import torch
import sys
from fxpmath import Fxp
import numpy as np
x = Fxp(-5.25789021)
print(x.info())
# network = sys.argv[1]
network = "trained-models/cornell-randsplit-rgbd-grconvnet3-drop1-ch32/epoch_19_iou_0.98"
net = torch.load(network)
# net_q = torch.quantization.quantize_dynamic(net, qconfig_spec=None, dtype=torch.qint8, mapping=None, inplace=False)
# net.eval()
model_dict = net.state_dict()
# model_dict_q = net_q.state_dict()
# print ("model_dict = ",model_dict)
keys = model_dict.keys()
print("keys = ", keys)
conv1_weight = model_dict['conv1.weight']
# conv1_weigdt_fixed = model_dict_q['conv1.weight']
# print("origin = ",conv1_weight[0][0][0])
# print("quantize = ",conv1_weigdt_fixed[0][0][0])
# print ("conv1_weight = ",conv1_weight.size())
# bn1_weight = model_dict["bn1.weight"]
# print ("bn1_weight = ",bn1_weight)
for key in model_dict:
import numpy as np
x = np.array([[1 / 3, 1 / 3]])
y = np.array([[1 / 3], [1 / 3]])
print("X original = ", x)
print("Y original = ", y)
print("-----------")
x_fxp = Fxp(x, signed=False, n_word=20, n_frac=15)
y_fxp = Fxp(y, signed=False, n_word=20, n_frac=15)
print("X fxp =", x_fxp.get_val())
print("Y fxp =", y_fxp.get_val())
print("-----------")
print(x_fxp.info(verbose=3))
print("-----------")
print("Dot product without scaling = ", x_fxp.get_val().dot(y_fxp.get_val()))
print("Dot prod scaled = ",
Fxp(x_fxp.get_val().dot(y_fxp.get_val()), n_word=20, n_frac=15))
print("Dot prod without specifications = ",
Fxp(x_fxp.get_val().dot(y_fxp.get_val())))
dot_fpx = Fxp(None, signed=True, n_word=20, n_frac=15)
dot_fpx.equal(x_fxp().dot(y_fxp()))
print("Dot with new variable = ", dot_fpx())