Skip to content

[GSK-2424] Improve a bit FFHQ structure #15

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 11 commits into from
Jan 4, 2024
Merged

[GSK-2424] Improve a bit FFHQ structure #15

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
23de9b5
Improve a bit FFHQ structure
Hartorn Jan 2, 2024
b18ec1b
Re-organize the batching a bit
Hartorn Jan 3, 2024
62907d1
Fix indexing for batching
Hartorn Jan 3, 2024
1a4c7a0
Output for notebook
Hartorn Jan 3, 2024
382315f
Fix batching for prediction
Hartorn Jan 3, 2024
c78a235
fixed predict, reproducing main
rabah-khalek Jan 3, 2024
566d1fb
Update loreal_poc/dataloaders/base.py
rabah-khalek Jan 4, 2024
a9cac74
Update loreal_poc/dataloaders/base.py
rabah-khalek Jan 4, 2024
9480fa9
Update loreal_poc/dataloaders/base.py
rabah-khalek Jan 4, 2024
6df14bd
Merge pull request #16 from Giskard-AI/fix-batching
rabah-khalek Jan 4, 2024
8836c95
pdm format
rabah-khalek Jan 4, 2024
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Re-organize the batching a bit
  • Loading branch information
@Hartorn
Hartorn committed Jan 3, 2024
commit b18ec1bdf78c8d13c766dfd57de15b0ee71ee34b
136 changes: 27 additions & 109 deletions examples/criteria1_partial_faces.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

99 changes: 93 additions & 6 deletions examples/ex1_draw_landmarks.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -184,9 +184,83 @@
},
{
"cell_type": "code",
"execution_count": 10,
"execution_count": 6,
"metadata": {},
"outputs": [],
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"LEFT_EYE\n",
"NMEMean:0.03064837061887285\n",
"NMEs:[0.03639644 0.02607136 0.02883627 0.02288638 0.0390514 ]\n",
"MEMean:6.205209482416183\n",
"MEStd:3.7323675345101868\n",
"NMEMean:0.03064837061887285\n",
"NMEStd:0.006134046369089966\n",
"Es:[[ nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" 5.4613892 1.78589613 5.436992 6.19588936 5.81028786 3.90036639\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan]\n",
" [ nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" 3.18053911 7.35603983 3.61900553 12.1698484 3.20499641 2.34261051\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan]\n",
" [ nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" 16.76414522 4.6575879 7.59055736 13.538641 9.41119509 6.31696699\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan]\n",
" [ nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" 4.09730899 2.86064765 3.59209187 5.20874284 0.60418623 4.28686925\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan]\n",
" [ nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" 11.6476298 2.34872646 6.37660105 9.59164079 10.75729009 6.04159515\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan nan nan nan nan\n",
" nan nan]]\n"
]
}
],
"source": [
"prediction = model.predict(ds, facial_part=FacialParts.LEFT_EYE.value)\n",
"marks = ds.all_marks\n",
Expand All @@ -198,7 +272,7 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 7,
"metadata": {
"collapsed": false,
"jupyter": {
Expand All @@ -215,14 +289,27 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 8,
"metadata": {
"collapsed": false,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [],
"outputs": [
{
"ename": "UnboundLocalError",
"evalue": "cannot access local variable '_image' where it is not associated with a value",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mUnboundLocalError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[8], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mdraw_marks\u001b[49m\u001b[43m(\u001b[49m\u001b[43mimage\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m[\u001b[49m\u001b[43mground_truth_landmarks\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mloreal_landmarks\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mgreen\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mred\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\n",
"File \u001b[0;32m~/Work/loreal-poc/loreal_poc/visualisation/draw.py:38\u001b[0m, in \u001b[0;36mdraw_marks\u001b[0;34m(image, list_of_marks, colors, with_text, squares)\u001b[0m\n\u001b[1;32m 36\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(image, np\u001b[38;5;241m.\u001b[39mndarray):\n\u001b[1;32m 37\u001b[0m _image \u001b[38;5;241m=\u001b[39m Image\u001b[38;5;241m.\u001b[39mfromarray(cv2\u001b[38;5;241m.\u001b[39mcvtColor(image, cv2\u001b[38;5;241m.\u001b[39mCOLOR_BGR2RGB))\n\u001b[0;32m---> 38\u001b[0m image_display \u001b[38;5;241m=\u001b[39m \u001b[43m_image\u001b[49m\u001b[38;5;241m.\u001b[39mcopy()\n\u001b[1;32m 39\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m marks, color, _with_text, square \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(\n\u001b[1;32m 40\u001b[0m list_of_marks,\n\u001b[1;32m 41\u001b[0m colors \u001b[38;5;129;01mor\u001b[39;00m [\u001b[38;5;28;01mNone\u001b[39;00m] \u001b[38;5;241m*\u001b[39m \u001b[38;5;28mlen\u001b[39m(list_of_marks),\n\u001b[1;32m 42\u001b[0m with_text \u001b[38;5;129;01mor\u001b[39;00m [\u001b[38;5;28;01mNone\u001b[39;00m] \u001b[38;5;241m*\u001b[39m \u001b[38;5;28mlen\u001b[39m(list_of_marks),\n\u001b[1;32m 43\u001b[0m squares \u001b[38;5;129;01mor\u001b[39;00m [\u001b[38;5;28;01mNone\u001b[39;00m] \u001b[38;5;241m*\u001b[39m \u001b[38;5;28mlen\u001b[39m(list_of_marks),\n\u001b[1;32m 44\u001b[0m ):\n\u001b[1;32m 45\u001b[0m image_display \u001b[38;5;241m=\u001b[39m _add_marks(image_display, marks, color\u001b[38;5;241m=\u001b[39mcolor, with_text\u001b[38;5;241m=\u001b[39m_with_text, square\u001b[38;5;241m=\u001b[39msquare)\n",
"\u001b[0;31mUnboundLocalError\u001b[0m: cannot access local variable '_image' where it is not associated with a value"
]
}
],
"source": [
"draw_marks(image, [ground_truth_landmarks, loreal_landmarks], [\"green\", \"red\"])"
]
Expand Down Expand Up @@ -251,7 +338,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.6"
"version": "3.11.7"
}
},
"nbformat": 4,
Expand Down
26 changes: 13 additions & 13 deletions examples/ex2_draw_partial_landmarks.ipynb
View file
Open in desktop

Large diffs are not rendered by default.

49 changes: 41 additions & 8 deletions examples/ex3_calculate_euc_dists.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
"cells": [
{
"cell_type": "code",
"execution_count": null,
"execution_count": 1,
"metadata": {
"collapsed": false,
"jupyter": {
Expand All @@ -19,28 +19,61 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 2,
"metadata": {},
"outputs": [],
"outputs": [
{
"data": {
"text/plain": [
"array([4. , 1. , 2.23606798, 2.23606798])"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"a_min_b = A - B\n",
"np.sqrt(np.einsum(\"ij,ij->i\", a_min_b, a_min_b))"
]
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 3,
"metadata": {},
"outputs": [],
"outputs": [
{
"data": {
"text/plain": [
"array([4. , 1. , 2.23606798, 2.23606798])"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"np.linalg.norm(A - B, axis=1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 4,
"metadata": {},
"outputs": [],
"outputs": [
{
"data": {
"text/plain": [
"array([4. , 1. , 2.23606798, 2.23606798])"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"d_all = np.sqrt(np.einsum(\"ij->i\", (A - B) ** 2))\n",
"d_all"
Expand All @@ -63,7 +96,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.6"
"version": "3.11.7"
}
},
"nbformat": 4,
Expand Down
93 changes: 83 additions & 10 deletions examples/ex4_tests_and_metrics.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
"cells": [
{
"cell_type": "code",
"execution_count": null,
"execution_count": 1,
"metadata": {
"collapsed": false,
"jupyter": {
Expand Down Expand Up @@ -37,7 +37,7 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 2,
"metadata": {
"collapsed": false,
"jupyter": {
Expand All @@ -51,9 +51,20 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 3,
"metadata": {},
"outputs": [],
"outputs": [
{
"data": {
"text/plain": [
"'cpu'"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import torch\n",
"\n",
Expand All @@ -75,7 +86,7 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 4,
"metadata": {
"collapsed": false,
"jupyter": {
Expand All @@ -89,23 +100,85 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 5,
"metadata": {
"collapsed": false,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [],
"outputs": [
{
"data": {
"text/html": [
"\n",
" <h4><span style=\"color:green;\">✓</span> Test \"NME_mean\" succeeded</h4>\n",
" <p>Description: Mean of normalised mean Euclidean distances across images</p>\n",
" <p>Metric: <b>0.0623</b> (threshold = 1)</p>\n",
" \n",
" <p>Prediction time: 13.1 s.</p>\n",
" "
],
"text/plain": [
"\n",
" Test \"NME_mean\" succeeded\n",
" Description: Mean of normalised mean Euclidean distances across images\n",
" Metric: 0.0623 (threshold = 1)\n",
" \n",
" Prediction time: 13.1 s.\n",
" "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"Test(metric=NMEMean, threshold=1).run(model, ds)"
]
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 6,
"metadata": {},
"outputs": [],
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/bazire/Work/loreal-poc/.venv/lib/python3.11/site-packages/face_alignment/api.py:147: UserWarning: No faces were detected.\n",
" warnings.warn(\"No faces were detected.\")\n",
"/Users/bazire/Work/loreal-poc/loreal_poc/tests/performance.py:47: RuntimeWarning: Mean of empty slice\n",
" mes = np.nanmean(es, axis=1)\n"
]
},
{
"data": {
"text/html": [
"\n",
" <h4><span style=\"color:green;\">✓</span> Test \"NME_mean\" succeeded</h4>\n",
" <p>Description: Mean of normalised mean Euclidean distances across images</p>\n",
" <p>Metric: <b>0.3962</b> (threshold = 1)</p>\n",
" Prediction fail rate: 0.4\n",
" <p>Prediction time: 18.74 s.</p>\n",
" "
],
"text/plain": [
"\n",
" Test \"NME_mean\" succeeded\n",
" Description: Mean of normalised mean Euclidean distances across images\n",
" Metric: 0.3962 (threshold = 1)\n",
" Prediction fail rate: 0.4\n",
" Prediction time: 18.74 s.\n",
" "
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from loreal_poc.marks.facial_parts import FacialParts\n",
"from loreal_poc.dataloaders.wrappers import CroppedDataLoader, CachedDataLoader\n",
Expand All @@ -132,7 +205,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.6"
"version": "3.11.7"
}
},
"nbformat": 4,
Expand Down
Loading