search

nateemma / strategies

Custom trading strategies using the freqtrade framework
329 stars 88 forks source link

NNBC Crash #13

Closed aloksaurabh closed 1 year ago

aloksaurabh commented 1 year ago

System Information

alok@kaliSSD:/media/alok/New_Volume/kali/conda/freqtrade$ sudo lshw -short
H/W path              Device          Class          Description
================================================================
                                      system         ASUS TUF Gaming A15 FA506IU_FA506IU
/0                                    bus            FA506IU
/0/0                                  memory         64KiB BIOS
/0/a                                  memory         64GiB System Memory
/0/a/0                                memory         32GiB SODIMM DDR4 Synchronous Unbuffered (Unregistered) 3200 MHz (0.3 ns)
/0/a/1                                memory         32GiB SODIMM DDR4 Synchronous Unbuffered (Unregistered) 3200 MHz (0.3 ns)
/0/c                                  memory         512KiB L1 cache
/0/d                                  memory         4MiB L2 cache
/0/e                                  memory         8MiB L3 cache
/0/f                                  processor      AMD Ryzen 9 4900H with Radeon Graphics
/0/100                                bridge         Renoir/Cezanne Root Complex
/0/100/0.2                            generic        Renoir/Cezanne IOMMU
/0/100/1.1                            bridge         Renoir PCIe GPP Bridge
/0/100/1.1/0                          display        TU116M [GeForce GTX 1660 Ti Mobile]
/0/100/1.1/0.1        card0           multimedia     TU116 High Definition Audio Controller
/0/100/1.1/0.1/0      input12         input          HDA NVidia HDMI/DP,pcm=3
/0/100/1.1/0.1/1      input13         input          HDA NVidia HDMI/DP,pcm=7
/0/100/1.1/0.1/2      input14         input          HDA NVidia HDMI/DP,pcm=8
/0/100/1.1/0.1/3      input15         input          HDA NVidia HDMI/DP,pcm=9
/0/100/1.1/0.2                        bus            TU116 USB 3.1 Host Controller
/0/100/1.1/0.2/0      usb1            bus            xHCI Host Controller
/0/100/1.1/0.2/1      usb2            bus            xHCI Host Controller
/0/100/1.1/0.3                        bus            TU116 USB Type-C UCSI Controller
/0/100/2.1                            bridge         Renoir/Cezanne PCIe GPP Bridge
/0/100/2.1/0          eth0            network        RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller
/0/100/2.2                            bridge         Renoir/Cezanne PCIe GPP Bridge
/0/100/2.2/0          wlan0           network        RTL8822CE 802.11ac PCIe Wireless Network Adapter
/0/100/2.3                            bridge         Renoir/Cezanne PCIe GPP Bridge
/0/100/2.3/0          /dev/nvme0      storage        PCIe SSD
/0/100/2.3/0/0        hwmon2          disk           NVMe disk
/0/100/2.3/0/2        /dev/ng0n1      disk           NVMe disk
/0/100/2.3/0/1        /dev/nvme0n1    disk           2048GB NVMe disk
/0/100/2.3/0/1/1      /dev/nvme0n1p1  volume         1907GiB Windows NTFS volume
/0/100/2.4                            bridge         Renoir/Cezanne PCIe GPP Bridge
/0/100/2.4/0          /dev/nvme1      storage        SPCC M.2 PCIe SSD

My setup

OS Type:    linux, Version: Linux-6.0.0-kali6-amd64-x86_64-with-glibc2.36
    python:     ['3.10.0 | packaged by conda-forge | (default, Nov 20 2021, 02:24:10) [GCC 9.4.0]']
    sklearn:    1.2.1
    tensorflow: 2.11.0, devices:[PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU'), PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
    keras:      2.11.0
    pytorch:    1.13.1
    lightning:  1.6.5
    darts:      0.23.1

Lookahead:  12  candles ( 1.0  hours)

ONE/USDT
    Compressed data 78 -> 64 (features)
    #training samples: 151455  #buys: 6068  #sells: 6072
    Loading existing model (/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/models/NNBC_fbb/NNBC_fbb_Transformer_Buy.h5)...
    Model is already trained
    Loading existing model (/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/models/NNBC_fbb/NNBC_fbb_Transformer_Sell.h5)...
    Model is already trained
    predicting buys...
    predicting sells...

WTC/USDT
    Compressed data 78 -> 64 (features)
    #training samples: 151455  #buys: 5918  #sells: 4420
    Model is already trained
    Model is already trained
    predicting buys...
    predicting sells...

Output

*** top 5 stats grouped by filename ***
/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/profiler.py:0: size=416 B, count=1, average=416 B
/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/NNBC.py:0: size=408 B, count=1, average=408 B

*** top 10 stats ***
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py:2388: size=430 MiB (+430 MiB), count=7 (+7), average=61.4 MiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/talib/__init__.py:27: size=41.7 MiB (+41.7 MiB), count=53 (+53), average=806 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/frame.py:12016: size=32.1 MiB (+32.1 MiB), count=40 (+40), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/dtypes/cast.py:1434: size=22.5 MiB (+22.5 MiB), count=28 (+28), average=822 KiB
<__array_function__ internals>:180: size=3304 KiB (+3304 KiB), count=48 (+48), average=68.8 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py:549: size=3288 KiB (+3288 KiB), count=25 (+25), average=132 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:562: size=3287 KiB (+3287 KiB), count=6 (+6), average=548 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimes.py:987: size=3287 KiB (+3287 KiB), count=4 (+4), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/algorithms.py:1251: size=3287 KiB (+3287 KiB), count=4 (+4), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/abc.py:123: size=2394 KiB (+2394 KiB), count=23960 (+23960), average=102 B

*** top 10 stats ***
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py:2388: size=591 MiB (+591 MiB), count=9 (+9), average=65.6 MiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/talib/__init__.py:27: size=41.7 MiB (+41.7 MiB), count=52 (+52), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/frame.py:12016: size=32.1 MiB (+32.1 MiB), count=40 (+40), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/dtypes/cast.py:1434: size=22.5 MiB (+22.5 MiB), count=28 (+28), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py:549: size=11.2 MiB (+11.2 MiB), count=30 (+30), average=384 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:562: size=6575 KiB (+6575 KiB), count=12 (+12), average=548 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimes.py:987: size=4931 KiB (+4931 KiB), count=6 (+6), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/algorithms.py:1251: size=4931 KiB (+4931 KiB), count=6 (+6), average=822 KiB
<__array_function__ internals>:180: size=3304 KiB (+3304 KiB), count=51 (+51), average=64.8 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexes/range.py:202: size=3287 KiB (+3287 KiB), count=4 (+4), average=822 KiB

Crash

1

*** Trace for largest memory block - (72 blocks, 5916631.5 Kb) ***
  File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/commands/optimize_commands.py", line 58
    backtesting.start()
  File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/optimize/backtesting.py", line 1344
    min_date, max_date = self.backtest_one_strategy(strat, data, timerange)
  File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/optimize/backtesting.py", line 1246
    preprocessed = self.strategy.advise_all_indicators(data)
  File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/strategy/interface.py", line 1318
    return {pair: self.advise_indicators(pair_data.copy(), {'pair': pair}).copy()
  File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/strategy/interface.py", line 1318
    return {pair: self.advise_indicators(pair_data.copy(), {'pair': pair}).copy()
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/generic.py", line 6368
    data = self._mgr.copy(deep=deep)
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 670
    res._consolidate_inplace()
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 1871
    self.blocks = _consolidate(self.blocks)
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 2329
    merged_blocks, _ = _merge_blocks(
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 2388
    new_values = new_values[argsort]

2

*** top 10 stats ***
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py:2388: size=6218 MiB (+6218 MiB), count=84 (+84), average=74.0 MiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/array_algos/take.py:158: size=6002 MiB (+6002 MiB), count=189 (+189), average=31.8 MiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:562: size=61.0 MiB (+61.0 MiB), count=114 (+114), average=548 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/algorithms.py:1577: size=59.4 MiB (+59.4 MiB), count=74 (+74), average=821 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexes/range.py:202: size=57.8 MiB (+57.8 MiB), count=72 (+72), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/talib/__init__.py:27: size=41.7 MiB (+41.7 MiB), count=52 (+52), average=822 KiB
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/frame.py:12016: size=32.1 MiB (+32.1 MiB), count=40 (+40), average=822 KiB
/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/optimize/backtesting.py:390: size=30.5 MiB (+30.5 MiB), count=420483 (+420483), average=76 B
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py:1914: size=28.9 MiB (+28.9 MiB), count=1261404 (+1261404), average=24 B
/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:438: size=25.7 MiB (+25.7 MiB), count=210241 (+210241), average=128 B

*** Trace for largest memory block - (76 blocks, 6083852.5 Kb) ***
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexing.py", line 1292
    return self._getbool_axis(key, axis=axis)
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexing.py", line 1093
    return self.obj._take_with_is_copy(inds, axis=axis)
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/generic.py", line 3902
    result = self._take(indices=indices, axis=axis)
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/generic.py", line 3886
    new_data = self._mgr.take(
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 978
    return self.reindex_indexer(
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 751
    new_blocks = [
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 752
    blk.take_nd(
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py", line 880
    new_values = algos.take_nd(
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/array_algos/take.py", line 117
    return _take_nd_ndarray(arr, indexer, axis, fill_value, allow_fill)
  File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/array_algos/take.py", line 158
    out = np.empty(out_shape, dtype=dtype)

*** top 5 stats grouped by filename ***
/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/profiler.py:0: size=416 B, count=1, average=416 B
/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/NNBC.py:0: size=408 B, count=1, average=408 B

I am sorry i am a bit new to ML. I understand python and bash well. Could you advise what is happening ?

nateemma commented 1 year ago

other people have found that this strategy has a memory leak problem due to TensorFlow on some linux variants, but we haven't been able to find a workaround yet. I am currently writing replacements for all of the NNBC strategies that should use less memory - I think NNTC.py is already on github, and it is the equivalent of the strategy you are using here, so maybe try that?

Thanks,

Phil

On Thu, Feb 23, 2023 at 7:56 PM aloksaurabh @.***> wrote:

System Information

@.***:/media/alok/New_Volume/kali/conda/freqtrade$ sudo lshw -short H/W path Device Class Description

                                  system         ASUS TUF Gaming A15 FA506IU_FA506IU

/0 bus FA506IU /0/0 memory 64KiB BIOS /0/a memory 64GiB System Memory /0/a/0 memory 32GiB SODIMM DDR4 Synchronous Unbuffered (Unregistered) 3200 MHz (0.3 ns) /0/a/1 memory 32GiB SODIMM DDR4 Synchronous Unbuffered (Unregistered) 3200 MHz (0.3 ns) /0/c memory 512KiB L1 cache /0/d memory 4MiB L2 cache /0/e memory 8MiB L3 cache /0/f processor AMD Ryzen 9 4900H with Radeon Graphics /0/100 bridge Renoir/Cezanne Root Complex /0/100/0.2 generic Renoir/Cezanne IOMMU /0/100/1.1 bridge Renoir PCIe GPP Bridge /0/100/1.1/0 display TU116M [GeForce GTX 1660 Ti Mobile] /0/100/1.1/0.1 card0 multimedia TU116 High Definition Audio Controller /0/100/1.1/0.1/0 input12 input HDA NVidia HDMI/DP,pcm=3 /0/100/1.1/0.1/1 input13 input HDA NVidia HDMI/DP,pcm=7 /0/100/1.1/0.1/2 input14 input HDA NVidia HDMI/DP,pcm=8 /0/100/1.1/0.1/3 input15 input HDA NVidia HDMI/DP,pcm=9 /0/100/1.1/0.2 bus TU116 USB 3.1 Host Controller /0/100/1.1/0.2/0 usb1 bus xHCI Host Controller /0/100/1.1/0.2/1 usb2 bus xHCI Host Controller /0/100/1.1/0.3 bus TU116 USB Type-C UCSI Controller /0/100/2.1 bridge Renoir/Cezanne PCIe GPP Bridge /0/100/2.1/0 eth0 network RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller /0/100/2.2 bridge Renoir/Cezanne PCIe GPP Bridge /0/100/2.2/0 wlan0 network RTL8822CE 802.11ac PCIe Wireless Network Adapter /0/100/2.3 bridge Renoir/Cezanne PCIe GPP Bridge /0/100/2.3/0 /dev/nvme0 storage PCIe SSD /0/100/2.3/0/0 hwmon2 disk NVMe disk /0/100/2.3/0/2 /dev/ng0n1 disk NVMe disk /0/100/2.3/0/1 /dev/nvme0n1 disk 2048GB NVMe disk /0/100/2.3/0/1/1 /dev/nvme0n1p1 volume 1907GiB Windows NTFS volume /0/100/2.4 bridge Renoir/Cezanne PCIe GPP Bridge /0/100/2.4/0 /dev/nvme1 storage SPCC M.2 PCIe SSD

My setup

OS Type: linux, Version: Linux-6.0.0-kali6-amd64-x86_64-with-glibc2.36 python: ['3.10.0 | packaged by conda-forge | (default, Nov 20 2021, 02:24:10) [GCC 9.4.0]'] sklearn: 1.2.1 tensorflow: 2.11.0, devices:[PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU'), PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')] keras: 2.11.0 pytorch: 1.13.1 lightning: 1.6.5 darts: 0.23.1

Lookahead: 12 candles ( 1.0 hours)

ONE/USDT Compressed data 78 -> 64 (features)

training samples: 151455 #buys: 6068 #sells: 6072

Loading existing model (/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/models/NNBC_fbb/NNBC_fbb_Transformer_Buy.h5)...
Model is already trained
Loading existing model (/media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/models/NNBC_fbb/NNBC_fbb_Transformer_Sell.h5)...
Model is already trained
predicting buys...
predicting sells...

WTC/USDT Compressed data 78 -> 64 (features)

training samples: 151455 #buys: 5918 #sells: 4420

Model is already trained
Model is already trained
predicting buys...
predicting sells...

Output

top 5 stats grouped by filename /media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/profiler.py:0: size=416 B, count=1, average=416 B /media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/NNBC.py:0: size=408 B, count=1, average=408 B

top 10 stats /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py:2388: size=430 MiB (+430 MiB), count=7 (+7), average=61.4 MiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/talib/init.py:27: size=41.7 MiB (+41.7 MiB), count=53 (+53), average=806 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/frame.py:12016: size=32.1 MiB (+32.1 MiB), count=40 (+40), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/dtypes/cast.py:1434: size=22.5 MiB (+22.5 MiB), count=28 (+28), average=822 KiB

<__array_function__ internals>:180: size=3304 KiB (+3304 KiB), count=48 (+48), average=68.8 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py:549: size=3288 KiB (+3288 KiB), count=25 (+25), average=132 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:562: size=3287 KiB (+3287 KiB), count=6 (+6), average=548 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimes.py:987: size=3287 KiB (+3287 KiB), count=4 (+4), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/algorithms.py:1251: size=3287 KiB (+3287 KiB), count=4 (+4), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/abc.py:123: size=2394 KiB (+2394 KiB), count=23960 (+23960), average=102 B *** top 10 stats *** /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py:2388: size=591 MiB (+591 MiB), count=9 (+9), average=65.6 MiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/talib/__init__.py:27: size=41.7 MiB (+41.7 MiB), count=52 (+52), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/frame.py:12016: size=32.1 MiB (+32.1 MiB), count=40 (+40), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/dtypes/cast.py:1434: size=22.5 MiB (+22.5 MiB), count=28 (+28), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py:549: size=11.2 MiB (+11.2 MiB), count=30 (+30), average=384 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:562: size=6575 KiB (+6575 KiB), count=12 (+12), average=548 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimes.py:987: size=4931 KiB (+4931 KiB), count=6 (+6), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/algorithms.py:1251: size=4931 KiB (+4931 KiB), count=6 (+6), average=822 KiB <__array_function__ internals>:180: size=3304 KiB (+3304 KiB), count=51 (+51), average=64.8 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexes/range.py:202: size=3287 KiB (+3287 KiB), count=4 (+4), average=822 KiB Crash 1 *** Trace for largest memory block - (72 blocks, 5916631.5 Kb) *** File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/commands/optimize_commands.py", line 58 backtesting.start() File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/optimize/backtesting.py", line 1344 min_date, max_date = self.backtest_one_strategy(strat, data, timerange) File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/optimize/backtesting.py", line 1246 preprocessed = self.strategy.advise_all_indicators(data) File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/strategy/interface.py", line 1318 return {pair: self.advise_indicators(pair_data.copy(), {'pair': pair}).copy() File "/media/alok/New_Volume/kali/conda/freqtrade/freqtrade/strategy/interface.py", line 1318 return {pair: self.advise_indicators(pair_data.copy(), {'pair': pair}).copy() File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/generic.py", line 6368 data = self._mgr.copy(deep=deep) File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 670 res._consolidate_inplace() File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 1871 self.blocks = _consolidate(self.blocks) File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 2329 merged_blocks, _ = _merge_blocks( File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 2388 new_values = new_values[argsort] 2 *** top 10 stats *** /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py:2388: size=6218 MiB (+6218 MiB), count=84 (+84), average=74.0 MiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/array_algos/take.py:158: size=6002 MiB (+6002 MiB), count=189 (+189), average=31.8 MiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:562: size=61.0 MiB (+61.0 MiB), count=114 (+114), average=548 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/algorithms.py:1577: size=59.4 MiB (+59.4 MiB), count=74 (+74), average=821 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexes/range.py:202: size=57.8 MiB (+57.8 MiB), count=72 (+72), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/talib/__init__.py:27: size=41.7 MiB (+41.7 MiB), count=52 (+52), average=822 KiB /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/frame.py:12016: size=32.1 MiB (+32.1 MiB), count=40 (+40), average=822 KiB /media/alok/New_Volume/kali/conda/freqtrade/freqtrade/optimize/backtesting.py:390: size=30.5 MiB (+30.5 MiB), count=420483 (+420483), average=76 B /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py:1914: size=28.9 MiB (+28.9 MiB), count=1261404 (+1261404), average=24 B /home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/arrays/datetimelike.py:438: size=25.7 MiB (+25.7 MiB), count=210241 (+210241), average=128 B *** Trace for largest memory block - (76 blocks, 6083852.5 Kb) *** File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexing.py", line 1292 return self._getbool_axis(key, axis=axis) File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/indexing.py", line 1093 return self.obj._take_with_is_copy(inds, axis=axis) File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/generic.py", line 3902 result = self._take(indices=indices, axis=axis) File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/generic.py", line 3886 new_data = self._mgr.take( File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 978 return self.reindex_indexer( File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 751 new_blocks = [ File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/managers.py", line 752 blk.take_nd( File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/internals/blocks.py", line 880 new_values = algos.take_nd( File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/array_algos/take.py", line 117 return _take_nd_ndarray(arr, indexer, axis, fill_value, allow_fill) File "/home/alok/miniconda3/envs/freqtrade-conda/lib/python3.10/site-packages/pandas/core/array_algos/take.py", line 158 out = np.empty(out_shape, dtype=dtype) *** top 5 stats grouped by filename *** /media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/profiler.py:0: size=416 B, count=1, average=416 B /media/alok/New_Volume/kali/conda/freqtrade/user_data/strategies/binance/NNBC.py:0: size=408 B, count=1, average=408 B I am sorry i am a bit new to ML. I understand python and bash well. Could you advise what is happening ? — Reply to this email directly, view it on GitHub , or unsubscribe . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>