Upload 3 files

app.py

@@ -0,0 +1,128 @@
+import gradio as gr
+from utils import Kline_predict_plot, back_test_plot
+
+
+with gr.Blocks() as generator:
+    with gr.Tab("English"):
+        gr.Markdown("# Available Markets: Stocks(Main China, HongKong, USA and so on) / Cryptocurrencies")
+        # gr.Markdown("Powered by AI")
+        with gr.Tab("Predictor"):
+            # gr.Markdown("Most 5 symbols at once can be put in `Symbols` box.")
+            once_max_num = 5
+            gr.Interface(
+                Kline_predict_plot,
+                [
+                    gr.Dropdown(["model_predict_sk2", ], label="Predict Model", value='model_predict_sk2'),
+                    gr.Textbox(label="Symbols", lines=1,
+                               placeholder=f"Enter symbols(can't more than {once_max_num} at once), "
+                                           f"such as '000001.SH, btc-usd, AAPL, 18010.hk'",
+                    ),
+                    gr.Textbox(label="Your Opinion", lines=2,
+                               placeholder=f"Enter your base opinion for next trade day if you have("
+                                           f"such as 'I think it would be a big up and I would long absolutely' or "
+                                           f"'I shall sell a few'); or just leave here blank.",
+                               ),
+                    gr.Dropdown(
+                        ["1hour", "1day", "1week", "1month"], label="Frequency Scale", value='1day', info="Kline scale."
+                    ),
+                    gr.Slider(label="Feature Saturation", minimum=2, maximum=10, value=3, step=1),
+                    gr.Slider(label="Sensitivity", minimum=2, maximum=20, value=5, step=1),
+                    gr.Checkbox(label="Completed?", value=True, info="Is the latest bar completed?"),
+                    gr.Slider(label="Show Bars", minimum=16, maximum=100, value=21, step=1),
+                    gr.Number(label="Once Max", value=once_max_num, visible=False)
+                ],
+                [
+                    gr.Textbox(label=f"Ignored Symbols: not available，too new or more than {once_max_num} symbols."),
+                    gr.Gallery(label="Bars Figure", show_label=True, columns=[1]),
+                ],
+                examples=[
+                    ["model_predict_sk2", "000001.sh,600000.sh,000001.sz", "", "1day", 3, 5, True, 21, 5],
+                    ["model_predict_sk2", "aapl,goog", "There would be a big up, so long", "1week", 5, 15, True, 21, 5],
+                    ["model_predict_sk2", "btc-usd,eth-btc", "A big down is coming", "1day", 3, 3, False, 21, 5],
+                    ["model_predict_sk2", "btc-usd,000001.sh,1810.hk goog", "", "1day", 3, 10, False, 30, 5],
+                ],
+            )
+        with gr.Tab("Back Tester"):
+            # gr.Markdown("Only 1 symbol at once can be put in `Symbols` box.")
+            gr.Interface(
+                back_test_plot,
+                [
+                    gr.Dropdown(["model_predict_sk2", ], label="Predict Model", value='model_predict_sk2'),
+                    gr.Textbox(label="Symbols", placeholder=f"Enter 1 Symbol, such as 'btc-usd'"),
+                    gr.Dropdown(
+                        ["1hour", "1day", "1week", "1month"], label="Frequency Scale", value='1day', info="Kline scale."
+                    ),
+                    gr.Slider(label="Feature Saturation", minimum=2, maximum=10, value=3, step=1),
+                    gr.Slider(label="Sensitivity", minimum=2, maximum=20, value=5, step=1),
+                    gr.Slider(label="Test Bars", minimum=10, maximum=200, value=100, step=1),
+                ],
+                [gr.Gallery(label="Profit Figure", show_label=True, columns=[1])],
+                examples=[
+                    ["model_predict_sk2", "000001.sh", "1day", 3, 5, 20],
+                    ["model_predict_sk2", "btc-usd", "1day", 3, 10, 30],
+                    ["model_predict_sk2", "aapl", "1week", 5, 15, 80],
+                ],
+            )
+    with gr.Tab("中文"):
+        gr.Markdown("# 可用市场: 股票(A股、港股、美股等) / 加密货币")
+        with gr.Tab("预测器"):
+            # gr.Markdown("`交易标的` 框中一次最多可同时输入5个标的")
+            once_max_num = 5
+            gr.Interface(
+                Kline_predict_plot,
+                [
+                    gr.Dropdown(["model_predict_sk2", ], label="预测模型", value='model_predict_sk2'),
+                    gr.Textbox(label="交易标的", lines=2,
+                               placeholder=f"请输入要预测的交易标的(一次最多不能超过{once_max_num}个), "
+                                           f"例如输入： \"000001.SH, btc-usd, AAPL, 18010.hk\"",
+                    ),
+                    gr.Textbox(label="主观观点", lines=2,
+                               placeholder=f"请输入你的主观多空观点(例如输入： \"我认为后市会大涨，我要全仓做多\" 或者 "
+                                           f"\"我认为应该少量卖出一部分\"); 如果没有明确的多空观点，此处无需输入"),
+                    gr.Dropdown(
+                        ["1hour", "1day", "1week", "1month"], label="K线级别", value='1day',
+                        info="选择小时K线、日K线、周K线或者月K线"
+                    ),
+                    gr.Slider(label="特征饱和度", minimum=2, maximum=10, value=3, step=1),
+                    gr.Slider(label="敏感度", minimum=2, maximum=20, value=5, step=1),
+                    gr.Checkbox(label="已完成?", value=True, info="最新的K线是否已经完成?"),
+                    gr.Slider(label="展示K线数", minimum=16, maximum=100, value=21, step=1),
+                    gr.Number(label="最多预测数", value=once_max_num, visible=False)
+                ],
+                [
+                    gr.Textbox(label=f"有误交易标的: 不可用、太新或者一次预测数大于{once_max_num}个"),
+                    gr.Gallery(label="K线图", show_label=True, columns=[1]),
+                ],
+                examples=[
+                    ["model_predict_sk2", "000001.sh,600000.sh,000001.sz", "", "1day", 3, 5, True, 21, 5],
+                    ["model_predict_sk2", "aapl,goog", "我认为后市会大涨，我要全仓做多", "1week", 5, 15, True, 21, 5],
+                    ["model_predict_sk2", "btc-usd,eth-btc", "后市很可能会大幅下跌", "1day", 3, 3, False, 21, 5],
+                    ["model_predict_sk2", "btc-usd,000001.sh,1810.hk goog", "", "1day", 3, 10, False, 30, 5],
+                ],
+            )
+        with gr.Tab("回测器"):
+            # gr.Markdown("Only 1 symbol at once can be put in `Symbols` box.")
+            gr.Interface(
+                back_test_plot,
+                [
+                    gr.Dropdown(["model_predict_sk2", ], label="预测模型", value='model_predict_sk2'),
+                    gr.Textbox(label="交易标的", placeholder=f"请输入一个交易标的, 例如输入：\"btc-usd\""),
+                    gr.Dropdown(
+                        ["1hour", "1day", "1week", "1month"], label="K线级别", value='1day',
+                        info="选择小时K线、日K线、周K线或者月K线"
+                    ),
+                    gr.Slider(label="特征饱和度", minimum=2, maximum=10, value=3, step=1),
+                    gr.Slider(label="敏感度", minimum=2, maximum=20, value=5, step=1),
+                    gr.Slider(label="回测K线数", minimum=10, maximum=200, value=100, step=1),
+                ],
+                [gr.Gallery(label="回测曲线", show_label=True, columns=[1])],
+                examples=[
+                    ["model_predict_sk2", "000001.sh", "1day", 3, 5, 20],
+                    ["model_predict_sk2", "btc-usd", "1day", 3, 10, 30],
+                    ["model_predict_sk2", "aapl", "1week", 5, 15, 80],
+                ],
+            )
+
+
+generator.queue(max_size=1).launch()
+

requirements.txt

@@ -0,0 +1 @@
+mplfinance

utils.py

@@ -0,0 +1,92 @@
+import io
+import os
+import json
+import requests
+import pandas as pd
+import mplfinance as mpf
+import matplotlib.pyplot as plt
+from PIL import Image
+
+
+def generate_pil_img():
+    img_buf = io.BytesIO()
+    plt.savefig(img_buf, format='png')
+    return Image.open(img_buf)
+
+
+def plot_Klines_fig(bars, code, precision=0.0001):
+    fig, axes = mpf.plot(bars, type='candle', mav=(5, 10), volume=True, title=f'\n{code}',
+                         figsize=(12, 6), returnfig=True)
+    fixed_txt = 'Predict\nNext Bar'
+    for ci in ['Open', 'High', 'Low', 'Close', 'Volume']:
+        fixed_txt += f'\n{ci[0]}:{"{:.3e}".format(bars.iloc[-1][ci])}'
+    hl_chg = bars['High'].iloc[-1] / bars['Low'].iloc[-1] - 1.0
+    axes[0].annotate(fixed_txt, textcoords='axes fraction', xytext=(1.006, -0.37), color='r',
+                     bbox=dict(facecolor='gray'),
+                     xy=(len(bars) - 1, max(bars['Low'].min(), bars.iloc[-1]['Low'] * (1 - hl_chg / 4))),
+                     arrowprops=dict(facecolor='r', width=5))
+    return generate_pil_img()
+
+
+def plot_txt(text):
+    plt.figure(figsize=(12, 6))
+    plt.text(0.2, 0.8, text, size=12, alpha=0.8)
+    plt.axis('off')
+    return generate_pil_img()
+
+
+def Kline_predict_plot(model_id, symbols, user_opinion, frequency, feature_saturation, sensitivity,
+                       latest_bar_completed, show_bars, once_max):
+    params = {
+        'codes_str': symbols,
+        'user_opinion': user_opinion,
+        'frequency': frequency,
+        'feature_saturation': feature_saturation,
+        'latest_completed': latest_bar_completed,
+        'sensitivity': sensitivity,
+        'show_bars': show_bars,
+        'once_max': once_max,
+    }
+    headers = {'Content-Type': 'application/json'}
+    response = requests.post(os.environ[model_id] + 'predict', data=json.dumps(params), headers=headers)
+    result = response.json()
+    coins_not_available = result['codes_remove']
+    klines_data = result['klines_data']
+    figs = []
+    for codei, vi in klines_data.items():
+        bars = pd.DataFrame(vi)
+        bars.index = bars.index.map(pd.Timestamp)
+        figs.append(plot_Klines_fig(bars, codei))
+    return ','.join(coins_not_available) if len(coins_not_available) > 0 else 'No one', figs
+
+
+def back_test_plot(model_id, symbols, frequency, feature_saturation, sensitivity, test_bars):
+    params = {
+        'codes_str': symbols,
+        'frequency': frequency,
+        'feature_saturation': feature_saturation,
+        'sensitivity': sensitivity,
+        'test_bars': test_bars,
+    }
+    headers = {'Content-Type': 'application/json'}
+    response = requests.post(os.environ[model_id] + 'back_test', data=json.dumps(params), headers=headers)
+    result = response.json()
+    test_result = result['test_result']
+    if isinstance(test_result, str):
+        return [plot_txt(test_result)]
+    else:
+        assert isinstance(test_result, dict), "`test_result` type is wrong."
+        keys_list = list(test_result.keys())
+        code, code_name = keys_list if len(keys_list[0]) < len(keys_list[1]) else keys_list[::-1]
+        labels, profits = pd.Series(test_result[code]), pd.Series(test_result[code_name])
+        labels.index, profits.index = labels.index.map(pd.Timestamp), profits.index.map(pd.Timestamp)
+        plt.figure()
+        (profits + 1.0).cumprod().plot(
+            label=f'profits => win_ratio:{round((profits > 0).mean() * 100, 2)}%')
+        (labels + 1.0).cumprod().plot(
+            label=f'{code} => up_ratio:{round((labels > 0).mean() * 100, 2)}%', linestyle='--')
+        plt.title(code_name)
+        plt.legend()
+        plt.grid()
+        return [generate_pil_img()]
+