2025-03-17 02:49:17,843 - INFO - Starting live trading demo for ETH/USDT on 1m timeframe
2025-03-17 02:49:17,844 - INFO - Using model: models/trading_agent_best_pnl.pt
2025-03-17 02:49:17,847 - INFO - Exchange initialized with standard CCXT: mexc
2025-03-17 02:49:17,848 - INFO - Fetching initial data for ETH/USDT
2025-03-17 02:49:18,537 - ERROR - Error fetching OHLCV data: mexc {"code":700002,"msg":"Signature for this request is not valid."}
2025-03-17 02:49:18,537 - WARNING - No initial data received
2025-03-17 02:49:18,537 - ERROR - Failed to fetch initial data. Exiting.
2025-03-17 02:50:45,182 - INFO - Starting live trading demo for ETH/USDT on 1m timeframe
2025-03-17 02:50:45,182 - INFO - Using model: models/trading_agent_best_pnl.pt
2025-03-17 02:50:45,182 - INFO - Using mock data for demo mode (no API keys required)
2025-03-17 02:50:45,182 - INFO - Generating mock data for ETH/USDT (1m)
2025-03-17 02:50:45,189 - INFO - Generated 1000 mock candles
2025-03-17 02:50:45,217 - INFO - Using GPU: NVIDIA GeForce RTX 4060 Laptop GPU
2025-03-17 02:50:46,501 - WARNING - Failed to load with weights_only=True: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 02:50:46,566 - WARNING - Failed with safe_globals: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy.dtype was not an allowed global by default. Please use `torch.serialization.add_safe_globals([dtype])` or the `torch.serialization.safe_globals([dtype])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 02:50:46,623 - ERROR - Error in live trading: Error(s) in loading state_dict for DQN:
	size mismatch for fc1.weight: copying a param with shape torch.Size([384, 40]) from checkpoint, the shape in current model is torch.Size([256, 64]).
	size mismatch for fc1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for lstm.weight_ih_l0: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.weight_hh_l0: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.bias_ih_l0: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for lstm.bias_hh_l0: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for lstm.weight_ih_l1: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.weight_hh_l1: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.bias_ih_l1: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for lstm.bias_hh_l1: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for attention.in_proj_weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
	size mismatch for attention.in_proj_bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([768]).
	size mismatch for attention.out_proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for attention.out_proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for fc2.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for fc3.weight: copying a param with shape torch.Size([192, 384]) from checkpoint, the shape in current model is torch.Size([128, 256]).
	size mismatch for fc3.bias: copying a param with shape torch.Size([192]) from checkpoint, the shape in current model is torch.Size([128]).
	size mismatch for value_stream.weight: copying a param with shape torch.Size([1, 192]) from checkpoint, the shape in current model is torch.Size([1, 128]).
	size mismatch for advantage_stream.weight: copying a param with shape torch.Size([4, 192]) from checkpoint, the shape in current model is torch.Size([3, 128]).
	size mismatch for advantage_stream.bias: copying a param with shape torch.Size([4]) from checkpoint, the shape in current model is torch.Size([3]).
	size mismatch for transformer_encoder.layers.0.self_attn.in_proj_weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
	size mismatch for transformer_encoder.layers.0.self_attn.in_proj_bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([768]).
	size mismatch for transformer_encoder.layers.0.self_attn.out_proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for transformer_encoder.layers.0.self_attn.out_proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.linear1.weight: copying a param with shape torch.Size([2048, 384]) from checkpoint, the shape in current model is torch.Size([2048, 256]).
	size mismatch for transformer_encoder.layers.0.linear2.weight: copying a param with shape torch.Size([384, 2048]) from checkpoint, the shape in current model is torch.Size([256, 2048]).
	size mismatch for transformer_encoder.layers.0.linear2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.self_attn.in_proj_weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
	size mismatch for transformer_encoder.layers.1.self_attn.in_proj_bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([768]).
	size mismatch for transformer_encoder.layers.1.self_attn.out_proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for transformer_encoder.layers.1.self_attn.out_proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.linear1.weight: copying a param with shape torch.Size([2048, 384]) from checkpoint, the shape in current model is torch.Size([2048, 256]).
	size mismatch for transformer_encoder.layers.1.linear2.weight: copying a param with shape torch.Size([384, 2048]) from checkpoint, the shape in current model is torch.Size([256, 2048]).
	size mismatch for transformer_encoder.layers.1.linear2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
2025-03-17 02:50:46,625 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 236, in run_live_demo
    agent.load(args.model)
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\main.py", line 1776, in load
    self.policy_net.load_state_dict(checkpoint['policy_net'])
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\nn\modules\module.py", line 2581, in load_state_dict
    raise RuntimeError(
RuntimeError: Error(s) in loading state_dict for DQN:
	size mismatch for fc1.weight: copying a param with shape torch.Size([384, 40]) from checkpoint, the shape in current model is torch.Size([256, 64]).
	size mismatch for fc1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for lstm.weight_ih_l0: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.weight_hh_l0: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.bias_ih_l0: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for lstm.bias_hh_l0: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for lstm.weight_ih_l1: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.weight_hh_l1: copying a param with shape torch.Size([1536, 384]) from checkpoint, the shape in current model is torch.Size([1024, 256]).
	size mismatch for lstm.bias_ih_l1: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for lstm.bias_hh_l1: copying a param with shape torch.Size([1536]) from checkpoint, the shape in current model is torch.Size([1024]).
	size mismatch for attention.in_proj_weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
	size mismatch for attention.in_proj_bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([768]).
	size mismatch for attention.out_proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for attention.out_proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for fc2.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for fc2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for ln2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for fc3.weight: copying a param with shape torch.Size([192, 384]) from checkpoint, the shape in current model is torch.Size([128, 256]).
	size mismatch for fc3.bias: copying a param with shape torch.Size([192]) from checkpoint, the shape in current model is torch.Size([128]).
	size mismatch for value_stream.weight: copying a param with shape torch.Size([1, 192]) from checkpoint, the shape in current model is torch.Size([1, 128]).
	size mismatch for advantage_stream.weight: copying a param with shape torch.Size([4, 192]) from checkpoint, the shape in current model is torch.Size([3, 128]).
	size mismatch for advantage_stream.bias: copying a param with shape torch.Size([4]) from checkpoint, the shape in current model is torch.Size([3]).
	size mismatch for transformer_encoder.layers.0.self_attn.in_proj_weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
	size mismatch for transformer_encoder.layers.0.self_attn.in_proj_bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([768]).
	size mismatch for transformer_encoder.layers.0.self_attn.out_proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for transformer_encoder.layers.0.self_attn.out_proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.linear1.weight: copying a param with shape torch.Size([2048, 384]) from checkpoint, the shape in current model is torch.Size([2048, 256]).
	size mismatch for transformer_encoder.layers.0.linear2.weight: copying a param with shape torch.Size([384, 2048]) from checkpoint, the shape in current model is torch.Size([256, 2048]).
	size mismatch for transformer_encoder.layers.0.linear2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.0.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.self_attn.in_proj_weight: copying a param with shape torch.Size([1152, 384]) from checkpoint, the shape in current model is torch.Size([768, 256]).
	size mismatch for transformer_encoder.layers.1.self_attn.in_proj_bias: copying a param with shape torch.Size([1152]) from checkpoint, the shape in current model is torch.Size([768]).
	size mismatch for transformer_encoder.layers.1.self_attn.out_proj.weight: copying a param with shape torch.Size([384, 384]) from checkpoint, the shape in current model is torch.Size([256, 256]).
	size mismatch for transformer_encoder.layers.1.self_attn.out_proj.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.linear1.weight: copying a param with shape torch.Size([2048, 384]) from checkpoint, the shape in current model is torch.Size([2048, 256]).
	size mismatch for transformer_encoder.layers.1.linear2.weight: copying a param with shape torch.Size([384, 2048]) from checkpoint, the shape in current model is torch.Size([256, 2048]).
	size mismatch for transformer_encoder.layers.1.linear2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm1.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm1.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm2.weight: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).
	size mismatch for transformer_encoder.layers.1.norm2.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([256]).

2025-03-17 02:52:12,557 - INFO - Starting live trading demo for ETH/USDT on 1m timeframe
2025-03-17 02:52:12,558 - INFO - Using model: models/trading_agent_best_pnl.pt
2025-03-17 02:52:12,558 - INFO - Using mock data for demo mode (no API keys required)
2025-03-17 02:52:12,558 - INFO - Generating mock data for ETH/USDT (1m)
2025-03-17 02:52:12,565 - INFO - Generated 1000 mock candles
2025-03-17 02:52:12,607 - INFO - Extracted model architecture: state_size=40, action_size=4, hidden_size=384, lstm_layers=2, attention_heads=4
2025-03-17 02:52:12,636 - INFO - Using GPU: NVIDIA GeForce RTX 4060 Laptop GPU
2025-03-17 02:52:13,909 - WARNING - Failed to load with weights_only=True: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 02:52:13,973 - WARNING - Failed with safe_globals: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy.dtype was not an allowed global by default. Please use `torch.serialization.add_safe_globals([dtype])` or the `torch.serialization.safe_globals([dtype])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 02:52:14,032 - INFO - Model loaded from models/trading_agent_best_pnl.pt
2025-03-17 02:52:14,032 - INFO - Model loaded successfully
2025-03-17 02:52:14,035 - INFO - Starting live trading simulation...
2025-03-17 02:52:19,117 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:52:19,118 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:52:19,118 - INFO - Continuing after error...
2025-03-17 02:52:29,139 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:52:29,140 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:52:29,140 - INFO - Continuing after error...
2025-03-17 02:52:39,157 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:52:39,157 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:52:39,158 - INFO - Continuing after error...
2025-03-17 02:52:49,176 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:52:49,177 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:52:49,177 - INFO - Continuing after error...
2025-03-17 02:52:59,196 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:52:59,196 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:52:59,196 - INFO - Continuing after error...
2025-03-17 02:53:09,220 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:53:09,220 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:53:09,220 - INFO - Continuing after error...
2025-03-17 02:53:19,244 - ERROR - Error in live trading loop: not enough values to unpack (expected 4, got 3)
2025-03-17 02:53:19,245 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 367, in run_live_demo
    next_state, reward, done, info = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)

2025-03-17 02:53:19,245 - INFO - Continuing after error...
2025-03-17 02:53:53,471 - INFO - Starting live trading demo for ETH/USDT on 1m timeframe
2025-03-17 02:53:53,472 - INFO - Using model: models/trading_agent_best_pnl.pt
2025-03-17 02:53:53,472 - INFO - Using mock data for demo mode (no API keys required)
2025-03-17 02:53:53,472 - INFO - Generating mock data for ETH/USDT (1m)
2025-03-17 02:53:53,479 - INFO - Generated 1000 mock candles
2025-03-17 02:53:53,520 - INFO - Extracted model architecture: state_size=40, action_size=4, hidden_size=384, lstm_layers=2, attention_heads=4
2025-03-17 02:53:53,552 - INFO - Using GPU: NVIDIA GeForce RTX 4060 Laptop GPU
2025-03-17 02:53:54,887 - WARNING - Failed to load with weights_only=True: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 02:53:54,958 - WARNING - Failed with safe_globals: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy.dtype was not an allowed global by default. Please use `torch.serialization.add_safe_globals([dtype])` or the `torch.serialization.safe_globals([dtype])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 02:53:55,016 - INFO - Model loaded from models/trading_agent_best_pnl.pt
2025-03-17 02:53:55,017 - INFO - Model loaded successfully
2025-03-17 02:53:55,019 - INFO - Starting live trading simulation...
2025-03-17 02:54:24,295 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:54:54,484 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:55:24,631 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:55:54,809 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:56:24,987 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:56:55,157 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:57:25,288 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:57:55,450 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:58:25,571 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:58:55,733 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:59:25,898 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 02:59:55,196 - INFO - Starting live trading demo for ETH/USDT on 1m timeframe
2025-03-17 02:59:55,196 - INFO - Using model: models/trading_agent_best_pnl.pt
2025-03-17 02:59:55,200 - INFO - Exchange initialized with standard CCXT: mexc
2025-03-17 02:59:55,200 - INFO - Fetching initial data for ETH/USDT
2025-03-17 02:59:55,844 - ERROR - Error fetching OHLCV data: mexc {"code":700002,"msg":"Signature for this request is not valid."}
2025-03-17 02:59:55,844 - WARNING - No initial data received
2025-03-17 02:59:55,844 - ERROR - Failed to fetch initial data. Exiting.
2025-03-17 02:59:56,090 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:00:26,253 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:00:56,413 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:01:26,591 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:01:56,732 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:02:26,890 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:02:57,233 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:03:27,392 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:03:57,555 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:04:27,713 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:04:57,867 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:05:28,019 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:05:58,171 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:06:28,323 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:06:58,510 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:07:28,695 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:07:58,884 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:08:29,079 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:08:59,295 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:09:29,452 - ERROR - Error creating chart: Line2D.set() got an unexpected keyword argument 'type'
2025-03-17 03:40:19,333 - INFO - Starting live trading demo for ETH/USDT on 1m timeframe
2025-03-17 03:40:19,333 - INFO - Using model: models/trading_agent_best_pnl.pt
2025-03-17 03:40:19,336 - INFO - Exchange initialized with standard CCXT: mexc
2025-03-17 03:40:19,336 - INFO - Fetching initial data for ETH/USDT
2025-03-17 03:40:24,689 - INFO - Fetched 500 candles for ETH/USDT (1m)
2025-03-17 03:40:24,693 - INFO - Initialized environment with 500 candles
2025-03-17 03:40:24,728 - INFO - Extracted model architecture: state_size=64, action_size=4, hidden_size=384, lstm_layers=2, attention_heads=4
2025-03-17 03:40:24,748 - INFO - Using GPU: NVIDIA GeForce RTX 4060 Laptop GPU
2025-03-17 03:40:25,795 - ERROR - Error loading model: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 03:40:25,797 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\main.py", line 1819, in load
    checkpoint = torch.load(path, map_location=self.device)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\serialization.py", line 1470, in load
    raise pickle.UnpicklingError(_get_wo_message(str(e))) from None
_pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.

2025-03-17 03:40:25,797 - WARNING - Failed to load model with weights_only=True: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
2025-03-17 03:40:25,860 - ERROR - Error loading model: 'str' object has no attribute '__module__'
2025-03-17 03:40:25,863 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 280, in run_live_demo
    agent.load(args.model)
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\main.py", line 1819, in load
    checkpoint = torch.load(path, map_location=self.device)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\serialization.py", line 1470, in load
    raise pickle.UnpicklingError(_get_wo_message(str(e))) from None
_pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL numpy._core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([scalar])` or the `torch.serialization.safe_globals([scalar])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\main.py", line 1819, in load
    checkpoint = torch.load(path, map_location=self.device)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\serialization.py", line 1462, in load
    return _load(
           ^^^^^^
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\serialization.py", line 1964, in _load
    result = unpickler.load()
             ^^^^^^^^^^^^^^^^
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\_weights_only_unpickler.py", line 334, in load
    elif full_path in _get_user_allowed_globals():
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\popov\miniforge3\Lib\site-packages\torch\_weights_only_unpickler.py", line 144, in _get_user_allowed_globals
    module, name = f.__module__, f.__name__
                   ^^^^^^^^^^^^
AttributeError: 'str' object has no attribute '__module__'. Did you mean: '__mod__'?

2025-03-17 03:40:25,863 - WARNING - Failed with safe_globals: 'str' object has no attribute '__module__'
2025-03-17 03:40:25,919 - INFO - Model loaded from models/trading_agent_best_pnl.pt
2025-03-17 03:40:25,920 - INFO - Model loaded successfully
2025-03-17 03:40:25,925 - INFO - Starting live trading simulation...
2025-03-17 03:40:26,348 - INFO - Fetched 1 candles for ETH/USDT (1m)
2025-03-17 03:40:26,406 - ERROR - Error in live trading loop: too many values to unpack (expected 3)
2025-03-17 03:40:26,406 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 370, in run_live_demo
    next_state, reward, done = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: too many values to unpack (expected 3)

2025-03-17 03:40:26,406 - INFO - Continuing after error...
2025-03-17 03:40:31,926 - INFO - Fetched 1 candles for ETH/USDT (1m)
2025-03-17 03:40:31,933 - ERROR - Error in live trading loop: too many values to unpack (expected 3)
2025-03-17 03:40:31,933 - ERROR - Traceback (most recent call last):
  File "D:\DEV\workspace\REPOS\git.d-popov.com\ai-kevin\crypto\gogo2\run_live_demo.py", line 370, in run_live_demo
    next_state, reward, done = env.step(action)
    ^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: too many values to unpack (expected 3)

2025-03-17 03:40:31,933 - INFO - Continuing after error...