"[<matplotlib.lines.Line2D at 0x7f54db400a58>,\n",
"[<matplotlib.lines.Line2D at 0x7f98546c2e48>,\n",
" <matplotlib.lines.Line2D at 0x7f54db400b70>,\n",
" <matplotlib.lines.Line2D at 0x7f98546ca0b8>,\n",
" <matplotlib.lines.Line2D at 0x7f54db3ad438>]"
" <matplotlib.lines.Line2D at 0x7f98546ca8d0>]"
]
]
},
},
"execution_count": 21,
"execution_count": 12,
"metadata": {},
"metadata": {},
"output_type": "execute_result"
"output_type": "execute_result"
},
},
...
@@ -3371,7 +3378,7 @@
...
@@ -3371,7 +3378,7 @@
},
},
{
{
"cell_type": "code",
"cell_type": "code",
"execution_count": 37,
"execution_count": 13,
"metadata": {
"metadata": {
"hideCode": true,
"hideCode": true,
"hidePrompt": true
"hidePrompt": true
...
@@ -3380,21 +3387,22 @@
...
@@ -3380,21 +3387,22 @@
{
{
"data": {
"data": {
"text/plain": [
"text/plain": [
"array([[0.00061726]])"
"array([[1620.05083396]])"
]
]
},
},
"execution_count": 37,
"execution_count": 13,
"metadata": {},
"metadata": {},
"output_type": "execute_result"
"output_type": "execute_result"
}
}
],
],
"source": [
"source": [
"reg_low.coef_"
"C_low=1/reg_low.coef_\n",
"C_low"
]
]
},
},
{
{
"cell_type": "code",
"cell_type": "code",
"execution_count": 38,
"execution_count": 14,
"metadata": {
"metadata": {
"hideCode": true,
"hideCode": true,
"hidePrompt": true
"hidePrompt": true
...
@@ -3406,18 +3414,19 @@
...
@@ -3406,18 +3414,19 @@
"array([3.12808083])"
"array([3.12808083])"
]
]
},
},
"execution_count": 38,
"execution_count": 14,
"metadata": {},
"metadata": {},
"output_type": "execute_result"
"output_type": "execute_result"
}
}
],
],
"source": [
"source": [
"reg_low.intercept_"
"L_low=reg_low.intercept_\n",
"L_low"
]
]
},
},
{
{
"cell_type": "code",
"cell_type": "code",
"execution_count": 39,
"execution_count": 15,
"metadata": {
"metadata": {
"hideCode": true,
"hideCode": true,
"hidePrompt": true
"hidePrompt": true
...
@@ -3426,21 +3435,22 @@
...
@@ -3426,21 +3435,22 @@
{
{
"data": {
"data": {
"text/plain": [
"text/plain": [
"array([[0.00286134]])"
"array([[349.48639641]])"
]
]
},
},
"execution_count": 39,
"execution_count": 15,
"metadata": {},
"metadata": {},
"output_type": "execute_result"
"output_type": "execute_result"
}
}
],
],
"source": [
"source": [
"reg_high.coef_"
"C_high=1/reg_high.coef_\n",
"C_high"
]
]
},
},
{
{
"cell_type": "code",
"cell_type": "code",
"execution_count": 40,
"execution_count": 16,
"metadata": {
"metadata": {
"hideCode": true,
"hideCode": true,
"hidePrompt": true
"hidePrompt": true
...
@@ -3452,13 +3462,14 @@
...
@@ -3452,13 +3462,14 @@
"array([4.77022699])"
"array([4.77022699])"
]
]
},
},
"execution_count": 40,
"execution_count": 16,
"metadata": {},
"metadata": {},
"output_type": "execute_result"
"output_type": "execute_result"
}
}
],
],
"source": [
"source": [
"reg_high.intercept_"
"L_high=reg_high.intercept_\n",
"L_high"
]
]
},
},
{
{
...
@@ -3468,7 +3479,234 @@
...
@@ -3468,7 +3479,234 @@
"hidePrompt": true
"hidePrompt": true
},
},
"source": [
"source": [
"While the lag time seems to be more or less the same in the two categories (3 to 5 seconds according to the regressions), the coeff"
"La latence semble concorder dans les deux catégories (3 à 5 secondesn selon la taille des fichiers). Le coefficient linéaire passe de 350 à 1600 selon la catégorie, montrant les limites du modèle linéaire dans ce cas."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We now perform the same operations with the second set of data"
"Cette connexion semble avoir une latence plus élevée, de l'ordre de 100ms. Le coefficient change drastiquement selon la taille des données, et est même négatif pour les données de taille importantes."
