push to github

app.py

@@ -0,0 +1,30 @@
+import pandas as pd
+import streamlit as st
+from streamlit_autorefresh import st_autorefresh
+from raindrop import make_raindrop_chart
+
+st.set_page_config(layout="wide", page_title="Raindrop Charts")
+tickers = pd.read_csv("tickers.csv")
+st.title("Raindrop Charts using Yfinance, Streamlit, & Plotly")
+
+today = pd.Timestamp.now()
+date = st.sidebar.date_input(label="Date Range", value=today)
+company = st.sidebar.selectbox(label="Company", options=tickers["Company"])
+vwap_margin = st.sidebar.number_input(label="VWAP Margin", value=0.1, step=0.01, min_value=0., format="%.2f")
+frequency = st.sidebar.number_input(label="Bin Size (minutes)", value=30, step=1, min_value=5, max_value=60)
+ticker = tickers.loc[tickers["Company"] == company, "Ticker"].values[0]
+if pd.Timestamp(date) >= pd.Timestamp.now().floor("d"):
+    count = st_autorefresh(interval=5000, limit=100, key="fizzbuzzcounter")
+raindrop_chart, vwap_open, vwap_close, ohlc = make_raindrop_chart(
+    ticker=ticker,
+    start=date.strftime("%Y-%m-%d"),
+    end=(date + pd.Timedelta(days=1)).strftime("%Y-%m-%d"),
+    interval="1m",
+    frequency_value=frequency,
+    margin=vwap_margin
+)
+col1, col2, col3 = st.columns(3)
+col1.metric("VWAP (Current vs Previous)", f"{str(vwap_close)}$", f"{str(vwap_close - vwap_open)}$")
+col2.metric("Current Prices (Close vs Open)", f"{str(ohlc['Close'])}$", f"{str(ohlc['Close'] - ohlc['Open'])}$")
+col3.metric("Last Update", str(pd.Timestamp.now().floor("s")))
+st.plotly_chart(raindrop_chart, use_container_width=True)

raindrop.py

@@ -0,0 +1,152 @@
+import pandas as pd
+import yfinance as yf
+import plotly.graph_objects as go
+from plotly.subplots import make_subplots
+
+
+def make_raindrop_chart(
+    ticker: str = "AAPL",
+    start: str = "2022-01-10",
+    end: str = "2022-01-11",
+    interval: str = "5m",
+    frequency_unit: str = "m",
+    frequency_value: int = 30,
+    margin: float = 0.1
+) -> go.Figure:
+    df = yf.download(
+        tickers=ticker,
+        start=start,
+        end=end,
+        interval=interval,
+    ).reset_index()
+    df["Typical"] = df[["Open", "High", "Low", "Close"]].sum(axis=1)/4
+    df["QTY*PX"] = df["Volume"] * df["Typical"]
+
+    grouping_frequency = pd.Timedelta(frequency_value, unit=frequency_unit)
+    split_frequency = pd.Timedelta(grouping_frequency.total_seconds() / 2, unit="s")
+
+    ohlc = df.groupby(pd.Grouper(key="Datetime", freq=grouping_frequency)).agg(
+        Open=("Open", "first"),
+        High=("High", "max"),
+        Low=("Open", "min"),
+        Close=("Open", "last"),
+        Volume=("Volume", "sum")
+    )
+    ohlc = ohlc.query("Volume > 0").reset_index()
+    df["Split"] = df.groupby(pd.Grouper(key="Datetime", freq=split_frequency)).ngroup()
+    df["Split"] = df["Split"] % 2
+    df["Datetime"] = df["Datetime"].dt.floor(grouping_frequency)
+    volume_divider = df["Volume"].max()/1000
+
+    fig = make_subplots(
+        rows=3,
+        cols=1,
+        row_heights=[0.45, 0.45, 0.1],
+        shared_xaxes=True,
+        vertical_spacing=0.01
+    )
+    showlegend = True
+
+    for period, period_df in df.groupby("Datetime"):
+        vwap_df = period_df.groupby("Split", as_index=False)[["Volume", "QTY*PX"]].sum()
+        vwap_df = vwap_df.query("Volume > 0").reset_index()
+        if not vwap_df.empty:
+            vwap_df["VWAP"] = vwap_df["QTY*PX"] / vwap_df["Volume"]
+            color = "blue"
+            if len(vwap_df.index) > 1:
+                vwap_open, vwap_close = vwap_df.loc[0, "VWAP"], vwap_df.loc[1, "VWAP"]
+                if (vwap_close - vwap_open) > margin:
+                    color = "green"
+                elif (vwap_open - vwap_close) > margin:
+                    color = "red"
+            period_df["Volume"] = period_df["Volume"].div(volume_divider).round()
+            for split, split_df in period_df.groupby("Split"):
+                is_pre_split = split == 0
+                # We multiply each row by volume of the trade. we multiply by 10 since smallest volume is 0.1
+                split_df = split_df.loc[split_df.index.repeat(split_df["Volume"])]
+                fig.add_trace(
+                    go.Violin(
+                        x=split_df["Datetime"],
+                        y=split_df["Typical"],
+                        side="negative" if is_pre_split else "positive",
+                        name="Raindrop",
+                        legendgroup="Raindrop",
+                        showlegend=showlegend,
+                        line=dict(color=color),
+                        spanmode="hard",
+                        scalegroup=str(period),
+                        scalemode="count",
+                        points=False,
+                        hoverinfo="y",
+                        hoveron="violins",
+                        meanline=dict(color="white"),
+                    ),
+                    row=1,
+                    col=1
+                )
+                showlegend = False
+
+    fig.add_trace(
+        go.Candlestick(
+            x=ohlc["Datetime"],
+            open=ohlc["Open"],
+            high=ohlc["High"],
+            low=ohlc["Low"],
+            close=ohlc["Close"],
+            name="OHLC",
+            decreasing_line_color="red",
+            increasing_line_color="green",
+        ),
+        row=2,
+        col=1
+    )
+
+    ohlc["BarColor"] = ohlc.apply(lambda x: "green" if x["Open"] < x["Close"] else "red", axis=1)
+    showlegend = True
+    for color, sub_ohlc in ohlc.groupby("BarColor"):
+        fig.add_trace(
+            go.Bar(
+                x=sub_ohlc["Datetime"],
+                y=sub_ohlc["Volume"],
+                marker=dict(color=color),
+                legendgroup="Volume",
+                name="Volume",
+                showlegend=showlegend,
+                hovertemplate=None,
+                texttemplate="%{y:.2s}"
+            ),
+            row=3,
+            col=1
+        )
+        showlegend = False
+
+    fig.update_xaxes(
+        rangeslider_visible=False,
+        row=2
+    )
+    fig.update_xaxes(
+        dtick=1000 * grouping_frequency.total_seconds(),
+        showgrid=True,
+        title="Datetime",
+        row=3
+    )
+    fig.update_xaxes(
+        rangebreaks=[dict(bounds=[16, 9.5], pattern="hour")],
+    )
+    fig.update_yaxes(title="Price", row=1)
+    fig.update_yaxes(title="Price", row=2)
+    fig.update_yaxes(title="Volume", row=3)
+    fig.update_layout(
+        title=dict(text=ticker),
+        violingap=0,
+        violingroupgap=0,
+        template="plotly_dark",
+        height=800,
+        uirevision="uirevision"
+    )
+    return fig, vwap_open, vwap_close, ohlc.to_dict("records")[-1]
+
+
+if __name__ == "__main__":
+    raindrop = make_raindrop_chart()[0]
+    raindrop.show("browser")

requirements.txt

@@ -0,0 +1,5 @@
+pandas~=1.3
+plotly~=5.5
+streamlit~=1.3
+streamlit-autorefresh==0.0.1
+yfinance~=0.1

tickers.csv

@@ -0,0 +1,104 @@
+Company,Ticker
+Activision Blizzard,ATVI
+Adobe Inc.,ADBE
+Advanced Micro Devices,AMD
+Alexion Pharmaceuticals,ALXN
+Align Technology,ALGN
+Alphabet Inc. (Class A),GOOGL
+Alphabet Inc. (Class C),GOOG
+Amazon.com,AMZN
+Amgen,AMGN
+Analog Devices,ADI
+ANSYS,ANSS
+Apple Inc.,AAPL
+"Applied Materials, Inc.",AMAT
+ASML Holding,ASML
+Autodesk,ADSK
+"Automatic Data Processing, Inc.",ADP
+Baidu,BIDU
+Biogen,BIIB
+BioMarin Pharmaceutical Inc.,BMRN
+Booking Holdings,BKNG
+Broadcom Inc.,AVGO
+Cadence Design Systems,CDNS
+CDW,CDW
+Cerner Corporation,CERN
+"Charter Communications, Inc.",CHTR
+Check Point Software Technologies Ltd.,CHKP
+Cintas Corporation,CTAS
+Cisco Systems,CSCO
+Citrix Systems,CTXS
+Cognizant Technology Solutions Corporation,CTSH
+Comcast Corporation,CMCSA
+Copart,CPRT
+Costco Wholesale Corporation,COST
+CSX Corporation,CSX
+DexCom,DXCM
+DocuSign,DOCU
+"Dollar Tree, Inc.",DLTR
+eBay Inc.,EBAY
+Electronic Arts,EA
+Exelon Corporation,EXC
+Expedia Group,EXPE
+"Facebook, Inc.",FB
+Fastenal Company,FAST
+"Fiserv, Inc.",FISV
+Fox Corporation (Class A),FOXA
+Fox Corporation (Class B),FOX
+"Gilead Sciences, Inc.",GILD
+IDEXX Laboratories,IDXX
+"Illumina, Inc.",ILMN
+Incyte Corporation,INCY
+Intel Corporation,INTC
+Intuit,INTU
+Intuitive Surgical,ISRG
+JD.com,JD
+Keurig Dr Pepper Inc.,KDP
+KLA Corporation,KLAC
+Kraft Heinz Co,KHC
+Lam Research,LRCX
+Liberty Global (Class A),LBTYA
+Liberty Global (Class C),LBTYK
+Lululemon Athletica,LULU
+Marriott International,MAR
+Maxim Integrated Products,MXIM
+MercadoLibre,MELI
+Microchip Technology,MCHP
+"Micron Technology, Inc.",MU
+Microsoft Corporation,MSFT
+Moderna,MRNA
+Mondelēz International,MDLZ
+Monster Beverage Corporation,MNST
+"NetEase, Inc.",NTES
+Netflix,NFLX
+Nvidia Corporation,NVDA
+NXP Semiconductors N.V.,NXPI
+"O'Reilly Automotive, Inc.",ORLY
+PACCAR Inc.,PCAR
+"Paychex, Inc.",PAYX
+PayPal,PYPL
+"PepsiCo, Inc.",PEP
+Pinduoduo Inc.,PDD
+QUALCOMM,QCOM
+Regeneron Pharmaceuticals,REGN
+Ross Stores Inc.,ROST
+Seagen Inc.,SGEN
+"Sirius XM Radio, Inc.",SIRI
+"Skyworks Solutions, Inc.",SWKS
+Splunk,SPLK
+Starbucks Corporation,SBUX
+"Synopsys, Inc.",SNPS
+T-Mobile US,TMUS
+"Take-Two Interactive, Inc.",TTWO
+"Tesla, Inc.",TSLA
+Texas Instruments,TXN
+Trip.com Group,TCOM
+Ulta Beauty,ULTA
+Verisign,VRSN
+Verisk Analytics,VRSK
+Vertex Pharmaceuticals,VRTX
+"Walgreen Boots Alliance, Inc.",WBA
+"Workday, Inc.",WDAY
+"Xcel Energy, Inc.",XEL
+"Xilinx, Inc.",XLNX
+Zoom Video Communications,ZM