File size: 11,362 Bytes
a4f8e39
 
 
 
 
41714f8
a4f8e39
 
 
b60a104
41714f8
a4f8e39
41714f8
 
 
b60a104
 
 
 
41714f8
 
 
b60a104
 
a4f8e39
 
41714f8
a4f8e39
41714f8
 
 
a4f8e39
 
 
 
 
 
 
41714f8
 
 
b60a104
 
 
 
 
41714f8
 
 
 
 
 
 
b60a104
 
41714f8
 
b60a104
 
 
 
 
 
 
 
 
 
 
 
 
 
41714f8
b60a104
 
 
 
 
 
 
 
 
 
 
 
 
41714f8
b60a104
 
 
 
 
 
 
 
 
 
 
41714f8
 
 
a4f8e39
41714f8
a4f8e39
 
 
 
 
 
 
 
 
 
0be8fa0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a4f8e39
 
41714f8
a4f8e39
 
0be8fa0
a4f8e39
 
 
 
b60a104
a4f8e39
 
 
 
 
 
b60a104
a4f8e39
41714f8
 
 
b60a104
 
 
 
 
 
a4f8e39
b60a104
 
 
a4f8e39
41714f8
b60a104
 
41714f8
 
b60a104
 
 
b2a4427
41714f8
b60a104
 
 
a4f8e39
b60a104
 
 
a4f8e39
41714f8
 
b60a104
 
 
 
a4f8e39
b60a104
41714f8
b60a104
41714f8
 
a4f8e39
b60a104
 
a4f8e39
41714f8
 
 
b60a104
 
 
 
 
41714f8
b60a104
41714f8
b60a104
 
41714f8
 
b60a104
 
 
41714f8
b60a104
41714f8
a4f8e39
41714f8
 
 
b60a104
41714f8
 
 
 
b60a104
 
41714f8
 
 
a4f8e39
b60a104
41714f8
 
b60a104
 
 
 
 
 
 
 
41714f8
b60a104
a4f8e39
41714f8
 
 
b60a104
a4f8e39
 
b60a104
 
 
 
a4f8e39
 
 
b60a104
 
 
 
 
 
41714f8
b60a104
 
 
 
a4f8e39
b60a104
 
 
41714f8
b60a104
 
 
 
 
 
 
 
 
 
 
41714f8
b60a104
 
 
 
 
 
 
41714f8
 
 
 
 
a4f8e39
 
 
b60a104
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
import warnings
warnings.filterwarnings("ignore")

import gradio as gr
import numpy as np
import pandas as pd
import yfinance as yf
import matplotlib.pyplot as plt

from pandas.tseries.frequencies import to_offset
from gluonts.dataset.common import ListDataset

# --- Moirai 2.0 via Uni2TS ---
# Make sure your requirements install Uni2TS from GitHub:
# git+https://github.com/SalesforceAIResearch/uni2ts.git
try:
    from uni2ts.model.moirai2 import Moirai2Forecast, Moirai2Module
except Exception as e:
    raise ImportError(
        "Moirai 2.0 not found in your Uni2TS install.\n"
        "Ensure requirements.txt includes:\n"
        "  git+https://github.com/SalesforceAIResearch/uni2ts.git\n"
        f"Original error: {e}"
    )

MODEL_ID = "Salesforce/moirai-2.0-R-small"
DEFAULT_CONTEXT = 1680  # from Moirai examples, but we clamp to series length

# ----------------------------
# Model loader (single instance)
# ----------------------------
_MODULE = None
def load_module():
    global _MODULE
    if _MODULE is None:
        _MODULE = Moirai2Module.from_pretrained(MODEL_ID)
    return _MODULE

# ----------------------------
# Shared forecasting core
# ----------------------------
def _future_index(last_idx: pd.Timestamp, freq: str, horizon: int) -> pd.DatetimeIndex:
    off = to_offset(freq)
    start = last_idx + off
    return pd.date_range(start=start, periods=horizon, freq=freq)

def _run_forecast_on_series(
    y: pd.Series,
    freq: str,
    horizon: int,
    context_hint: int,
    title: str,
):
    if len(y) < 50:
        raise gr.Error("Need at least 50 points to forecast.")

    ctx = int(np.clip(context_hint or DEFAULT_CONTEXT, 32, len(y)))
    target = y.values[-ctx:].astype(np.float32)
    start_idx = y.index[-ctx]

    ds = ListDataset([{"start": start_idx, "target": target}], freq=freq)

    module = load_module()
    model = Moirai2Forecast(
        module=module,
        prediction_length=int(horizon),
        context_length=ctx,
        target_dim=1,
        feat_dynamic_real_dim=0,
        past_feat_dynamic_real_dim=0,
    )
    predictor = model.create_predictor(batch_size=32)  # device handled internally

    forecast = next(iter(predictor.predict(ds)))
    if hasattr(forecast, "mean"):
        yhat = np.asarray(forecast.mean)
    elif hasattr(forecast, "quantile"):
        yhat = np.asarray(forecast.quantile(0.5))
    elif hasattr(forecast, "samples"):
        yhat = np.asarray(forecast.samples).mean(axis=0)
    else:
        yhat = np.asarray(forecast)

    yhat = np.asarray(yhat).ravel()[:horizon]
    future_idx = _future_index(y.index[-1], freq, horizon)
    pred = pd.Series(yhat, index=future_idx, name="prediction")

    # Plot
    fig = plt.figure(figsize=(10, 5))
    plt.plot(y.index, y.values, label="history")
    plt.plot(pred.index, pred.values, label="forecast")
    plt.title(title)
    plt.xlabel("Time"); plt.ylabel("Value"); plt.legend(); plt.tight_layout()

    out_df = pd.DataFrame({"date": pred.index, "prediction": pred.values})
    return fig, out_df

# ----------------------------
# Ticker helpers
# ----------------------------
def fetch_series(ticker: str, years: int) -> pd.Series:
    """Fetch daily close prices and align to business-day frequency."""
    data = yf.download(
        ticker,
        period=f"{years}y",
        interval="1d",
        auto_adjust=True,
        progress=False,
        threads=True,
    )
    if data is None or data.empty:
        raise gr.Error(f"No price data found for '{ticker}'.")

    col = "Close" if "Close" in data.columns else ("Adj Close" if "Adj Close" in data.columns else None)
    if col is None:
        raise gr.Error(f"Unexpected columns from yfinance: {list(data.columns)}")

    if isinstance(data.columns, pd.MultiIndex):
        if ticker in data[col].columns:
            s = data[col][ticker]
        else:
            s = data[col].iloc[:, 0]
    else:
        s = data[col]

    y = s.copy()
    y.name = ticker
    y.index = pd.DatetimeIndex(y.index).tz_localize(None)

    # Business-day index; forward-fill holidays
    bidx = pd.bdate_range(y.index.min(), y.index.max())
    y = y.reindex(bidx).ffill()

    if y.isna().all():
        raise gr.Error(f"Only missing values for '{ticker}'.")
    return y

def forecast_ticker(ticker: str, horizon: int, lookback_years: int, context_hint: int):
    ticker = (ticker or "").strip().upper()
    if not ticker:
        raise gr.Error("Please enter a ticker symbol (e.g., AAPL).")
    if horizon < 1:
        raise gr.Error("Forecast horizon must be at least 1.")
    y = fetch_series(ticker, lookback_years)
    return _run_forecast_on_series(y, "B", horizon, context_hint, f"{ticker} β€” forecast (Moirai 2.0 R-small)")

# ----------------------------
# CSV helpers
# ----------------------------
def _read_csv_columns(file_path: str) -> pd.DataFrame:
    try:
        df = pd.read_csv(file_path)
    except Exception:
        df = pd.read_csv(file_path, sep=None, engine="python")
    return df

def _coerce_numeric_series(s: pd.Series) -> pd.Series:
    s = pd.to_numeric(s, errors="coerce")
    return s.dropna().astype(np.float32)

def build_series_from_csv(file, value_col: str, date_col: str, freq_choice: str):
    """
    Returns (series y with DateTimeIndex, freq string).
    - If date_col is provided: parse dates and infer/align frequency.
    - If NO date_col: create a synthetic date index using freq_choice (default to 'D' if auto/blank).
    """
    if file is None:
        raise gr.Error("Please upload a CSV file.")

    # Gradio file object handling (v4/v5)
    path = getattr(file, "name", None) or getattr(file, "path", None) or (file if isinstance(file, str) else None)
    if path is None:
        raise gr.Error("Could not read the uploaded file path.")

    df = _read_csv_columns(path)
    if df.empty:
        raise gr.Error("Uploaded file is empty.")

    # Value column selection
    value_col = (value_col or "").strip()
    if value_col:
        if value_col not in df.columns:
            raise gr.Error(f"Value column '{value_col}' not found. Available: {list(df.columns)}")
        vals = _coerce_numeric_series(df[value_col])
    else:
        numeric_cols = [c for c in df.columns if pd.api.types.is_numeric_dtype(df[c])]
        if numeric_cols:
            vals = _coerce_numeric_series(df[numeric_cols[0]])
        else:
            vals = _coerce_numeric_series(df.iloc[:, 0])

    if vals.empty or len(vals) < 10:
        raise gr.Error("Not enough numeric values after parsing (need at least 10).")

    date_col = (date_col or "").strip()
    freq_choice_norm = (freq_choice or "").strip().upper()

    if date_col:
        if date_col not in df.columns:
            raise gr.Error(f"Date column '{date_col}' not found. Available: {list(df.columns)}")
        dt = pd.to_datetime(df[date_col], errors="coerce")
        mask = dt.notna() & vals.notna()
        dt = pd.DatetimeIndex(dt[mask]).tz_localize(None)
        vals = vals[mask]

        if len(vals) < 10:
            raise gr.Error("Too few valid rows after parsing date/value columns.")

        # Sort & dedupe index BEFORE inferring/aligning freq
        order = np.argsort(dt.values)
        dt = dt[order]
        vals = vals.iloc[order].reset_index(drop=True)

        y = pd.Series(vals.values, index=dt, name=value_col or "value").copy()
        y = y[~y.index.duplicated(keep="last")].sort_index()

        # Choose frequency
        if freq_choice_norm and freq_choice_norm != "AUTO":
            freq = freq_choice_norm
        else:
            inferred = pd.infer_freq(y.index)
            if inferred:
                freq = inferred
            else:
                weekday_ratio = (y.index.dayofweek < 5).mean()
                freq = "B" if weekday_ratio > 0.95 else "D"

        # Align to chosen frequency
        y = y.asfreq(freq, method="ffill")

    else:
        # No date column: build synthetic index
        freq = "D" if (not freq_choice_norm or freq_choice_norm == "AUTO") else freq_choice_norm
        idx = pd.date_range(start="2000-01-01", periods=len(vals), freq=freq)
        y = pd.Series(vals.values, index=idx, name=value_col or "value").copy()

    if y.isna().all():
        raise gr.Error("Series is all-NaN after processing.")
    return y, freq

def forecast_csv(file, value_col: str, date_col: str, freq_choice: str, horizon: int, context_hint: int):
    y, freq = build_series_from_csv(file, value_col, date_col, freq_choice)
    return _run_forecast_on_series(y, freq, horizon, context_hint, f"Uploaded series β€” forecast (freq={freq})")

# ----------------------------
# UI
# ----------------------------
with gr.Blocks(title="Moirai 2.0 β€” Time Series Forecast (Research)") as demo:
    gr.Markdown(
        """
# Moirai 2.0 β€” Time Series Forecast (Research)
Use **Salesforce/moirai-2.0-R-small** (via Uni2TS) to forecast either a stock ticker *or* a generic CSV time series.

> **Important**: Research/educational use only. Not investment advice. Model license: **CC-BY-NC-4.0 (non-commercial)**.
        """
    )

    with gr.Tab("By Ticker"):
        with gr.Row():
            ticker = gr.Textbox(label="Ticker", value="AAPL", placeholder="e.g., AAPL, MSFT, TSLA")
            horizon_t = gr.Slider(5, 120, value=30, step=1, label="Forecast horizon (steps)")
        with gr.Row():
            lookback = gr.Slider(1, 10, value=5, step=1, label="Lookback window (years of history)")
            ctx_t = gr.Slider(64, 5000, value=1680, step=16, label="Context length")
        run_t = gr.Button("Run forecast", variant="primary")
        plot_t = gr.Plot(label="History + Forecast")
        table_t = gr.Dataframe(label="Forecast table", interactive=False)
        run_t.click(forecast_ticker, inputs=[ticker, horizon_t, lookback, ctx_t], outputs=[plot_t, table_t])

    with gr.Tab("Upload CSV"):
        gr.Markdown(
            "Upload a CSV with either (1) a **date/time column** and a **value column**, "
            "or (2) just a numeric value column (then choose a frequency, or leave **auto** to default to **D**)."
        )
        with gr.Row():
            file = gr.File(label="CSV file", file_types=[".csv"])
        with gr.Row():
            date_col = gr.Textbox(label="Date/time column (optional)", placeholder="e.g., date, timestamp")
            value_col = gr.Textbox(label="Value column (optional β€” auto-detects first numeric)", placeholder="e.g., value, close")
        with gr.Row():
            freq_choice = gr.Dropdown(
                label="Frequency",
                value="auto",
                choices=["auto", "B", "D", "H", "W", "M", "MS"],
                info="If no date column, 'auto' defaults to D (daily)."
            )
        with gr.Row():
            horizon_u = gr.Slider(1, 500, value=60, step=1, label="Forecast horizon (steps)")
            ctx_u = gr.Slider(32, 5000, value=512, step=16, label="Context length")
        run_u = gr.Button("Run forecast on CSV", variant="primary")
        plot_u = gr.Plot(label="History + Forecast (CSV)")
        table_u = gr.Dataframe(label="Forecast table (CSV)", interactive=False)
        run_u.click(
            forecast_csv,
            inputs=[file, value_col, date_col, freq_choice, horizon_u, ctx_u],
            outputs=[plot_u, table_u],
        )

if __name__ == "__main__":
    demo.launch()