confidence: {result.confidence:.2%}")
print(f"Outlook:    {result.outlook:.4f}")
print(f"Bullish:    {result.scores['bullish']:.4f}")
print(f"Bearish:    {result.scores['bearish']:.4f}")
print(f"Words:      {result.word_count}")

Expected Output:

Label:      very_bullish
Confidence: 80.05%
Outlook:    0.9257
Bullish:    0.4917
Bearish:    0.2981
Words:      6

Response Object: AnalysisResult

Field	Type	Description
label	str	"very_bullish", "bullish", "neutral", "bearish", "very_bearish"
confidence	float	Model confidence (0.0 to 1.0)
outlook	float	Overall sentiment score (0.0 to 1.0)
scores	dict	Detailed scores: bullish, bearish, neutral, confusion
word_count	int	Number of words analyzed
matched_count	int	Sentiment words matched

Helper Properties:

• result.is_bullish → True if label contains "bullish"
• result.is_bearish → True if label contains "bearish"
• result.is_neutral → True if label is "neutral"

---

🔄 Compare Eras — Cross-Era Analysis

POST /analyze/compare-eras — Documentation

Analyze how the same text is interpreted across different market eras. Essential for understanding semantic drift and backtesting strategies.

from discourses import Discourses

# Initialize client
client = Discourses(api_key="your-api-key")

# Compare sentiment across eras
result = client.compare_eras(
    text="This stock is going to the moon! Diamond hands!",
    eras=["primitive", "meme", "present"]
)

# View per-era results
print("Era-by-Era Sentiment:")
print("-" * 50)
for era, data in result.results.items():
    label = data['classification']['label']
    outlook = data['scores']['outlook']
    print(f"  {era:12} {label:14} (outlook: {outlook:.4f})")

# Check semantic drift
print()
print(f"Drift Direction: {result.drift['direction']}")
print(f"Drift Magnitude: {result.drift['magnitude']:.4f}")
print(f"Peak Era:        {result.drift['peak_era']}")

Expected Output:

Era-by-Era Sentiment:
--------------------------------------------------
  primitive    bullish        (outlook: 0.6912)
  meme         very_bullish   (outlook: 0.9933)
  present      very_bullish   (outlook: 0.9977)

Drift Direction: positive_shift
Drift Magnitude: 0.3065
Peak Era:        present

💡 Notice how "diamond hands" and "to the moon" have much stronger bullish signals in the meme and present eras compared to primitive!

Response Object: CompareResult

Field	Type	Description
results	dict	Per-era analysis results
drift	dict	Semantic drift metrics
meta	dict	Processing metadata

Drift Fields:

Field	Type	Description
direction	str	"positive_shift", "negative_shift", or "stable"
magnitude	float	Size of the drift (0.0 to 1.0)
min_era	str	Era with lowest sentiment
peak_era	str	Era with highest sentiment

---

⚡ Batch — High-Volume Processing

POST /analyze/batch — Documentation

Efficiently analyze multiple texts in a single request. Perfect for processing feeds, historical data, or real-time streams.

from discourses import Discourses

# Initialize client
client = Discourses(api_key="your-api-key")

# Texts to analyze (with custom IDs)
texts = [
    {"id": "post_1", "text": "Diamond hands! This is going to the moon"},
    {"id": "post_2", "text": "Bearish on this one, expecting a pullback"},
    {"id": "post_3", "text": "Strong growth with excellent outlook ahead"},
]

# Batch analyze with specific era
result = client.batch(texts=texts, era="meme")

# Process results
print("Batch Sentiment Analysis")
print("=" * 55)
for post_id, data in result.results.items():
    label = data['classification']['label']
    outlook = data['scores']['outlook']
    emoji = "🟢" if "bullish" in label else "🔴" if "bearish" in label else "⚪"
    print(f"{emoji} {post_id}: {label:14} (outlook: {outlook:.4f})")

print()
print(f"Processed: {result.meta['texts_processed']}")
print(f"Failed:    {result.meta['texts_failed']}")

Expected Output:

Batch Sentiment Analysis
=======================================================
🟢 post_1: bullish        (outlook: 0.6735)
🟢 post_2: very_bullish   (outlook: 0.9353)
🟢 post_3: very_bullish   (outlook: 0.9257)

Processed: 3
Failed:    0

Response Object: BatchResult

Field	Type	Description
results	dict	Results keyed by your custom IDs
meta	dict	Processing metadata

Meta Fields:

Field	Type	Description
era	str	Era used for analysis
texts_processed	int	Successfully processed count
texts_failed	int	Failed count
processing_time_ms	int	Total processing time

---

Era Selection Guide

Use Case	Recommended Era
Modern social media sentiment	present
Meme stock / crypto analysis	meme
Traditional financial news	primitive
Understanding semantic drift	compare_eras()

# Modern analysis
client.analyze(text, era="present")

# Meme-era analysis (WSB, crypto culture)
client.analyze(text, era="meme")

# Traditional financial language
client.analyze(text, era="primitive")

# Compare across all eras
client.compare_eras(text, eras=["primitive", "meme", "present"])

---

Error Handling

The SDK provides typed exceptions for robust error handling:

from discourses import (
    Discourses,
    AuthenticationError,
    RateLimitError,
    ValidationError,
    DiscoursesError,
)

client = Discourses(api_key="your-api-key")

try:
    result = client.analyze("Market analysis text")
    
except AuthenticationError:
    # Invalid or expired API key
    print("Check your API key at https://discourses.io/dashboard")
    
except RateLimitError as e:
    # Too many requests
    print(f"Rate limited. Retry after {e.retry_after} seconds")
    
except ValidationError as e:
    # Invalid input (empty text, too long, etc.)
    print(f"Invalid input: {e.message}")
    
except DiscoursesError as e:
    # Catch-all for other API errors
    print(f"API error: {e}")

---

Links

📄	Whitepaper Academic research & methodology	📖	Documentation Full API reference
🔬	Methodology How era-calibration works	🔑	Dashboard Get your API key
💬	Issues Report bugs & requests	📦	PyPI Package & versions

---

License

MIT License — see LICENSE for details.

---

_{Built with ❤️ by discourses.io}