Fixed
Override Logic Mismatches: Fixed critical Enum-to-String type mismatch in apply_trend_override that was silencing the "Safety Valve" logic. Data Pipeline Failures: Injected robust error handling and type checking in market_analyst.py to identify why RegimeDetector receives invalid data (causing "UNKNOWN" regimes). Gemini 404 Errors: Removed invalid/deprecated model names causing 404s
This commit is contained in:
swj.premkumar
parent
a6e4c9b770
commit
e88a01d0ea
11
CHANGELOG.md
11
CHANGELOG.md
|
|
@ -2,6 +2,17 @@
|
|||
|
||||
All notable changes to the **TradingAgents** project will be documented in this file.
|
||||
|
||||
## [Unreleased] - 2026-01-11
|
||||
|
||||
### Added
|
||||
- **Gemini 2.0 & 3.0 Support**: Updated `cli/utils.py` to support `gemini-2.0-flash`, `gemini-2.5-flash-lite`, `gemini-2.5-pro`, `gemini-3-flash-preview` and `gemini-3-pro-preview` models.
|
||||
- **Console Debugging**: Added explicit console print statements for critical "Smoking Gun" debug traces in `market_analyst.py` and `trading_graph.py`.
|
||||
|
||||
### Fixed
|
||||
- **Override Logic Mismatches**: Fixed critical Enum-to-String type mismatch in `apply_trend_override` that was silencing the "Safety Valve" logic.
|
||||
- **Data Pipeline Failures**: Injected robust error handling and type checking in `market_analyst.py` to identify why `RegimeDetector` receives invalid data (causing "UNKNOWN" regimes).
|
||||
- **Gemini 404 Errors**: Removed invalid/deprecated model names causing 404s.
|
||||
|
||||
## [Unreleased] - 2026-01-10
|
||||
|
||||
### Added
|
||||
|
|
|
|||
49
README.md
49
README.md
|
|
@ -87,10 +87,57 @@ Our framework decomposes complex trading tasks into specialized roles. This ensu
|
|||
- Continuously evaluates portfolio risk by assessing market volatility, liquidity, and other risk factors. The risk management team evaluates and adjusts trading strategies, providing assessment reports to the Portfolio Manager for final decision.
|
||||
- The Portfolio Manager approves/rejects the transaction proposal. If approved, the order will be sent to the simulated exchange and executed.
|
||||
|
||||
<p align="center">
|
||||
<img src="assets/risk.png" width="70%" style="display: inline-block; margin: 0 2%;">
|
||||
</p>
|
||||
|
||||
### Decision Logic & "Safety Valve" Architecture
|
||||
|
||||
The following diagram illustrates the complete decision-making flow, including the critical **"Don't Fight the Tape" Safety Valve** which intercepts and overrides `SELL` decisions for hyper-growth stocks in strong trends.
|
||||
|
||||
```mermaid
|
||||
graph TD
|
||||
Start([Start Propagate]) --> Init[Initialize State]
|
||||
Init --> Analysts{Analyst Agents}
|
||||
|
||||
subgraph Data_Collection
|
||||
Analysts -->|Fetch| Market[Market Analyst]
|
||||
Analysts -->|Fetch| Social[Social Analyst]
|
||||
Analysts -->|Fetch| News[News Analyst]
|
||||
Analysts -->|Fetch| Fund[Fundamentals]
|
||||
end
|
||||
|
||||
Market -->|Regime: Trending/Volatile| GraphState
|
||||
Social -->|Sentiment Score| GraphState
|
||||
News -->|Macro Signals| GraphState
|
||||
Fund -->|Growth Metrics| GraphState
|
||||
|
||||
GraphState --> Debate{Debate Phase}
|
||||
Debate -->|Bull Thesis| Bull[Bull Researcher]
|
||||
Debate -->|Bear Thesis| Bear[Bear Researcher]
|
||||
|
||||
Bull --> Trader[Trader Agent]
|
||||
Bear --> Trader
|
||||
|
||||
Trader -->|Propose Trade| Risk[Risk Manager]
|
||||
Risk -->|Refine/Approve| FinalState[Final Logic State]
|
||||
|
||||
FinalState --> SafetyValve{"🛡️ SAFETY VALVE\n(Trend Override)"}
|
||||
|
||||
SafetyValve --> Check1["Price > 200 SMA?"]
|
||||
SafetyValve --> Check2["Revenue Growth > 30%?"]
|
||||
SafetyValve --> Check3["Regime != BEAR?"]
|
||||
|
||||
Check1 & Check2 & Check3 -->|ALL TRUE + Action=SELL| Override([🛑 BLOCK SELL -> FORCE HOLD])
|
||||
Check1 & Check2 & Check3 -->|ELSE| Pass([✅ Pass Through])
|
||||
|
||||
Override --> Execution[Execute Order]
|
||||
Pass --> Execution
|
||||
|
||||
style SafetyValve fill:#f96,stroke:#333,stroke-width:2px
|
||||
style Override fill:#f00,stroke:#fff,stroke-width:2px,color:#fff
|
||||
style Pass fill:#0f0,stroke:#333,stroke-width:2px
|
||||
```
|
||||
|
||||
## Installation and CLI
|
||||
|
||||
### Installation
|
||||
|
|
|
|||
|
|
@ -0,0 +1,182 @@
|
|||
# OPERATIONAL DOCTRINE: THE SURVIVAL-MOMENTUM PROTOCOL
|
||||
|
||||
**TO:** All Trading Agents, Analysts, and Portfolio Managers
|
||||
**FROM:** System Architect / Risk Control
|
||||
**DATE:** January 11, 2026
|
||||
**SUBJECT:** OPERATIONAL DOCTRINE: THE SURVIVAL-MOMENTUM PROTOCOL
|
||||
|
||||
---
|
||||
|
||||
## THE CORE MANDATE
|
||||
|
||||
We are not value investors. We are not momentum chasers. **We are Survivalists.**
|
||||
|
||||
Our goal is to **capture Alpha during paradigm shifts while guaranteeing survival during regime collapses.** We achieve this by adhering to a rigid hierarchy of logic that prioritizes **Hard Data** over Narrative and **Trend** over Opinion.
|
||||
|
||||
---
|
||||
|
||||
## I. THE HIERARCHY OF TRUTH
|
||||
|
||||
In the event of a conflict between agents or data sources, this hierarchy governs the decision:
|
||||
|
||||
1. **Hard Code Overrides (The Safety Valves):** If `Price > 200SMA` and `Growth > 30%`, the system **CANNOT** sell, regardless of the Analyst’s fear.
|
||||
2. **Mathematical Regime (The Context):** The output of the `RegimeDetector` (Volatility + ADX) is the law. If the math says **TRENDING_UP**, the LLM cannot hallucinate "Uncertainty."
|
||||
3. **Fundamental Data (The Fuel):** Revenue Growth, FCF Margins, and Insider Activity are facts. Narratives about "future potential" are opinions.
|
||||
4. **LLM Synthesis (The Narrative):** The Analyst's prose is the last filter, not the first.
|
||||
|
||||
---
|
||||
|
||||
## II. THE MOMENTUM EXCEPTION ("Don't Fight the Tape")
|
||||
|
||||
Value traps look cheap; Momentum rockets look expensive. **We do not short innovation.**
|
||||
|
||||
### The Rule
|
||||
Valuation multiples (P/E, P/S) are **irrelevant** if:
|
||||
1. The Asset is in a **Confirmed Uptrend** (Price > 200 SMA).
|
||||
2. The Asset is in **Hyper-Growth** (Revenue Growth > 30%).
|
||||
3. The Market Regime is **Bullish/Momentum**.
|
||||
|
||||
### The Consequence
|
||||
In this state, we **HOLD**. We do not "take profits" because a stock is "too high." We only sell when the Trend breaks or the Growth slows.
|
||||
|
||||
---
|
||||
|
||||
## III. THE SURVIVAL PRIORITY ("Don't Catch Knives")
|
||||
|
||||
When the regime shifts to **VOLATILE** or **TRENDING_DOWN**:
|
||||
|
||||
1. **Valuation Matters Instantly:** 100x P/S is a death sentence in a downtrend.
|
||||
2. **Insider Selling is a Siren:** If insiders sell into a downtrend, we exit immediately.
|
||||
|
||||
### The Rule
|
||||
If expected value is negative and the trend is broken, we **liquidate**. We do not "average down." We do not "buy the dip" on broken structural stories.
|
||||
|
||||
### Sizing
|
||||
In **Unknown/Volatile** regimes, position sizing must be reduced to ensure no single failure threatens the portfolio.
|
||||
|
||||
---
|
||||
|
||||
## IV. THE RELATIVE STRENGTH DOCTRINE
|
||||
|
||||
We judge assets not in a vacuum, but against the Tide.
|
||||
|
||||
* If the Market (SPY) is Flat/Choppy and the Asset is Trending Up, this is **Alpha**. We press the advantage.
|
||||
* If the Market is Up and the Asset is Flat, this is **Weakness**. We cut the laggard.
|
||||
|
||||
---
|
||||
|
||||
## V. EXECUTION DISCIPLINE
|
||||
|
||||
1. **Binary Thinking is the Enemy:** Rarely is the answer "Sell 100%" or "Buy 100%." We scale out of risks and scale into strength.
|
||||
2. **No Hallucinations:** We do not invent "SG&A explosions" to justify fear. We verify data against the source.
|
||||
3. **The Stop Loss:** A Stop Loss is not a suggestion; it is a mechanism of **survival**. It must be respected above all conviction.
|
||||
|
||||
---
|
||||
|
||||
## SUMMARY
|
||||
|
||||
* We are aggressive when the math supports velocity.
|
||||
* We are cowardly when the math signals destruction.
|
||||
* We do not have "feelings" about stocks. We have parameters.
|
||||
|
||||
**Execute.**
|
||||
|
||||
---
|
||||
|
||||
# USER MANUAL: LOGIC & RULES
|
||||
|
||||
This document details the operational logic of the Survival-Momentum Protocol. It translates the high-level doctrine into specific algorithmic rules, hard-coded overrides, and stress-test scenarios.
|
||||
|
||||
This is the "User Manual" for the machine you have built.
|
||||
|
||||
## 1. THE RULES (The Logic Engines)
|
||||
These are the fundamental laws programmed into the `RegimeDetector` and `MarketAnalyst`.
|
||||
|
||||
### Rule A: The "Price is Truth" Law
|
||||
* **Concept:** Fundamental data (Earnings, P/E) is lagging (past). Price action is leading (future).
|
||||
* **The Code Logic:**
|
||||
* **IF** `RegimeDetector` calculates **TRENDING_UP** (based on ADX > 25 + Positive Returns),
|
||||
* **THEN** the system ignores traditional valuation warnings like "Overbought RSI" or "High P/E."
|
||||
* **Why:** In a mania, "Overbought" stays overbought for months. Selling early is a failure.
|
||||
|
||||
### Rule B: The "Insider Veto" Law
|
||||
* **Concept:** Insiders know more than the algorithm.
|
||||
* **The Code Logic:**
|
||||
* **IF** Net Insider Activity is **Negative (Selling) > $50M** in the last quarter,
|
||||
* **AND** Stock Price is **Below the 50-day SMA**,
|
||||
* **THEN** Buy signals are **Disabled**.
|
||||
* **Why:** Smart money selling into a downtrend is the ultimate "Get Out" signal.
|
||||
|
||||
### Rule C: The "Relative Strength" Filter
|
||||
* **Concept:** Don't buy a boat that is sinking while the tide is rising.
|
||||
* **The Code Logic:**
|
||||
* **IF** SPY (Broad Market) is **TRENDING_UP**,
|
||||
* **BUT** Target Asset is **SIDEWAYS** or **TRENDING_DOWN**,
|
||||
* **THEN** The asset is flagged as **WEAKNESS**.
|
||||
* **Action:** The Trader must prefer Leaders (Stocks matching or beating SPY regime) over Laggards.
|
||||
|
||||
## 2. THE OVERRIDES (The Hard Gates)
|
||||
These are the Python functions in `trading_graph.py` that physically block the LLM from executing a bad decision.
|
||||
|
||||
### Override 1: The "Don't Fight the Tape" (The PLTR Fix)
|
||||
* **Trigger:** The Analyst LLM tries to **SELL** or **SHORT**.
|
||||
* **The Check:**
|
||||
1. Is Price > 200-day Simple Moving Average (SMA)?
|
||||
2. Is Revenue Growth > 30% YoY?
|
||||
3. Is Market Regime **TRENDING_UP** or **BULL**?
|
||||
* **The Intervention:** If **ALL TRUE**, the system effectively "slaps the hand" of the Trader.
|
||||
* **Result:** Order converted from **SELL** to **HOLD**.
|
||||
* **Log Output:** `🛑 TREND OVERRIDE TRIGGERED: Cannot short hyper-growth in uptrend.`
|
||||
|
||||
### Override 2: The "Falling Knife" Guard (The Zoom/Peloton Fix)
|
||||
* **Trigger:** The Analyst LLM tries to **BUY** the dip.
|
||||
* **The Check:**
|
||||
1. Is Price < 200-day SMA? (Downtrend)
|
||||
2. Is Market Regime **VOLATILE** or **TRENDING_DOWN**?
|
||||
3. Is Valuation > 50x P/S?
|
||||
* **The Intervention:** If **ALL TRUE**, the system blocks the Buy.
|
||||
* **Result:** Order converted from **BUY** to **WAIT**.
|
||||
* **Log Output:** `🛑 SAFETY VALVE TRIGGERED: Valuation too high for broken trend.`
|
||||
|
||||
## 3. SAMPLE SCENARIOS (Stress Tests)
|
||||
Here is how the system handles specific market environments compared to a standard "Value" or "Momentum" bot.
|
||||
|
||||
### Scenario A: The "Rocket Ship" (e.g., NVIDIA in 2023 / PLTR Now)
|
||||
* **The Setup:** Stock is up 200%. P/E ratio is 150x. Everyone on CNBC says it's a bubble.
|
||||
* **The Value Investor Bot:** Sells immediately. "Overvalued."
|
||||
* **The Human Trader:** Panic sells to lock in profits, then cries as it doubles again.
|
||||
* **YOUR SYSTEM:**
|
||||
* **Regime:** Detects **TRENDING_UP** (High Volatility is accepted via Momentum Exception).
|
||||
* **Analyst:** Screams "Valuation Risk!"
|
||||
* **Override:** Checks Growth > 30% + Price > 200SMA.
|
||||
* **Decision:** **HOLD**.
|
||||
* **Outcome:** You ride the bubble until the trend actually breaks.
|
||||
|
||||
### Scenario B: The "Tech Crash" (e.g., ZOOM in 2022)
|
||||
* **The Setup:** Stock was $500, now $400. P/S is still 80x. Revenue growth slows from 300% to 40%.
|
||||
* **The "Dip Buyer" Bot:** Buys. "It's cheap compared to last month!"
|
||||
* **YOUR SYSTEM:**
|
||||
* **Regime:** Detects **TRENDING_DOWN** (Price < SMA, ADX High).
|
||||
* **Analyst:** "Fundamentals still look okay, maybe a buy?"
|
||||
* **Override:** Checks Price < 200SMA + Valuation (80x P/S) > Limit.
|
||||
* **Decision:** **SELL / AVOID**.
|
||||
* **Outcome:** You exit at $400 before it goes to $60.
|
||||
|
||||
### Scenario C: The "Choppy Market" (e.g., SPY in 2015)
|
||||
* **The Setup:** Market is flat. Volatility is low. No clear trend.
|
||||
* **The Momentum Bot:** Gets chopped up (Buy high, sell low) repeatedly.
|
||||
* **YOUR SYSTEM:**
|
||||
* **Regime:** Detects **SIDEWAYS** or **MEAN_REVERTING**.
|
||||
* **Indicator Selector:** Switches logic. Instead of using breakouts, it uses Bollinger Bands or RSI Mean Reversion.
|
||||
* **Decision:** Buy at Support, Sell at Resistance.
|
||||
* **Outcome:** Capital preservation during noise.
|
||||
|
||||
---
|
||||
|
||||
## SUMMARY OF DOCTRINE
|
||||
|
||||
* **In Bull Markets:** We trust the Trend. Valuation is ignored.
|
||||
* **In Bear Markets:** We trust the Math. Valuation is everything.
|
||||
* **In Uncertainty:** We trust Cash.
|
||||
|
||||
**This architecture ensures you never miss a bubble, but you never hold the bag when it pops.**
|
||||
16
cli/utils.py
16
cli/utils.py
|
|
@ -139,11 +139,14 @@ def select_shallow_thinking_agent(provider) -> str:
|
|||
("Claude Opus 4.5 (Thinking) - Premier reasoning with extended thinking", "claude-opus-4-5-thinking"),
|
||||
],
|
||||
"google": [
|
||||
("Gemini 2.5 Flash-Lite - Cost efficiency and low latency", "gemini-2.5-flash-lite"),
|
||||
("Gemini 1.5 Flash - Cost efficiency and low latency", "gemini-1.5-flash"),
|
||||
("Gemini 2.0 Flash - Next generation features and speed", "gemini-2.0-flash"),
|
||||
("Gemini 2.5 Flash Lite - Cost efficiency and low latency", "gemini-2.5-flash-lite"),
|
||||
("Gemini 2.5 Flash - Next generation features, speed, and thinking", "gemini-2.5-flash"),
|
||||
("Gemini 2.0 Flash Exp - Next generation features and speed", "gemini-2.0-flash-exp"),
|
||||
("Gemini 3.0 Flash - Next generation features, speed, and thinking", "gemini-3-flash-preview"),
|
||||
("Gemini 3.0 Pro - Adaptive thinking, cost efficiency", "gemini-3-pro-preview"),
|
||||
("Gemini 2.5 Pro", "gemini-2.5-pro")],
|
||||
("Gemini 1.5 Pro - High reasoning capability", "gemini-1.5-pro"),
|
||||
],
|
||||
"openrouter": [
|
||||
("Meta: Llama 4 Scout", "meta-llama/llama-4-scout:free"),
|
||||
("Meta: Llama 3.3 8B Instruct - A lightweight and ultra-fast variant of Llama 3.3 70B", "meta-llama/llama-3.3-8b-instruct:free"),
|
||||
|
|
@ -199,10 +202,15 @@ def select_deep_thinking_agent(provider) -> str:
|
|||
("Claude Opus 4.5 (Thinking) - Premier reasoning with extended thinking", "claude-opus-4-5-thinking"),
|
||||
],
|
||||
"google": [
|
||||
("Gemini 1.5 Flash - Cost efficiency and low latency", "gemini-1.5-flash"),
|
||||
("Gemini 2.0 Flash - Next generation features and speed", "gemini-2.0-flash"),
|
||||
("Gemini 2.5 Flash Lite - Cost efficiency and low latency", "gemini-2.5-flash-lite"),
|
||||
("Gemini 2.5 Flash - Next generation features, speed, and thinking", "gemini-2.5-flash"),
|
||||
("Gemini 2.0 Flash Exp - Next generation features and speed", "gemini-2.0-flash-exp"),
|
||||
("Gemini 2.5 Pro - High reasoning capability", "gemini-2.5-pro"),
|
||||
("Gemini 3.0 Flash - Next generation features, speed, and thinking", "gemini-3-flash-preview"),
|
||||
("Gemini 3.0 Pro - Adaptive thinking, cost efficiency", "gemini-3-pro-preview"),
|
||||
("Gemini 2.5 Pro", "gemini-2.5-pro"),
|
||||
("Gemini 1.5 Pro - High reasoning capability", "gemini-1.5-pro"),
|
||||
],
|
||||
"openrouter": [
|
||||
("DeepSeek V3 - a 685B-parameter, mixture-of-experts model", "deepseek/deepseek-chat-v3-0324:free"),
|
||||
|
|
|
|||
|
|
@ -69,6 +69,8 @@ fi
|
|||
|
||||
# 3. Start the Trading Agents
|
||||
echo "🚀 Starting Trading Agents..."
|
||||
# Note: Debug print() statements will appear in the terminal
|
||||
# Rich library's Live display handles the animated UI
|
||||
python3 -m cli.main
|
||||
|
||||
# 4. Open Reports
|
||||
|
|
|
|||
|
|
@ -51,6 +51,10 @@ This folder contains unit tests and verification scripts to validate the functio
|
|||
* **Purpose:** Integration test for regime detection within the broader graph context.
|
||||
* **Usage:** `python tests/verify_regime_integration.py`
|
||||
|
||||
11. **`test_finance_args.py`**
|
||||
* **Purpose:** Verifies robustness of financial tools against extraneous LLM arguments and type mismatches.
|
||||
* **Usage:** `python tests/test_finance_args.py`
|
||||
|
||||
## How to Run
|
||||
|
||||
Ensure your virtual environment is activated:
|
||||
|
|
|
|||
|
|
@ -0,0 +1,54 @@
|
|||
|
||||
import os
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import MagicMock, patch
|
||||
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
os.environ["OPENAI_API_KEY"] = "sk-dummy"
|
||||
|
||||
from tradingagents.agents.analysts.market_analyst import create_market_analyst
|
||||
|
||||
def test_market_analyst_logic():
|
||||
print("🚀 TESTING MARKET ANALYST LOGIC")
|
||||
|
||||
# Mock state
|
||||
state = {
|
||||
"company_of_interest": "PLTR",
|
||||
"trade_date": "2026-01-11",
|
||||
"market_report": "",
|
||||
"messages": []
|
||||
}
|
||||
|
||||
# Mock LLM since create_market_analyst requires it
|
||||
mock_llm_main = MagicMock()
|
||||
market_analyst_node = create_market_analyst(mock_llm_main)
|
||||
|
||||
# 1. Mock get_stock_data to FAIL
|
||||
with patch("tradingagents.agents.analysts.market_analyst.get_stock_data") as mock_tool:
|
||||
mock_tool.invoke.side_effect = Exception("API Timeout")
|
||||
|
||||
# We also need to mock the LLM because it's called
|
||||
# But wait, the LLM is inside the node logic? No, the node calls llm.invoke
|
||||
# Actually market_analyst_node takes (state, name, llm) or creates it?
|
||||
# It's a partial. `create_market_analyst_node` returns the function.
|
||||
# But `market_analyst.py` has `market_analyst_node(state, name, llm)`.
|
||||
|
||||
# Wait, how is it defined?
|
||||
# def market_analyst_node(state, name, llm=None): ...
|
||||
|
||||
# I need to pass a mock LLM
|
||||
mock_llm = MagicMock()
|
||||
mock_llm.invoke.return_value.content = "Mock Report"
|
||||
|
||||
print("\n[TEST] 1. Simulating Data Fetch Crash...")
|
||||
try:
|
||||
# The inner function only takes 'state'
|
||||
result = market_analyst_node(state)
|
||||
regime = result.get("market_regime")
|
||||
print(f"Result: {regime}")
|
||||
except Exception as e:
|
||||
print(f"CRASHED: {e}")
|
||||
|
||||
if __name__ == "__main__":
|
||||
test_market_analyst_logic()
|
||||
|
|
@ -0,0 +1,42 @@
|
|||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
|
||||
from tradingagents.dataflows.y_finance import get_income_statement, get_balance_sheet, get_cashflow
|
||||
|
||||
def verify_relaxed_types():
|
||||
print("🔍 VERIFYING RELAXED TYPE VALIDATION...")
|
||||
|
||||
ticker = "AAPL"
|
||||
# This dictionary mimics the user's error case where LLM passed reasoning in the date field
|
||||
bad_date_arg = {"reason": "Analyzing trends", "date": "2026-01-01"}
|
||||
|
||||
try:
|
||||
print("Testing get_income_statement with DICT in curr_date...")
|
||||
try:
|
||||
# This should NOT assume it's a string anymore.
|
||||
res = get_income_statement(ticker, curr_date=bad_date_arg)
|
||||
print("✅ get_income_statement ACCEPTED dict in curr_date (no Pydantic/Type Error)")
|
||||
except Exception as e:
|
||||
print(f"❌ get_income_statement FAILED: {e}")
|
||||
|
||||
print("Testing get_balance_sheet with DICT in curr_date...")
|
||||
try:
|
||||
res = get_balance_sheet(ticker, curr_date=bad_date_arg)
|
||||
print("✅ get_balance_sheet ACCEPTED dict in curr_date")
|
||||
except Exception as e:
|
||||
print(f"❌ get_balance_sheet FAILED: {e}")
|
||||
|
||||
print("Testing get_cashflow with DICT in curr_date...")
|
||||
try:
|
||||
res = get_cashflow(ticker, curr_date=bad_date_arg)
|
||||
print("✅ get_cashflow ACCEPTED dict in curr_date")
|
||||
except Exception as e:
|
||||
print(f"❌ get_cashflow FAILED: {e}")
|
||||
|
||||
except Exception as e:
|
||||
print(f"⚠️ General Error: {e}")
|
||||
|
||||
if __name__ == "__main__":
|
||||
verify_relaxed_types()
|
||||
|
|
@ -21,8 +21,11 @@ def test_google_api():
|
|||
test_models = [
|
||||
"gemini-1.5-pro",
|
||||
"gemini-1.5-flash",
|
||||
"gemini-2.0-flash",
|
||||
"gemini-2.0-flash-exp",
|
||||
"gemini-2.5-flash",
|
||||
"gemini-2.5-flash-lite",
|
||||
"gemini-2.5-pro",
|
||||
"gemini-3-flash-preview",
|
||||
"gemini-3-pro-preview",
|
||||
]
|
||||
|
|
|
|||
|
|
@ -0,0 +1,66 @@
|
|||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
|
||||
import pandas as pd
|
||||
from io import StringIO
|
||||
from tradingagents.agents.utils.agent_utils import get_stock_data
|
||||
from tradingagents.engines.regime_detector import RegimeDetector
|
||||
|
||||
def verify_macro_regime():
|
||||
print("🌍 VERIFYING MACRO ENVIRONMENT DETECTION (PLTR CONTEXT)...")
|
||||
|
||||
# We test SPY as proxy for "Broader Regime"
|
||||
macro_ticker = "SPY"
|
||||
print(f"\n1. Fetching Proxy Data for Broader Market ({macro_ticker})...")
|
||||
|
||||
try:
|
||||
# Fetch 1 year of data
|
||||
raw_data = get_stock_data.invoke({
|
||||
"symbol": macro_ticker,
|
||||
"start_date": "2025-01-01",
|
||||
"end_date": "2026-01-11",
|
||||
"format": "csv"
|
||||
})
|
||||
|
||||
if "Error" in raw_data or "No data" in raw_data:
|
||||
print(f"❌ MACRO FAIL: Could not fetch data for {macro_ticker}. Result: {raw_data[:100]}")
|
||||
return
|
||||
|
||||
df = pd.read_csv(StringIO(raw_data), comment='#')
|
||||
|
||||
if 'Close' not in df.columns:
|
||||
# Try case insensitive fallback
|
||||
col_map = {c.lower(): c for c in df.columns}
|
||||
if 'close' in col_map:
|
||||
df.rename(columns={col_map['close']: 'Close'}, inplace=True)
|
||||
|
||||
if 'Close' in df.columns and len(df) > 10:
|
||||
print(f"✅ Data fetched: {len(df)} rows")
|
||||
|
||||
# DETECT REGIME
|
||||
print(f" Running RegimeDetector for {macro_ticker}...")
|
||||
regime, metrics = RegimeDetector.detect_regime(df['Close'])
|
||||
|
||||
print(f"\n📊 BROADER MARKET REGIME ({macro_ticker}): {regime.value}")
|
||||
print(f" Volatility: {metrics['volatility']:.2%}")
|
||||
print(f" Trend Strength (ADX): {metrics['trend_strength']:.2f}")
|
||||
print(f" Hurst Exponent: {metrics['hurst_exponent']:.2f}")
|
||||
print(f" Overall Return: {metrics.get('overall_return', 0):.2%}")
|
||||
|
||||
if regime.value == "UNKNOWN":
|
||||
print("⚠️ Warning: Regime is UNKNOWN (likely insufficient history or data quality)")
|
||||
else:
|
||||
print("✅ MACRO REGIME SUCCESSFULLY IDENTIFIED.")
|
||||
|
||||
else:
|
||||
print(f"❌ MACRO FAIL: insufficient data columns or rows. Cols: {df.columns.tolist()}")
|
||||
|
||||
except Exception as e:
|
||||
print(f"❌ MACRO FAIL: Exception: {e}")
|
||||
import traceback
|
||||
traceback.print_exc()
|
||||
|
||||
if __name__ == "__main__":
|
||||
verify_macro_regime()
|
||||
|
|
@ -0,0 +1,78 @@
|
|||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
|
||||
import pandas as pd
|
||||
from io import StringIO
|
||||
from tradingagents.agents.utils.agent_utils import get_stock_data
|
||||
from tradingagents.engines.regime_detector import RegimeDetector, MarketRegime
|
||||
|
||||
def verify_momentum_exception():
|
||||
print("🚀 VERIFYING PLTR MOMENTUM EXCEPTION...")
|
||||
|
||||
ticker = "PLTR"
|
||||
start_date = "2024-01-01"
|
||||
end_date = "2025-01-11"
|
||||
|
||||
print(f"Fetching PLTR data ({start_date} to {end_date})...")
|
||||
|
||||
try:
|
||||
raw_data = get_stock_data.invoke({
|
||||
"symbol": ticker,
|
||||
"start_date": start_date,
|
||||
"end_date": end_date,
|
||||
"format": "csv"
|
||||
})
|
||||
|
||||
if "Error" in raw_data:
|
||||
print(f"❌ DATA FETCH ERROR: {raw_data}")
|
||||
return
|
||||
|
||||
df = pd.read_csv(StringIO(raw_data), comment='#')
|
||||
|
||||
# Helper cleaning
|
||||
if 'Close' not in df.columns:
|
||||
col_map = {c.lower(): c for c in df.columns}
|
||||
if 'close' in col_map:
|
||||
df.rename(columns={col_map['close']: 'Close'}, inplace=True)
|
||||
|
||||
if 'Close' not in df.columns:
|
||||
print("❌ 'Close' column missing.")
|
||||
return
|
||||
|
||||
prices = df['Close']
|
||||
print(f"✅ Data Loaded: {len(prices)} rows.")
|
||||
|
||||
# Calculate Volatility Manually to confirm test conditions
|
||||
returns = prices.pct_change().dropna()
|
||||
recent_returns = returns.tail(60) # Default window
|
||||
volatility = recent_returns.std() * (252 ** 0.5)
|
||||
|
||||
print(f"🧐 ACTUAL VOLATILITY (60d): {volatility:.2%}")
|
||||
|
||||
# RUN DETECTOR
|
||||
regime, metrics = RegimeDetector.detect_regime(prices)
|
||||
|
||||
print(f"\n📊 DETECTED REGIME: {regime.value.upper()}")
|
||||
print(f" Volatility: {metrics['volatility']:.2%}")
|
||||
print(f" Trend Strength (ADX): {metrics['trend_strength']:.2f}")
|
||||
print(f" Cumulative Return (Window): {metrics['cumulative_return']:.2%}")
|
||||
|
||||
if regime == MarketRegime.TRENDING_UP:
|
||||
if metrics['volatility'] > 0.40:
|
||||
print("✅ SUCCESS: MOMENTUM EXCEPTION ACTIVATED! (High Vol + Strong Trend = TRENDING_UP)")
|
||||
else:
|
||||
print("ℹ️ Standard TRENDING_UP (Volatility below threshold).")
|
||||
elif regime == MarketRegime.VOLATILE:
|
||||
print("❌ FAILURE: Still classified as VOLATILE despite strong trend.")
|
||||
else:
|
||||
print(f"❌ UNEXPECTED REGIME: {regime.value}")
|
||||
|
||||
except Exception as e:
|
||||
print(f"❌ TEST EXCEPTION: {e}")
|
||||
import traceback
|
||||
traceback.print_exc()
|
||||
|
||||
if __name__ == "__main__":
|
||||
verify_momentum_exception()
|
||||
|
|
@ -0,0 +1,68 @@
|
|||
|
||||
import sys
|
||||
import os
|
||||
from pathlib import Path
|
||||
from unittest.mock import MagicMock, patch
|
||||
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
os.environ["OPENAI_API_KEY"] = "sk-dummy"
|
||||
|
||||
from tradingagents.agents.analysts.market_analyst import create_market_analyst
|
||||
|
||||
def test_regime_end_to_end():
|
||||
print("🚀 TARGET: Verify Regime Error Propagation (Foolproof Test)")
|
||||
|
||||
# 1. Setup Market Analyst with LLM
|
||||
mock_llm = MagicMock()
|
||||
market_analyst_func = create_market_analyst(mock_llm)
|
||||
|
||||
# 2. Define State
|
||||
state = {
|
||||
"company_of_interest": "PLTR",
|
||||
"trade_date": "2026-01-11",
|
||||
"messages": []
|
||||
}
|
||||
|
||||
print("\n[SCENARIO 1] Fatal Crash in Tool Setup (Previously Silent)")
|
||||
# We force a crash by mocking get_stock_data to Raise ERROR immediately
|
||||
# BUT wait, the 'tools' list in market_analyst.py uses the actual imported functions.
|
||||
# To cause a crash within the 'try' block but BEFORE the 'except', we can mock datetime or something fundamental.
|
||||
# Or simpler: We mock 'get_stock_data' to be something that crashes when 'invoke' is called.
|
||||
|
||||
with patch("tradingagents.agents.analysts.market_analyst.get_stock_data") as mock_tool:
|
||||
# Make the tool invoke raise a standard Exception
|
||||
mock_tool.invoke.side_effect = RuntimeError("Simulated API Explosion")
|
||||
mock_tool.name = "get_stock_data" # Ensure it has a name so we don't crash on .name access
|
||||
|
||||
# Run Node
|
||||
result = market_analyst_func(state)
|
||||
|
||||
regime = result.get("market_regime")
|
||||
print(f"Outcome Regime: '{regime}'")
|
||||
|
||||
if "Simulated API Explosion" in regime:
|
||||
print("✅ SUCCESS: Error caught and propagated!")
|
||||
elif "Fatal" in regime:
|
||||
print("✅ SUCCESS: Fatal error caught!")
|
||||
else:
|
||||
print(f"❌ FAILURE: Got '{regime}' instead of Error.")
|
||||
|
||||
print("\n[SCENARIO 2] Silent Import Failure Simulation")
|
||||
# simulating if the internal logic fails drastically (e.g. tools list error)
|
||||
# We will mock 'get_stock_data' to NOT have a .name attribute.
|
||||
# This causes the list comprehension [t.name for t in tools] to CRASH.
|
||||
|
||||
with patch("tradingagents.agents.analysts.market_analyst.get_stock_data") as mock_tool_broken:
|
||||
del mock_tool_broken.name # This forces AttributeError at line 98
|
||||
|
||||
result_crash = market_analyst_func(state)
|
||||
regime_crash = result_crash.get("market_regime")
|
||||
print(f"Outcome Regime: '{regime_crash}'")
|
||||
|
||||
if "Fatal" in regime_crash:
|
||||
print("✅ SUCCESS: Previously silent crash is now CAUGHT!")
|
||||
else:
|
||||
print(f"❌ FAILURE: Crash was swallowed? Got: {regime_crash}")
|
||||
|
||||
if __name__ == "__main__":
|
||||
test_regime_end_to_end()
|
||||
|
|
@ -0,0 +1,41 @@
|
|||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
|
||||
# FIX for API Key Error
|
||||
import os
|
||||
os.environ["OPENAI_API_KEY"] = "sk-dummy"
|
||||
|
||||
from tradingagents.graph.trading_graph import TradingAgentsGraph
|
||||
from enum import Enum
|
||||
|
||||
class MarketRegime(Enum):
|
||||
TRENDING_UP = "trending_up"
|
||||
VOLATILE = "volatile"
|
||||
BEAR = "bear"
|
||||
|
||||
def verify_override_logic():
|
||||
print("🚀 VERIFYING OVERRIDE LOGIC FIX...")
|
||||
|
||||
graph = TradingAgentsGraph(selected_analysts=["market"])
|
||||
|
||||
# Test Case: PLTR Scenario
|
||||
hard_data = {
|
||||
"current_price": 185.0,
|
||||
"sma_200": 153.0,
|
||||
"revenue_growth": 0.62, # 62%
|
||||
"status": "OK"
|
||||
}
|
||||
decision = "SELL 50% because valuation is insane."
|
||||
|
||||
# 1. The Nightmare Type (Enum)
|
||||
print("\n[TEST] 1. Passing Raw Enum Object (MarketRegime.TRENDING_UP)")
|
||||
output = graph.apply_trend_override(decision, hard_data, MarketRegime.TRENDING_UP)
|
||||
|
||||
# 2. The String
|
||||
print("\n[TEST] 2. Passing String ('trending_up')")
|
||||
output = graph.apply_trend_override(decision, hard_data, "trending_up")
|
||||
|
||||
if __name__ == "__main__":
|
||||
verify_override_logic()
|
||||
|
|
@ -0,0 +1,104 @@
|
|||
|
||||
import sys
|
||||
import os
|
||||
import json
|
||||
from pathlib import Path
|
||||
from unittest.mock import MagicMock
|
||||
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
|
||||
# Set Dummy Key to bypass potential import checks (YFinance doesn't need it, but LangChain might)
|
||||
os.environ["OPENAI_API_KEY"] = "sk-dummy"
|
||||
|
||||
from tradingagents.agents.analysts.market_analyst import create_market_analyst
|
||||
from langchain_core.messages import AIMessage
|
||||
from langchain_core.runnables import Runnable
|
||||
|
||||
class SpyLLM(Runnable):
|
||||
"""A Mock LLM that behaves like a LangChain Runnable and prints inputs."""
|
||||
def __init__(self):
|
||||
self.captured_prompt = None
|
||||
|
||||
def bind_tools(self, tools):
|
||||
# Return self so the chain 'prompt | llm.bind_tools(tools)' works
|
||||
print(f"\n🔗 SPY LLM: Tools bound successfully: {[t.name for t in tools]}")
|
||||
return self
|
||||
|
||||
def invoke(self, input_val, config=None, **kwargs):
|
||||
# input_val might be a PromptValue or list of messages
|
||||
if hasattr(input_val, "to_messages"):
|
||||
messages = input_val.to_messages()
|
||||
else:
|
||||
messages = input_val
|
||||
|
||||
print(f"\n🕵️ SPY LLM RECEIVED INPUT (Type: {type(input_val).__name__})")
|
||||
|
||||
# Analyze the input messages
|
||||
for msg in messages:
|
||||
content = getattr(msg, "content", "")
|
||||
role = getattr(msg, "type", "unknown")
|
||||
|
||||
debug_lower = content.lower()
|
||||
idx = debug_lower.find("regime detected")
|
||||
|
||||
if idx != -1:
|
||||
self.captured_prompt = content
|
||||
print("\n[SYSTEM PROMPT CAPTURED]")
|
||||
print("="*60)
|
||||
# Print 200 chars around the hit
|
||||
start = max(0, idx - 100)
|
||||
end = min(len(content), idx + 300)
|
||||
print(f"...{content[start:end]}...")
|
||||
print("="*60)
|
||||
elif role == "system":
|
||||
print(f"\n[SYSTEM MSG ({len(content)} chars)]: {content[:50]}... [SEARCH FAILED] ...{content[-50:]}")
|
||||
print(f" -> 'regime detected' index: {idx}")
|
||||
|
||||
# Return a valid AIMessage to satisfy the node logic
|
||||
return AIMessage(content="[SPY LLM]: I have received the prompt. Data analysis complete.", tool_calls=[])
|
||||
|
||||
def pipe(self, other):
|
||||
# Support pipe operator if needed
|
||||
return self
|
||||
|
||||
def verify_pltr_analyst_flow():
|
||||
print("🚀 STARTING PLTR PIPELINE AUDIT")
|
||||
print(" Goal: Run tools, Detect Regime, Verify Prompt Construction.")
|
||||
|
||||
# 1. Setup Spy
|
||||
spy_llm = SpyLLM()
|
||||
market_analyst_node = create_market_analyst(spy_llm)
|
||||
|
||||
# 2. Define State
|
||||
state = {
|
||||
"company_of_interest": "PLTR",
|
||||
"trade_date": "2026-01-11", # Future date to test Simulation Logic too?
|
||||
"messages": []
|
||||
}
|
||||
|
||||
print(f"\n📊 INPUT STATE: Ticker={state['company_of_interest']}, Date={state['trade_date']}")
|
||||
|
||||
# 3. Execution (Real Tool usage, Mocked LLM)
|
||||
print("\n... Running Node Logic (Fetching Data via YFinance)...")
|
||||
try:
|
||||
result = market_analyst_node(state)
|
||||
|
||||
print("\n✅ NODE EXECUTION COMPLETE")
|
||||
print("\n📋 FINAL STATE OUTPUT:")
|
||||
print(f" - Market Regime: {result.get('market_regime')}")
|
||||
print(f" - Broad Market: {result.get('broad_market_regime')}")
|
||||
print(f" - Volatility Score: {result.get('volatility_score')}")
|
||||
|
||||
metrics = result.get('regime_metrics', {})
|
||||
print(f" - Key Metrics: {json.dumps(metrics, indent=2)}")
|
||||
|
||||
if result.get('market_regime', '').startswith("UNKNOWN"):
|
||||
print("\n❌ FAILED: Regime is UNKNOWN. Check logs for warnings.")
|
||||
else:
|
||||
print("\n✅ PASSED: Regime detected successfully.")
|
||||
|
||||
except Exception as e:
|
||||
print(f"\n💥 CRASHED: {e}")
|
||||
|
||||
if __name__ == "__main__":
|
||||
verify_pltr_analyst_flow()
|
||||
|
|
@ -0,0 +1,75 @@
|
|||
|
||||
import sys
|
||||
import os
|
||||
import pandas as pd
|
||||
from unittest.mock import MagicMock, patch
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.append(str(Path(__file__).parent.parent))
|
||||
os.environ["OPENAI_API_KEY"] = "sk-dummy"
|
||||
|
||||
from tradingagents.agents.analysts.market_analyst import create_market_analyst
|
||||
|
||||
def verify_regime_flow():
|
||||
print("🚀 VERIFYING DATA FLOW TO REGIME DETECTOR...")
|
||||
|
||||
# Mock LLM
|
||||
mock_llm = MagicMock()
|
||||
mock_llm.bind_tools.return_value.invoke.return_value.tool_calls = []
|
||||
mock_llm.bind_tools.return_value.invoke.return_value.content = "Analysis Report"
|
||||
|
||||
# Create Node
|
||||
analyst_node = create_market_analyst(mock_llm)
|
||||
|
||||
# Mock State
|
||||
state = {
|
||||
"company_of_interest": "PLTR",
|
||||
"trade_date": "2026-01-11",
|
||||
"messages": []
|
||||
}
|
||||
|
||||
# Mock CSV Return
|
||||
mock_csv = """Date,Close
|
||||
2025-01-01,100
|
||||
2025-01-02,105
|
||||
2025-01-03,110
|
||||
2025-01-04,115
|
||||
2025-01-05,120
|
||||
2025-01-06,125
|
||||
"""
|
||||
|
||||
with patch("tradingagents.agents.analysts.market_analyst.get_stock_data") as mock_tool:
|
||||
# Side effect to return mocked CSV
|
||||
mock_tool.invoke.return_value = mock_csv
|
||||
mock_tool.name = "get_stock_data"
|
||||
|
||||
# Also need to mock SPY data or it will try to fetch it
|
||||
# Actually SPY fetch is inside a try/except so if it fails it's fine, but let's mock it to be clean
|
||||
# Wait, get_stock_data is called twice. Once for Ticker, once for SPY.
|
||||
# We can use side_effect with a function to return based on input
|
||||
|
||||
def side_effect(args):
|
||||
symbol = args.get("symbol")
|
||||
if symbol == "SPY":
|
||||
return mock_csv # Spy data
|
||||
return mock_csv # Target data
|
||||
|
||||
mock_tool.invoke.side_effect = side_effect
|
||||
|
||||
print("\n[TEST] Calling Market Analyst Node...")
|
||||
|
||||
# We rely on the internal LOGGER to print our debug message:
|
||||
# "DEBUG: Passing prices to detector. Type: <class 'pandas.core.series.Series'>, Length: 6"
|
||||
|
||||
result = analyst_node(state)
|
||||
|
||||
print(f"\n[RESULT] Market Regime: {result.get('market_regime')}")
|
||||
print(f"[RESULT] Volatility: {result.get('volatility_score')}")
|
||||
|
||||
if "UNKNOWN" not in result.get("market_regime", "UNKNOWN"):
|
||||
print("✅ SUCCESS: Regime detected successfully.")
|
||||
else:
|
||||
print("❌ FAILURE: Regime is still UNKNOWN.")
|
||||
|
||||
if __name__ == "__main__":
|
||||
verify_regime_flow()
|
||||
|
|
@ -3,6 +3,8 @@ import time
|
|||
import json
|
||||
from tradingagents.agents.utils.agent_utils import get_fundamentals, get_balance_sheet, get_cashflow, get_income_statement, get_insider_sentiment, get_insider_transactions
|
||||
from tradingagents.dataflows.config import get_config
|
||||
from tradingagents.utils.logger import app_logger as logger
|
||||
|
||||
|
||||
|
||||
from tradingagents.utils.anonymizer import TickerAnonymizer
|
||||
|
|
@ -50,7 +52,7 @@ def create_fundamentals_analyst(llm):
|
|||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
|
||||
logger.info(f"Fundamentals Analyst Prompt: {prompt}")
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
|
||||
result = chain.invoke(state["messages"])
|
||||
|
|
|
|||
|
|
@ -10,174 +10,245 @@ from tradingagents.engines.regime_detector import RegimeDetector, DynamicIndicat
|
|||
from tradingagents.utils.anonymizer import TickerAnonymizer
|
||||
import pandas as pd
|
||||
from io import StringIO
|
||||
from io import StringIO
|
||||
from datetime import datetime, timedelta
|
||||
from tradingagents.utils.logger import app_logger as logger
|
||||
|
||||
|
||||
# Initialize anonymizer (shared instance appropriate here or inside)
|
||||
def create_market_analyst(llm):
|
||||
|
||||
def market_analyst_node(state):
|
||||
current_date = state["trade_date"]
|
||||
# Re-initialize or reload anonymizer state
|
||||
anonymizer = TickerAnonymizer()
|
||||
real_ticker = state["company_of_interest"]
|
||||
ticker = anonymizer.anonymize_ticker(real_ticker)
|
||||
|
||||
# NOTE: We continue to use 'ticker' variable name but it now holds 'ASSET_XXX'
|
||||
|
||||
# REGIME DETECTION LOGIC
|
||||
regime_val = "UNKNOWN"
|
||||
# Initialize default panic state
|
||||
regime_val = "UNKNOWN (Fatal Node Failure)"
|
||||
metrics = {}
|
||||
optimal_params = {}
|
||||
regime_context = "REGIME DETECTION FAILED or DATA UNAVAILABLE"
|
||||
broad_market_regime = "UNKNOWN"
|
||||
volatility_score = 0.0
|
||||
|
||||
report = "Market Analysis failed completely."
|
||||
tool_result_message = state["messages"]
|
||||
|
||||
try:
|
||||
# Calculate start date (1 year lookback for robust regime detection)
|
||||
dt_obj = datetime.strptime(current_date, "%Y-%m-%d")
|
||||
start_date = (dt_obj - timedelta(days=365)).strftime("%Y-%m-%d")
|
||||
|
||||
# Fetch data for regime detection using the anonymized ticker
|
||||
raw_data = get_stock_data.invoke({
|
||||
"symbol": real_ticker,
|
||||
"start_date": start_date,
|
||||
"end_date": current_date,
|
||||
"format": "csv"
|
||||
})
|
||||
# Re-initialize or reload anonymizer state
|
||||
anonymizer = TickerAnonymizer()
|
||||
real_ticker = state["company_of_interest"]
|
||||
ticker = anonymizer.anonymize_ticker(real_ticker)
|
||||
|
||||
# Parse data
|
||||
if isinstance(raw_data, str) and len(raw_data.strip()) > 50 and "Error" not in raw_data and "No data" not in raw_data:
|
||||
# Parse data (Standardized CSV format with # comments)
|
||||
df = pd.read_csv(StringIO(raw_data), comment='#')
|
||||
|
||||
# Handle case-insensitive 'Close' column
|
||||
if 'Close' not in df.columns:
|
||||
# Try to find a column that matches 'close' case-insensitively
|
||||
col_map = {c.lower(): c for c in df.columns}
|
||||
if 'close' in col_map:
|
||||
df.rename(columns={col_map['close']: 'Close'}, inplace=True)
|
||||
|
||||
# Clean index/date
|
||||
if 'Date' in df.columns:
|
||||
df['Date'] = pd.to_datetime(df['Date'])
|
||||
df.set_index('Date', inplace=True)
|
||||
|
||||
# Sort by date
|
||||
df.sort_index(inplace=True)
|
||||
|
||||
# Check for sufficient data
|
||||
# Ensure 'Close' column exists after potential renaming
|
||||
if 'Close' in df.columns:
|
||||
price_data = df['Close']
|
||||
else:
|
||||
price_data = pd.Series([]) # Empty series if 'Close' column is not found
|
||||
# NOTE: We continue to use 'ticker' variable name but it now holds 'ASSET_XXX'
|
||||
|
||||
print(f"DEBUG: Regime Detection - Ticker: {real_ticker}, Rows: {len(price_data)}")
|
||||
# REGIME DETECTION LOGIC
|
||||
regime_val = "UNKNOWN (Start)"
|
||||
optimal_params = {}
|
||||
regime_context = "REGIME DETECTION FAILED or DATA UNAVAILABLE"
|
||||
|
||||
# ... [Existing Logic] ...
|
||||
try:
|
||||
# Calculate start date (1 year lookback for robust regime detection)
|
||||
dt_obj = datetime.strptime(current_date, "%Y-%m-%d")
|
||||
start_date = (dt_obj - timedelta(days=365)).strftime("%Y-%m-%d")
|
||||
|
||||
# 1. Fetch data for TARGET ASSET
|
||||
raw_data = get_stock_data.invoke({
|
||||
"symbol": real_ticker,
|
||||
"start_date": start_date,
|
||||
"end_date": current_date,
|
||||
"format": "csv"
|
||||
})
|
||||
|
||||
if not price_data.empty and len(price_data) >= 10:
|
||||
# Detect Regime
|
||||
regime, metrics = RegimeDetector.detect_regime(price_data)
|
||||
optimal_params = DynamicIndicatorSelector.get_optimal_parameters(regime)
|
||||
regime_val = regime.value
|
||||
volatility_score = metrics.get("volatility", 0.0)
|
||||
# 2. Fetch data for BROAD MARKET (SPY)
|
||||
try:
|
||||
spy_data_raw = get_stock_data.invoke({
|
||||
"symbol": "SPY",
|
||||
"start_date": start_date,
|
||||
"end_date": current_date,
|
||||
"format": "csv"
|
||||
})
|
||||
|
||||
# Construct Context String
|
||||
regime_context = f"MARKET REGIME DETECTED: {regime_val}\n"
|
||||
regime_context += f"METRICS: {json.dumps(metrics)}\n"
|
||||
regime_context += f"RECOMMENDED STRATEGY: {optimal_params.get('strategy', 'N/A')}\n"
|
||||
regime_context += f"RECOMMENDED INDICATORS: {json.dumps(optimal_params)}\n"
|
||||
regime_context += f"RATIONALE: {optimal_params.get('rationale', '')}"
|
||||
if isinstance(spy_data_raw, str) and len(spy_data_raw.strip()) > 50 and "Error" not in spy_data_raw:
|
||||
df_spy = pd.read_csv(StringIO(spy_data_raw), comment='#')
|
||||
# Basic cleaning for SPY
|
||||
if 'Close' not in df_spy.columns:
|
||||
col_map = {c.lower(): c for c in df_spy.columns}
|
||||
if 'close' in col_map:
|
||||
df_spy.rename(columns={col_map['close']: 'Close'}, inplace=True)
|
||||
|
||||
if 'Close' in df_spy.columns and len(df_spy) > 10:
|
||||
spy_regime, _ = RegimeDetector.detect_regime(df_spy['Close'])
|
||||
broad_market_regime = spy_regime.value
|
||||
except Exception as e_spy:
|
||||
logger.warning(f"Broad Market (SPY) detection failed: {e_spy}")
|
||||
|
||||
|
||||
# Parse TARGET data
|
||||
if isinstance(raw_data, str) and len(raw_data.strip()) > 50 and "Error" not in raw_data and "No data" not in raw_data:
|
||||
# Parse data (Standardized CSV format with # comments)
|
||||
df = pd.read_csv(StringIO(raw_data), comment='#')
|
||||
|
||||
# Handle case-insensitive 'Close' column
|
||||
if 'Close' not in df.columns:
|
||||
# Try to find a column that matches 'close' case-insensitively
|
||||
col_map = {c.lower(): c for c in df.columns}
|
||||
if 'close' in col_map:
|
||||
df.rename(columns={col_map['close']: 'Close'}, inplace=True)
|
||||
|
||||
# Clean index/date
|
||||
if 'Date' in df.columns:
|
||||
df['Date'] = pd.to_datetime(df['Date'])
|
||||
df.set_index('Date', inplace=True)
|
||||
|
||||
# Sort by date
|
||||
df.sort_index(inplace=True)
|
||||
|
||||
# Check for sufficient data
|
||||
# Ensure 'Close' column exists after potential renaming
|
||||
if 'Close' in df.columns:
|
||||
price_data = df['Close']
|
||||
else:
|
||||
price_data = pd.Series([]) # Empty series if 'Close' column is not found
|
||||
|
||||
if not price_data.empty and len(price_data) >= 5:
|
||||
# DEBUG INJECTION FOR MARKET ANALYST
|
||||
try:
|
||||
debug_msg = f"DEBUG: Passing prices to detector. Type: {type(price_data)}, Length: {len(price_data)}"
|
||||
logger.info(debug_msg)
|
||||
print(f"\n[CONSOLE] {debug_msg}")
|
||||
|
||||
regime, metrics = RegimeDetector.detect_regime(price_data)
|
||||
|
||||
# Handle Enum or String return
|
||||
if hasattr(regime, "value"):
|
||||
regime_val = regime.value
|
||||
else:
|
||||
regime_val = str(regime)
|
||||
|
||||
optimal_params = DynamicIndicatorSelector.get_optimal_parameters(regime)
|
||||
volatility_score = metrics.get("volatility", 0.0)
|
||||
|
||||
except Exception as e_det:
|
||||
err_msg = f"CRITICAL: Detector Call Failed. Data Snippet: {str(price_data.head())}. Error: {e_det}"
|
||||
logger.critical(err_msg)
|
||||
print(f"\n[CONSOLE] {err_msg}")
|
||||
regime_val = "UNKNOWN (Detector Failed)"
|
||||
metrics = {"volatility": 0.0}
|
||||
optimal_params = {}
|
||||
|
||||
# Construct Context String (Enhanced)
|
||||
regime_context = f"MARKET REGIME DETECTED: {regime_val}\n"
|
||||
regime_context += f"BROAD MARKET CONTEXT (SPY): {broad_market_regime}\n"
|
||||
regime_context += f"METRICS: {json.dumps(metrics)}\n"
|
||||
regime_context += f"RECOMMENDED STRATEGY: {optimal_params.get('strategy', 'N/A')}\n"
|
||||
regime_context += f"RECOMMENDED INDICATORS: {json.dumps(optimal_params)}\n"
|
||||
regime_context += f"RATIONALE: {optimal_params.get('rationale', '')}"
|
||||
else:
|
||||
msg = f"Insufficient price data for {ticker}. Len: {len(df)}"
|
||||
logger.warning(msg)
|
||||
regime_val = "UNKNOWN (Insufficient Data)"
|
||||
else:
|
||||
print(f"WARNING: Insufficient price data for {ticker}. Columns: {list(df.columns)}, Len: {len(df)}")
|
||||
else:
|
||||
print(f"WARNING: Market data retrieval failed for regime detection for {ticker}. Data snippet: {str(raw_data)[:100]}")
|
||||
except Exception as e:
|
||||
print(f"WARNING: Regime detection failed for {ticker}: {e}")
|
||||
|
||||
tools = [
|
||||
get_stock_data,
|
||||
get_indicators,
|
||||
]
|
||||
|
||||
system_message = (
|
||||
"""ROLE: Quantitative Technical Analyst.
|
||||
CONTEXT: You are analyzing an ANONYMIZED ASSET (ASSET_XXX).
|
||||
CRITICAL DATA CONSTRAINT:
|
||||
1. All Price Data is NORMALIZED to a BASE-100 INDEX starting at the beginning of the period.
|
||||
2. "Price 105.0" means +5% gain from start. It does NOT mean $105.00.
|
||||
3. DO NOT hallucinate real-world ticker prices. Treat this as a pure mathematical time series.
|
||||
|
||||
DYNAMIC MARKET REGIME CONTEXT:
|
||||
{regime_context}
|
||||
|
||||
TASK: Select relevant indicators and analyze trends.
|
||||
Your role is to select the **most relevant indicators** for the DETECTED REGIME ({regime_val}).
|
||||
The goal is to choose up to **8 indicators** that provide complementary insights without redundancy.
|
||||
|
||||
INDICATOR CATEGORIES:
|
||||
|
||||
Moving Averages:
|
||||
- close_50_sma: 50 SMA: A medium-term trend indicator. Usage: Identify trend direction and serve as dynamic support/resistance. Tips: It lags price; combine with faster indicators for timely signals.
|
||||
- close_200_sma: 200 SMA: A long-term trend benchmark. Usage: Confirm overall market trend and identify golden/death cross setups. Tips: It reacts slowly; best for strategic trend confirmation rather than frequent trading entries.
|
||||
- close_10_ema: 10 EMA: A responsive short-term average. Usage: Capture quick shifts in momentum and potential entry points. Tips: Prone to noise in choppy markets; use alongside longer averages for filtering false signals.
|
||||
|
||||
MACD Related:
|
||||
- macd: MACD: Computes momentum via differences of EMAs. Usage: Look for crossovers and divergence as signals of trend changes. Tips: Confirm with other indicators in low-volatility or sideways markets.
|
||||
- macds: MACD Signal: An EMA smoothing of the MACD line. Usage: Use crossovers with the MACD line to trigger trades. Tips: Should be part of a broader strategy to avoid false positives.
|
||||
- macdh: MACD Histogram: Shows the gap between the MACD line and its signal. Usage: Visualize momentum strength and spot divergence early. Tips: Can be volatile; complement with additional filters in fast-moving markets.
|
||||
|
||||
Momentum Indicators:
|
||||
- rsi: RSI: Measures momentum to flag overbought/oversold conditions. Usage: Apply 70/30 thresholds and watch for divergence to signal reversals. Tips: In strong trends, RSI may remain extreme; always cross-check with trend analysis.
|
||||
|
||||
Volatility Indicators:
|
||||
- boll: Bollinger Middle: A 20 SMA serving as the basis for Bollinger Bands. Usage: Acts as a dynamic benchmark for price movement. Tips: Combine with the upper and lower bands to effectively spot breakouts or reversals.
|
||||
- boll_ub: Bollinger Upper Band: Typically 2 standard deviations above the middle line. Usage: Signals potential overbought conditions and breakout zones. Tips: Confirm signals with other tools; prices may ride the band in strong trends.
|
||||
- boll_lb: Bollinger Lower Band: Typically 2 standard deviations below the middle line. Usage: Indicates potential oversold conditions. Tips: Use additional analysis to avoid false reversal signals.
|
||||
- atr: ATR: Averages true range to measure volatility. Usage: Set stop-loss levels and adjust position sizes based on current market volatility. Tips: It's a reactive measure, so use it as part of a broader risk management strategy.
|
||||
|
||||
Volume-Based Indicators:
|
||||
- vwma: VWMA: A moving average weighted by volume. Usage: Confirm trends by integrating price action with volume data. Tips: Watch for skewed results from volume spikes; use in combination with other volume analyses.
|
||||
|
||||
- Select indicators that provide diverse and complementary information. Avoid redundancy (e.g., do not select both rsi and stochrsi). Also briefly explain why they are suitable for the given market context. When you tool call, please use the exact name of the indicators provided above as they are defined parameters, otherwise your call will fail. Please make sure to call get_stock_data first to retrieve the CSV that is needed to generate indicators. Then use get_indicators with the specific indicator names. Write a very detailed and nuanced report of the trends you observe. Do not simply state the trends are mixed, provide detailed and finegrained analysis and insights that may help traders make decisions."""
|
||||
+ """ Make sure to append a Markdown table at the end of the report to organize key points in the report, organized and easy to read."""
|
||||
)
|
||||
|
||||
prompt = ChatPromptTemplate.from_messages(
|
||||
[
|
||||
(
|
||||
"system",
|
||||
"You are a helpful AI assistant, collaborating with other assistants."
|
||||
" Use the provided tools to progress towards answering the question."
|
||||
" If you are unable to fully answer, that's OK; another assistant with different tools"
|
||||
" will help where you left off. Execute what you can to make progress."
|
||||
" If you or any other assistant has the FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** or deliverable,"
|
||||
" prefix your response with FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** so the team knows to stop."
|
||||
" You have access to the following tools: {tool_names}.\n{system_message}"
|
||||
"For your reference, the current date is {current_date}. The company we want to look at is {ticker}",
|
||||
),
|
||||
MessagesPlaceholder(variable_name="messages"),
|
||||
msg = f"Market data retrieval failed for {ticker}. Snippet: {str(raw_data)[:100]}"
|
||||
logger.warning(msg)
|
||||
regime_val = "UNKNOWN (Data Fetch Error)"
|
||||
except Exception as e:
|
||||
logger.warning(f"Regime detection failed for {ticker}: {e}")
|
||||
regime_val = f"UNKNOWN (Error: {str(e)})"
|
||||
|
||||
tools = [
|
||||
get_stock_data,
|
||||
get_indicators,
|
||||
]
|
||||
)
|
||||
|
||||
system_message = (
|
||||
f"""ROLE: Quantitative Technical Analyst.
|
||||
CONTEXT: You are analyzing an ANONYMIZED ASSET (ASSET_XXX).
|
||||
CRITICAL DATA CONSTRAINT:
|
||||
1. All Price Data is NORMALIZED to a BASE-100 INDEX starting at the beginning of the period.
|
||||
2. "Price 105.0" means +5% gain from start. It does NOT mean $105.00.
|
||||
3. DO NOT hallucinate real-world ticker prices. Treat this as a pure mathematical time series.
|
||||
|
||||
DYNAMIC MARKET REGIME CONTEXT:
|
||||
{regime_context}
|
||||
|
||||
TASK: Select relevant indicators and analyze trends.
|
||||
Your role is to select the **most relevant indicators** for the DETECTED REGIME ({regime_val}).
|
||||
The goal is to choose up to **8 indicators** that provide complementary insights without redundancy.
|
||||
|
||||
INDICATOR CATEGORIES:
|
||||
|
||||
Moving Averages:
|
||||
- close_50_sma: 50 SMA: A medium-term trend indicator. Usage: Identify trend direction and serve as dynamic support/resistance. Tips: It lags price; combine with faster indicators for timely signals.
|
||||
- close_200_sma: 200 SMA: A long-term trend benchmark. Usage: Confirm overall market trend and identify golden/death cross setups. Tips: It reacts slowly; best for strategic trend confirmation rather than frequent trading entries.
|
||||
- close_10_ema: 10 EMA: A responsive short-term average. Usage: Capture quick shifts in momentum and potential entry points. Tips: Prone to noise in choppy markets; use alongside longer averages for filtering false signals.
|
||||
|
||||
MACD Related:
|
||||
- macd: MACD: Computes momentum via differences of EMAs. Usage: Look for crossovers and divergence as signals of trend changes. Tips: Confirm with other indicators in low-volatility or sideways markets.
|
||||
- macds: MACD Signal: An EMA smoothing of the MACD line. Usage: Use crossovers with the MACD line to trigger trades. Tips: Should be part of a broader strategy to avoid false positives.
|
||||
- macdh: MACD Histogram: Shows the gap between the MACD line and its signal. Usage: Visualize momentum strength and spot divergence early. Tips: Can be volatile; complement with additional filters in fast-moving markets.
|
||||
|
||||
Momentum Indicators:
|
||||
- rsi: RSI: Measures momentum to flag overbought/oversold conditions. Usage: Apply 70/30 thresholds and watch for divergence to signal reversals. Tips: In strong trends, RSI may remain extreme; always cross-check with trend analysis.
|
||||
|
||||
Volatility Indicators:
|
||||
- boll: Bollinger Middle: A 20 SMA serving as the basis for Bollinger Bands. Usage: Acts as a dynamic benchmark for price movement. Tips: Combine with the upper and lower bands to effectively spot breakouts or reversals.
|
||||
- boll_ub: Bollinger Upper Band: Typically 2 standard deviations above the middle line. Usage: Signals potential overbought conditions and breakout zones. Tips: Confirm signals with other tools; prices may ride the band in strong trends.
|
||||
- boll_lb: Bollinger Lower Band: Typically 2 standard deviations below the middle line. Usage: Indicates potential oversold conditions. Tips: Use additional analysis to avoid false reversal signals.
|
||||
- atr: ATR: Averages true range to measure volatility. Usage: Set stop-loss levels and adjust position sizes based on current market volatility. Tips: It's a reactive measure, so use it as part of a broader risk management strategy.
|
||||
|
||||
Volume-Based Indicators:
|
||||
- vwma: VWMA: A moving average weighted by volume. Usage: Confirm trends by integrating price action with volume data. Tips: Watch for skewed results from volume spikes; use in combination with other volume analyses.
|
||||
|
||||
- Select indicators that provide diverse and complementary information. Avoid redundancy (e.g., do not select both rsi and stochrsi). Also briefly explain why they are suitable for the given market context. When you tool call, please use the exact name of the indicators provided above as they are defined parameters, otherwise your call will fail. Please make sure to call get_stock_data first to retrieve the CSV that is needed to generate indicators. Then use get_indicators with the specific indicator names. Write a very detailed and nuanced report of the trends you observe. Do not simply state the trends are mixed, provide detailed and finegrained analysis and insights that may help traders make decisions."""
|
||||
+ """ Make sure to append a Markdown table at the end of the report to organize key points in the report, organized and easy to read."""
|
||||
)
|
||||
|
||||
prompt = ChatPromptTemplate.from_messages(
|
||||
[
|
||||
(
|
||||
"system",
|
||||
"You are a helpful AI assistant, collaborating with other assistants."
|
||||
" Use the provided tools to progress towards answering the question."
|
||||
" If you are unable to fully answer, that's OK; another assistant with different tools"
|
||||
" will help where you left off. Execute what you can to make progress."
|
||||
" If you or any other assistant has the FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** or deliverable,"
|
||||
" prefix your response with FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** so the team knows to stop."
|
||||
" You have access to the following tools: {tool_names}.\n{system_message}"
|
||||
"For your reference, the current date is {current_date}. The company we want to look at is {ticker}",
|
||||
),
|
||||
MessagesPlaceholder(variable_name="messages"),
|
||||
]
|
||||
)
|
||||
|
||||
prompt = prompt.partial(system_message=system_message)
|
||||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
|
||||
prompt = prompt.partial(system_message=system_message)
|
||||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
logger.info(f"Market Analyst Prompt: {prompt}")
|
||||
|
||||
try:
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
result = chain.invoke(state["messages"])
|
||||
if len(result.tool_calls) == 0:
|
||||
report = result.content
|
||||
tool_result_message = [result]
|
||||
except Exception as e_llm:
|
||||
logger.error(f"ERROR: Market Analyst LLM and Tool use failed: {e_llm}")
|
||||
report = f"Market Analysis failed due to LLM error. Regime Context: {regime_context}"
|
||||
tool_result_message = state["messages"] # No new message
|
||||
|
||||
result = chain.invoke(state["messages"])
|
||||
|
||||
report = ""
|
||||
|
||||
if len(result.tool_calls) == 0:
|
||||
report = result.content
|
||||
|
||||
except Exception as e_fatal:
|
||||
logger.critical(f"CRITICAL ERROR in Market Analyst Node: {e_fatal}")
|
||||
regime_val = f"UNKNOWN (Fatal Crash: {str(e_fatal)})"
|
||||
report = f"Market Analyst Node crashed completely: {e_fatal}"
|
||||
|
||||
return {
|
||||
"messages": [result],
|
||||
"messages": tool_result_message,
|
||||
"market_report": report,
|
||||
"market_regime": regime_val,
|
||||
"market_regime": regime_val, # PLTR Regime (e.g., TRENDING_UP)
|
||||
"regime_metrics": metrics,
|
||||
"volatility_score": volatility_score
|
||||
"volatility_score": volatility_score,
|
||||
"broad_market_regime": broad_market_regime # SPY Regime (e.g., SIDEWAYS)
|
||||
}
|
||||
|
||||
|
||||
return market_analyst_node
|
||||
|
|
|
|||
|
|
@ -3,6 +3,8 @@ import time
|
|||
import json
|
||||
from tradingagents.agents.utils.agent_utils import get_news, get_global_news
|
||||
from tradingagents.dataflows.config import get_config
|
||||
from tradingagents.utils.logger import app_logger as logger
|
||||
|
||||
|
||||
|
||||
from tradingagents.utils.anonymizer import TickerAnonymizer
|
||||
|
|
@ -54,7 +56,7 @@ def create_news_analyst(llm):
|
|||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
|
||||
logger.info(f"News Analyst Prompt: {prompt}")
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
result = chain.invoke(state["messages"])
|
||||
|
||||
|
|
|
|||
|
|
@ -3,6 +3,8 @@ import time
|
|||
import json
|
||||
from tradingagents.agents.utils.agent_utils import get_news
|
||||
from tradingagents.dataflows.config import get_config
|
||||
from tradingagents.utils.logger import app_logger as logger
|
||||
|
||||
|
||||
|
||||
from tradingagents.utils.anonymizer import TickerAnonymizer
|
||||
|
|
@ -47,7 +49,7 @@ def create_social_media_analyst(llm):
|
|||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
|
||||
logger.info(f"Social Media Analyst Prompt: {prompt}")
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
|
||||
result = chain.invoke(state["messages"])
|
||||
|
|
|
|||
|
|
@ -61,8 +61,10 @@ class AgentState(MessagesState):
|
|||
]
|
||||
fundamentals_report: Annotated[str, "Report from the Fundamentals Researcher"]
|
||||
|
||||
# regime data
|
||||
# regime data
|
||||
market_regime: Annotated[str, "Current Market Regime (e.g. VOLATILE, TRENDING_UP)"]
|
||||
broad_market_regime: Annotated[str, "Broad Market Context (e.g. SPY Regime)"]
|
||||
regime_metrics: Annotated[dict, "Metrics used to determine regime"]
|
||||
volatility_score: Annotated[float, "Current Volatility Score"]
|
||||
|
||||
|
|
|
|||
|
|
@ -1,4 +1,4 @@
|
|||
from typing import Annotated
|
||||
from typing import Annotated, Any
|
||||
from datetime import datetime
|
||||
from dateutil.relativedelta import relativedelta
|
||||
import yfinance as yf
|
||||
|
|
@ -303,7 +303,8 @@ def get_stockstats_indicator(
|
|||
def get_balance_sheet(
|
||||
ticker: Annotated[str, "ticker symbol of the company"],
|
||||
freq: Annotated[str, "frequency of data: 'annual' or 'quarterly'"] = "quarterly",
|
||||
curr_date: Annotated[str, "current date (not used for yfinance)"] = None
|
||||
curr_date: Annotated[Any, "current date (not used for yfinance)"] = None,
|
||||
**kwargs
|
||||
):
|
||||
"""Get balance sheet data from yfinance."""
|
||||
try:
|
||||
|
|
@ -333,7 +334,8 @@ def get_balance_sheet(
|
|||
def get_cashflow(
|
||||
ticker: Annotated[str, "ticker symbol of the company"],
|
||||
freq: Annotated[str, "frequency of data: 'annual' or 'quarterly'"] = "quarterly",
|
||||
curr_date: Annotated[str, "current date (not used for yfinance)"] = None
|
||||
curr_date: Annotated[Any, "current date (not used for yfinance)"] = None,
|
||||
**kwargs
|
||||
):
|
||||
"""Get cash flow data from yfinance."""
|
||||
try:
|
||||
|
|
@ -363,7 +365,8 @@ def get_cashflow(
|
|||
def get_income_statement(
|
||||
ticker: Annotated[str, "ticker symbol of the company"],
|
||||
freq: Annotated[str, "frequency of data: 'annual' or 'quarterly'"] = "quarterly",
|
||||
curr_date: Annotated[str, "current date (not used for yfinance)"] = None
|
||||
curr_date: Annotated[Any, "current date (not used for yfinance)"] = None,
|
||||
**kwargs
|
||||
):
|
||||
"""Get income statement data from yfinance."""
|
||||
try:
|
||||
|
|
@ -414,7 +417,7 @@ def get_insider_transactions(
|
|||
return f"Error retrieving insider transactions for {ticker}: {str(e)}"
|
||||
def get_fundamentals(
|
||||
ticker: Annotated[str, "ticker symbol of the company"],
|
||||
curr_date: Annotated[str, "current date (not used for yfinance info)"] = None
|
||||
curr_date: Annotated[Any, "current date (not used for yfinance info)"] = None
|
||||
):
|
||||
"""Get fundamental data from yfinance ticker.info."""
|
||||
try:
|
||||
|
|
|
|||
|
|
@ -26,62 +26,83 @@ class RegimeDetector:
|
|||
@staticmethod
|
||||
def detect_regime(prices: pd.Series, window: int = 60) -> Tuple[MarketRegime, Dict]:
|
||||
"""
|
||||
Detect current market regime.
|
||||
|
||||
Args:
|
||||
prices: Price series (must have at least 'window' data points)
|
||||
window: Lookback period for regime detection
|
||||
|
||||
Returns:
|
||||
(regime, metrics) tuple where metrics contains diagnostic info
|
||||
Determines the market regime based on Volatility, ADX, and Returns.
|
||||
INCLUDES 'MOMENTUM EXCEPTION' for high-growth stocks.
|
||||
"""
|
||||
if len(prices) < window:
|
||||
raise ValueError(f"Need at least {window} data points, got {len(prices)}")
|
||||
|
||||
# Calculate regime metrics
|
||||
returns = prices.pct_change().dropna()
|
||||
recent_returns = returns.tail(window)
|
||||
|
||||
# 1. Volatility (annualized)
|
||||
volatility = recent_returns.std() * np.sqrt(252)
|
||||
|
||||
# 2. Trend strength (ADX approximation)
|
||||
trend_strength = RegimeDetector._calculate_trend_strength(prices.tail(window))
|
||||
|
||||
# 3. Mean reversion tendency (Hurst exponent)
|
||||
hurst = RegimeDetector._calculate_hurst_exponent(prices.tail(window))
|
||||
|
||||
# 4. Directional bias (Cumulative Return)
|
||||
# We check both the specific window and the broader history to capture leaders in consolidation
|
||||
window_return = (prices.iloc[-1] / prices.iloc[-window]) - 1
|
||||
full_history_return = (prices.iloc[-1] / prices.iloc[0]) - 1
|
||||
|
||||
# Classify regime
|
||||
metrics = {
|
||||
"volatility": volatility,
|
||||
"trend_strength": trend_strength,
|
||||
"hurst_exponent": hurst,
|
||||
"cumulative_return": window_return,
|
||||
"overall_return": full_history_return
|
||||
}
|
||||
|
||||
# Decision tree for regime classification - Prioritize Trend & Momentum
|
||||
# If ADX > 25, it's trending. We use the broader return to confirm if it's a leader.
|
||||
if trend_strength > 25:
|
||||
if window_return > 0 or full_history_return > 0.10: # Up on window OR strong long-term momentum
|
||||
regime = MarketRegime.TRENDING_UP
|
||||
else:
|
||||
regime = MarketRegime.TRENDING_DOWN
|
||||
elif full_history_return > 0.30: # Massive long-term momentum overrides Hurst/Volatility
|
||||
regime = MarketRegime.TRENDING_UP
|
||||
elif volatility > 0.80: # High volatility threshold for individual tech stocks
|
||||
regime = MarketRegime.VOLATILE
|
||||
elif not np.isnan(hurst) and hurst < 0.45: # Tighter mean reversion check
|
||||
regime = MarketRegime.MEAN_REVERTING
|
||||
else:
|
||||
regime = MarketRegime.SIDEWAYS
|
||||
|
||||
return regime, metrics
|
||||
try:
|
||||
if len(prices) < window:
|
||||
# Fallback for short history
|
||||
if len(prices) > 10:
|
||||
window = len(prices) - 1
|
||||
else:
|
||||
return MarketRegime.SIDEWAYS, {}
|
||||
|
||||
# 1. Calculate Metrics
|
||||
# We use existing helper methods but adapt the call signature slightly if needed
|
||||
# The user provided logic assumes 'market_data' DataFrame but we take 'prices' Series
|
||||
# We will adapt the user's logic to work with the Series input or reconstruct DataFrame if needed
|
||||
# Actually, standardizing on the existing helper methods is safer, but implementing the LOGIC FLOD is key.
|
||||
|
||||
# Reconstruct helpers calls based on existing class structure
|
||||
|
||||
# Volatility
|
||||
returns = prices.pct_change().dropna()
|
||||
recent_returns = returns.tail(window)
|
||||
volatility = recent_returns.std() * np.sqrt(252)
|
||||
|
||||
# ADX
|
||||
trend_strength = RegimeDetector._calculate_trend_strength(prices.tail(window))
|
||||
|
||||
# Hurst
|
||||
hurst = RegimeDetector._calculate_hurst_exponent(prices.tail(window))
|
||||
|
||||
# Simple Price Return
|
||||
start_price = prices.iloc[-window]
|
||||
end_price = prices.iloc[-1]
|
||||
price_change_pct = (end_price - start_price) / start_price
|
||||
|
||||
# Full history return (keeping from previous logic as extra metric)
|
||||
full_history_return = (prices.iloc[-1] / prices.iloc[0]) - 1
|
||||
|
||||
# 2. DEFINE THRESHOLDS
|
||||
VOLATILITY_THRESHOLD = 0.40 # 40% Annualized Volatility
|
||||
ADX_STRONG_TREND = 25.0
|
||||
|
||||
# Metrics dict
|
||||
metrics = {
|
||||
"volatility": volatility,
|
||||
"trend_strength": trend_strength,
|
||||
"hurst_exponent": hurst,
|
||||
"cumulative_return": price_change_pct,
|
||||
"overall_return": full_history_return
|
||||
}
|
||||
|
||||
# 3. THE LOGIC CASCADE
|
||||
|
||||
# 🛑 CRITICAL FIX: THE MOMENTUM EXCEPTION
|
||||
# If stock is volatile BUT going up strongly, it is BULLISH, not VOLATILE.
|
||||
if volatility > VOLATILITY_THRESHOLD:
|
||||
if price_change_pct > 0 and trend_strength > ADX_STRONG_TREND:
|
||||
# "High Beta Breakout"
|
||||
return MarketRegime.TRENDING_UP, metrics
|
||||
else:
|
||||
# "Crashing / Chopping"
|
||||
return MarketRegime.VOLATILE, metrics
|
||||
|
||||
# Standard Logic for Lower Volatility
|
||||
if trend_strength > ADX_STRONG_TREND:
|
||||
regime = MarketRegime.TRENDING_UP if price_change_pct > 0 else MarketRegime.TRENDING_DOWN
|
||||
return regime, metrics
|
||||
|
||||
# Mean Reverting Logic
|
||||
if hurst < 0.4:
|
||||
return MarketRegime.MEAN_REVERTING, metrics
|
||||
|
||||
return MarketRegime.SIDEWAYS, metrics
|
||||
|
||||
except Exception as e:
|
||||
print(f"Regime Detection Error: {e}")
|
||||
return MarketRegime.SIDEWAYS, {"error": str(e)}
|
||||
|
||||
@staticmethod
|
||||
def _calculate_trend_strength(prices: pd.Series) -> float:
|
||||
|
|
@ -90,6 +111,10 @@ class RegimeDetector:
|
|||
|
||||
Returns value 0-100, where >25 indicates strong trend.
|
||||
"""
|
||||
prices = prices.dropna()
|
||||
if len(prices) < 14:
|
||||
return 0.0
|
||||
|
||||
high = prices.rolling(2).max()
|
||||
low = prices.rolling(2).min()
|
||||
|
||||
|
|
@ -105,11 +130,19 @@ class RegimeDetector:
|
|||
|
||||
# Smooth with 14-period EMA
|
||||
atr = pd.Series(tr).ewm(span=14, adjust=False).mean()
|
||||
plus_di = 100 * pd.Series(plus_dm).ewm(span=14, adjust=False).mean() / atr
|
||||
minus_di = 100 * pd.Series(minus_dm).ewm(span=14, adjust=False).mean() / atr
|
||||
|
||||
# Avoid division by zero
|
||||
atr = atr.replace(0, np.nan).ffill().fillna(1e-9)
|
||||
|
||||
# Reconstruct Series with correct index to align with ATR
|
||||
plus_di = 100 * pd.Series(plus_dm, index=prices.index).ewm(span=14, adjust=False).mean() / atr
|
||||
minus_di = 100 * pd.Series(minus_dm, index=prices.index).ewm(span=14, adjust=False).mean() / atr
|
||||
|
||||
# ADX
|
||||
dx = 100 * abs(plus_di - minus_di) / (plus_di + minus_di)
|
||||
# Handle DX NaNs (caused by 0 division if +DI and -DI are both 0)
|
||||
dx = dx.fillna(0)
|
||||
|
||||
adx = dx.ewm(span=14, adjust=False).mean()
|
||||
|
||||
return adx.iloc[-1] if not pd.isna(adx.iloc[-1]) else 0.0
|
||||
|
|
|
|||
|
|
@ -41,6 +41,7 @@ class Propagator:
|
|||
"sentiment_report": "",
|
||||
"news_report": "",
|
||||
"market_regime": "UNKNOWN",
|
||||
"broad_market_regime": "UNKNOWN",
|
||||
"volatility_score": 0.0,
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -11,6 +11,9 @@ from langchain_anthropic import ChatAnthropic
|
|||
from langchain_google_genai import ChatGoogleGenerativeAI
|
||||
|
||||
from langgraph.prebuilt import ToolNode
|
||||
from datetime import datetime
|
||||
from tradingagents.utils.logger import override_logger as logger
|
||||
|
||||
|
||||
from tradingagents.agents import *
|
||||
from tradingagents.default_config import DEFAULT_CONFIG
|
||||
|
|
@ -232,8 +235,17 @@ class TradingAgentsGraph:
|
|||
raw_llm_decision = final_state["final_trade_decision"]
|
||||
|
||||
# Apply Technical Override (Don't Fight the Tape)
|
||||
regime_val = final_state.get("market_regime", "UNKNOWN").upper()
|
||||
print(f"\n🔍 [DEBUG] APPLYING OVERRIDE: Regime='{regime_val}', Growth={self.hard_data.get('revenue_growth', 'N/A')}")
|
||||
# Handle Enum vs String robustly
|
||||
raw_regime = final_state.get("market_regime", "UNKNOWN")
|
||||
if hasattr(raw_regime, "value"):
|
||||
regime_val = raw_regime.value
|
||||
else:
|
||||
regime_val = str(raw_regime)
|
||||
regime_val = regime_val.upper().strip()
|
||||
|
||||
msg = f"🔍 [DEBUG] APPLYING OVERRIDE: Regime='{regime_val}', Growth={self.hard_data.get('revenue_growth', 'N/A')}"
|
||||
logger.info(msg)
|
||||
print(f"\n[CONSOLE] {msg}")
|
||||
|
||||
overridden_decision = self.apply_trend_override(
|
||||
raw_llm_decision,
|
||||
|
|
@ -344,8 +356,14 @@ class TradingAgentsGraph:
|
|||
# Fetch 300 days of history to ensure we can calculate 200 SMA
|
||||
start_date = (dt_obj - timedelta(days=450)).strftime("%Y-%m-%d")
|
||||
|
||||
# FIX: Handle Future Simulation Dates
|
||||
# YFinance errors if end_date is in the future relative to today
|
||||
today = datetime.now()
|
||||
actual_end_date = min(dt_obj, today).strftime("%Y-%m-%d")
|
||||
|
||||
ticker_obj = yf.Ticker(ticker.upper())
|
||||
history = ticker_obj.history(start=start_date, end=trade_date)
|
||||
# Use actual_end_date instead of trade_date if trade_date is future
|
||||
history = ticker_obj.history(start=start_date, end=actual_end_date)
|
||||
|
||||
metrics = {
|
||||
"current_price": 0.0,
|
||||
|
|
@ -363,7 +381,7 @@ class TradingAgentsGraph:
|
|||
return metrics
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error fetching hard data for {ticker} override: {e}")
|
||||
logger.error(f"Error fetching hard data for {ticker} override: {e}")
|
||||
return {"status": "ERROR", "error": str(e)}
|
||||
|
||||
def apply_trend_override(self, trade_decision_str: str, hard_data: Dict[str, Any], regime: str) -> Any:
|
||||
|
|
@ -372,9 +390,16 @@ class TradingAgentsGraph:
|
|||
Prevents the system from shorting high-growth winners during a Bull Market.
|
||||
"""
|
||||
if hard_data.get("status") != "OK":
|
||||
logger.info(f"DEBUG OVERRIDE: Failed due to Hard Data Status: {hard_data.get('status')}, Error: {hard_data.get('error')}")
|
||||
return trade_decision_str
|
||||
|
||||
regime = str(regime).strip().upper()
|
||||
# Robust Enum Extraction (Double Lock)
|
||||
if hasattr(regime, "value"):
|
||||
regime_val = regime.value
|
||||
else:
|
||||
regime_val = str(regime)
|
||||
|
||||
regime_val = regime_val.upper().strip()
|
||||
|
||||
price = hard_data["current_price"]
|
||||
sma_200 = hard_data["sma_200"]
|
||||
|
|
@ -387,28 +412,41 @@ class TradingAgentsGraph:
|
|||
is_hyper_growth = growth > 0.30
|
||||
|
||||
# 3. Supportive Regime (Protect leaders unless it's a clear TRENDING_DOWN regime)
|
||||
is_bear_regime = regime in ["TRENDING_DOWN", "BEAR", "BEARISH"]
|
||||
# Note: If regime is 'VOLATILE' or 'UNKNOWN', is_bear_regime is False -> Override Logic ACTIVATES.
|
||||
is_bear_regime = regime_val in ["TRENDING_DOWN", "BEAR", "BEARISH"]
|
||||
is_bull_regime = not is_bear_regime
|
||||
|
||||
msg_override = f"DEBUG OVERRIDE: Price={price}, SMA={sma_200}, Growth={growth}, Regime='{regime_val}'"
|
||||
logger.info(msg_override)
|
||||
print(f"[CONSOLE] {msg_override}")
|
||||
logger.info(f"DEBUG CHECK: Technical={is_technical_uptrend}, Growth={is_hyper_growth}, BullRegime={is_bull_regime}")
|
||||
|
||||
# 4. Trigger Override if trying to SELL a leader in a bull market
|
||||
if is_technical_uptrend and is_hyper_growth and is_bull_regime:
|
||||
# We check if the decision string contains SELL or STRONG_SELL
|
||||
# (llm output is usually messy text, so we check for the verdict)
|
||||
decision_upper = trade_decision_str.upper()
|
||||
if "SELL" in decision_upper:
|
||||
print(f"\n🛑 TREND OVERRIDE TRIGGERED for {self.ticker}")
|
||||
print(f" Reason: Stock (${price:.2f}) is > 200SMA (${sma_200:.2f}) and Growth is {growth:.1%}")
|
||||
print(f" Action 'SELL' blocked. Converting to 'HOLD'.\n")
|
||||
|
||||
return {
|
||||
"action": "HOLD",
|
||||
"quantity": 0,
|
||||
"reasoning": (
|
||||
allowed_action = "HOLD"
|
||||
reasoning = (
|
||||
f"OVERRIDE: System attempted to short a Hyper-Growth stock ({growth:.1%}) "
|
||||
f"above its 200-day trend (${sma_200:.2f}) in a Bull regime. "
|
||||
f"Original Decision: {trade_decision_str[:100]}..."
|
||||
),
|
||||
)
|
||||
|
||||
logger.warning(f"🛑 TREND OVERRIDE TRIGGERED for {self.ticker}")
|
||||
print(f"\n[CONSOLE] 🛑 TREND OVERRIDE TRIGGERED for {self.ticker}")
|
||||
logger.warning(f" Reason: Stock (${price:.2f}) is > 200SMA (${sma_200:.2f}) and Growth is {growth:.1%}")
|
||||
logger.warning(f" Action 'SELL' blocked. Converting to '{allowed_action}'.")
|
||||
|
||||
return {
|
||||
"action": allowed_action,
|
||||
"quantity": 0,
|
||||
"reasoning": reasoning,
|
||||
"confidence": 1.0
|
||||
}
|
||||
else:
|
||||
logger.info("DEBUG OVERRIDE: Conditions met, but decision was NOT 'SELL'. No action needed.")
|
||||
else:
|
||||
logger.info("DEBUG OVERRIDE: Conditions NOT met. Passive.")
|
||||
|
||||
return trade_decision_str
|
||||
|
|
|
|||
|
|
@ -0,0 +1,39 @@
|
|||
|
||||
import logging
|
||||
import os
|
||||
from pathlib import Path
|
||||
|
||||
def setup_logger(name: str, log_file: str = "trading_agents.log", level=logging.INFO):
|
||||
"""Function to setup a logger; can be called multiple times for different loggers."""
|
||||
|
||||
# Check if this logger already exists to avoid duplicate handlers
|
||||
logger = logging.getLogger(name)
|
||||
if logger.handlers:
|
||||
return logger
|
||||
|
||||
logger.setLevel(level)
|
||||
|
||||
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
|
||||
|
||||
# Console Handler (Optional - Commented out to keep CLI clean)
|
||||
# ch = logging.StreamHandler()
|
||||
# ch.setFormatter(formatter)
|
||||
# logger.addHandler(ch)
|
||||
|
||||
# File Handler
|
||||
try:
|
||||
# Create logs directory if it doesn't exist?
|
||||
# For now, keep in root or specific path.
|
||||
# Using current working directory for simplicity as requested.
|
||||
file_handler = logging.FileHandler(log_file)
|
||||
file_handler.setFormatter(formatter)
|
||||
logger.addHandler(file_handler)
|
||||
except IOError as e:
|
||||
print(f"Error setting up logger file handler: {e}")
|
||||
|
||||
return logger
|
||||
|
||||
# Create main system logger
|
||||
app_logger = setup_logger("TradingAgents", "agent.log")
|
||||
# Create override specific logger
|
||||
override_logger = setup_logger("OverrideLogic", "agent.log")
|
||||
Loading…
Reference in New Issue